137 - My Light Bulb Moment: Building an inexpensive mains current limiter

[@_cheesestraws_]

I really like the "two pattress boxes back to back" approach, that's really neat. I will use that technique myself.

[@CliveChamberlain946]

At 7:22 yes, most certainly a habitually hardened kur-AM-mick socket, but I like the sir-AM-mick type 😉😅

[@bobbyohanesian4228]

Excellent need to make one thank you

[@Peter_S_]

This item was a part of the workbench of every 1950s and 1960s television and amplifier repairman. It should still be a standard tool for anybody repairing old electronics or playing with old machines. This device will stop an old RIFA capacitor from causing damage when testing an old machine. When we switched to LED lighting, I bought a few 200W and 300W bulbs for just this use.

[@cambridgemart2075]

We were using it in the 1980s and I still have one on my shelf now.

[@abizarlakdawalla4519]

Another option is to add an AC socket in place of the lamp socket into which a high wattage resistive load can be plugged, such as a clothes iron or a resistance heater. That allows you to test higher wattage DUT.

[@ovalwingnut]

The Best "Dim Bulb" explanation I've ever heard. Just saying. Good job. 500 points awarded 💡 p.s. I love the font you selected for your label. Quite modern :O) Cheers from the U.S.A.

[@t1d100]

Agreed

[@mikebarton3218]

Sadly, the corner radii are not standard between different manufacturers so unless you stick to one manufacturer you can get the mismatch that you had with your socket. It wouldn’t bother most people, indeed many would not even notice, but details like that drive me nuts. A very absorbing video nonetheless and thank you for sharing it.

[@kwpctek9190]

Medium base bulbs may be disappearing, this project can be more flexible with 4 x 25w C7 candelabra bulbs (or 40w) with 3 of them switchable, giving it a nice range of 25 to 100 watts (or 40 to 160). This is also called a E12 base socket and 4 would be easier to mount and take about the same space.

[@ivanr5315]

Thanks! Very informative. In the UK, for most DUTs what do you suggest as bulb rating? 60/100W should be enough?

[@TheHWcave]

It really depends what kind of DUTs land on your bench, or more accurately their vintage. Things with valves will probably need more. So far I have used 100W but all I am repairing is "solid state" (transistors or ICs). Possible exceptions: Class-A audio transistor amps or RF transmitters / tranceivers may need more that 100W but I have not had one of those on my bench.

[@johncoops6897]

The formula is shown in the video from about 3:00 onwards.

[@ivanr5315]

​@@TheHWcavethanks, clear! I think I might make a version that can accommodate two bulbs to keep it flexible so if one bulb is not enough I can add another one in parallel. Thanks

[@TheHWcave]

@@ivanr5315 Yes, that would a good solution

[@davestorr6764]

This has made me wonder: For initial test at power-up, what are the benefits of using the current-limit (incandescent bulb) approach v’s using the voltage-limit (variac) approach? Love the channel, this is brain food 👍

[@TheHWcave]

It depends (as always). If a device is really quite old and has lots of big caps, it is useful to run it for a bit a lower voltage from the variac (and with the light-bulb current limiter) so the caps can re-format (rebuild the insulating layer). A variac by itself is not limiting the current. A dead short would trip the variac even at 30V or so. If the device isn't quite as antique, the light bulb is the only thing I am using at the first power-up. So (IMHO) the light bulb is always used, the variac occasionally and in addition.

[@johncoops6897]

INFO About the "normal current" for a Device Under Test (DUT)... The manufacturer's rating label will show the maximum current (Amps) that a device draws. Thus is the higher when eithee from starting up (eg motors) or at full power, and this can differ greatly from the "wattage" which is usually the normal running current. The maximum current rating may be an instantaneous value, but it all depends on the device, and if you don't understand that characteristic then you shouldn't be testing things anyway 😮 Therefore the light bulb wattage should initially match the DUT rating for "Watts". The worst that can happen is that the DUT cannot get started, and the light bulb stays at full brightness. That's thw whole idea of having s current limit irrespective of low voltage or AC Mains. A simple AC Power meter is very worthwhile as well. A cheap plug-in model like a "killawatt" will be OK, or use a AC panel meter from Ebay etc.

[@TheOriginalMKK]

Clever

[@Crimsomnia]

I heard halogen bulbs would be ok for this purpose if incandescents are not available. Is there any truth to this claim?

[@TheHWcave]

I don't know. Its likely it would work. After all, they are resistive loads like a normal filament bulb. It would be interesting to test but I don't have any to play with. You could also try to find a suitably big wire-wound resistor, say 500 Ohm 150W for 240V mains. The main problem is that you could not see what is going on without hooking a voltmeter up in parallel to the resistor. It also lacks the PTC characteristics of a light bulb but it would limit the current and prevent or at least reduce disasters in the test objects.

[@beanmeup9902]

That does NOT look like a standard light switch . . . far from it . . . it appears to be a three-way light switch.

[@TheHWcave]

I can't remember, I just took what my local DIY place offers. You only need a standard SPST but a three-way light switch can be used that way.

[@beanmeup9902]

@@TheHWcave Brilliant . . . watching some more of your videos . . . Bravo!

[@bertoid]

My version of this sort of thing: kzbin.info/www/bejne/mYbcXmuYrM98lbs

[@TheHWcave]

Very nice. I like the way you protected the bulbs. The only drawback I can see is that it makes it a bit harder to see bulbs that just glow unless you look from the top at the bulbs inside.

[@bertoid]

Having the panel meter there to show wattage, reduces the need to see the bulbs. But it's currently connected before the bulbs, so it's not showing the voltage at the load. I should change that, but then I'd probably also need to separate the meter's supply input and voltage sense input. One day...