#131

[@RalphBacon]

Due to technical issues (alright, user error) I had to re-upload this video with a corrected soundtrack. Doh! If you previously commented on this video please do so again (apart from the audio glitch) as all the comments were deleted when I replaced the video. My apologies to everyone concerned.

[@RalphBacon]

@@2000jago And I was unaware of that but someone else has indicated this too. Problem solved. Well, one of the problems!

[@boblewis5558]

Hi Ralph, just a note of caution/awareness for you and your viewers: SOME relay manufacturers, notably SOME Chinese ones rate their relays OVERALL i.e. the 10A rating you have MAY be only 5A per contact set so 8A actual vs 5A rating on one set would be a problem and POSSIBLY the reason for the over heating. Sometimes too, the Chinese relays save money by using bright chromed steel rather than copper or phosphor bronze contact levers and that can cause similar issues at high sustained currents due to their higher resistance. It's always a good idea (best practice) to double up contacts anyway but I think you're right, Triac zero x-over controlled relays are a much better bet for power switching. It's a good idea also to ensure a separate, additional, manual two pole isolation switch in series with loads for absolute safety under maintenance or fault conditions etc.

[@DrexProjects]

Ya, 5A per side. That's what I was going to comment. You win...

[@RalphBacon]

I came to the same conclusion, Bob, (or, at least, educated guess this 10A might be an overall rating ) which is why the relay DP switches are now paralleled in my fuse box (aka consumer unit). But I'm not happy with that relay (via Amazon, from China, would you believe) especially as I just watched the Big Clive video linked to by another commenter and that resistor is just waiting to overheat. I will change it in due course. The last one lasted a few months so I better get it done sooner rather than later. SSR might do the job. A very high rated one.

[@boldford]

@@RalphBacon I'm not so sure there is any benefit of paralleling the contacts. In reality the two will never open at exactly the same time. They may open only a millisecond or two apart, however the one that opens last is the one that will arc.

[@Roy_Tellason]

@@Willam_J I was also going to post something to this effect. When I saw that relay terminal having moved out of position the first thing that came to mind for me was not arcing, which is going to ultimately have other effects such as poor contact, eventually leading to no contact, but high resistance at the relay terminal where it plugs into that socket. I don't know if it's the metal that the relay connection points are made out of, or the metal that the socket contacts are made of, but I'd be suspicious of both! Particularly if they came from asian sources, as opposed to the ones you mention. Perhaps a bit of investigation with a magnet might be revealing. Or noting the temperature of those connection points after the unit has been ON for a period of time. I got my arduino stuff from banggood, and they work, but when it comes to power-handling components I will NOT go there! And yes, it's sometimes difficult to get the good stuff, but well worth it.

[@tomr1107]

Thank you. I used the snubber to remove a thump sound from my cars speaker system. I needed to add a high amp relay and direct power from battery in order to power additional gauges and boost controller. After installing the distributor block and relay, the cars amps would thump the speakers on shut down. Thank you for the clear explanation.

[@RalphBacon]

Glad it helped, Tom, thanks for sharing.

[@Mixbag]

How did you do this please? My car speakers doing the ecacly the same on power on/ off

[@johnjeremias9437]

Amazing video! Just what I was looking for, thanks. Only missed the snubber circuit diagram and a visit from Benny. On the same note, many household steam irons stop functioning due of same spark-problem. The contact surface of the thermostat temperature control braker-contacts looses its conductivity. Fine sand-papering the contactcs will return you an otherwise perfect iron. I therefore presume that these irons are lacking a snubber absorbing circuit in their design or simply dont have it due to component cost savings. Keep the videos comming. I have enjoying all your uploads

[@RalphBacon]

You are doubtless correct about irons missing a snubber (it saves 3c from each iron). Glad you like the videos!

[@nickcalladine]

Possible expansion based on solid state : add a microcontroller with a web server, thermostatic control and pir / radar motion sensor. Turn it on and off via the phone / web interface / send a message. If manually turned on by switch or by app ,set status and monitor temperature and pir PIR hasnt been activated for 20 or xx minutes, auto switch off Set Status / Send a Message Monitor Temperature (so creating a thermostat) so create a cut off temperature, if left on and warm outside or that when the room is heating up past a room limit due to empty room Turn on / off over wifi via your phone Food for thought :). Keep up the good work and finding it rather interesting !

[@RalphBacon]

That's a good list, and something I had aspired to by using MQTT and my Raspberry Pi but time ran away with me. But I'm still keen to connect my Sonoff up, in parallel to that relay (so it overrides it) so I can turn on the heating via my phone / Alexa from inside the house. We shall see what I end up with. I suspect it will evolve over time. Thanks for the ideas!

[@nickcalladine]

@@RalphBacon thanks for the learning.. wish I had your skills :)

[@Tinker001]

@@RalphBacon Don't bother with the Sonoff. Put Tasmota on an ESP01 & control your solid-state relay itself. You can have your switch signal the ESP.

[@mrmrlee]

Hi, I just bought one of those compact Chinese washing machines, I love the simple design as it's basically a mechanical timer alternately powering 2 relays to change the direction of the agitator motor. However with the lights out in the room it puts on quite a spark show visible through the translucent plastic cabinet!

[@RalphBacon]

Wow! Sounds like those relays will eventually fail as the contact area becomes burnt out.

[@spyder000069]

In my experiments with powering a space heater I found: The melting in your relay was possibly also due to higher resistance as the contacts wore. Not all relays are created equal. Some have substantially longer contact cycle life ratings so you have to check the spec sheet. The sparking was always when the contacts were released causing the wear to the contacts. In my project I used zero crossing detection and timed how long it took for the relay to open up and then let it open at a point where the voltage was lowest stopping much of the arcing. Of course relays being mechanical meant it wasn't always timed right but they were consistent enough and I got 3-5 times more life out of the contacts with that method. I ran tests where the mcu just cycled the relay under load constantly and counted cycles and then at intervals (after days) measured resistance until I felt like the relays service life ended and compared to the zero cross method.

[@RalphBacon]

Interesting. I have an SSR that I wanted to try but compared to a standard relay it is *huge* especially with the heatsink. That would be a good candidate for zero-volt crossover switch off to prevent pops and other RFI in your speakers. I'm going to use a snubber in my next heater workshop project (already under way) in the hope that helps, as it is purely resistive but high current.

[@otoauto2257]

I have a project that uses mechanical relay to turn on off AC power circuit at exactly the zero-cross point of voltage. The control circuit just consisted an operational amplifier and some discrete components. Let's visit mine kzbin.info/www/bejne/j3_ddqF7pr-Dfqs to watch the video. Thank you

[@MrEdwardhartmann]

One other thing about the Chinese SSRs. Big Clive took one apart (I think it was even the same manufacturer that you showed) - it was rated at 25A, but only had a 12A triac inside, so, if you are going to get one of those be aware. I assume that the relays are a knockoff and not really the ones made by the manufacturer that is printed on the relay.

[@RalphBacon]

I watched that video, Edward, at the prompting of another viewer. As Clive said, it might be OK for hobby use, if you half the rating (from 25A down to 12A) and then give it some headroom to stop it getting stupidly hot. But I'm not so sure what I'm going to do now. We shall just have to wait.

[@jimb032]

Ralph...why did you use that little ice cube relay instead of a contactor? I wouldn't use one of those for anything more than control circuitry...

[@RalphBacon]

Well, James, it did say 10A @240V on the spec-sheet so I thought I was safe. sigh. It probably means 10A TOTAL current across both switches; as I've now pararlleled the relay switches (both now take 4A) it will probably be OK, but I'm going to investigate SSRs (basically, TRIACs).

[@kbouwman64]

@@RalphBacon I have used hundreds and hundreds of these ice cube relays over the last 30 years and soon learned like James Kerns Jr. that they are very useful for control circuit and best avoided for power loads over an amp or so. This is true for the reputable brands as well as the cheep Chinese generics.

[@wonton8983]

5 years late, but we had similar problems with permanently on relays getting hot. Your relays rating says 240V AC, but if you look at the rating written inside the relay on the coil it says 220V AC.

[@RalphBacon]

I'm not convinced that whether it's 220V or 240V that this is the root cause of the problem. The snubber eliminates the spark entirely which erodes the contacts (or makes them weld together).

[@robinrixon5241]

Just to say, I use one of the solid state relays to switch power to my monitors and speakers, using a USB connection from the PC as a switch - my BIOS offers the option of USB on/off while on standby, and this is enough to provide a signal to turn my monitors off. BUT - as others have said, these relays are often quoted at higher currents than they can really handle, so buy with caution - AND - make sure you get one with the plastic cover!

[@Catstorm99]

I do exactly the same but my bios wouldn't turn off the USB ports but a little USB add in card sorted that.

[@RalphBacon]

I tell you, this is a minefield. Overrated relays and SSRs, things that go bang and flash, I will have to re-evaluate how I proceed.

[@PenguinFach]

Couple of things that I think could be clarified. 1. If you have a damaged relay contact don't swap to the other contact because the mechanical damage can cause the relay to stick open or closed. You should also replace the relay base too because it will have heat damage. 2. It is always better to use a mains rated contactor for this application rather than a relay because they are better rated for thus sort of application. 3. Don't snub across the contacts: snub the load. The former can lead to a dangerous condition because it leaks current continuously via the capacitor which wastes power, shortens the snubber life and component degradatiin can lead to a device becoming live unexpectedly. 4. Although your load is resistive, don't discount the inductance of all the wiring and cable between the relay and load. Though small it is still enough to initiate contact arcing. 5. The MOV has a limited life, probably a few years but not indefinite. On you primarily resistive load it will last quite a while but on an inductive load where the snubber capacitor may not be well matched, the MOV may be tripped more often and it will eventually die.

[@RalphBacon]

All good points, thanks for sharing 👍

[@nukiepoo]

2 things about your 10A relay: 1)the snubber diode, while protecting the transistor, also slows down the release of the contact armature. Not good for spark arresting 2) whenever you have non-inductive arcing in a 2-pole relay, you are MUCH better off putting the two poles in series instead of the parallel configuration you used. Why? Because when in series the overall contact distance and opening rate doubles - which is what you really want to aid in minimizing contact arcing. If you look at many motor contactors, you’ll find almost without exception that the contacts are in series pairs

[@RalphBacon]

Two poles in series is a good idea, but two poles in parallel spreads the load, too.

[@analoghardwaretops3976]

​@@RalphBaconthe series is better when contacts are carrying current for shorter duration..i.e large off to on time..& @ high repetition rates.. Parallel is better for relatively longer "on" runtimes..but equal current share is no gurantee.

[@kamaledirisinghe]

Ralph, very informative video and I learned a lot by watching it. Good luck to you for more uploads.

[@RalphBacon]

Thanks, Kamal, glad you found it interesting. Great to hear from you and thanks for posting.

[@Gaark]

Referring to Gary Bringhursts comment earlier: I think it was a Matthias Wandel video I saw where he used a SSR to control his hot water cylinder (to make it use power only at off peak times), and the thing turned to a crisp after a month or so although it was rated well above the 10-15 amps the cylinder used. He showed the remains and all that was left was the guts of a triac and the aluminium back plate. [I wonder if having it mounted on top of the water heater put it out of the normal operating temperature though]

[@RalphBacon]

I watched that video and it was a doozy. He was lucky indeed that he didn't start a building fire. I shall tkae the lessons learned from that before I install anything. You're probably right about the mounting site, I mean, it can get up to 60 -70 Celsius inside a hot water heater cupboard.

[@xenofonlive]

Hey ! I use DC-DC relay!How to connect the snubber?I think must connect it to output load on relay right? One cable to + one to - ? I read this snubber works for both AC or DC 5-400V and inductive load less than 1000W.I can't see in the video exactly how you connected it!

[@RalphBacon]

Wey hey! If your load is inductive, connect to the relay contact on the COM connection and ground, not across the contacts. If that doesn't work (you still get sparks) connect across the two relay contacts, COM and NO or NC, whichever is being opened when you turn the appliance off. As it's DC, there's no danger of keeping the load on a live state by doing this.

[@logsnaath]

How is the snubber module connected? Is it between the external relay contacts NC/NO and COM?

[@RalphBacon]

Yes, that's how I set it up initially and it stopped the sparking across the relay points.

[@chrisarmstrong8198]

The heater element is usually a coil of nichrome wire and this has some inductance. The energy stored in an inductance is proportional to the current squared, and it is this energy that is dissipated in the spark.

[@RalphBacon]

And it can create quite a spark too, as my relay testifies!

[@fasousa4798]

Thank you. I have seen this being wired on the A1A2 solenoid terminals but you are wiring it. On the load terminals. I understand the principle. I have a 3 phase contactor that carries no more than 25A per phase. The contactor is controlled by a Shelly 1. How should I wire a RC snubber to protect this assembly. This contactor serves my garage and I want to use it as you use your main switch but with a Shelly control thus remotely via app on phone. What's your thoughts on that? Thanks

[@RalphBacon]

The question for you is: do you actually get a spark when disconnecting (or connecting) the Shelly contactor? I must be honest and say that I would have expected Shelly to have included some sort of spark suppression within their (mostly enclosed) switches (particularly on the Pro series)! Before adding any kind of snubber, a swift email to them would probably be in order. FYI I don't have any kind of snubber on my manual Kill Switch in my workshop (a heavy duty cooker-style switch); it's only on the resistive heater controller that I have included one to protect the relay contacts.

[@fasousa4798]

@@RalphBacon Probably i was not clear enough on my comment. The shelly1 is only controlling the heavy duty contactor. Shelly does sell a RC snubber but that has to do with disconnecting relays that can cause current spikes that may cause disruptions on some scenarios i guess. On my case and since the contactor will always carry around 10-20A i would like to prevent arcing so to protect the contactor and thus the circuit it leads to. Should i add a snubber like you did and if so how to if it is a three phase contactor? just on the coil A1A2 terminals?

[@RalphBacon]

On three phase I would expect a separate snubber on each phase but then again I have not done this and don't want to mislead you. You need to find someone with this sort of experience (try emailing shelly and see what they say).

[@shubhamkhatri9159]

I am unable to understand that why this rc snubber is not working when I am using half hp water pump only at load and using this Rc snubber to control relay contact arc. Do you have answer please describe.

[@RalphBacon]

Have you wired the snubber *across* the relay contacts (not in parallel)? This is required because you have an inductive load that will generate a lot of back EMF. BEWARE: the water pump will NOT be fully isolated if you do this as the snubber will continue to supply (a small amount of) power to it. You really need another switch or relay in series that you can manually or programmatically switch off a few seconds after the other one.

[@shubhamkhatri9159]

@@RalphBacon I did the same as you did in the video and I am using 370 watt water pump would you please tell me is this snubber is for my water pump or not because in the connection I made no mistake.

[@RalphBacon]

If you are using the same snubber as me but without success you probably need a heavier duty one. Have a look at some of the large component manufacturers to see what they have; then you can see whether you can get it any cheaper (but safe too) from closer to home.

[@lamehunter]

love listening to you Ralph, great channel!! keep up the good work

[@RalphBacon]

Thank you! Will do!

[@emf9]

I know this is an old video, but any suggestions for how to calculate the frequency? I am trying to control a 1/2 hp 120v 60hz motor with a relay and am trying to calculate the snubbing needed.

[@RalphBacon]

The frequency is not important. It's not the mains frequency that we are trying to quench here. It's the collapsing field around the motor that produces a high voltage (and hence spark). So any standard snubber will work (check out Digispark, RS Components, Farnell, Mouser - they all stock them)

[@Rcbeacon]

A couple of thoughts. A SSR will dissipate power, get hot. 10A and a 2-volt drop is 20 watts, similar to a soldering iron. This will destroy even a 50A SSR if the heat has nowhere to go. I have seen those clear relays (Omron) with contacts stuck closed when passing quite low currents but frequent operation. The small arcing eventually made an uneven contact surface that can stick. I have also seen 50A relays fail more often than you would expect when carrying 20A. The black enclosed screw-to-panel type. The relay may reliably carry 8A, but the base contacts also need to be up to the task. A slightly higher resistance contact can get hot and eventually fail burning the surrounding plastic. The point in the cycle where the contact opens will have a big effect on the contact arcing. I have seen and used capacitors across the relay contact to extend relay life. But the capacitor does pass a small current on AC. These days I try to avoid relays if possible. Setting up a cycle test with an Arduino is always fun.

[@RalphBacon]

Since I've put a snubber circuit on this relay it's been "OK" - at least, it hasn't yet failed but it's not the solution I'm looking for. I'll probably revisit this in a few weeks before the cold weather sets in as I'd like permanent solution that I can also control using my Sonoff 4-channel, Alexa controlled unit. Thanks for the information in your post, most interesting (but makes me all the more determined to get it sorted!).

[@dalenassar9152]

Did you put the snubber across the relay coil or across the contacts?? THANKS MUCH for the great video!!

[@RalphBacon]

No, not across the coil, that's the microcontroller side. The snubber goes across the contacts, or at least from one of them to neutral. Across the relay coil, ensure you put a reverse biased (cathode to +ve) diode or your microcontroller will quickly get destroyed by the back emf. Relay modules normally have them included.

[@OsoPolarClone]

Great video as always. A great video for a couple of projects I am working on.

[@RalphBacon]

Glad it made you think about your projects, Bruce, I shall be looking at SSRs for high power stuff soon too, now that I know they are available.

[@kennmossman8701]

I suspect that the heater may have a high inrush current which is an issue. If it is a 'bulb' or resisitive type these often have a very inrush current which lasts until they have heated up. If that is the case a typical solution is to limit the current initially - same approach is done with tube and some transistor amplifiers.

[@RalphBacon]

Although given that no more than 13A can ever be supplied (before the fuse trips) if the relay could handle that amount it should not be damaged either, right? I suspect the relay I'm using is borderline.

[@kennmossman8701]

@@RalphBacon "A cold incandescent lamp has a low resistance but as the tungsten filament heats the resistance increases dramatically. After an incandescent heat lamp is powered on the filament takes only one tenth of a second to reach over 6,000 degrees Fahrenheit. During this time the resistance increases to 15 times that of the cold filament." per www.ametherm.com/blog/inrush-current/limiting-heat-lamp-inrush-current/ I doubt it would trip the breaker as the overload period is very short and the fuse is more slow-blow in nature.

[@engrjolo1631]

It doesnt work for me. I connected the rc module in parellel with the relay contact but arcing during switching off still occurs..

[@RalphBacon]

It depends on what you have connected, Duino Ino. Some you connect in parallel and some you connect across the switch (but this means the circuit remains 'live' even when switched off). Inductive loads require this method. As mine was a heater without a fan it was purely resistive so wiring the snubber across the relay worked well.

[@bullrunhunter6489]

On a 3 phase contactor would it be prudent to use a snubber on each through contact terminal? I see others only use one however I imagine a voltage spike can occur on the other hot lines as well. I have one for the 24v DC side of contactor and also on the 208v line 1.

[@RalphBacon]

I've never used 3-phase, Braiden, but as they are line (hot) too I suspect you are totally correct; each line will require its own snubber. Makes sense.

[@serj2008]

Hello, Will this rc snubber help me out with my gpio 5v relay module to not get fried by a 220v solenoid valve? Thanks a lot!

[@RalphBacon]

To protect the Arduino, you need to ensure THREE things: 1. The relay coil must work (energise) with about 20mA (the maximum rule-of-thumb current for a single pin, although you can push it to 40mA after which the magic smoke will escape) 2. If the above is not enough use a simple 2N7000 MOSFET (simple NPN BJT if you really must) 3. A 1N4001 diode (or similar) must be connected (in reverse) across the relay coil. So the CATHODE goes to the positive side. This will ensure no nasty high voltages generated from the coil as it switches off reaches the Arduino pin. The snubber in this video is to prevent arcing in the relay as it switches off (if there is any arcing). A tiny amount is acceptable; but if the load is large (several amps) you may get quite a flash which will damage the relay after a while. But it won't do anything to protect the Arduino as it is fitted on the "hot" side of the relay (not the coil side).

[@serj2008]

@@RalphBacon I don't have an arduino, I have a raspberry pi 4 connected to a 4 relay module optically, and one of the relays is switching on/off a 220volt solenoid valve. As i understand, when this 220v solenoid is switched off, a very high negative inductive load is going back to the relay module, specifically frying the contacts. This is what I'm trying to resolve with the rc snubber. Will it suffice?

[@RalphBacon]

Yes, that's the whole point of the snubber - but if you don't see a spark (if the case is transparent) you don't have an issue. If you are just trying to be cautious then connect it up and it will do the job. Inductive loads (such as a solenoid) means you have to connect it up across the relay contacts - meaning the solenoid is never truly disconnected from the 240v - be cautious.

[@BerndFelsche]

Watch out for heat sink requirements on the SSRs. They can be OK if you have a naked metal panel surface in your switch box but read the data sheets carefully.

[@RalphBacon]

Ah yes, Bend, I notice that Banggood and others also sell an aluminium heatsink for these SSRs. The problem that introduces is that it all gets a bit bulky, negating some of the solid state advantages. The list of things to consider goes on and on.

[@zaprodk]

@@RalphBacon Combine SSR with mechanical relay. Problem solved and it's used a lot in the industry this way. This way the SSR takes the grunt of the "spark" as we might call it, and the mechanical relay carries the load. Safe, no extra heat etc.

[@TheDiverJim]

So if you want to control an AC motor (ceiling fan) do you need the diode or the snubber or both?

[@RalphBacon]

Neither. The conventional way of controlling a ceiling fan is to put a capacitor rated at 450V or higher with a value of about 1.5uF - 2.5uF in series with the live (hot) wire. Unfortunately, unless you spend days on some complicated maths, it's all a bit trial and error. You can switch between different capacitors but you have to be very, very careful that you don't electrocute yourself via that switch. That's why ceiling fans have dangling cords with the switch well away from human contact.

[@TheDiverJim]

@@RalphBacon so the sonoff light switch and similar sonoff relay I’m using shouldn’t be used for controlling my ceiling fans?

[@davidgrey943]

Hi, Ralph, does the snubber network go in parallel with contacts or does it sit between the contacts and neutral. I have 14 to put in once I find which is the best way to wire the snubber network. Snubbers is something I haven.t dealt with before.

[@RalphBacon]

Both! If you have an INDUCTIVE load (eg motor, fan, something with a coil) then you must wire this up ACROSS the switch - but that means the circuit is NEVER fully isolated and you can still get a shock from touching parts that you thought had been switched off. If you have a RESISTIVE load (heater element without a fan, or perhaps just a tiny fan) then wire it up in parallel to the HOT side of the supply. The load is then still isolated when you switch off. Reasoning? An inductive load will generate a high voltage (aka back EMF) when the power is turned off. As the magnetic field collapses the voltage generated by the coil/windings can be significant and will jump the gap of a switch - hence wiring it ACROSS the switch allows it to dissipate but with the caveat above. A resistive load does nothing when switched off but as the switch contacts move apart the air becomes ionised and can still conduct electricity as a spark for a few milliseconds. By wiring it up across the hot side of the switch, the electricity will not conduct through the air. If you do have an inductive load you are "snubbing" put a big warning on the switch that the load is NOT fully isolated and can still kill! I've seen some designs where a secondary switch is placed in series - switch off one (with the snubber) then the other. The secondary switch can even be automated with a microcontroller!

[@devilvortex1]

Snubber are delay circuit true? The resistor act like a pump.

[@RalphBacon]

Definition: A snubber is a circuit that acts as a protection by limiting or stopping (snub) voltage spikes i.e. switching voltage amplitude and its rate of rise, therefore reducing the total power dissipation. It operates by clamping the voltage spikes but does not alter the ring frequency of the system. www.allumiax.com/blog/significance-of-snubber-circuit-analysis-in-power-systems

[@ANUNAYSINHA610]

can you please put the circuit diagram of how you connected the snubber. in parallel to the relay right? but then wont there be leakage current? Thank you

[@RalphBacon]

If you are using this for an _inductive_ load (eg a motor) then you must wire it across the switch, and *yes* there will be leakage. It keeps the circuit live even though you have switched off, which is why I don't recommend it. If the load is not inductive (eg heating element) then wire it from the live side of the switch to the other side of the load (usually Neutral).

[@ANUNAYSINHA610]

@@RalphBacon okay thank you

[@SergiyGryb]

Hi Ralph, please, explain how did you connect the snabber module: is it parallel to the relay contacts? or in sequence to the relay contacts and a load?

[@RalphBacon]

For a resistive load (like my non-fan heater) it is in parallel to the load - that is from the relay OUTPUT contact and to ground. If it were an inductive load (like a large motor or fan) it might have to be wired across the relay contacts which is NOT safe as the motor would then be permanently "live" - although it is often done.

[@SergiyGryb]

Thank you, Ralph! Clear now. I am looking for a solution to limit an initial current for a resistive load (like heater), something like a soft start for electric instruments. That would certainly prolong life of a switching relay. Just curious if soft start modules would efficiently work with a dumb resistive load? Any idea?

[@RalphBacon]

Varistors are sometimes used in switched mode power supplies. But maybe not very high power. There's a lot of information here: sound.whsites.net/project39.htm

[@michaelbreslin9116]

Would putting both switches in series, from the relay make a dfference/help? If one switch sticks or arcs a bit, the other one will hopefully make the complete break.

[@RalphBacon]

That is my hope, Michael, because that is exactly what I have done (for now). So both switches of the relay are in parallel, thus sharing the load (about 4A each, which if they cannot sustain means it really is a rubbish relay!). One day I hope to get this whole power switch upgraded - but to what, I don't know yet.

[@michaelbreslin9116]

Ralph S Bacon. Yes I saw the bit, you put them in parallel, but was wondering if they might be more effective in series.

[@RalphBacon]

Then each relay switch would have to switch the full 8A load still, thus doubling my chances that they would fail. By putting them in parallel we share the load (and the risk of failure). The same principle is used in power MOSFETs - if you have a MOSFET only capable of, say, 5A, by paralleling two you can source 10A power. Often used in power amps too. Make sense?

[@michaelbreslin9116]

@@RalphBacon . I think when MOSFETs are used in the parallel configuration, it is to share the load, once connected. They are faster than mechanical switches too, so this will help with inrush, and cut off instances. But, the relays in parallel, when the first one disconnects, it effectively has made no difference, since the second one is still connected, and hence it will always be the slower one that takes the hit, and cause possible arcing. I'm only thinking this through logically, and have no real experience of doing this.

[@RalphBacon]

The speed, in this sort of scenario, of a relay disconnecting, is of little importance. Both will drop out within a millisecond or two of each other and still spark (probably) but at least they are sharing the load exactly as you describe above with the MOSFETs. A snubber may be still required but at least they should not melt (as mine did).

[@stevewolftests]

Thank you. Great explanation. I always trying to learn. One point I do wish that people who do these helpfull videos would do is at some point show a wiring diagram for example in your case how you wired up the whole thing, with snubber. When I look at your table its hard to see. I always struggle when people us descriptions rather than a diagram. Even at the end you could say "oh here is how I wired it". Sorry just me. Thanks for video it is informative.

[@RalphBacon]

Great suggestion! I'll do this in future. In the meantime if you go to the GitHub for this video (github.com/RalphBacon/Relays-Snubbers-SSRs ) you'll find a pdf with a wonderful circuit diagram for you to peruse!

[@stevewolftests]

@@RalphBacon thanks.

[@darrenchapman7203]

This type of relay fault is very common in the make/brake relays in grid tie inverters even to the point of melting the solder on the pcb. Many household electricians don't know a lot about electronics so inverters often end up as scrap when $20 worth of relays would have fixed it..I live off grid in the southeast of south Australia and have repaired several inverters picked up from the local scrap metal yard , my SMA sunny island inverter happily provides a grid and controls the grid tie inverters based on the load.

[@RalphBacon]

So, Darren, the lessons to be learned seems to be: * use a well known brand of relay; * don't skimp on the ratings; * don't buy Chinese unless certified; * consider SSRs (along with all the above); * enclose it all in a safe (fireproof) housing; Did I miss anything here? Anyone?

[@Tinker001]

@@RalphBacon Fuses EVERYWHERE! :P

[@ajayshokal9014]

hello sir, I am trying to drive 48VDC motor through Microcontroller, its working fine but at some times arcing problem is occured. Connection- Relay contacts, NO- Phase, COMM- 10amp contactor(A1), GND- A2. when relay is on contactor will pass the 440V signal to a converter , then converter will convert 440v AC into 48VDC. And then this 48VDC is given to DC motor through DC contactor. I have tried 0.1uf AC capacitor series with 100ohm resistor accross 10Amp contactor but not worked, what will be solution for this?

[@RalphBacon]

Try putting the snubber _across_ the relay contacts rather than parallel on the non-hot side. This is not a good idea for AC mains project as the circuit will be _always_ live because the snubber capacitor supplies current and voltage but for 48VDC it should be fine.

[@ajayshokal7193]

@@RalphBacon sir I tried a snubber with value of C= 0.1uf 310VAC, R=100ohm 1watt but not worked parallel across the AC contactor.

[@Hasitier]

I have a 7A garden pump running on a Sonoff 16A with the internal Relais switching it. As the pump has a boot up and auto off sequence the Relais will never switch on or off while the motor itself runs (except for a power failure maybe) but I might think about adding a solid state Relais after the Sonoff maybe?

[@RalphBacon]

Interesting, as I was going to use my Sonoff to ALSO control my heating, in parallel with the relay I just described. In this way I can get Alexa to switch on my workshop heater from inside the house. But this whole episode has made me pause and reconsider. A 16A Sonoff relay SHOULD be OK for my heater (and your pump) but you just never know. Perhaps I'll get the Sonoff to turn on a 50A SSR in just the way you're thinking of doing with your pump. Hmm....

[@Hasitier]

Ralph S Bacon if you have a Relais or SSR to switch the final load (the heater) you could just add a Sonoff SV (Low voltage version) or maybe just an esp8266 with the tasmota firmware to make the bridge from Alexa or mqtt to an output pin that than switches the Relais or SSR (maybe a Transistor or Mosfet is needed to get the needed voltage).

[@RalphBacon]

Indeed, I had considered that (and was my original intention) but time is always against me. My Sonoff is ready to go, already configured with Alexa and everything. But now I'm nervous connecting that across the mains...

[@Hasitier]

Ralph S Bacon why not just triggering your new Relais or maybe SSR with the output of the Sonoff? So it sees nearly no load

[@Tinker001]

@@RalphBacon The simple ESP8266 route is pretty easy & reliable. I have a bunch of ESP01s set up with relays here. If I wake up in the middle of the night & need to use the washroom, I say "Alexa, I have to pee" & she turns on 3 small LED lights between the bedroom & the bathroom for me.

[@smartteg7381]

Hi please can I use one spark absorption circuit for multiple relays

[@RalphBacon]

Not unless the relays are wired in parallel; otherwise it's one snubber per relay (they are cheap enough).

[@j1952d]

Suspect you're first relay melt-down was caused by the arcing indirectly, thus: arcing contacts get pitted and mucky, making them high(er) resistance. The resistance causes heating ( that P = I x V again, which can also be written P = I x I x R, or P = V x V / R), hence melt-down. Your double-pole version would fail eventually - notice that it's only arcing on one side - that's because of the tiny time difference between the 2 sides opening. When the first side gets too pitted and mucky to take the current, the other pole will take over, arc, and you're back to square 1. One or both sides would melt down.

[@RalphBacon]

I'll certainly address this problem for sure in my new workshop, possible with a SSR.

[@squalazzo]

how gets connected the snubber, in parallel with the relay?

[@RalphBacon]

The snubber is connected from the switched side of the relay and to neutral, so in parallel with the heater. It didn't do anything by putting it in parallel with the relay and the danger is that it keeps the heater "live" even when the relay is switched on.

[@Roy_Tellason]

@@RalphBacon This doesn't sound right. I've never seen a snubber connected in any way but across the contacts where you are trying to supress the spark!

[@3adelz90]

I am now confused. I was about to manufacture a relay board but you said we have to add a snubber circuit ! I will order Omron 10A relays from Mouser and will control lightings. I only have a free wheeling diode but no snubber. Shall I add them ? And how do I make the connections ? Thanks.

[@RalphBacon]

No, you don't NEED a snubber circuit. It removes the flash as I demonstrated that will give a longer life to the relay contacts, as they don't get pitted over time. However, I would add one - see the PDF file I included in my Github for this video, it explains where you need to connect it (it differs depending on whether the load is inductive or not). One final caveat, if you connect it across the relay contacts (that is, in parallel) then the "off" side of the circuit is never truly off - there will be leakage current via the snubber that will keep it "live". One reason why they are not always used.

[@3adelz90]

@@RalphBacon Great. Thanks alot for the explaination. I spent 3 hours watching your videos today :D Great channel indeed.

[@RalphBacon]

Wow! You are either very interested or have time on your hands! Seriously, though, I'm glad you find them interesting.

[@3adelz90]

Actually both. We have holiday today and I woke up at 6 am watching your videos :D keep it up.

[@williammiller7543]

Good video. Could you explain why 2 different connections in you PDF and please give some examples of the calculations. Why one across the points and the second parallel to the load. Do you need BOTH? The calcs look straight forward for resistave loads, but I'm trying to use it to turn a freezer on/off for Dry Aging meats. (Controlling Temp and Humidity.) Anyway the freezer is 120VAC and claims 7.1AMPS. Can you provide some rules of thumb for motors (compressor motor on Freezer).

[@RalphBacon]

The two types of connections are for inductive and resistive loads. As my heater is purely resistive I placed the snubber across the actual heater. This is safer than across the switch or relay as that keeps the component at mains/household voltage level even when the switch is off. I'm not qualified to comment further, William, but read the comments here and ensure you have a highly rated component that won't fail down the line in some spectacular manner. Enclosing it in a metal container is best but then you have insulation to worry about. And it should all be fused, of course. It seems that SSRs are best placed for inductive loads like your freezer (they get switched on at zero volts so no inrush current or RFI) but like me with an 8A load, your 7A load should be controlled by a device many times that rating, probably 30A or more. I'll be doing a video about that in the future.

[@davaxtechnology5139]

What can i use to stop spark on my relay when it engages 60V, i bought this snubber and it sparks anyway, it sparks when connect not disconnect

[@RalphBacon]

If it sparks when the relay closes, connect the snubber _across_ the relay contacts. WARNING! This means the circuitry the other side is never truly isolated (you've bridged the on-off relay contacts with the snubber) so you might want to consider an additional switch (if this works for you).

[@davaxtechnology5139]

@@RalphBacon thanks I found ssr relay which i will use Polyt Pqssr-pd Input 3-32vdc output 5a 200vdc So i will try him

[@wendellbriancamarines3135]

Hello Ralph... this vid has really helped me with what i should do with the arc on my relay.. same relay that u have... havent connected the RC yet... still trying to figure out how and where u connected it in parallel... saw a link u commented on one of those who asked for the diagram but i cant open the link in github... if i may ask how it was connected... id really appreaiate it...

[@RalphBacon]

The link in the GitHub for this video points to this pdf that shows the connections: github.com/RalphBacon/Relays-Snubbers-SSRs/blob/master/spark_suppression.pdf Opened just fine. If you have problems let me know!

[@wendellbriancamarines3135]

@@RalphBacon thanks a lot ralph! I hope the snubber does it work... havent recieved them yet but now that i have a diagram feom a person that personally used it, i can assure myself... big thanks! And goodluck to future projects!

[@davaxtechnology5139]

I got OMRON G2R2 DC12V 5A 250 V AC 5A 30 V DC Can i use this relay to switch 90V DC 2A because when i open datasheet it says it can switch 380 VAC, 125 VDC I dont understand why they dont type on relay max AC and DC voltage ? which is now wrong datasheet or what does it say on relay ? Im confused .

[@RalphBacon]

I'm confused too. On the relay is says 30V DC, but the datasheet says 125V DC? At how much current? That is a high voltage for a little relay to switch.

[@davaxtechnology5139]

@@RalphBacon and that on most relays so i need now look for every relay datasheet I need for electric bike just measured it is 85VDC and it pulls 0.25A

[@RalphBacon]

So fairly low current, but the overall power is going to be higher. Keep looking, and be sure. You can always email the manufacturer if in doubt, especially if UK or USA based.

[@davaxtechnology5139]

@@RalphBacon the current stays the same around 0.25 A It is just the ignition for lights so low current

[@rohit032]

I saw this video and brought same snubbers from aliexpress....but sadly in my case Spark is still present even after adding snubber... can you suggest any other snubber can work...? load is 1kw heater....

[@RalphBacon]

Does the heater contain a fan? My heater is purely resistive (no fan or other inductive loads). If you have an inductive load (eg a fan as well as a heating element) then the snubber must go across the switch but be WARNED that the heater will be permanently connected to live mains circuitry (albeit at very low power, but enough to give you a shock).

[@rohit032]

@@RalphBacon thanks for replying... Heater does not contain fan....can heater have some inductive component?? It's flexible cord heater...

[@RalphBacon]

How have you wired up the two wires of the snubber, Rohit?

[@ولدزايد-ذ2غ]

When released contacts in arduino 5v relay LCD become freeze what safety components is good to stop this types of problems

[@RalphBacon]

Sounds like you are either getting back-EMF from the coil (you do have a reverse polarity diode across the coil, right?) or its coming from the relay contacts - how are you powering up your project? Same PSU? If so, try a separate PSU for the relay contacts to whatever that is powering.

[@ولدزايد-ذ2غ]

@@RalphBacon i put relay contacts to open 💡 bulb inside oven and button when released LCD becomes freezer bulb use 250v LCD use arduino nano to control heating elements inside oven

[@WireWeHere]

A drawback to Solid State Relays is their tendency to fail closed. The addition of a secondary overtemp device is prudent and required for many certification organizations. Cutting most SSR ampacity ratings in half is a good start.

[@RalphBacon]

Indeed. Relays do the opposite (mainly, unless the contacts weld themselves together due to overload). 😟

[@richardgrier4721]

I've always called them DIN rail(s), and I'm in the US. I've never heard the alternate designation until today.

[@barryg41]

Been using DIN rails for 30 years. Lol

[@RalphBacon]

Drat, Expedia told me wrong! I reckoned that as DIN (Deutsche Industrie Norm) was a European standard it might not have made it across the pond so I queried it. That's what Expedia told me. I stand corrected. On a related note, do you guys refer to "mains voltage" or "household voltage" to describe the incoming electricity to your houses? In the UK and Australia it's always the former. Once again, I'm told that "mains" has no meaning in the States.

[@richardgrier4721]

We call it either line voltage or mains voltage. I'd rate this at 6o--40, respectively.

[@RalphBacon]

@@richardgrier4721 At least "mains voltage" is a meaningful term then. "Line voltage" here (UK) is what a qualified electrician might say but us DIY-ers just say "(the) mains". Good to know, thanks for the update, Richard.

[@nhikoid]

Re cheap SSR's - I tried using one ( went through three of them ) to switch dump loads on/off ( wind turbine ). One of the SSR's melted. The other seems to read 77v ( ?? ) on output regardless of input signal volts ( 1-32v). Oddly doing a second reading with my multi meter read 242v (expected). The third one seem to leak a few milliamps on output that fooled by voltage sensors. So, back to good old EMR relays. The Fotek lookalikes on ebay/Ali surely can't be real Foteks?. They ( and similar ) are 10 times the price on Farnell/RS etc...

[@RalphBacon]

I'm not surprised by this. An SSR I have, rated at 40A, has a 10A MOSFET inside. For anything mains voltage related we, as hobbyists, will have to pay the going rate in DigiKey, RS Components and the like, if we want components rated at their stated value. Not cheap!

[@wannasri4308]

Sir, my house water pump died. Everytime i push the reset buttons, there is spark inside the relay. Does that means the relay switch damaged?

[@RalphBacon]

Not necessarily, Wan Nasri. If the relay is closing (or opening) and there is a spark it might just mean that there is electricity flowing in the relay. Check whether the relay is switching the electricity as expected.

[@alanknight7867]

Hi Even cheaper SSR = 2 resistors , BTA24 ( 5 for 99p on Ebay ) + MOC3023 ( 10 for 99p on Ebay ) But you will need a heatsink ( but then you need a heatsink for the big SSR's too ) Alan

[@RalphBacon]

I'll check it out, Alan, thanks for letting me know.

[@RikHaanen]

Could a solid state "relay" do a better job?

[@RalphBacon]

It would not spark, but SSRs are big and get very warm. And let's not forget all those SSRs rated at, say, "20A" that have an 8A triac inside 😮

[@RikHaanen]

@@RalphBacon Thanks for the info. In other words you don't trust them. That is good to know

[@youdonotknowmyname9663]

I personally only use these kinds of relays (the ones with the transparent "case") for "control signals", not for switching actual loads. And when choosing a relay, I always use one that can handle double the current that I will need. But that's just my personal preference ...

[@RalphBacon]

I agree with your overrating for relays - the problem is finding one that fits and is not some cheapo thing that will fail 10 minutes after fitting it! This should be OK (it lasted a long time in my old workshop) but I will be doing something different this time.

[@youdonotknowmyname9663]

@@RalphBacon Yes, space requirements are always the the biggest problem ... And I know that buying things "cheap" is always a big temptation, but I have a personal "rule": If it is safety relevant, don't buy something cheap, instead buy from a reputable manufacturer. (So, for power relays that would be Siemens, ABB, Omron, ..., yes the products from those manufacturers are expensive as hell, but you can trust them.) No, I am not sponsored by those companies ;-)

[@victimovtalent6036]

Sir,heating element is coil similar with inductor coil,could be hundred or thousand turn made of particular wire

[@RalphBacon]

Well, yes, the heating element might be wound around a ceramic former, but it will not act as an inductor (not usually, but I guess it needs testing).

[@big10p]

Thanks for the vid. I couldn't see how you're wiring the snubber to the relay, though. I'm currently using a relay to switch on a CRT monitor and mains LED light in an arcade machine, but am getting contact sparks when switching ON. I have Live and Neutral connected through the relay (with a 12vdc coil), and was advised to connect a snubber module across both switches (1 snubber for LIve; another for Neutral), but still get sparks when switching ON. Any help with this would be greatly received. :)

[@RalphBacon]

The snubber is either connected across the relay contacts (usually COM/NO) for inductive loads, or across the hot Live wire and Neutral for resistive loads. However, I have noticed that sometimes it "just works" the other way around. Be aware that if connecting a snubber across the relay contacts then the circuit it still connected to the Line/Live input via the snubber and is not isolated; ideally an additional on/off switch should be fitted before the relay. But you mention sparking when switching ON, not OFF. I'm wondering whether that is switch bounce causing the sparking, in which case a single snubber should still do the trick, connected between the L and N connections of the relay.

[@big10p]

@@RalphBacon Wow, thanks for the quick reply! I will try connecting a single snubber across the relay L and N connections, but should I do this on the mains side, or the load side? Or doesn't it matter? I'm guessing the LED strip light I'm switching is a resistive load, and the CRT monitor (which has a degauss coil that engages every time it's switched on) is an inductive load. Is this likely to cause any problems? Thanks again for your help on this!

[@gpTeacher]

Very instructive Ralph. Thanks!

[@RalphBacon]

Glad you liked it, gpTeacher, it has caused a lot of comment about the safety of relays from China!

[@grimreboot]

Very good demonstration :).

[@RalphBacon]

Thank you! Cheers!

[@JaviepalFordring1]

Excelent Vídeo! But how can I connect my Snubber? Only in parallel with my Relay? I plan to use with a water pump and another circuit for a fan.

[@RalphBacon]

Rather than me trying to talk you through this, look at the following .pdf which gives various scenarios (eg inductive load vs resistive load) and will show you what you need to do: www.bpesolutions.com/bpemanuals/Snubber.pdf

[@JaviepalFordring1]

@@RalphBacon wow!! that a great documentation, thanks!!

[@JaviepalFordring1]

@@RalphBacon remeber that i wanted to conected this to a water pump?, well, i did follow the documentation, and buyed the same Snubber, but my relay end up dead, then i tried only the snubber with a normal Switch, but one of my cables burn, nothing more burned only the conection between the 2 cables, i only replace them, and still see sparks, i don't know if the snubber is not working, or I don't know how to follow instructions. I conected my snubber from the first cable that goes to the relay, to the second cable that goes to the relay. Do you think, maybe a solid state relay solve my problem maybe?.

[@ahk501]

Would a TVS diode work for the same ?

[@RalphBacon]

Not sure, but why would you want to do this (instead of using a 'proper' snubber setup)?

[@boblewis5558]

One other point worthy of consideration is that even though they are "mainly" resistive, heaters of all sorts usually if not always involve coils of resistance wire, and in the case of kettles and immersion heaters and the like, a fair capacitance too. It is entirely possible (indeed probable) that there is quite an inductive (or more accurately? ... reactive) load to handle for both relays AND SSRs. Maybe zero crossing for minimum power detection should be employed instead? Easy to implement? Right now off the top of my head I can't think of an easy, simple, reliable low cost way. Anyone have some ideas?

[@RalphBacon]

I was kinda hoping, Bob, that SSRs would automatically include zero volt crossing - the smaller 2A SSR I demoed recently certainly did have, but I know nothing about these larger 20A+ ones - I will have to look at the spec sheet. Interesting about what you say about coiled heaters. Even if the inductance was low (for a coil), the large current flowing would boost that. Hmm.

[@boblewis5558]

@@RalphBacon yes, you're absolutely right, most SSRs WILL include zero VOLT crossing (if they're well designed they will) BUT my point was about REACTIVE power, especially when inductive loads are involved where the current phase-lags the voltage. At zero volts the SSR will switch off normally (more accurately, will no longer turn on during the next phase of the voltage drive) but there may still be a following phase-lagging current and it's usually the current flowing being suddenly stopped that creates bigger problems. A zero current switching regime or minimum reactive power regime may prove more effective and beneficial. voltage spikes are relatively easy to handle at the low(ish) switching voltages involved but current spikes can cause big problems. I plan (at SOME) point to run some tests using a bidirectional Hall effect current monitor to try zero CURRENT crossover control instead to see what, if any, improvements or issues arise.

[@eladnachmias6627]

Thanks for the video! Now two years later, do you have any more insights about the RC snubber?

[@RalphBacon]

Insights? Well, they do stop sparking across the contacts. The relay I was trying to protect remained working up until I moved house and did not melt like the previous one did. The only thing I don't like (and this applies to my new Smart Heater Controller too), wiring it _across_ the switch or relay contacts means the circuit is never truly isolated which could be a danger.

[@eladnachmias6627]

@@RalphBacon thanks for the reply. Do you have a concrete example what could go wrong because of the lack of insulation you described?

[@RalphBacon]

If you try and service a machine that is switched "off" but has a snubber wired across the contacts to prevent arcing, then the machine is still connected to the mains, albeit now via a capacitor and a resistor - still enough to give you a jolt. Might not kill you but still unpleasant, not to mention dangerous. In this case you need another series-wired switch or a switch-operated relay (so it opens/closes when a voltage is applied via that switch. Current goes through switch _and_ relay but only the one that opens first needs the snubber).

[@eladnachmias6627]

@@RalphBacon Okay now I see the problem. Yes I agree it's not safe; especially for automated home use, where someone might think the power is off and it's safe replacing a light bulb for example. I have installed a series fuse that protects each relay channel and can be used as a switch like you described. The double relay solution you suggested sounds like it solves the problem, but unfortunately it's costly in price and space (especially for multiple channel switching system). Also, if you want it to operate automatically, some sort of electrical circuit is needed for the delayed switch off of the secondary relay. BTW, maybe a simple capacitor over the control for the secondary relay's coil is enough for this desired delayed second switch-off effect.

[@eladnachmias6627]

@@RalphBacon Maybe instead of connecting the 2ndary relay purely in series with the entire circuit, one could connect it only over the sub-circuit of the RC snubber and the end-device. In that case, the main operating current won't have to pass trough it, and maybe a cheaper slim 6A relay could be sufficient for this aim.

[@freemusicscore7281]

Hi Thank you for an excellent video and instructive. Hope you can help me with using snubber. I have SSR-40 DA (similar to the white relay in the video but rated 40A). I used Arduino digital pin to power the 3 - 32VDC. I have no problem switching ON/OFF the 2.5A Pump with 220VAC. I also added a mechanical switch in series with the Relay Contact just for emergency stop for the pump. I didn't use soft starter for this just directly connect the L and N of Pump to the contacts of the Relay. If I switch off the mechanical switch it restarts the Arduino. I am suspecting that switching off creates a spark which causes Arduino to restart. I also tried to disconnect the pump and use a small light bulb (AC) and the problem still persist when turning the emergency switch to off. Could you give advise if Snubber Circuit could be used to remove the spark. Thank you.

[@RalphBacon]

You say the in-series switch causes the Arduino to restart? Does it restart when using the relay too? I'm assuming you have a flyback diode across the relay coil? If you're going to connect a snubber circuit to prevent sparks from an inductive load (eg a motor) then the snubber goes across the relay contacts. (With a lightbulb it is in parallel with the relay hot wire and neutral). However, this means the motor (or other load) is never fully isolated from the mains voltage, so is unsafe to work on. A series switch like you mention is a great idea, but would be even better if controlled from the Arduino a second or so after the relay has been switched off - no "forgetting" to do it then!

[@freemusicscore7281]

​@@RalphBacon Thank you for the immediate reply. I used solid state relay controlled by digital pin from Arduino and no problem with switching on or off the pump with relay (Arduino doesn't restart). Only when I turn off the emergency switch then the Arduino restarts. The emergency switch is so I can manually stop the pump from running in case Arduino cannot switch off the relay contacts or in case some sort of problem in coding happens causing the pump to run continuously. As for the snubber circuit, I will take consideration that the pump won't be isolated if I connect it parallel to the contact.

[@RalphBacon]

I see. Perhaps you need a snubber across the emergency switch too?

[@amoghjain]

your video was very helpful in helping me understand how to drive inductive load from relays! Can you please post a circuit daigram of the whole setup? that would be very helpful!

[@RalphBacon]

If you look at the pdf in my Github for this video (github.com/RalphBacon/Relays-Snubbers-SSRs/blob/master/spark_suppression.pdf) you will immediately see the two ways of wiring up the snubber (whether for inductive or resistive loads).

[@amoghjain]

@@RalphBacon Thankyou ! now that I watched the video again, I understand the wiring! You had it connected across the load! Thanks!

[@jeffbluejets2626]

Hi Ralph, Couple of instances now where Arduino pro mini is used around mains switching devices where the mains device really upsets the Arduino. Any 5v line filtering does nothing. First time was with an optical tacho project connected to a machine with a series mains motor which I overcame by powering from batteries in the end. Now this has cropped up again where I have an induction motor with centrifugal start switch ( driven via a mains relay) making my counter readouts on 2 x 16x2 LCD screens jump the count and hense my measurements being useless. Isolation has not cured it this time though so thought about your snubbers and will give them a go. Not really hopeful though and considering having to re-mod my machine. Both machines used for construction of parts in miniature petrol four stroke engines. One is a cam grinder and the other is for grinding crankshafts.

[@RalphBacon]

If interference of some kind is upsetting the Arduino then it _can_ be cured, Jeff. Put a scope on the power supply rails first, as that's the most likely cause. Filtering and smoothing will eliminate this cause. RFI can be tamed by putting a shield around the Arduino board (you've likely seen such shields around TV tuner modules) and connected to ground. Wires coming into the Arduino (and going out of) can act as conduits for more interference and may have to be shielded (single screened cable will probably be enough). Sounds like overkill but if interference is there is must be eliminated.

[@JohnSmithZen]

The heater might draw more/less current once it gets fully heated up. Maybe you should have let it run longer before doing the spark test. Also I wonder if the first one overheated because the sparks caused oxidation of the relay contacts which increased their resistance.

[@RalphBacon]

The resistance of a heating element is normally higher the hotter it gets (otherwise thermal runaway would result in it melting). So _less_ current is drawn the longer it's on (within a few minutes). Regarding the oxidisation, you could well be right. My new, improved system has not had any trouble.

[@Ciiads]

Good work 😊👍.

[@RalphBacon]

Thank you! Cheers!

[@bastiannenke9613]

Good video, found it while searching for those small snubber PCBs. Wasn't sure if they're really working, recently got a lot of triacs extremely cheap and those modules seem nice for testing, not sure if I'll do proper PCBs at some point.

[@RalphBacon]

Glad it was useful to your investigations, Bastian.

[@Roy_Tellason]

In addition to the couple of places where I've commented on specific points that were being made here, I have one other issue that I need to touch on. That snubber network with the MOV on it is a real hazard. I believe in the usefulness of those parts, and in fact have a number of outlet strips and other things containing them around here, for "surge supression". The point to understand about those devices, though, is that they'll only take a limited number of "surges". After which point they fail. BY SHORTING! Which is why every "surge protected" outlet strip I have here has either a thermal fuse inside of it, or a circuit breaker, to deal with that evenuality. This is also why you sometimes see such devices with an indicator (LED?) that shows "protected" which, when it's not lit, means that the device has reached its limit and blown the fuse. I would *never* wire up that snubber network you show with the MOV in there and not include at least a fuse inline with it.

[@RalphBacon]

That's all good advice, Roy. In the UK, it's not possible to connect a consumer item to the mains electricity without the plug having a fuse, something not true of many other countries. Also, all UK plugs offer the Earth connection (whether it is used or not is another matter). Connecting items directly to the ring mains is a different matter and would benefit from an in-line fuse (or other protection). Thanks for highlighting this important safety advice.

[@Karl_Levine]

My comment will probably get lost in the crowd, But watching this and then reading the comments still left the question of what's the safe proper way to prevent these spikes n the relay contacts for DC loads, someone mentioned these MOV unsafe in these snubber circuits so I'm wondering if someone can advise. I'd like to switch sensitive electronics such as a few mcus & ssds & sensors that total around let's say 8A, safely and spark free. Thanks.

[@RalphBacon]

Nope, not lost in the crowd! You say you are switching 8A, Karl? What voltage are you talking about here? AC or DC? The higher the voltage the more likely you will get a spark, although you _can_ get a spark from relay contacts at low voltages and high currents too. Additionally, if you are driving inductive loads (eg transformers) your chance of a spark is greater, because as the field in the transformer collapses it will generate a very high voltage across the contacts. Because it is a high voltage is will "jump" the gap. The snubber circuit I showed is fine for non-inductive loads when wired _in parallel_ with the relay contacts. With inductive loads you will probably have to wire it _across_ the relay contacts which means that the "off" part of the circuit stays live if switching AC, as the capacitor allows a tiny current through it. Let's not forget the essential protection diode across the relay coil too, as it will dissipate the high voltage generated from the relay coil on switch off. Let me know what you are switching here.

[@Karl_Levine]

Thank you Ralph, I'm switching 5 & 12v DC, the loads consist of a, few Arduinos sensors leds, spinning HDDS & electronic SSDs.

[@RalphBacon]

@@Karl_Levine Have you checked you're really getting sparking at the relay? At this voltage I'd be surprised if any major sparking would occur. And a tiny spark is not going to damage anything (unlike the huge flash when my 240V AC heater switched off). If you are getting some sparking, use the snubber (or similar) that I showed. It should be absolutely fine. Connect it between the hot relay connection and ground. Another way to switch your power without any arcing at all is to use a MOSFET (basically a solid state relay but much more compact). They could handle that sort of power in their sleep. Check out videos #172 and #173 (kzbin.info/www/bejne/bICrkI2feZerpas and kzbin.info/www/bejne/iWPcYqJ3i9KrmqM ).

[@Karl_Levine]

Ralph thank you. Yes seen it spark twice which when disconnecting, and I'm thinking it's probably more than I can afford to risk considering it's powering these boards and hdds. Relays are switching positive 5v - 12v range So wondering if you would use that specific snubber board you show on it and also if you think it should be wired between each positive NO terminal & ground. And also wondering if you have any worries about that MOV.

[@RalphBacon]

I certainly don't have any concerns about the Metal Oxide Varistor (MOV) in this configuration. The energy it will dissipate will be tiny compared to a mains-powered configuration. Connect the snubber between the NO and GND and if that has no effect between the COM(mon) and GND. It can only be the inductive HDDs causing the back EMF voltage, I would have thought. Have you thought any more about using a MOSFET solution? Given the (low) voltage in this circuit it's what I would do. No moving parts, no arcing. Just ensure you put a protection diode across the device, if there really is some soft of inductive voltage being generated by your HDDs. Just a suggestion.

[@lezbriddon]

snubber, so if the (r.c) are in series, where does the varistor connect?

[@RalphBacon]

It also connects across the two mains wires, Lez. It does nothing normally (it has a high resistance) but should a high level spike appear the resistance drops and it can absorb the very quick pulse.

[@Roy_Tellason]

@@RalphBacon I see one problem with that little snubber board. When MOVs absorb transients, they gradually degrade, until they fail. And when they do, the fail _shorted_, which is why ever instance where I have seen them properly used has either a fuse or a circuit breaker inline with the device...

[@MrBobWareham]

Ralph, you should always de-rate a relay in a DIN rail relay holder by one third as the contacts only push together if you want to use a ten amp relay for 2kw load hard wire it to the cabling for safety and ad a snubber across the contacts to stop contact arcing I always do when running motors or anything else a good choice ready made Roxburgh - RE1201 is the one I use and comes potted with two wires for connection across the relay points all the best Bob-UK

[@RalphBacon]

It's amazing, Bob, how many people are aware of the dangers of cheap (fake) relays. Now I am too. I'll be changing that relay to the 30A one or better still an SSR rated at 30A+ on a heatsink too. Belt and braces and all that...

[@whitefields5595]

Ralph, Another method is to get rid of the relay altogether and use a TRIAC with a zero-crossing detector. That is what is inside the 'solid state relay'. You can get clever and use the Arduino interrupt function to tune the firing point if you want to separate the zero-crossing detection from the Triac. Which ever way you do it the triac will switch at the point at which the mains crosses zero so no load switch and no spark. Be careful on inductive or capacitive loads though as the voltage and current are not in phase. You could also phase-fire the triac to control the heat output too. You need to do a triac video

[@RalphBacon]

A previous video of mine did cover this somewhat but I didn't know you could get those super high power SSRs for a reasonable price (although now I'm beginning to wonder about the safety aspect). If I go down that route, I think for an 8A heater I should go for a 30A+ SSR, to allow for the Chinese exaggerations!

[@whitefields5595]

@@RalphBacon Ralph, If you buy Chinese triacs, then its back on the Naughty Step, if you live that long that is! I suggest you use the BTA16 600-B I use this UK supplier of ST Electronics (Netherlands) triacs www.bitsbox.co.uk/index.php?main_page=index&cPath=140_167_170 You will need an opto-coupler too, with or without zero-crossing detection (MOC 3021/3041) Depends if you want to phase-fire using interrupts on the Arduino

[@RalphBacon]

I'll compare that with what is on offer from our Chinese warehouses. It looks like it could be a lot safer inasmuch that known components are being used.

[@hoggif]

Triac and zero crossing swithes *on* at zero point. That avoids high dv/dt and current spikes at turn on. Great with resistive loads. Any triac switches off near zero crossing when current has dropped to minimum holding current. With resistive loads relays can suffer from high turn on current spikes. You can avoid this by measuring mains voltage and turning on near zero crossing. Sometimes this can cause contact welding. The problem here is arching when turning off at high current point (ie. high voltage with resistive load). The solution to this too is to measure mains voltage and turn off near zero crossing. Actual point of turn-off is probably a couple milliseconds before zero crossing because of delay in relay. A brute force solution could also be using a large snubber network across relay contacts. Turning relay off with current flowing (with some finite inductance in wires) means the current continues to flow some way. It will make voltage spikes large enough to arc over if nothing else is available (like snubber network). (SSRs can be a real pain to dissipate the heat unless you have plenty of metal surfaces to use around! As neat as they usually are, heating with large currents is easily a huge problem! Something like triac in it can drop perhaps 1V which would make 8W at 8 amps. That needs a very big heat sink!)

[@iqbalmann4534]

I ordered this so you put it across the contacts? Not the cool right?

[@RalphBacon]

Put it across the contacts only if: 1) the load is inductive (eg a motor or fan) 2) it doesn't do the job if you connect on the hot side of the contact and neutral. Remember that if you do connect across the contacts then _your load remains LIVE even when the relay is open_ because the capacitor in the snubber conducts electricity. Not much, but enough to give you a jolt.

[@iqbalmann4534]

Thank you so much for your time . My project is a dc coil switching dc also Very inductive solenoid for hydraulics. Could I use a diode as well?

[@socjha6796]

My heater run even though the temp set to zero..After a while, the contactor or maybe the relay spark and it stop running. Later on, I turn the temp to 80C. and it run normally (I think). I don’t know why it spark.

[@RalphBacon]

It sparks because it is disconnecting an inductive load. As the field in the motor collapses it induces a large back EMF which you see as a spark across the relay contacts (because high voltages do that sort of thing). No need to worry unduly unless the spark seems unusually large!

[@tomgeorge3726]

Hi, Ralph. can I offer this scenario for the failure of your relay. Each time your relay opens it was producing an arc. That arc, burns the contact surface causing; 1. Pitting of the contact surface. 2. A deposit of carbon and burn contact on the contact surface. Eventually that contact surface has developed a voltage drop due to loss of contact area caused by pitting and increased contact resistance due to carbon and burnt deposit build up. The heat generated cooks the NO contact enough for heat to conduct down the terminal and melt the relay base. I looks like the fan in the heater is your inductive load, do you have a power monitor box like Big Clive to see what the power factor is?

[@RalphBacon]

A very plausible explanation Tom. But whatever the reason, I'm hoping to eliminate the (potential) overload by splitting the load between two sides of the relay (done) and by fitting a subber (next on my list). BTW my heater has no fan, it is pure convection (hence the big fan on the wall pushing all the warm air from one side of the workshop to the other! I think this will be part of my heating evolution too, at some point in the future...

[@keesouwehand4209]

Hello Ralph, have you considered the possibility of an inrush current being present at the startup of your device?

[@RalphBacon]

Yes, Kees, it's very possible that the heater element is nichrome so will have a lower resistance when cold (hence what you're suggesting about the inrush current) which rapidly drops as the element heats up (which it does very quickly). So it might take 10A or even higher just for a few seconds at the start of the heating cycle, thus overloading that relay. Not enough to damage it immediately but over the 4 months I've used it... especially as it failed during colder weather = more frequent heating cycles. If that is the case, then by parallelling the relay switches it should survive, but as Big Clive has shown the mains operated relay has other problems with the LED resistor... Sigh. I will have to fix that whatever.

[@allexandruprroca6713]

thank you a lot , solved my vacum cleaner switch arching problem

[@RalphBacon]

Great stuff!

[@raym9691]

Could you please do a more in-depth video on the switch system you are using I'd like to do something similar in my shop thanks and good work !

[@RalphBacon]

Can you just clarify what you mean by switch system, Ray - you mean for this heater or my workshop in general? How my workshop is wired up?

[@raym9691]

@@RalphBacon I mean how you have the master switch set up to turn off all of your outlets in your workshop. Thank you for your response!

[@RalphBacon]

I tell you what, Ray, I'll draw it out and put it up in my next video's GitHub, how about that? Keep a lookout for it. If it's missing give me a nudge!

[@RalphBacon]

Actually, I've uploaded it to this video's GitHub. It's a simplified diagram (it doesn't include the wall heaters thermostat and relay etc) if you need all that give me shout and I will draw it out in more detail. The version I've drawn out is what I first started out with for several months. github.com/RalphBacon/Workshop-Heater-Repair-Using-Arduino-Nano

[@geordiegrumble4584]

Hi Ralph. I suspect that your relay problem is caused by a loose connection in the socket, and is causing arcing and overheating. The sparking contacts in this scenario would be incapable of causing this particular damage. It's nothing to do with the relay! Keep up the good work!

[@RalphBacon]

There are so many possibilities on this one, Geordie Grumble, I will have to investigate each in turn and regularly check it to see whether it's still happening, I guess. Thanks for posting, nice to hear from you.

[@FryGuyNS]

Watch the cheap SSR's a lot of knock-offs on Aliexpress, Banggood, eBay and Amazon. Got burned too many times, I will only buy from Digikey, Mouser, Jameco or similar electronic component supplier. They work but are at best half the rated power on the sticker. There are SSR's with many input voltage ranges. Can be AC or DC or both, low voltage and full AC rated voltage inputs. They require a heatsink, running near its rated power they get warm. Good idea to size well above expected load to keep them cool. Good SSR's generally are in the 40-70 USD range for 20-35 amp.

[@RalphBacon]

Quite so, Jonathon. In my first exploration of SSRs, that was the price I was encountering, hence why I did not go down that route. When I discovered the Chinese clones I was pleasantly surprised but I've been made aware now that they could cause big problems, especially if run anywhere near their stated current value. I'm pausing in my eagerness to obtain one until I determine whether it's all worth it (I would need a 30-40A relay for my 8A heater just to be safe).

[@FryGuyNS]

@@RalphBacon I've destructively disassembled several of them they are built well internally. Good isolation, fully potted. They have the zero-cross circuitry and optoisolator you'd expect just the SCR is way undersized. If you do use an SSR in a project keep in mind they can fail stuck on.

[@RalphBacon]

I've ordered one now (massively overrated for what I need) PLUS a heatsink that I might not even need. At the very least it will provide some interesting video points! Actually, I'm not sure it will even fit inside my consumer unit (fuse box) with that heatsink, I will have to wait 2-3 weeks before it turns up (Chinese New Year not helping me here). Thanks for your evaluation, seems to mirror pretty much what others have discovered. Now I wonder why they didn't use the correct triac; for a few pennies more they could have had a reliable product.

[@FryGuyNS]

@@RalphBacon The only thing that makes sense about why they don't use the correct Triac component, that I heard about is that after a production run employees will produce extra units from lower grade seconds parts after hours to sell for extra money. This would explain to me why many of them look exactly like legitimate parts for most of them. Just may have small differences between them. It's a cultural difference I don't think I'll ever understand why it's done. I just wished that they wouldn't be so misleading on the specifications, even if the proper specifications were listed its still a good value for the cost. Maybe that's more of the reason why I don't understand it.

[@itsnotfar]

Nice video, thanks

[@RalphBacon]

Glad you liked it!

[@kennmossman8701]

You are aware of switch bounce - same happens with a relay; it doesn't 100% close; there is a transition state and bounce which can cause sparking.

[@RalphBacon]

Indeed Kenn, that is also a cause of sparking althbough when switching something OFF it is usually more noticeable.

[@thomasvnl]

SSR's have a drawback in that they use triacs and get hot. Also the SSR's that are supplied by a lot of sellers on eBay, banggood, etc. are cheap knock offs that are not rated what it says on the label, so be careful using them on your mains Ralph!

[@RalphBacon]

Yes, Thomas, that much is becoming very clear to me (regarding knock-offs, substandard components etc). I will have to be very careful. I would like my workshop to remain intact, without scorch marks (or worse).

[@andymouse]

Hi Ralph....Yep, solid state for me too.

[@RalphBacon]

I shall certainly be checking them out. But I don't want them exploding either. See comment above from Master of None 68, and watch that video. Worked fine for FOUR MONTHS before exploding.

[@andymouse]

@@RalphBacon Check them out, yes of course, but we used to use them in our electron microscopes from low end to very expensive instruments and I never heard of a failure, however this was some time ago and they weren't cheap then, I think its a case of being generous with your ratings and dont buy knockoffs from our Chinese buddies when dealing with mains and potential damage to life and property...……..with this in mind I think they are worth a serious looksie...….dont no about you and fellow Arduinite's those see through mains relays look kind of scary!

[@ronen124]

good job Sir Ralph ✔😀👓

[@RalphBacon]

Thank you Ronen!

[@Saskaburner]

This looks like goofy in several ways. #1 A snubber is for an inductive load not a resistive load like a heater. You could have used it on the coil circuit of the relay but not the load side contacts. #2 How did you get a spark after the wire was pulled of the snubber leaving an open circuit to the load? It would seem there are things not shown on this video. A proper wiring diagram would be useful.

[@RalphBacon]

Goofy it may look, but you must have experienced a light switch giving flashes when switching off a light? Even more obvious when switching off a heavy load such as a kettle or water heater. All of which are resistive not inductive. In this Real LIfe example, my resistive heater is switched via a relay and flashes each time it switches off. With the snubber attached, as shown in the video, it no longer flashes.

[@juanrivera4938]

Hey Ralph, awesome video! the best I've seen by far :), can i bother u with something I cannot find anywhere please? I'm designing my own pcb where I want to protect 3x75W fluorescent lights with a 19A inductive load with 1 relay 30A 220V but i cant find the Schematic anywheeeereeeeeee, you gave the components which is the biggest part, do u have them anywhere by any chance? thanks! and good luck! ill watch ur other vid, create a patreon!

[@RalphBacon]

I'm confused, Juan! You are "protecting" 3x75W fluorescent lamps... OK... with a 19A inductive load...??? from what? Describe to me what it is you are doing and I'll see if I can assist.

[@juanrivera4938]

@@RalphBacon Thanks for the super fast reply! we are using these ballast: www.osram.com/ecat/QUICKTRONIC%20PROFESSIONAL%205-ECG%20for%20FL%20and%20CFL,%20not%20dimmable-ECG%20for%20fluorescent%20and%20compact%20fluorescent%20lamps-Electronic%20Control%20Gears%20for%20Lamps-Digital%20Systems/com/en/GPS01_1027993/ZMP_1008329/ as you can see, they have an Inrush current of 40 A (i made a mistake before, they are 3x75W normal consumption, 3x40A inductive load not 3x19A) I'm trying to figure out two things. 1, the schematic of the module of the video and 2, how many of this electronic ballast I can put per snubber. Currently, I'm using 1 relay per ballast but the PCB is huge, I would love to be able to use -hopefully- one single relay for all 3, if more better. thanks! pd: I'm developing www.ivy.cl, everything is 100% designed and built by us and we are a Startup in Chile :), if I can help u with UV-C let me know!

[@RalphBacon]

Let's just restate what the snubber does. As you switch off the power (to the relay, which itself supplies power to the load - lamps in your case) the air across the relay points conducts electricity as the air has become ionised. As the gap gets wider (as the relay opens the contacts more fully) it cannot continue to conduct electricity so the sparking stops. To suppress the sparking (which can eventually damage the relay contacts) a snubber is placed either from the hot relay contact and neutral or, in the case of inductive loads, across the relay contacts themselves. Placing the snubber directly across the relay contacts is not ideal as the circuit is not fully off; indeed the wiring could still be considered hot even though the current will be limited by the capacitor/resistor of the snubber. See bit.ly/snubber1 but the arcing should be gone. This arrangement may not be acceptable for commercial products without additional safety measures (such as a further relay contact - in series with the other relay contacts - that is switched off after a second or so, thus isolating the power to the circuit). If you are working with standard mains voltages (120v / 240v) then the appropriate snubber will be suitable for whatever load (current) your draw from the circuit, so a single, suitably rated relay with one snubber should work. But that inrush current looks very high; have to thought about limiting that with an NTC device(such as a very low resistance thermistor that initially has a higher resistance to limit the inrush current - but as it gets warm it reduces the resistance to an almost imperceptible amount)?

[@juanrivera4938]

​@@RalphBacon I was just studying NTC and came across ur snubber video that seems like an easy solution to my issue. Studying further your idea in NTCs I came across a wall. Every calculator or datasheet of NTCs need a Capacitance value to calculate the total energy and therefore the size of the NTC. www.ametherm.com/blog/inrush-current/calculate-inrush-current-three-steps any ideas on how I can calculate that for the electronic ballast? It's not in the datasheet or my google findings. Thanks!

[@RalphBacon]

Have a look at one of the major electronics suppliers (Eg Mouser, RS Components, Digikey) and see what they sell. The don't always tell you the capacitor or resistor values but typical values are 100Ω resistor and 0.22μF cap (all rated for the voltage you are using). I don't know of any formula to calculate this.

[@eladnachmias6627]

Have anyone came across such RC snubber with multiple channels?

[@RalphBacon]

You mean to cater for several switches?

[@eladnachmias6627]

@@RalphBacon exactly. I'm looking for a single circuit/module with multiple RC snubbers connected in parallel, rather than connecting multiple separated RC snubber modules (like the one in the video) in parallel.

[@RalphBacon]

I'm afraid I've never seen them. Given that they are quite chunky, and involve mains voltages, I'm thinking that using multiple ones in parallel is probably the safest thing to do.

[@joew4202]

Why didn’t you show yourself actually wiring the snubber???

[@RalphBacon]

Too slow, and frustrating to my viewers, to watch me fumble about putting in two wires!

[@joew4202]

@@RalphBacon schematic at least? The reason I watched this was to learn how to wire a snubber.

[@RalphBacon]

There are only two wires; one goes to N (neutral) the other to L (line) but it depends whether you wire it BEFORE, AFTER or ACROSS the switch. Wiring ACROSS the switch is sometimes necessary with inductive loads (motors etc) but means that the equipment is never truly isolated by the relay any more. An additional isolation switch is a good investment in such a situation. Wiring BEFORE or AFTER the contacts depends on whether the current is flowing because there is energy remaining in the circuit (as I explained, in an inductive load) or whether it is the current flowing through ionised air. I'd try "before" first, then "after" wiring.

[@darude1420]

Wow that spark at 11:35 was so good, an advert appeared

[@RalphBacon]

Hmm. I don't allow mid-roll adverts so I guess it was one of those things at the bottom of the screen?

[@darude1420]

@@RalphBacon I think it was. Don't worry, the timing was just unbelievably spot on. Really enjoyed your content in this video!

[@ch2o2]

... is it really just an ohmic load without an inductive component in your heating system? The whole cable system up to the heating itself forms an inductance, albeit a small one, and resistive heating elements are also usually built in the form of coiled resistance wires, which are wrapped around a ceramic body - this also means a (minimal, but at least) inductive load. It would be very informative if one could look at the voltage curve at the contacts with an oscilloscope in order to be able to precisely identify any voltage peaks that may occur. Of course, the intensity of the contact spark also depends on the point in time at which the sinusoidal oscillation is switched - at a zero crossing there is of course no contact fire and the contact load is maximum at the zenith of the oscillation. In the past, in order to avoid contact fire, one had to use RC elements parallel to the contacts. The varistor should help to cut the voltage peaks - is there a circuit structure somewhere? In addition, it is very interesting how sympathetic and methodical you make your contributions and clips ..respect and greetings :)

[@RalphBacon]

Whilst technically you're correct inasmuch that everything can present some sort of inductive load, the examples you present will not affect the sparking at all and can be treated as a straightforward resistive load. It's just motors of some kind that are truly inductive for this project.

[@whatsthisnow101]

Hello you might be better with a contractor and using the spare poles as additional contacts to break your circuit i.e. wire the poles in series with your load, this will increase your contact life. Also use relays and parts from a reputable supplier i.e. Siemens, ABB, Schneider etc. and not Aliexpress. Your load is feeding a heater and therefore the load is resistive and you will have no back EMF, also as it is a AC supply the current zero crossings will help to extinguish arcing. We feed inductive and resistive loads much higher than this using contractors without issue. Relays we reserve purely for control only with a contactor feeding any load. Also certain contactors have a higher SIL reliability score, and spare poles wired in series will further increase the risk of contact weld.

[@RalphBacon]

Well, this has been mentioned, I believe, by others but if we are to be really pedantic about this, consider what a contactor really is: What is a electrical contactor? A contactor is an electrical device which is used for switching an electrical circuit on or off. *It is considered to be a special type of relay.*

[@analoghardwaretops3976]

​@@RalphBacon relays are usually used in control circuitry..though it's contacts may be "suitably rated" for your requirements..bear in mind... 1)contacts are slow moving in either open-to-close ..or close-to-open. 2)a relatively small gap between open contacts 3)a 10A (ac) contact is usually safe for 10A inrush sustained for few 8-10 elect cycles during which contact bounce should settle down. 4) a closed contact carrying current when being "open circuited" or turning off , the contacts require a higher breaking/separating force to pull the moving contact away from the stationary one..and its spring force may reduce over time specially if it is subject to excessive heating. 5)hence relays are usually wìred to switch on/ off Contactors that are supposed to handle the power/load circuit current.. 6)Contactors have the make/ break contacts that are of fast snapaction type and not sluggish. 7)each contactor NO/NC moving contacts are isolated @ both ends in open circuit mode & there's a sufficiently large gap when in open circuit. this reduces sparking to minimum when in operation.

[@Ed19601]

Interesting. I found them a tad cheaper at Aliexpress (74 cts + 12 CTS S&H), looking forward to your video in a year

[@RalphBacon]

That's if I still have a workshop in a year's time, Ed, rather than a smouldering pile of charred wood...

[@Ed19601]

@@RalphBacon let's put our faith in Chinese technology

[@RalphBacon]

@@Ed19601 ROFL best joke of the year, better than any Brexit joke. Note in my library: "The post-apocalyptical fiction section has now been moved to Current Affairs".

[@MrEdwardhartmann]

I think someone has already said this, but if you are going to run a solid state relay continuously, you need a really big heat sink. Matthias Wandel did a video of an SSR on his hot water heater that burnt up, but he was drawing 22Amps. He did have the SSR mounted to a metal plate, but it was not a real heat sink. kzbin.info/www/bejne/fIecpWR9e7uVnrc

[@RalphBacon]

I saw that video too, Edward, boy was he lucky. If I ever go down that route I will de-rate the relay to about half its value, add a heatsink and stand well back! Sometimes it's not worth the risk, especially as his worked for FOUR MONTHS before nearly burning his house down.

[@jamesmclaughlinprimitivele4587]

Triac is better since it only cuts out when current is off.

[@RalphBacon]

Well, wait for my video on my SSR before making your mind up!

[@ARIZACI]

O minik rölelere güvenipte asla bir ısıtıcı bağlamam! En azından 40a bir kontaktör olmasını tercih ederim. Minik rölelerin üzerinde 8a 10a yazdığına kanmayın, tamam anlık olarak 8a 10a yük çekebilir ama sürekli bu yüklerin altında kısa sürede ısınıp eriyecektir. Yangın bile çıkarabilir!

[@RalphBacon]

My Turkish is a bit rusty but I think you said: I would never connect a heater without relying on those tiny relays! I would prefer to have at least a 40a contactor. Don't be fooled by the words 8a and 10a written on the tiny relays, okay, they can handle 8a and 10a loads momentarily, but they will heat up and melt in a short time under these continuous loads. It might even start a fire! And I totally agree, most of the time I de-rate most mains electrical items from China by about half. However, in this instance, I believe I bought them from a reputable UK supplier, who I hope has done some due diligence and has ensured these are up to the job.More expensive but, as you say, we don't want any fires.