Looking for any advice on fixing a broken circuit

[Nixsnow]

Hi there,

I've been trying for a little while now to try fix a broken power box switch/circuit/thing? I'm by no means much of an adept at fixing electronics, as most of them look quite imposing, but I'm trying to fix a broken light for my mom, and after testing a few of the outer components (like the bulbs, the wiring etc.) all of them seemed to be working, so i decided to take apart the power box and the circuit didn't look as complex as some of the others i usually see, so i figured I'd try my hand at fixing what I could. I started just trying to clean up all the accumulated hair and dust as that usually seems to be the problem with other things I try my hand at repairing, but no luck there. So i delved a little deeper (and spent a lot of time on YouTube learning how to use a multi-meter) and I've kind of identified the problem, but i can't quite isolate the source or figure out what to replace/repair, so any advice would be super helpful.

I've attached a picture of the circuit with some labels just in case anything wasn't obvious at a glance, and if you need any more info on a certain part or something i'd be happy to add it.

Basically it's a "dual light", where one of the light bulbs turns on with an on/off switch, and the other one is on a kind of sliding-dimmer, where if you slide it up the light gets brighter. The simple on/off light works just fine, but the dimmer one is non-functional.

When I tested the power points, I get the normal mains voltage reading on the power output for the normal on/off light, and normal readings for the power input, but on the dimmer output i get nothing. I did another test where i put the positive pin of the multi-meter on the mains live power, and the negative pin on the "N" of the dimmer power out, and i got a normal reading, but testing "N" on the mains power in and "L" on the dimmer out i get nothing.

I tested what i could on the circuit with the multi-meter, and so far all i could identify was this:

I tested the continuity on the sliding dimmer switch, and i only get a beep when it's pushed all the way into the on position.
I tested the fuse with continuity and it got a beep.
I tried testing the resistors and capacitors, but i don't get anything, but i also think i read somewhere that you won't get an accurate reading while it's still connected to the circuit board, so i can't be sure, but i don't want to pull anything off just yet with my amateur soldering skills before i have no other options yet as i don't want to break any already-working parts in my ignorance.
I tested what i think is an inductor(?), the copper circle, and i do get a beep from that.

Then i tried testing the transistor, and i thought that might be the culprit, as most of the guides i saw on testing them said you need to put the red or black on the left side, and the other in the middle, then switching them around and seeing if you get a drop anywhere, which i tried and didn't see a drop at all. However then i tested it again, but i put the red on the far left, and the black on the far right (totally ignoring the middle pin) and get a drop both ways around, so I'm not sure if that means it's working, as all the tutorials spoke about the middle pin, but i get no reaction from that whatsoever.

Lastly, i tried checking what i assume is a diode (the little blue one right above the transistor) and i don't get a beep from that at all, so i thought that might be the problem, but I'm not sure if it falls into the category of "things you cant test whilst still connected to the circuit board" so now I'm at a bit of a loss which one to try replace, and so i found myself here.

I've also included a picture of the backside, just to make sure there isn't any obvious solder-burns or something but I think for the most part everything looks mostly okay, but i might be missing something obvious. If anybody has any ideas of what could be the problem, or if you'd like me to test more stuff and get back to you I'd be happy to do whatever you need. I'd really like to try fix it as there's no replacement unit available anywhere, and it's quite a sentimental light so I'd love to be able to repair it if possible. Sorry if i haven't included any information that should be really obvious, but if you need anything more please let me know, and thanks in advance for any help 

[james_s]

That looks like a standard light dimmer. The "transistor" is a triac, the little blue thing is a diac, the inductor is for filtering noise to reduce interference to radio reception caused by the dimmer. I'd expect the circuit is similar or identical to that used in basic wall dimmers. The most common problem with dimmers is a lamp blows and the resulting arc inside the bulb kills the triac, but usually they fail shorted so the light is just on full brightness. Is that a thermal fuse glued in the center of the inductor connected by those black wires? Check and see if that's open.

[Nixsnow]

So just before dying a bulb did blow, although after replacing it there was no response even at full brightness. I'm not sure if this gives anymore insight, but weirdly enough i tried using an LED bulb that ended up being incorrect (wasn't a dimmable one) and for some reason it did light up a little, but nowhere near the brightness it should've been. Normal halogen bulbs didn't work whatsoever.

I tried testing the continuity between those two black wires and get no response, so i'm not sure if that confirms the thermal fuse has an issue, but if i try and pull it out is there a best way to do it? I'm not sure what this white stuff is, so i'm not sure if i should try use a heat gun or just try rip it out, do you have any advice?

Thanks again for the help, i know it's probably a super simple question given all the expertise here, but i really appreciate the assistance

[Jwillis]

Hello and welcome to the forum.

Well you got a good start. You are correct that you don't get an accurate reading testing components in circuit. So you should test the following out of circuit.

The little blue thing you called a diode is a DIAC . Its a type of diode for AC (Alternating Current). You can google DIAC to get a better understanding of how they work.
Diacs tend to fail shorted. So with your MultiMeter(DMM) set to Diode Mode, put positive probe to one end and negative probe to other end. If you can a reading of OL (open line)your good so far. Reverse the probes and test again. You should get OL (open line) . The  Diac is OK. If you get any voltage reading across the DIAC either way then it is faulty.

The transistor thing is call A TRIAC. Google TRIAC to get a better understanding of how they work. Also take note of the letters and numbers printed on it and post them here so we can see which TRIAC it is. Or you can search the letters and numbers for a DATASHEET and post that here. Either way we will need to find A data sheet for it eventually for testing and possible  replacement if required.
The TRIAC is attached to the heat sink with a rivet. but it looks like the TRIAC it will come off with the heat sink. No need to take off the rivet just yet.

With your DMM set to diode again . Place the positive lead to the first pin which should be the GATE. Place the negative probe to the second pin and note the voltage on your DMM . You should get the the same voltage with probes reversed. Post your results here. With the probes connected to second and third pins you should get OL just like the DIAC. 

The big coil inductor and yellow capacitor are for suppressing noise and voltage spikes.

To test the slide POTENTIOMETER and resistors , set the DMM to resistance ( ). This is measured as OHMS . The Brown, Red, Orange, Gold should read around 12K Ohms. The Brown Green Yellow Gold resistor should read around 150K ohms. If your not sure post readings here but I don't think there is anything wrong with those. Or the Red capacitor. The yellow capacitor also looks OK.

I hope tings work out for you and maybe some other people can help you as well. Good Luck.

[james_s]

I bet that is a thermal fuse, and it was probably damaged by the overcurrent that occurred when the bulb blew. I would probably try bypassing that and see if that fixes it.

[MrAl]

Hi,

First thing, check the fuse. That's the white thing.

[james_s]

Hi,

First thing, check the fuse. That's the white thing.

From his original post "I tested the fuse with continuity and it got a beep."

[Ian.M]

@Nixsnow,

As others have requested, please confirm that the white ceramic fuse has continuity, testing either holder clip to holder clip with the external wiring disconnected, or out of circuit.

A continuity test only 'beeps' for low resistances so is often less than useful when testing many components.  OTOH it is often combined with a diode test range that shows the voltage across the part under test while injecting a small current, typically a fraction of a mA, which is very useful when testing diodes and other semiconductors that are expected to have a diode junction between particular pins.

With the board disconnected from all external wiring, re-test all components that you did not get a continuity 'beep' for, with the multimeter on a resistance range.  If it doesn't have auto-ranging start off on a 200K range and go up or down according to the reading.

In this case, you can test most of the circuit without disconnecting any components.  Unless faulty, the DIAC wont conduct till the voltage across it rises to about 30V, which is far far more than most meters put out on resistance ranges.  C2 is expected to read open circuit, as the 'sneak' path back through R1, the slider pot and the TRIAC is blocked by the TRIAC (unless faulty).  Therefore you should be able to measure the resistance of R1, and of the slider pot without the surrounding circuit interfering.   If either were open circuit or C2 was very leaky, the TRIAC would never turn on.  The slider pot should read only a few ohms when the slider is at the end with one pin, increasing smoothly to hundreds of K as you move it to the other end.  Any sudden jumps are cause for concern.

If you've got a DMM with a capacitance range it would be worth seeing what it reads for C1 and C2, and compare the readings with the values printed on their sides.  The capacitance readings should match the values +/- 20%, but C1 may have decreased in value due to culminative damage from mains spikes and a bit lower for it isn't cause for concern.

Assuming everything reads OK so far,  its pointing at a failed semiconductor, i.e. either the DIAC or TRIAC has gone open circuit, or the TRIAC's gate has been internally damaged so it is no longer capable of turning on.  You will need to desolder the TRIAC to proceed further, e.g. to test it can be triggered in a simple test  jig.  It may be simpler to take a chance and 'shotgun' it by replacing both the TRIAC and DIAC.

[Nixsnow]

Hi everyone,

Thanks so much for all the responses and the explanations, I read through everything everyone suggested, but I also figured it would be better to test things/take things off the board one at a time just because with my amateur skills I don't want to damage any other components unnecessarily, so i figured I'd go through things in order one by one.

I managed to get the thermal fuse out, and tested it with the MM on diode/continuity mode, and i got no continuity nor any voltage changes in either direction. I assume that means the fuse is blown? I've attached the image of it taken out if that helps to assess anything. I noticed the reply about just bypassing it, which i understand in theory but I honestly have no idea how to accomplish it on this circuit, and I'm also pretty worried about blowing myself up into the afterlife by doing something wrong and then plugging it into the mains, so if that is the correct way to go about fixing this, could anyone possibly elaborate a bit more on the method of doing so?

Once again, thanks everyone for all the help, i really appreciate everyone going so in-depth to help me with such a newbie thing, and if this doesn't happen to be the fix I'll go about trying all the other suggestions one by one as well.

[Ian.M]

Errr...... you are clipped onto the insulation, so of course it reads open.  Move the clips to the bare metal wire ends and try again! 

[Nixsnow]

Oh no, sorry i should've specified a bit more. I just placed it like that for the sake of testing the ends of the wires. The little tool i have with the alligator clips is entirely made of metal, so if i just touch both clips with my MM it beeps even if nothing is on it, so i figured it would probably not be accurate for testing anything via the clips themselves. I tested the fuse just on its own on the little green mat and got no response, and then when i placed it in the clips (where they're on the insulation) i tested the ends of the wire, not the alligator clips themselves. I'm not sure if I'm confusing something here, but i assume that should give an accurate result?

[Ian.M]

Ah, that explains it.  I thought the clips were on the ends of your multimeter test leads!

Yes that thermal fuse is bad.  However it may have been good before you desoldered it as the heat of soldering often blows them unless you clip heatsinks to the leads between the solder joint and the fuse body, or chill it well below zero with freezer spray.   You have to be pretty quick soldering them as well to minimise the heat buildup.   

To let you carry on testing, you can bypass it with a wire link, but it needs to be properly replaced before putting the lamp back into service, as its there to prevent a fire if the dimmer board overheats e.g. due to someone putting in too high a wattage bulb.

[Nixsnow]

Yeah i figured that could be a problem when i was doing it. I did test the wires coming from the fuse on the board before i pulled it, and got the same dead result, but I'm not sure if it being on the board influences that aspect.

I'm still a bit unsure about the whole process of a bypass, so just to make sure, I've attached a picture of the solder points. The parts in blue were where the fuse wire leads were soldered in. The fuse itself was glued into the inductor, and underneath the inductor is the red square on the board. I'm not totally sure, but I'm pretty sure that red solder point is where the inductor wires in, and I'm not sure if that's relevant at all. If i wanted to bypass it, would i basically just solder a little wire between those two points in the blue square and then plug everything back in and test if it works?

[Ian.M]

Yes, link the two pads inside the blue box and try it with a known good incandescent bulb of appropriate power rating for the fitting.  Do not attempt to use a LED or CFL bulb, as even so-called dimmable ones  often don't 'play well' with old TRIAC dimmers.

[james_s]

He said he tested the thermal fuse in circuit before soldering and it was still open. If a part shows open in circuit then it's open, other parts of the circuit can't make a part appear to be open. I've had thermal fuses like this fail quite a few times when there was no excessive heat, they are just fragile and can fail due to drops, vibration or overcurrent. I think it's pretty clear that the bulb blew and the resulting current surge blew the thermal fuse. It's a 125C part, should be easy to find a replacement. I'm not even sure why they have a thermal fuse in the inductor anyway, normally the inductor would just be sized such that it can handle as much current as the main fuse in the circuit can pass without overheating.

[Nixsnow]

So great news, it worked! I created an absolute soldering monstrosity of a wire between the two holes where the fuse was, and plugged everything back in and tested it, and the light worked (dimmed and everything!)
I guess the next problem now is that i have to replace the wire with a new fuse, however I have absolutely no idea how to identify the one i pulled out. The only things i can identify on it are "L2", a logo of some sort, "125 or 126 C°", and 007S. Is there any sort of generic fuse that is commonly available I could switch it out with? Or do i need to find some sort of specialty store and ask around perhaps?

Thanks again for all the tips, I was starting to think it might be an absolute nightmare trying to get this working, but you've all been great 

[Ian.M]

Measure its body dimensions and also tell us the current rating of the 'regular' 20mm ceramic fuse (in the fuse clips) to give us an idea of the thermal fuse's likely current rating, and we'll see if we can suggest one

You could try googling: "125°C" radial thermal fuse
and see if you can find one in a similar package.

[Nixsnow]

The other ceramic fuse (i assume you're talking about the white one with the silver caps) says "2A" on the board, and the writing on it reads "F2AH250V"

The dimensions of the broken one I've pulled out are:
5.91mm / 0.233inch width
6.93mm / 0.273inch height
2.83mm / 0.111inch thick

Just taking a bit of a shot in the dark, but would I be right in assuming I need a fuse rated for 2A, 250Volts and can withstand 150°? My country has a little less freely available components like this, so the one of the only places I found something similar was here:

https://www.mantech.co.za/ProductInfo.aspx?Item=15M1766

If i need something a bit different though i can try and ask around at any electrical supply stores, but if you need any more info just let me know.

[Ian.M]

So the odds are you need a 2A or 3A, 250V AC rated thermal fuse.  The 20mm fuse "F2AH250V" feeds both bulbs, which implies the thermal fuse carries no more than 1A.  That's the way you want it, because its a PITA if the thermal fuse blows on an overload, so you'd prefer the 20mm fuse to have a high chance of blowing first.

The fuse is rated to open at  125°C.  You need one close to that, e.g. between  120°C and 130°C. The 150°C one you found is unsuitable.  A radial one (both leads on one end) would be preferable, but you could probably make do with an axial one similar to the one you found, provided it has an insulating body and you sleeve its leads right up to the body.  Do *NOT* attempt to use any with a metal body!

[james_s]

Personally I'd probably just bypass it and not overload the dimmer by putting too big of a bulb in the lamp but I know that will make some nervous.

This one will probably work https://www.digikey.com/en/products/detail/optifuse/K2B-125/12170796

[Nixsnow]

I see, I think i get the gist of what you're saying. I'm going to try calling any nearby suppliers tomorrow to see if there's one close by, but otherwise I'll grab the one from the link @james_s provided. Thanks so much both of you

Just one last question. I assume i should ideally place the new fuse back inside the inductor like it was before, but after browsing around it doesn't seem like the electronic silicone adhesive is all too common around here. I can probably find it with a bit of searching around, but is that a super necessary thing to make sure i do? Just curious if other glues or something like a hot glue gun could accomplish the same thing or if it's a dealbreaker, as after getting it all working again I'd be a real idiot if i messed the whole thing up not doing a proper job securing it.

[Ian.M]

*DON'T* use hot-glue. If you use it on something that gets hot, it tends to melt, flow where you don't want it, then starts to carbonise, becoming conductive, and ruining any circuit or components in contact with it.

Many other adhesives are either unsuitable for high temperatures, or corrosive to copper, so a neutral cure silicone is probably the best choice here. 

If you cant get electronics grade silicone sealant easily, you can substitute automotive grade silicone sealant, provided its the sort that's safe for use on oxygen sensors.  Don't use black silicone as the pigment may be conductive carbon black. 

[Nixsnow]

Thanks for such a detailed explanation. Gonna try and get all the things i need tomorrow hopefully, and then will let you know how it goes if i manage to get everything assembled and working again.

[themadhippy]

. Gonna try and get all the things i need

get 2 fuses,that way you wont overheat the first one soldering it back in place and wont need the spare.If you only get one that sod murphy WILL strike

[james_s]

The type of glue really shouldn't be very critical in this application, any kind of silicone or latex caulk ought to do. The inductor windings are coated in insulating varnish, it shouldn't get particularly hot and can't exceed around 125C without blowing the thermal fuse so you shouldn't need any special high temperature adhesive should do. I know it's advised against, but I've used regular silicone bathroom calk on electronics lots of times and I never had any issues with corrosion, I've always let it cure out in the open though, not in a sealed enclosure.

Digikey shipping may be expensive in your part of the world so you might try some of the various Chinese sellers, thermal fuses are common parts.