Variacs DO short their windings

[@mikeselectricstuff]

I just had a look at the brush on my 10 amp variac. It's a 10x10x10mm cube of brass, with a small T-profile length of graphite set into it, about 10x5x5, the edge that contacts the winding being about 1.5x the width of the winding pitch. The graphite measures about 1.5 ohms in the winding-to-winding axis, and 0.3 ohms from the brass to the contact face - the whole length of the graphite insert is effectively all in parallel to keep the resistance to the brass low, while having enough resistance in the winding-to-winding distabnce to minimise heat from shorting turns. I know someone who tried replacing a broken variac brush with brass. it didn't end wel...

[@cheater00]

Hey Mike, have you ever popped into the synth-diy mailing list? I think you might find it interesting A lot of analog electronics design happening.

[@timjackson3954]

Likewise on my 2A one. Mine came with a spare brush, which I have never needed and occasionally find lying about the workshop (it's supposed to live in the bottom of the Variac cabinet). I notice that the windings are actually set slightly apart and embedded in epoxy so that the contact surface is quite flat and the 'dwell' is very short. I imagine if and when the leg of the T wears down it will be another story, but it will probably outlive me.

[@bigclivedotcom]

Someone else mentioned a tiny brass roller that rode across the windings. I tried measuring mine, but didn't see significant results with my ordinary meter. Maybe I should have tried my installation tester for a higher resistance test current.

[@cybermaus]

@@bigclivedotcom I have a heavy one, that has five ! ~1 cm carbon sliders side by side so with spacing the entire takeoff area is ~6 cm , using the height of the wingdings on the outer circumference. (for normal use I have a smaller 2Amp one though)

[@ketas]

i have second hand ones with copper cored graphite wheels, but they got stuck and have been rubbed flat on one side

[@Nono-hk3is]

This is my variac. There are many others like it, but this one is mine.

[@tisme1105]

Full Insulated Metal Jacket.

[@ucitymetalhead]

I picture electroboom with a crazy look in his eye and a big capacitor ready to blow.

[@virtualtools_3021]

This is me bottle o' scrumpy. There are many like it, but this one is... MINE! :D

[@clytle374]

Where do you get one?

[@rbmwiv]

I have never seen a variac like that. What’s the name and model? I’m interested in a low current one with dc possibility my variac will run up to 20 amps and I have a extension cord I made for testing ac powered stuff with an incandescent bulb wired in series on one leg. That way if there’s a short the bulb gets bright instead of seeing the magic smoke coming out of the equipment that you’re fixing. One last question do they still have that motorcycle race on The Isle of Man? I remember that when I was younger and thought it was awesome going around an island as fast as possible. Thanks for your help and time.

[@AndyFletcherX31]

I've wondered about the shorted turns in the past on variacs in the past. Thanks for actually measuring one.

[@jercos]

Years of armchair theorizing destroyed in under 10 minutes with a thermal camera and hopi.

[@LazerLord10]

Hmmm... I imagine that they can get away with shorting the windings because the voltage difference between the adjacent windings is very small. I'm commenting this before your measurement on the HOPI, but if there is a 80mm diameter variac and a 3mm wide contact, the maximum voltage across the contact would be 240 * 3/(3.14*80) = 2.8V. Obviously if it's shorted, the voltage is ~0V, but this shows the scale of things relative to 240V.

[@tiporari]

/\ this /\ Small voltage drop = small current dissipation. Good 'nuff engineering.

[@arthurmoore9488]

@Flat Sign Big Clive mentioned he checked for that, and it seemed to be normal carbon with the same resistance in all directions.

[@LazerLord10]

@Flat Sign He tested this in this video. Some fancier high-powered variacs might use this, but this one doesn't. Also, if there are multiple vertical contacts, which one is connected to the actual wiper?

[@billr3053]

@@LazerLord10 Depending on the physical preciseness of the contact - I would say the 1st wire that makes contact is the effective one.

[@hausaffe100]

the problem is not voltage but magnetic flux

[@ncot_tech]

I really like the direct relation between turning the dial and the analog meter moving. I've got too used to watching digital meters dance around as their electronics fuss over what the actual value should be.

[@spaceted3977]

My Variac is Digital, but the display and the dial are not at all accurate. !!! It's important to use a Volt Meter unless you want to blow up something !!!!

[@tncorgi92]

I remember as a kid having one of these for the lights on the Christmas tree, you could dial down the brightness when the room was dark. Of course being a kid I had to plug all sorts of things into it to see how they were affected.

[@RC-nq7mg]

That was likely a rheostat, not a variac. resistor as opposed to transformer. Rheostats were very common for lighting dimmers in the day.

[@nullerrno]

Found my variac in the e waste bin at my university. Not much was wrong with it. Fixed it up. Works like a charm.

[@MichaelBeeny]

As a young lad, many years ago, I had a model train speed controller. Imagine my delight when, after taking it to bits (as one does) I found it was in fact a variable transformer from about 6 volts to 25 ish. The variable part was an isolated secondary. Amazing for a toy. Needless to say it was never connected to the train again.

[@jercos]

I've seen a system much like that, including a fixed tap for 12v AC lighting! Very cool stuff.

[@MichaelBeeny]

@@jercos Yes, I remember that now, made by the German Märklin model railways. Much better than the Hornby stuff that used DC and a tacky resistance to control the speed.

[@whitcwa]

I have an old GE variable reluctance transformer. It has a rotatable core in a cylindrical stator. It creates variable voltage without the steps associated with brushed windings. It's much heavier than an equal powered Variac.

[@LakeNipissing]

The windings on these variacs aren't always with fine wire. I've got about a dozen old Statco 20 Amp, 0 to 120 VAC variacs, and the windings look like 14 GA. The carbon brush looks like it contacts more than one winding as well, but I guess it doesn't hurt it. Does dim the lights in the room for an instant when the power switch is first turned on, even with no load connected.

[@bigclivedotcom]

That sounds like the classic behaviour of a toroidal transformer.

[@Ranger_Kevin]

Yeah, I got my hands on an old Grundig RT5A combined Variac/Isolation transformer that can output up to 3,5A, and it actually has a "soft start" mechanism so it does not blow the breaker when you switch it on. It actually needs it, when I first got it I forgot to reset the power switch and turned it on at the outlet - and suddenly it was dark ^^ ) Built like a brick outhouse, really heavy and the wires on the secondary are really thick don't know what gauge, though. But must be able to take afair bit of heat. (www.radiomuseum.org/r/grundig_regel_trenn_transformato_2.html) Now I am kind of intrigued to pull the cover off again and look at it with my cheapo thermal imager.

[@swilwerth]

Magnetizing inrush current. It happens to all transformers, but it goes larger as the transformer size. Power transformers at substations makes crazy sounds when started because of this.

[@SigEpBlue]

My General Radio W20 Variac is probably very similar in construction, and yeah, its windings are about that size. It tends to pop the circuit breakers in my apartment if I energize it without a load. But it helps if I place a light resistive load on it, say a 40-W incandescent lamp, and the knob around 20-30%. FWIW, the large aluminum disk that rotates and holds the brushes is referred to in GenRad/IET documentation as the "radiator". The whole disk is electrically 'hot' during operation -- found that one out the hard way! 🤣

[@mysock351C]

That's how my 14 amp Staco Energy one is. Wound with coarse windings that are just ground flat on top, and the brush is the width of the core thickness. These things are ridiculously expensive, too. Got mine in new condition for free, but I think it was like 400-500 USD new. The sudden inrush is due to the core saturating since there is no magnetic field present at turn-on to resist the incoming AC. Turning the variac on near the zero crossing causes the windings to pretty much only present their DC resistance once saturation is reached, and draw LOTS of current until the field builds up. I have a large Denon receiver that does the same thing.

[@lwilton]

Yep, the brush is designed to bridge at least two turns. But it is carbon so it has some resistance, and thus it isn't a dead short, it is just a low resistance across a small voltage. The brush is designed to bridge turns so that it will work effectively as a rheostat and the output of the variac will not step in voltage, but will be a more or less smooth increase or decrease as the arm is moved. The rheostat effect comes about as the area of contact changes from one winding turn to the next. BTW, in place of that simple digital volt meter, you might want to look on eBay for a Peacefair model PZEM-021 or model PZEM-022 meter. This will do everything the Hopi will, and fits in about the same hole shape as one of those digital LED voltmeters. They claim to have a minimum input voltage of 80V because they are powered from the same line they are measuring. However, a little looking at the PC board will show you where you can cut a trace and separate the power input from the measuring voltage input. At that point it is fairly easy to adapt either one for use with a variac. They use the same chips as utility-supplied electric power meters, so they are pretty accurate. Search terms are "digital power watt meter monitor voltage voltmeter ammeter" and the like.

[@CollectiveSoftware]

It never occurred to me that connecting adjacent windings would do that, but it makes sense

[@joed3483]

Clive, you are amazing. I love your videos and you make complicated things very accessible and even friendly with your kind delivery. We use variacs in the music industry to get the “brown sound” by slightly lowering the voltage for guitars amps. I have wondered for years how these things worked. I knew it was “like a transformer” but couldn’t figure out how you would build something like this. You, as always, really connected the dots! Thank you.

I had the same incident with my 3D printer a few times. I discovered that adding a very thin layer of PVA glue to the bed helps with adhesion on those zones where bed level is not perfect and, at the same time, stops the plastic from fusing with the glass bed. To apply the PVA I use a generic Dealz (Poundland) spray bottle filled with Dealz liquid PVA lightened with distilled water (filtered is good enough) and a bit IPA, to help the water to evaporate faster. Everybody has a slightly different solution for this issues, so you'll get lots of suggestions, my recommendation would be to test and find the one that works better for you.

[@MisterMooo]

I uses Kapton/polyimide tape for my 3D printer or those flexible reusable plastic beds. No more scraping.

[@reme2493]

you don’t need to add anything to a glass bed to induce adhesion. I print fine on mine without anything.

[@jasonudall8614]

Prrit stick

@@reme2493 It depends a lot on the printer and the material you are using. Ideally, you don't need to add anything to the bed, you are right, but reality is a bit more complicated. I personally was a huge fan of the "nothing better than a perfectly clean bed", until I started doing more and more prints and started having more and more different issues. Some printers, specially the ones with bigger beds, almost always have issues with the bed not being perfectly flat, which affects levelling and adhesion, even when using ABL (that I strongly recommend). Please, keep in mind that I'm talking about machines on the price range of 300 to 500 €. On the other hand, some materials really love glass, specially PETG, but also some blends of PLA. In that case, the problem is the opposite, your piece will literally solder to the bed, that is what happened to Clive here and that has happen to me too. In my own experience, adding a layer of "something" between the glass and the piece that you are printing helps a lot. Of course, I'm talking about my own accumulated experience, that is only about 4 years, I'm no expert here, just another amateur. Every individual experience may and most likely will vary, that is why I'm suggesting to keep trying different techniques until you find the one that works the best for you, that is what really matters at the end of the day.

[@sullivanrachael]

I have heard of people adding a waft of aerosol hairspray. Adds an easily breakable bond.

[@jbuchana]

When I have a print adhere too well to the build plate, I unclamp the build plate and print from the printer and put them in the freezer for about half an hour, so far that's always let me get the print off without damaging the glass.

[@assassinlexx1993]

You tell it to chill out 🥶👍

[@waldevv]

I find my glass bed sometimes barely sticks so I use glue stick and that makes things way too hard to pry off. Hard to find a balance really, I used to have the flexible magnetic sheet that came with the printer but I was an idiot and scraped it with the scraping tool too much and it wore it down. I guess painters tape is something to try out next

[@Fridelain]

​@@waldevv Why not just buy another magnetic sheet?

[@jbuchana]

@@waldevv Painter's tape worked well for me with PLA, but with ABS, the tape would melt into the print and was near-impossible to remove. The other problem was that I had to reapply it regularly. Kapton tape worked better with abs, but it had to be reapplied often as well, and it was very hard to get it on without wrinkles. A sheet of glass and a glue stick has been the best so far, but sometimes the prints stick too well, then I put the whole sheet in the freezer, and the print pops right off.

[@SuperAWaC]

this is also what you are told to do with troublesome prints that come off of FDM printers that cost 5-6 figures

[@zh84]

My first acquaintance with this was when I took apart a model railway transformer. I had read about transformers in the Great World Encyclopaedia of Science, and it was immediately obvious how the wiper selected a different number of coils in the output stage. It was linear rather than circular, and there was a long streak on the copper windings where the wiper had been wiping!

[@kathyinwonderlandl.a.8934]

My brother used to take apart his Christmas presents of electric train set transformers so much so that he was given this Dr Frankenstein Chemistry set / welding kit instead next years because they were less costly.

[@kreynolds1123]

I was wondering how they constructed that linear variac you seen. Was the magnetic circuit completed like a square toroid of transformer steel plates or was there a long airgap with coils wound on group of rectangular transformer steal plates, but not completing the magnetic circuit?

[@andygozzo72]

@@kreynolds1123 that 'linear' thing may have been a high power variable resistor with sliding contact, not a 'variac' or transformer of any kind

[@kreynolds1123]

@@andygozzo72 Thank you. It is a little difficult finding many examples of linear autotransformers.

[@bennylloyd-willner9667]

I had a Märklin one and it was my first "lab equipment". I was around 9 and my dad had to get a resettable porcelain fuse in the fuse box in our apartment. The fuse blew so often when I set fire to steel wool and tried arc welding with sewing needles. It sure had some punch 😁

[@LazerLord10]

That glass delaminating thing is very common with PETG, but that was PLA. I had it happen at a small scale once. Maybe it was cheap glass?

[@mtootm]

I've had pla delaminate on PEI sheet. Temp wasn't high enough to get good bonding. Increased first layer temp and problem solved.

[@Chris_the_Muso]

I use cheap soda glass, and delaminating is never a problem because of the high coefficient of expansion. Once it cools the prints quite literally pop themselves off. Borosilicate glass is much more expensive and is normally recommended because it has a very small coefficient of expansion. I got 3mm glass from a local place for about $2. I got two for a spare, but I've not broken any of them in over a year. So, I'm thinking that the better the glass (eg borosilicate) and if the surface is textured (ie sandblasted) so the adhesion is better, that's when you are more likely to have problems if the print sticks really well and there's no contraction of the surface to help get it off. If I leave the bed to cool down completely I literally hear when the print releases. I've also had prints complete while I'm out, and come back to find they've popped off so forcefully they are on the other side of the bed.

[@mtootm]

@@Chris_the_Muso you should set up some targets and see how many you can pop.

[@00011theman]

Yep, I had it happen on PLA as well. Tore off a good chunk of the glass bed.

[@LazerLord10]

@@Chris_the_Muso haha, that's amazing. I use cut mirrors (I found some really thick 1/4 inch ones), and a smearing of sugar water to keep things attached. It works unreasonably well.

[@KarstenJohansson]

When I heard back in the 80's that Eddie Van Halen used one on his amp, I got a variac and put it on an old Yorkville Sound amp. Since it wasn't really meant for guitar specifically, it didn't have a distortion circuit. But the variac, when turned down, allowed me to crank the volume "up to 11" which overdrove the tubes something spectacular, turning that amp into an insta-Marshall. Seriously dangerous project, but it worked amazingly well.

[@jkobain]

Dearest Clive! I never regret when you digress in your videos, and I think many of us like the content you make. ^^

[@DrakkarCalethiel]

Finally we got to see your Variac. Cute small one, my 5kva one is a pain to move!

[@michaelmoorrees3585]

I've got a 2KVA (120V 17A), and it turns easily.

[@DrakkarCalethiel]

@@michaelmoorrees3585 It turns super easy, but moving it from one place to another is a pain. That beast is heavy. It also loves to pop breakers sometimes if I don't hook it up to the soft-start. :D

[@aulus4163]

@@DrakkarCalethiel You could attach some caster feet to it. Did the same with my 10kVA 60kg monstrosity : D works well

[@DrakkarCalethiel]

@@aulus4163 60kg 10kVA, what a beast! My 5kVA one is just around 16kg with all the control stuff. When I'm going to pair it with a couple beefy transformers on the output, everything will get mounted on a rollable platform. Still need to find some tranies that don't cost an arm and a leg new...

[@peterjameson321]

Thank you BC. I'd been aware of this for a very long time but it's good to have it highlighted by you in your usual professional way. My 8 amp variac dissipates 30 watts on standby and the brush has an heatsink which gets quite warm on standby. When you think of it, there are only a few volts between each turn of the winding so the current flowing as the brush shorts them out is minimal on a miniature variac but it gets significant on a large variac with a correspondingly large brush hence the brush's heatsink on my 8 amp one.

[@mikefochtman7164]

Used a variac in our shop. Rated for 15A and the windings were much more 'course' as it was about 14 AWG wire. As you turned the knob, you could 'feel' the bumping as the brush moved from one turn to the next. They are a sort of auto-transformer (uses the same winding for primary and secondary) with 'taps' on every turn of winding. Nice video. And some VERY nice FLIR photos.

[@-yeme-]

maybe would be interesting to slide a thin piece of plastic under the brush and see how much of the loss is due to the turns being shorted by it, and what remains

[@WizardTim]

I've always wondered this since photonicinduction started posting videos of his variacs and now I know. Although as I've always been warned not to accidentally short a normal toroidal transformer with the mounting lug in the middle I imagined there to be a fair bit more heating. I also have some of those ZnSe (Zinc Selenide) lenses for my FLIR and they do slightly affect the readings, it can go both ways due to temperature of the lens but it's usually reporting lower than reality by 1 or 2 degrees. The pure germanium ones don't suffer from this problem but are 3x the price.

[@EnergeticWaves]

A neighbor gave me one of those from probably the 50's. Big heavy thing. the wire has cracks but it still works. I've always wondered how it works, so thanks for showing me.

[@lewzero]

And this is why I love these videos, every once in a while you have things I'm just about to buy, and you dispel the mystery I wonder about. This time it was nothing to do with the video, you just happened to explain why the contacts work (they're bare, I know I'm ignorant, it's ok, I learn quick). Thanks for another great one, I'll keep watching long as you're making em.

[@digitalhare4516]

Love how you can tell how new the video is by how scuffed looking the desk and BigCliveDotCom sticker is.

[@bolwem50]

I've been wondering about this for decades, thanks for the answer. I guess the windings have a good thermal bond to the transformer iron, so the shorted turns don't overheat. As 99% of the coils are unshorted, the overall losses are low.

[@ScottGrammer]

I've used variacs for over 40 years and always wondered about this. Thanks for answering some questions for me!

[@750kv8]

The brush seems to create a virtual single turn secondary winding with the turns it shorts out. Much like the shading rings do in a shaded pole motor. That can muchly explain why they heat up just in a small spot. It's a secondary short circuit. The shading rings in a shaded pole motor also get rather hot.

[@Stuartrusty]

All these years using these things and I never considered that this was happening. I have had many experiences with variacs, not all of them pleasant. Working within electrical and mechanical engineering for 35+ years, looking at the back of that variac in the video reminded me of the 3 phase 415V ganged variac that I used to use in the Electrical Test Department of a local firm I used to work for. It was in a test rig for high power DC motor testing with the outputs of the variacs going to a 3 phase diode bridge rectifier. It was all open live connections mounted on a piece of thick paxolin board. The only protection from certain electrocution was a flap of cardboard over the front of the connections secured with gaffer tape. The whole rig was bolted to the side of the test bench and the floor, on top was a large wheel very reminiscent of a water gate valve. I did manage to get zapped occasionally on the output side of the motor test rig, think it was due to leakage current from the big diode bridge. The other one I remember was "The Puppy" as it was referred to. This was a large wooden box on wheels that contained a variac and high voltage step up isolation transformer and DC bridge rectifier. On the front was a large red "jewel" indicator (remember those anyone?) lamp that had a 15W pygmy lamp in it . It's purpose was a high voltage leakage current detector, I think it was 1200 volts maximum output. There was a long high voltage test lead came out the front with a pointed bakelite probe holder and brass probe tip (not insulated or retractable!). Of course it had gaffer tape ears, tongue and mouth, the variac pointer was the nose, and the indicator light was one of the eyes! Dangerous kit by today's standards and probably would send an HSE representative into palpitations! Also good to see an MK 655 D8 red mains plug still in use. Does it have "Hospital Property" stamped on it? I can remember my father (he was a hospital electrician) having one or two of these in his works toolbox. Used to see these on everything in the local hospitals, from ECG machines to electric floor polishers, vacuum cleaners etc.

[@godfreypoon5148]

I think it's more to do with the resistance of the brush in general. If you assume the brush to have no resistance and do the calculations, a 10A variac would dissipate hundreds of watts in a single shorted turn. (Assuming 1.2 ohms end-to-end resistance, i.e. 5% impedance, and about 240 turns. This gives 5 milliohms per turn, put 1V across that and you get 200W!) If you expect around the same impedance again in the overall brush (from winding to the output connection), and pull a figure out of the air to say that the turn-shorting resistance would be 25% of that, then you burn 3.33W in the brush and 55mW in the shorted turn. Which is bupkis. It doesn't make any sense that the brush doesn't look hot in your thermal camera images, so I may be totally wrong. Edit: When I say transformer impedance, I mean the one that ruins your voltage regulation... not the one to do with transmission line characteristic impedances etc. Edit 2: Ah yes, you tested your wee lil bugger with HOPI. I expect that figure is dominated by iron loss, just like any transformer.

[@bluevanman2008]

Interesting. I've got an old RS Components (I think) metal cased variac from the early~mid 1980s at work. I now have the desire to find it and open it up for a look-see :)

[@cbcdesign001]

Ahh, what a cute little Variac. I have one I call Big Bertha, over 50 years old with a rated current of 45A and still going strong. She is a beast!

[@railgap]

It's easy for anyone to see this fact just by looking at the brush. The brush is wider than 2 turn surfaces. Note also that they turn / mill the windings flat for the brush; this is obviously necessary for brush movement, and while it may increase the wire's contact area, it drastically reduces the circular cross-section. So the classic name brand units at least (Superior, Staco, and GenRad) always seemed to use quite large wire for a given current rating. I've always assumed it was to compensate for that very problem. And the cheap chinese units do NOT do this, I've noticed they all run hotter. There's a reason used 50A name-brand units (very much in demand among Tesla coil builders, among a handful of other uses) are still breathtakingly expensive; there's a boat-load of copper in them!

[@railgap]

The question is: WHY do they short the windings when clearly it is preventable? It's a current density problem, localized heating at the brushes. There isn't sufficient contact surface even after milling the wires to get the rated current into and out of the windings by using only one turn. So designers accepted the heating of one turn because it prevents even higher localized heating at the brushes.

[@roberttaylor2328]

@Railgap Esoterica, while I am a Luddite, this is the explanation that I have been waiting for. I hope that educated corroborators chime in... :-)

[@l3p3]

Interesting that you want to replace the analog meter with a digital one. I used to have digital ones but have the odd feeling, that it is much faster and more intuitive to adjust the voltage/current to a desired spot. That will be due to our instincts, we have a hardwired closed-loop in our brains for moving things to where we want it. There is no such instinctive support for 3 digits of numbers.

[@lwilton]

I added digital volt and amp meters to a variac that never had a readout. Except for the last few volts I never attempt to set a voltage from the meters, they are too slow. The variac has a 0-100 scale, and I've marked the actual output voltage at a couple of interesting points like 115V and 125V, so I can go directly there, or estimate some other voltage as a fraction of those voltages. Once I'm close I can be precise with the meter, if it matters.

[@Agent24Electronics]

It looks like a cheap eBay meter with a slow update rate too.

[@lwilton]

@@Agent24Electronics It might be relatively fast. I used eBay meters when I did my variac probably 10 years ago, and they update once or twice a second. Not wonderful, but not horrible either.

[@chrispomphrett4283]

I just like analogue meters. Another advantage is if the load is varying, a meter shows it better than wandering digits...

[@userPrehistoricman]

@@lwilton twice a second is slow. 3Hz is typical

[@peterjameson321]

Yet another great myth-busting video Clive. Thank you! There are many assumed theories about the graphite brushes on variacs, but it's obvious that they simply short out a few turns of the coil which after all amounts to only a few volts precisely as you say. I have an 8 amp variac and it dissipates 13 watts quiescent AND it has a finned heatsink on the graphite brush. That's where the quiescent power goes.

[@davida1hiwaaynet]

Very neat video! I've wondered about that same thing, seeing how a variac doesn't have any "dead" areas between windings as the control is swept across the entire scale. Makes me think of a power transformer with an "on load tap changer" which momentarily bridges out a segment of the winding during tap changes. Those have a more complex system than a variac, but the idea is similar. There is no off state between taps. A low-value resistor is used in series with each winding for a very brief moment during tap changes. I've got two Chinese variacs which have developed some dead spots and audible arcing. Should tear them down and inspect to see what went wrong. Both of them are less than 2 years old and followed the same path to failure. New out of box, there were many dead spots, cleared up by repeatedly sweeping the knob with power off. Then after several hours of use, dead spots reappeared at the more commonly used ranges such as 100V and the very low range used to check for signs of life in very old equipment before ramping up. Both of these now take very little effort to rotate the knob and I expect that winding varnish was on the end turns where the brush rides, causing the initial problem; and that now the brush has become stuck due to more gumming up from varnish.

[@robkennedy3000]

I can never see or hear about a Variac without thinking of Andy from photonicinduction cranking them silly in his loft. I hope the guys doing well and can make an appearance again. I miss his "I ain't havin' it attitude"

[@michaellynch1018]

Dave from EEVblog posted a community post here on youtube a while back saying he was in good health and working on some videos to be posted soon

[@robkennedy3000]

@@michaellynch1018 yeah,. I tried to follow it best I could a while back and saw his wife was in the UK finally. Last I saw was about wanting to make videos "soon". Guess he needed more time than he thought

[@Akbar_Friendly_in_Cherno]

I read about putting your build plate in the freezer to get stubborn, stuck prints unstuck for years before I tried it. Works every time. You even get an audible "pop" when it comes off. Build surfaces are consumables IMO.

[@jaydauro7539]

We used to carry a 20A Variac when I toured with a sound system (1980's). This would allow us to adjust the voltage at venues that were out of spec (for the console and other sensitive gear. Nowadays with better power supplies this is not as much of an issue.) For the connections to the windings there were actually 4 connections, 0%, about 10%, about 90% and 100%. We would connect the hot at 90% and the neutral at 0%. This would allow us to move the output wiper past the hot connection at 90% and boost the output voltage above the input. Very handy when we were at a venue with low voltage. Being from the US, we would also bring this variac along when touring in 240 volt lands, etc., allowing easy adjustment. (We would find some places where 240 was actually 260+, being able to tweak as needed was handy.

[@sacrificialrubber779]

Used those for many years on stage to be able to vary the output on singular lights! Quite handy bit of kit!👌🏻

[@jercos]

Much kinder to transmission lines than modern triac dimmers too... phase-fired control inherently has terrible power factor.

[@kreynolds1123]

The volt per winding is very very small so there's a very very small eddy current in two adjacent shorted windings. Another thing to consider: A carbon brush has more resistance than copper has. Two wires being shorted through a brush results in less "shorted" current than if the brush were made of copper. But, as pointed out, the brush may sometimes span more than two windings.

[@kreynolds1123]

@belly tripper Kindly, while "Were" is used in the second person singular and plural and first and third person plural, it is also used in the subjunctive mood to indicate unreal or hypothetical statements. www.grammarly.com/blog/was-vs-were/

[@kreynolds1123]

@belly tripper Past tense "Jack was early to work", vs subjunctive unreal or hypothetical "if the goose were to lay a golden egg."

[@kreynolds1123]

@belly tripper ok, roger that.👌 I'd like to propose that you brush up some. Chech out the section on 'The Subjunctive Mood with "Be" and "Were" ' www.merriam-webster.com/words-at-play/getting-in-the-subjunctive-mood. If you are to be a grammar nazi, you might want to be an informed grammar nazi. Please have a nice day.

[@kreynolds1123]

@belly tripper With all due respect, given your two statements ' your grammar is horrible, you mean ""if the brush **was** copper"" ', please understand that when one wants to criticize another person's grammar, one should know that an informed grammar nazi would not start either sentance with a lower case letter. Nor would they end either statement with either a comma or nothing. Nor would a grammar nazi use "" where a single " is required. Lighten up on people please. Putting them down on something does not lift you up. And when you are the one that falls short on the same measuring stick you used to measure another person, the heaping dish of crow pie doesn't taste very good. In the case that the last idiom is not common wherever you are, I'll share this from wikipedia "eating crow is a colloquial idiom, used in some English-speaking countries, that means humiliation by admitting having been proven wrong after taking a strong position.The crow is a carrion-eater that is presumably repulsive to eat in the same way that being proven wrong might be emotionally hard to swallow."

[@antanasv2642]

Clive, you will probably find that your digital meter does not measure below ~70V. You can modify it by adding a third wire, disconnect the supply from measured voltage. In my case it was removing one 0ohm resistor

[@bigclivedotcom]

I'll check that out to see where it taps its monitoring supply from.

[@kingsman428]

Oh that's just brilliant, there was me thinking about having a lazy week but instead, I've now got to take apart my Variac and examine it with my Seek 😁 camera.

[@kathyinwonderlandl.a.8934]

You’re a very wise and naughty man..I truly wish I understood what exactly you’re speaking of but I’m hoping to absorb things subconsciously in the end..

[@hls6925]

I've had mine over 40 years. Still use it! It's a DURATRAK type V5HMTF, rated at 240v 50Hz in, 0-270v 2A out

[@keyen3]

Always informative. Please consider making a video on the crimp connectors you used. Tips and tricks for making them? What kind of tools do you use? etc... Thanks

[@BjornV78]

That Digital panelmeter normally begins to work and show voltage from 60a65V, but you can hack it to show voltage from 0 to 240V. Lift one side of the brown capacitor and use that for measuring the voltage, use a separate 12V DC and connect this parallel to the black electrolytic capacitor. It's the same hack that is used on the Peacefair panelmeters.

[@video99couk]

It's worth reminding viewers that this does NOT provide insulation from the mains, it is not an isolation transformer, so needs to be used with care.

[@bigclivedotcom]

I've added that in the description.

[@thehobe2111]

Thank you for doing this expose'. I have always wondered about this exact phenomena. I guess the carbon brush is the "trick". A little lossy on the contact prevents high currents to flow in the shorted winding.

[@PaulB-xb5zx]

@bigclivedotcom I don't understand some of the stuff you talk about or show, but I still enjoy watching your videos. Thank you for not having music, I find it quite distracting when trying to watch/learn from some channels' videos. I know you mentions things on many videos during the process, but I'd love to see a general tips video for faults, issues, traps, etc, when making or fixing things. I think it would help the less experienced and knowledgeable people like me.

[@vectravi2008]

I used one to vary the speed of a fixed speed wall fan without having a buzzing noise that you get if you use a normal resistor in the circuit. Work fine and never had a problem.

[@WELLINGTON20]

Your almost at 1 million! But don’t change. Keep going around the same way your going on.

[@johnmurrell3175]

The brushes on large DC motors normally bridge several commutator segments as well, I believe that this improves the commutation and reduces the sparking by ramping the current up & down more slowly than if a brush just contacted a single segment. Brush manufacture is a black art ( pun intended) with a secret mixture of anthracite, other carbon clay & copper to get the correct resistance. One UK supplier had big problems when the Welsh anthrasite mine(s) closed, the replacement seemed to had sufficiently different characteristics to result in an increase in brush & commutator wear.

[@DCFusor]

I recently had a massive 15 amp / 120v one burn out just sitting there. I'd been using it to control a Harbor Freight spot welder, for which it was very useful. I forgot and left it energized for half an hour (spot welder inactive) and then there was that smell in the room - and about 3 turns at the position of the brush were black... The variac was collected by my Dad in the 50's, no idea of the actual vintage. Things like those stay in the family for generations, as many know, they don't grow on trees and the big ones cost real money. Luckily, we've been avid collectors...

[@Roy_Tellason]

I have one that I put into a box with several switches: A dual-pole on-off switch that interrupts both sides of the line, just in case. A switch that selects whether the high side goes to the end of the variac or to a tap, allowing me to "boost" the voltage to somewhere around 130V, for some tests. And a DPDT center-off switch that in one position puts a light bolb in series with the output, in the other position connects directly, and in the "off" position allows for the connection of a meter to some binding posts so I can measure the current. There are also neon indicators to tell me when the power is on and when I've blown the fuse. In a relatively recent acquisition of a pile of "stuff" I got another one. The rather distinctive knob that came with it is one that I've seen installed in premise wiring, as light dimmers. I had no idea that they were using variacs for that purpose.

[@bigclivedotcom]

Variacs were used as orchestra pit music stand lighting dimmers to avoid phase angle control noise near microphone cables.

[@jamesharmer9293]

My stepfather had several of these for the lights in his animation studio. He found that the lights lasted much longer if you wound them up gently and didn't run them at full power, which was important as the special bulbs were expensive. Of course he'd got them second hand and much used, so big blue sparks would shoot out the sides as you wound them up. Also they had to make sure that the lamps were always running at the same voltage throughout the shoot, since mains voltage varies from day to day, which would show up as a noticeable flicker when sped up by the animation.

[@misterhat5823]

I've always thought that was likely. But, so many say it's not a shorted turn. Glad someone actually proved it either way.

[@absalomdraconis]

I can think of ways to avoid shorting turns, but it'll always come down to either a different transformer style, major cost/size increases, or opens replacing shorts.

[@lwilton]

It depends on your definition of "shorted". The wiper is carbon, and carbon makes a resistor. So it isn't a dead short, it is a very low resistance resistor connecting a few turns, and the turns have a relatively low voltage difference between them, so it is a fairly small power loss.

[@misterhat5823]

@@lwilton You're really just splitting hairs. I could claim everything has resistance so there's no such thing as a short, but we both know that's absurd.

[@therealjammit]

I have a few variacs of different power capabilities. On the small one (1 amp) the contact brush shows no lengthwise/widthwise resistance differences, but the ones I have that are over 1 amp have a definite difference in resistance. Maybe the small ones are too small to notice the issue? I'd also like to mention that my taps are all terminated to one wire. Those wires are brought out, given a few twists, and soldered to.

My ZnSe 20mm (aliexpress, ~$5) lens attenuates the reading with about 2°C on the Seek Compact. Made my Seek *very* useful with SMD work.

[@robstorms]

I believe it is a actually an autoformer . A transformer would have at least two independent magnetically coupled windings and this has only one winding

[@robstorms]

@@lwilton "I know you are but what am I ?"

[@Basement-Science]

@@robstorms "Transformer" just doesnt tell you anything about how many windings something has. Autotransformer, however, does.

[@robstorms]

@@Basement-Science Just funnin with you ! Thanks for the clarification !

[@drussell_]

Yes, they short out the windings which is why it is nice when they are overbuilt like the old ones were, so that the thing can actually handle *real* usage... The new-style Chinese units have a very narrow pointy-cone brush to help reduce the strain on the undersized windings but will never hold up over extended usage, especially as the brush wears down and more windings come into play on the "shorted turn" section. There's a thread on the EEVblog about this, I'll see if I can find you a link...

[@wackowacko8931]

The most likely reason for the shorts is breakdown of lacquer on the windings of the Variac. Over time, and especially on tight windings the lacquer starts to crack, which causes shorts. Heat accelerates this. Buffing the top off the lacquer for a contact point doesn't help, but it is the most direct way of providing the varying voltage. Cracking of the lacquer over time caused the OEM car manufacturers to stop using ammeters in their cars and they switched to voltmeters (because the resistance was very big or would fail open circuit). The cracks in the turns of the D"Arsonval movement caused I(squared)R problems where the resistance got very small which, would cause the power consumption to go way up. This would melt down vehicle wiring harnesses (yes, even with bypass style ammeters). Tons of vehicles from the 40's up into the 70s were sent to the scrapyard because the wiring harness would melt down when the ammeters shorted out.

[@JDfromWitness]

I have a couple of very old 6kw autotransformers from an old theatrical dimmer board. The ends do terminate on one wire, but the brush does bridge more than one winding. Not sure what the brushes are made of (lost now) but I think there was some resistance across the brush itself being that it was carbon or some mixture of materials. Could be this micro-resistance is enough to not dead-short a winding. Otherwise, I would think it would heat up like an instant-on soldering gun.

[@BjornV78]

The carbon brush on a variac touch / shorts always 2 or 3 windings at the time, this "shorting" forms a small voltage divider, if the brush was only touching 1 winding, there was between each winding a small disconnect and you get arcing on each movement, also the output would be very "jumpy" and not smooth when rotating the variac. Due the contact of 3 windings at the time, there are no disconnects between each movement, because 1 winding in the middle of the 3 windings still keeps contact, the outer 2 windings are connecting and disconnecting during movement of the axle.

[@bradprimeaux8443]

What material did you use?. PETG is known to adhere too well and pull up chunks of the build plate. PLA and ABS don't have that issue. I use cheap hair spray to help ABS and PLA adhere and I also use it for PETG as a release agent so it can't destroy the glass.

[@bigclivedotcom]

I use PLA.

[@bradprimeaux8443]

@@bigclivedotcom I'm impressed it stuck that well. I can't get PLA to stick to bare glass ever. Try the aerosol hairspray trick. I use the cheap aquanet and it works well.

[@MB-st7be]

Unsmoothed rectified is useful for measuring very large inductances in the presence of DC... maybe you wanted to test chokes?

[@dantronics1682]

the one I have has the live connecter a small distance from the end, in other words the end is floating so when you wind it to max you get up to 260Vac with 220V going in. When you was playing around with those 047mf caps 2 weeks ago to vary the voltage on the led bulbs I was going to ask if you couldnt do it with a variac, I couldnt see you not having one of them

[@eDoc2020]

Mine has the same function (except it's 140 from 120). The live connection to enable the boost would probably connect to the coil right near Clive's mystery shorted winding. I'm guessing that's why.

[@seanet1310]

Clive, you need a three phase 20A, oil cooled Variac to play with.

[@bigclivedotcom]

First I need 3 phase.

[@misterhat5823]

We had one of those at the last two places I worked.

[@misterhat5823]

@@bigclivedotcom Can you even get three phase? 3 phase residential service is impossible in the US even if there's 3 phase on the pole outside. Is it the same on Isle of Man?

[@ellenorbjornsdottir1166]

@@bigclivedotcom There should be a way to make 3 phase.

[@NiHaoMike64]

@@ellenorbjornsdottir1166 A cheap way to get 3 phase with a lot of power is to hack a Prius inverter.

[@jercos]

The bridge rectifier is needed for the voltmeter, yes? Using a larger rectifier and adding terminal posts was presumably a nearly-free upgrade, while a capacitor would slow the meter response.

[@bigclivedotcom]

That's possibly part of the reason.

[@linswad]

I’ve never noticed before that the Hopi has Frequency misspelt as ‘Frequence’.

[@5roundsrapid263]

I think that’s French.

[@linswad]

@@5roundsrapid263 The other labels are all in English, even “Annual Power Consumption”, so I don’t see why that one would be in French.

[@ianhosier4042]

I have one too but it doesnt have a case, just have to be careful not to touch the brown coloured wires

[@jercos]

Deterrence against curious burglars!

[@poptartmcjelly7054]

Speaking of bigger variacs i do have a 20kW variac and it is quite different from what you have here. For starters it's not a toroidal transformer but a rectangular one, it's very long compared to its width. Instead of carbon brushes it has multiple carbon rollers and the way is works is a little different from how your variac works, because the rollers make contact on multiple places on the transformer. I can't remember exactly how it's constructed since i haven't looked at it for a while but i do remember that it does buzz quite loudly, which i guess is part and parcel of a big honkin' transformer.

[@muzikman2008]

I used to use variacs a lot where I used to work, we tested motors and used them for fault finding when mains equipment was faulty and we could ramp up the voltage gradually. Useful things. I figured out how they worked years ago, and always thought they shorted windings out because it was just designed that way, or 2 or 3 windings would deliver a better surface area for the carbon brush with less resistance/heating in that particular spot. Prob just a rough and ready design though. Who knows lol. We read too much into design sometimes.

[@ArtifexExMachina]

If you spray a thin layer of hairspray on the printer bed it will help with giving prints the right amount of adhesion for staying on put during printing but coming of easily once they are cooled. May require the bed to be heated, not sure. Using hairspray will also give prints a nice matte finish that better matches the finish of the rest of the print.

[@oldgrizzlygamer]

I really hope you show the installation of the new meter.

[@Debraj1978]

I also used to wonder on this point and thanks for confirming it. Other point that comes to my mind is, variac is sometime rated by VA. For example, if the variac is rated for 500VA. Q1: If we derive a low voltage from the variac, say 12V, will the variac now be able to supply 500/12 = 41A? Q2: The above may be true, but the copper windings may not be rated for 41A. Under such case, will the max current come into effect (to prevent burning of copper winding)?

[@bigclivedotcom]

It's usually a fixed maximum current for the full range. Whether just 1V or 100V.

[@dnwheeler]

Shorts caused by the wiper shouldn't matter at all - it just results in a few "dead" loops at the wiper position. Likewise, any other short would just kill a loop or two, and decrease the overall resistance of the entire coil (and affect the linearity of the output). I'm not sure why there would be hot spots unless the windings are uneven, so that a few windings are closer together (so the combined heat of the windings would influence each other). None of the shorts would put more current through any single section of wire.

[@FerralVideo]

I have one of these capable of 4A. has pretty beefy wire to it as a result. I received it loose, and installed it into an enclosure I found in a stream. Doesn't have a meter, but its faceplate voltage indications are accurate enough for government work. I can use my own Hopi on the output if I need higher accuracy. I tapped it to accept American 115v and output up to 280v at full scale, but I have to be careful with the current, as I could hit 4A on the input pretty easily when stepping it up that far. The idea of unsmoothed DC output is interesting! I might have to build a FULL BRIDGE RECTIFIER for mine. Probably as a module I can attach to the output cable...

[@Basement-Science]

Well since noone seems to mention it...: The reason it's not a big deal is simple. P=U^2/R, where U is the voltage of a single turn and R is its resistance. Both U and R are proportional to the number of turns you short out. But because it's U^2, the power dissipation becomes a lot bigger if you short more turns

[@misterhat5823]

Clive kinda did mention it. He showed the power measurement at no load. Just didn't mention why.

[@zambonidriver42]

I patiently waited for “Let’s take it to bits.” 😉

[@zambonidriver42]

I think you need shirts. That bearded outliney-sunglassey icon.... w your dotcom, and a couple choices of sayings. Don’t print them yourself, stock, and ship. Do some kind of third party host thing. Don’t want you tied to regular scrambles to the post.

[@PeterMilanovski]

What a great way to get high voltage DC! I was thinking about putting a bridge rectifier in a separate enclosure with a power cord to plug into the Variac. I was looking for a good way to test capacitors for DC leakage all the way up to their rated voltage... Since my Variac goes to 11 (300V AC) I'm going to be seeing 600V after the rectifier and I'm not sure which capacitor type would be great for that task... With 600V DC, you could also use it in vacuum tube amplifier testing and design... I have a 15A model so I have plenty of flexibility, even a voltage doubler could potentially give me adjustable 1200V DC.. It's not a bad idea when you consider the possibilities with one of these things, I could literally get rid of my 30V DC lab supply!!!

[@bigclivedotcom]

Keep in mind that the variac is not isolated from the mains supply.

[@PeterMilanovski]

@@bigclivedotcom oh yeah! It's very easy to forget that! I already have the best LCR Bridge available from eBay, I got it after watching a video by XRayTonyB did on how to use it! I had been watching it on eBay for years but never got it because it looked very complicated to use, the information it provides is like nothing else on another device anywhere near it's $100Au price tag, he has the older silver aluminium case version which needs external power, I got the newer black aluminium case version with the built-in rechargeable battery... What this meter doesn't tell you, you don't need to know! It's the bee's knees for capacitors, inductors and resistors... I just wanted to test a few different capacitors at their rated voltage in series with a uAmp meter to see how accurate the LCR Bridge is... Actually, this meter would be a worthy addition to your bench! I feel that there's still plenty of people who don't seem to understand capacitor failure modes, I recently watched a video from 12V vids and he was, so called, restoring a Denon amplifier. He's been in the repair business for a very long time and I saw him check the power supply filter capacitors with his Fluke multimeter for capacity, the capacitors came up with more capacity than they were actually rated for! He said that they are good! But I'm like WTF? When a capacitor leaks DC, it takes longer to reach full capacity and therefore giving the impression that it's an overachiever! He should know better! It was a painful video to watch! In the end, he called it done and ready to go back to the customer even though it's giving a triangle wave at 5khz and above with a sine wave input! He's not alone though... Some people just rely on ESR readings... It's not enough to show how bad it really is.. the LCR Bridge I'm talking about, it's something else, it really should be on everyone's bench... Without it, you really are just poking around in the dark with a metal stake... Oh and btw, I loved the group chat video with kangaroo Dave and the others... Hanging around and talking shite... That's entertainment...

[@frankowalker4662]

I wish I could afford a mains variac, but the prices are insane. I got a low voltage Irwin AC/DC power supply that uses a variac, 0-16V, £5 from a local market about 7 years ago. It's perfect.

[@chriswilson1853]

My dad got given a job lot of large variacs a few years ago. I think he gave most of them away to friends, one friend who was a professional photographer used one to control the brightness of the floodlamps in his studio.

[@chriswilson1853]

Also I forgot to add that whenever I plugged one of them in, there was a 50/50 chance that it would blow the main breaker. Not sure why.

[@frankowalker4662]

@@chriswilson1853 How odd. A faulty batch?

[@chriswilson1853]

@@frankowalker4662 I always assumed it was something to do with inrush current. They were pretty meaty units, more than 3kW I think.

[@frankowalker4662]

@@chriswilson1853 Yeah, at 3kw, that makes sense.

[@tomp016]

These are fantastic bits of kit, i like your idea of having the dc output aswell thats something i havent done to mine yet currently saving up for lots on contactors to modify mine further. I definately would reccomend earthing if possible as the main spindle on these can have the tendancy to become live in some instances not common but it can happen, i would also reccomend maybe having a little pannel fuse built in fuse after the output of these to protect the variac just encase of short or overload on the load side as in some instances you can quite easiy flash the windings in no time at all leaving you wondering what the hel just happened l i was a bit naughty once with some destructive testing and managed to get mine a tad hotter than desireable in no time at all at the time i didnt have a bigger variac which was surely called for however i am looking into putting in some kind of thermal protection on on the windings however same variac thanks to someone whoever owned my variac before me did flash the windings and deemed it dead i was lucky enough to get it for peanuts and repair it on my smaller 4amp variac onthe principle you have explained perfectly of that some of the windings do indeed short out when the wiper arm passes over them... Myself i use 20mm glass fuses in an unscrewable pannel fuse cutout i know i should really use ideally ceramic fuses because of the surge currents fuse design of glass ect ect but i use glass as theyre cheap as chips on ebay in packs of 100, on the larger one i use an appropriate sized mcb per phased variac. I would potentially reccomend using an rcd before the variac preferably rcd plug on smaller more portable ones they work quite well even if you decide to be naughty and feed some equipment too many volts deliberately for some destruction currently looking into going as far as putting in a dedicated socket for the larger variac via rcbo ... Sometimes throughout the year thanks to industrial upgrades ect you can buy three phase variacs for a fraction of the cost of a new single phase variac for absolute peanuts and perform a reconfiguration to turn a three phase one into single phase to gain an exponentially greater output cureent. I was lucky enough to get a 3 phase 20amp for 'special uses' i changed the configuration on mine so that essentially as i power each of the three up i can have maximum amperages up to 20 / 40 / 60amp outputs 0-275v as mine has a little bit of over run after the connections i think photonic induction made a superb video going into the ins and outs of modifications and menantioning the over run on these variacs some years ago... keep making the fab videos sorry about the waffle waffle :)

[@arbutuswatcher]

When I was in college I had the fortune of attending one of their liquidation sales. Typically, they had one of these every so often, when they wanted to replace lab equipment with new, & had to spend allocated dollars lest the get less the following year (politics). At any rate, I managed to procure a Variac & a Multi-tape Isolation Transformer. Combining these two made for some interesting experiments!

[@anthonyshiels9273]

In the 1970's I built a variable power supply using a variac rated at 10 Amps. It was used to run an electromagnet. Because I was a laboratory technician I was able to get items from RS Components then. Since I retired they are no longer in a position to fulfill any orders from me.

[@bigclivedotcom]

To buy from RS components you have to present yourself as a company. It's purely a box ticking exercise as they are simply not allowed to sell directly to the public at their trade counters for red tape reasons. If you tell them you are a company they will happily sell to you without needing proof, purely because you have ticked the box for them. You could call yourself Shieltronics.

[@graemezimmer604]

Thanks Clive, very interesting. I've always wondered about shorted turns on a variac.

[@dougaltolan3017]

The heating from the suspected shorted winding is more even and much hotter at the bottom. To me that suggests that the heating from the wiper contact is much more to do with contact resistance and the heat traveling along the wire.

[@mikehart2555]

I always wondered about this, too! If the brush was bigger then the distance between the windings, it would short turns. If the brush was narrower than the distance between the windings, there would be momentary drops to zero volts as the variac was adjusted. What I'm not seeing is how there is so little current being drawn because of the shorted turn. I would have thought that it would have been much higher than this. Remember all the fuss about when you mount a toroid you need to be careful about the mounting hardware not forming a shorted turn? It would be interesting to see if that's not so important after all.

[@bigclivedotcom]

The hardware creating a shorted turn on a toroid is a real thing.

[@Dutch3DMaster]

Not sure what the material is you printed in, but if it's something like PETG or ABS that is most definitely something that can happen with glass build plates, and I've read it can help to cool down the plate either naturally or by putting it in the freezer to have the model crimp lose. Isopropylalcohol is a very good cleaning agent I've come to learn for cleaning the buildplate, unfortunately it is impossible to obtain in The Netherlands at the moment, so I'll have to do with ketanotus alcohol for the moment (works, but not incredibly well). I've had models that were being stubborn from being released of the buildplate but upon the buildplate cooling down I heard a very loud crack and the model was moveable.

[@jercos]

"Denatured" alcohol is the term I see in the US, is ketanotus alcohol also typically >95% ethyl alcohol? Straight ketones (e.g., MEK/butanone) might work better if your ethyl alcohol source is 70%/"rubbing" alcohol.

[@Dutch3DMaster]

@@jercos I saw Ketanotus being sold with a 70 and 95% alcohol rating (I am unsure what the difference is, except for the fact that it's nearly impossible to obtain right now because those are also sold out...) Isopropyl is something I was unable to find at all, I couldn't find any store here that sells it, which might just be down to the fact they took it out of stock untill they restock the product to avoid people asking for it.

[@lImbus924]

I must say I'm not sure I understand the workings. I always thought these were massive variable resistors, therefore I found them a bit mysterious and never wanted to own one. I hear Clive clearly saying that this is not, so that part is explained now. But how is this a transformer? It only has one winding as far as I can see. Where is the second winding? Or is there an internal first winding and we are seeing an adjustable second winding? Or is it by definition a transformer because the coil core is not ceramic but ferrite and some effect makes it that it mechanically all works like a variable resistor with wiper etc. but because of the ferrite core it's working is different than a big variable resistor ?

[@lImbus924]

also, if it was a variable resistor, shorting the windings would not be an issues. perhaps that is why I never imagined they were something else than a big potentiometer?

[@bigclivedotcom]

It works inductively instead of with resistance. It acts as an autotransformer with taps on a single winding.

[@benbaselet2026]

Useful little thing is if you add turns after the input tap you can actually get a higher voltage out than what comes in :-) One must always remember that since there is only a single winding you are always directly connected to the mains input so a variac will not isolate you from the mains like a regular 2-winding transformer would so you may still need an isolation trafo to be safer when working with these.

[@tthomassims9005]

@@bigclivedotcom auto transformers do not provide isolation so relative to mains voltage they will supply mains voltage but relative to its own output the voltage is dropped I would love to have one as using a current reduction via an old tungston bulb just will not cut it sometimes you really do need voltage control I have a DC regulated power supply that can do this and restrict current by reducing voltage when the current reaches a set target

[@GigsTaggart]

Another way to think of it is just like a voltage divider, except with reactance (i.e. inductance) instead of (well, technically in addition to) resistance.

[@PaulOostindie]

I have worked on many variacs and even a 9Megawatt model used for adjusting a molten metal arc furnace (the secondary current is 30,000Amps) and also on 20 and 50 Megawatt transformers with OLTC Online Tap Changers. They all use very special carbon brushes with a fixed resistivity and you don't dare change the type of carbon or graphite on the points. The carbon has just enough resistance not to appreciably short between the windings but also not have too much resistance to cause a lot of heating from the main current. You see the same thing on DC motors that use graphite as brushes on the commutator. While an AC motor with a wound rotor that has solid slip rings use a more conductive graphite because they are not shorting windings. With motors the other side benefit is that graphite is slippery and saves wear on the commutator but it isn't the main reason it is used. So how do they change the conductivity of graphite? The grain size, the process that they make the graphite with, the compression density, impurities like clay and the firing temperature (about 1100 deg C) in hydrogen gas but the main method is to add metallic powder usually copper to reduce the resistance. A bit of copper drops the resistance by 80%. You can try making a replacement brush out of solid copper and you will find it doesn't last very long from all the arcing. I know I tried it.

[@rileyfenley522]

When printing on glass you should try putting down a thin layer of glue stick (PVA). It works very well and just washes off with a little warm water. The glue stick creates a interface layer that increases the hold but also makes it easier to remove when cooler and shouldn’t bind so well as to crack glass or mirror. You can print multiple prints over the same place you put the glue stick before removing and reapplying. A stick comes with every Prusa printer. You can also go with a removable flex sheet. You just flex the sheet and the print just virtually falls off. Depending on the surface type on the flex plate if you are having adhesion problems you can use the glue stick as well.

[@tthomassims9005]

I wounder if a sillicone sheet ontop of a lexan sheet would work as a flex sheet or sillicone ontop of glass without any adhesive to bond the 2 just lift the silicone sheet and plop the part falls off

[@stevetobias4890]

I always wondered if the insulation was removed from the upper ring of windings. Makes you wonder if it could arc, maybe that's why it's limited to 750mA. I guess if only a small part of the top of the wire was exposed it wouldn't come near to the neighbouring winding, but then as you said it is very thin wire, over time could it wear down the insulation?

[@bogywankenobi3959]

Clive: It would seem to me that the few turns that were shorted by the carbon brush would have no current through them and therefore should not get hot at all. It stands to untested reason that the voltage across them is also zero implying that there should be no heat dissipated in them, either. So any heat seen in those windings would need to be explained another way. I would therefore like to suggest that the heat on those windings is possibly due to resistance in the interface between the carbon and the windings. Observing it with the FLIR when it is first turned might just show the heat starting to manifest at the carbon/copper junction and then moving down the windings. I would like to see my theory put to the test. Also would it be possible to replace the carbon with a brass wiper? Or possibly a copper one? I can't help but believe that the carbon would have more resistance than either brass or copper. Or even aluminum. I hope I am not out of like by suggesting this. I love a good exercise in the application of first or fundamental principles as it relates to physics. Bogy.

[@bigclivedotcom]

A shorted turn still has current induced in it by the core it is wrapped round.

[@bogywankenobi3959]

@@bigclivedotcom Completely escaped my thinking. Thank you.

[@MaNNeRz91]

Use a glue stick on the build plate for the printer. It allows the print to stick firmly and releases easier when done. Or a flexible plate would work too

[@mtootm]

Hey Clive what build surface does your printer use? PEI sheets work great. ABS juice works great as an interface layer for all types of plastic. If you use ABS juice and it's causing the print to adhere extremely well, try PVA glue stick. My personal favorite adhesive is BedWeld. Works great on my PEI sheet with ABS, PLA, ASA, PETG, TPE. And of course waiting till it cools and contracts helps a ton.

[@bigclivedotcom]

Heated glass with a pattern of dots on it.

[@mtootm]

@@bigclivedotcom I see. My friend has a Creality Ender with a similar glass bed. They ask you not to use any adhesive on the textured side of the glass. You can flip it over and try different adhesives until you find the stuff that works.

[@jameslmorehead]

I have a couple variacs. One is 240 input with 5-240 output. Another is 115V input with 0-150V output. The third is 110v input and 6. 14. 24. and 0-40v output. Variacs are just basically an auto transformer with one tap that can move. Great bit of kit though.

[@spaceted3977]

I've just got my 2 Kw Digital Variac from Amazon !!! It's a super device, but it's nearly as heavy as my Royal Enfield Motorbike !!!! I shall be using it with a Residual Circuit Device, for fixing power tools, motors and a bit of electronics !!!! Mine arrived in perfect condition. It only has 13 amp Ac output 0 to 240v which is just what I wanted !!!! But if there's a short it should trip the RCD unlike a Microwave Transformer which just kills you instead !!!! If the Brush sparks like a Bumper Car Pole in a Fairground, you might need to fit a new Carbon Brush !!!!!!

[@bigclivedotcom]

Does it really go up to 2000V? If so, be very cautious, as depending on the circuitry it may not trip the RCD.