I checked the fuse — visual inspection indicates the wire is not burned through, fuse seems good. I simply get zero response when I press the power button.
Does anyone have any repair advice?
Time to hotwire it!
Any chance it is just a dead battery in the remote?
Neither the main power button on the unit nor the remote… will work to turn it on.
The main power button on the front is one of those things that’s kind of like a “blister pack”! I don’t know the technical term, but there is a thin compliant layer of material on the front, and it lets you press in on it to trigger the button. Same style as is commonly used on microwave ovens I think.
I am going to recheck the fuse using my multimeter — there is cause to not trust my visual inspection. The fuse could have failed in ways not easily visible.
Notes:
According to the owner’s manual, it’s a 2A fuse.
owner’s manual: https://m.media-amazon.com/images/I/81UqKwdl29L.pdf
Well, easy enough to hardware it and add a physical switch if the board popped.
Hopefully it is something obvious.
Checking the fuse is a good first step but even then usually the fuse having cleared means there’s something unhappy downstream. Also a good idea to double check it either on a different outlet or plugging something else into the outlet that it was using, just to make sure.
My next step would be to unplug it, gut it and take a few photos, specifically looking at the control panel assembly and maybe dismantling that as far as you’re comfortable…
I’d also be taking a look to see if replacement parts are available, again specifically the control panel assembly…
Main tell-tale things will be any kinda black/brown marks, corrosion, loose or burned wires, that kinda thing.
Fuse passes continuity check with multimeter!
Outlet checks out, powers another device plugged into it.
So, now I guess I’m looking at the inside, seeing if anything obvious sticks out!
Thanks, guys!
Look for indications of Magic Smoke being released… 
The manufacturer, WEN, does indeed sell the main PCB, for $30. I have sent them a support request just in case they may show me kindness on the thing failing only weeks after the 2-year warranty.
Also, I did get it dismantled and I got pics of both sides of the main PCB. I cannot really see any immediately obvious issues. I do see something on the PCB that looks like a small fuse.
@jono035 do you see anything in these pics that stands out to you? Any other advice?
Nothing immediately jumps out at me.
Single-sided PCBs are notorious for having cracked solder joints, though, so I’d take a closer look at some of those.
Specifically with ones like that, it looks like it could have a circular crack in a ring around the pin. The soldering all looks a little hit and miss which could cause that as well.
Basically you’re looking for anything that looks like this:
That resistor on the input looks like it has seen a fair bit of heat, along with the discoloured traces on the underside.
There’s also what looks like a mini voltage regulator which would be worth checking the output voltage of with the board plugged in. That starts to get a little sketchy, though, so I’d personally advise doing that only if you’ve got an isolating transformer or are confident working with mains voltages.
Beyond that, there’s not a whole lot there. The 3x TO-220 package devices are likely FETs, although the way the circuit is laid out is a little bit weird. They appear to have unrectified mains voltage across them which would conduct backwards via the body diode, so maybe they’re actually SCRs or something odd like a JFET… I feel like I’m missing something there. Either way, I suspect if they were the issue you’d either be popping the fuse or it would be beeping/flashing LEDs without the fan moving.
Pretty simple damn board all told, so there’s not a whole lot that should be going wrong with it. The fact that it’s doing nothing at all makes me think broken joint or open circuit resistor.
OK, thank you!!! I used my Vevor microscope and looked at all the solder joints and did not spot anything, although my eye is untrained so I may have missed something.
My current thinking now is to accept a free PCB if they give it, or just pay the $30 price if they don’t. Thank you again for looking!
Definitely check that discoloured resistor. That’s the next easiest step.
OK, will try that.
OK, my first test of the resistor was before desoldering it from the board, but given that my ohm test with my multimeter consistently told me 3.2 M ohms (note: M, not K) I’m thinking that resistor is indeed toast! I’m thinking I will check my stash of resistors and see if I have the right one, desolder old, solder in a new one, and see if I get success. Sound right?
That’s the general idea, although I’d expect that it’s a little unlikely you’d have a resistor that’s a correct replacement. It looks like a larger high power type, which would be consistent with the heat damage.
If you measure the dimensions it’d be possible to roughly guess the power rating based on that, but given that it has already been overheated once, I’d probably replace it with something higher rating. The main issue there being that it can be a challenge to get them to fit as they’re usually longer. What are the dimensions of the resistor?
At a guess I’d say it’s drawing 10-20mA which would be around 1-2W… Finding a 2W replacement would probably work, as would wiring a few 1W ones in parallel.
It’s also worth noting that the resistor could have just failed due to being under-rated for the application or due to a persistent high-line condition or similar causing it to run hotter than normal, or it could be due to something else having failed, leading to it drawing more current and the resistor eventually failing. With faults like this, all you can really do is fix the first and most obvious one then give it another try.
I’ve desoldered the resistor, and it reads as “OL” (overload). It’s about 7.7 mm long and at its thickest point it’s about 3.3 mm in diameter. It looks beefier than my stash of resistors. It’s way bigger (fatter) than the 1K resistors in my stash.
Reevaluating at this point: 1. I don’t have the right replacement, 2. If I go to the trouble to get the right replacement, it may not be all that’s wrong. 3. My time is worth something. 4. the $30 price on the new PCB sounds better and better.
All fair considerations, but perhaps also consider that this is a valuable learning experience and that if it’s a poorly specified component (which I suspect it could well be), then that’s likely to lead to a future failure of the new board.
If you don’t have the time to devote then by all means, a $30 replacement is the sensible approach.
My next step would be going to Digikey.com and using their parametric filters to find similar resistors. We know it’s a 1K 10% resistor and is ~3.3mm OD, ~7.7mm long.
That size doesn’t quite match up with common lengths, but the resistors are measured from where the metal terminals start and you might have measured the main body length, so I’d assume it’s maybe 9mm long.
That gives a couple hundred results:
Hopefully that link works, if you wanted to look for yourself.
The highest power is 2W so I’d think that’s a pretty reasonable assumption.
So for replacement, I’d probably try a 2W 1KR resistor.
Something like this:
Alternatively, if you wanted to give a bit of headroom or you could more easily find 1W resistors, you could use a pair, either 2x 2.2K to make 1.1K parallel or a 2.2K and a 1.86K, that kinda thing. 2x 2W resistors would give you 4W total so would run roughly half as hot as a single 2W, for instance.
If you wanted to do a quick test with resistors on hand and you’ve got 1/4W ones, you could do the same thing by using 8 or more 1/4W resistors in some combination of parallel/series to make an equivalent to a 2W one.
8x 8.2K resistors in parallel would be 1K with a 2W rating. 5x 2.7K resistors in parallel would make 510R and then 2 of those sets in series would make 1K with a 2.5W rating. 10x 100R in series would make 1K @ 2.5W, etc. Any of those options work. For them being in parallel you can just solder the first one in and then start adding extras ‘on top’, tacking the legs to the previous resistor. It’s not the ‘best’ solution, but it’d work well enough.
@jono035 Thank you SO much! I am awed at your knowledge.
