Can this be tested with multi-meter?

[billbyrd1945]

My wine cooler went crazy and I'm guessing at what to order. Does anyone know if this can be tested with nothing more than a multi-meter? The failure went like this: The LED panel on the door of the cooler starter blinking rapidly. When I checked with meter, the voltage going to the LED board was bouncing rapidly from a little over 1.0v to two other voltages below 1.0v (changing so fast I couldn't say for sure what they were). I put a 5.0v wall wart on it and it lit up properly. I've googled the IC to death with no answer to my question. To save you some squinting, It's SONIX SN8P2711APB.  Thank you for any help.

[shakalnokturn]

http://www.sonix.com.tw/article-en-1005-3118

[coromonadalix]

http://www.datasheetdir.com/SN8P2711APB+Microcontrollers-8bit
http://www.sonix.com.tw/article-en-1005-3118

what can be tested with an dmm, the i/o ports, if they output say 3.3 or 5 volts,   the Xin Xout pins are the crystal oscillator, in your case OSC1  its an resonator (does the same job) 

You have U1 on the board whos like an transistor, read the markings, is it an regulator or an voltage supervisor "watchdog"/ reset ic ?

You have to test the supplies going in or out the pcb, capacitors, solder quality ...

[GeoffreyF]

If I were to take a guess from what you write, I would say the Power supply.   The electrolytics look ok in the photo but if they are bulging at the ends, that may be a culprit.  If you can measure AC where there should be DC, that would confirm it more for me. 

[billbyrd1945]

" I would say the Power supply"
That sounds like something I'm capable of doing. I check for that and report back. Thanks

[billbyrd1945]

In response to the post about the power supplies, here's what I found: power supply black is putting out 7.6vdc, power supply white is putting out 3.3vdc. Is there any way to find out which one is correct and what to do about it. I would like to follow this through for learning purposes if nothing else. Thanks for pointing me to the power supplies.

[shakalnokturn]

Well... They both look like they are rated 12.5V DC so neither would be right 

You can go through the usual advice and check the electrolytics, it's usually a good thing to do (if you do don't neglect the smaller ones stuck against heatsinks and large warm resistors).
Personally, considering voltages, symptoms and aspect I'd be going straight to the large green capacitors on the primary side, the non-polarised ones.

[billbyrd1945]

I didn't even see those labels. You can tell how inexperienced I am. I will try to de-solder the caps. Thank you.

[GeoffreyF]

The Cap on the lower right - near the green and white piece of paper looks like it might have failed. The top looks swollen.

3.2 is a common voltage for CMOS devices.
7.6 may be right or it may be due to a failure. If you can find the rectifiers, a DVM can tell you if they are conducting differently in each direction. If not, they have failed. Those are pretty generic and can be replaced with any silicon diode with a reasonable current and PIV rating.  Go high on those if in doubt.

[billbyrd1945]

Thanks. I will check.
In the mean time, I de-soldered one of the large green caps to check it. Problem is, it doesn't have a marking I'm familiar with. It says 1U0K 250. Anyone have a clue. It measures 0.3nF with my multimeter.

[Yansi]

1uF 250Vdc.

[GeoffreyF]

The larger electrolytics are more likely to be your problem, as I prior described.

[billbyrd1945]

And I'm glad to know that because there's no mystery about their individual values. Working on removing the one you suggested as the most suspicious looking.

[shakalnokturn]

1microF reading at 0.3nF?
It's a surprise you're getting anything at all out of this power supply.
You may consider replacing these before trashing all the electrolytics.
IMHO...

[Yansi]

The larger electrolytics are more likely to be your problem, as I prior described.

No they very likely aren't

These mylar capacitors (talking about the 1U0K 250) are a known piece of shit. But they are cheap, so widely used.  The loss of capacitance is due to internal arcing in the capacitor. If you smash it into pieces with a hammer and take a piece of the internal foil,  looking through a light, ya'll see the foil is full of burnt holes.

By having a brief look on your power supply board, it seems like a bog standard push-pull forward converter with a half bridge stage of bipolar transistors and a base driving transformer (the smaller one of the two). The 1U0K 250 capacitor is in series with the primary, to strip off DC voltage from the main transformer's primary.

Replace this 1U0K 250V thingy with a new one (or at least one measuring good, taken for example from an old PC AT/X power supply - there are the same) and it will be good as a new.  I would then also suggest replacing all the electrolytics anyway.

Your PSU will be wired very similarly as an PC/AT PSU anyway, so here's  a schematic with the 1uF cap circled as a hint for your repair.

[billbyrd1945]

Wow! I've never had it so good! So much good feed back. I always feel like I'm being a PITA, but, at 73 I enjoy learning things I never thought I could. My biggest impedance to learning a lot in a short time span is desoldering. I can't keep up with the suggestions. Takes me forever to get one removed.

I previously posted 0.3nF. I was using my new Uni-T UT61E. Since then, my new Fluke 15b+ arrived. It measures that same mylar as 0.973uF! With that much difference I felt that my first UT61E measurement must have been user error. On retest, the UT61E shows 0.95, so close enough I guess.

My one and only electrolytic test is a rock-steady 789uF for a 2200uF with the Fluke and some bizarre reading on the UT61E. I'll get back to the desoldering struggle. Thanks to everyone.

[shakalnokturn]

I was wrong then...

[macboy]

Wow! I've never had it so good! So much good feed back. I always feel like I'm being a PITA, but, at 73 I enjoy learning things I never thought I could. My biggest impedance to learning a lot in a short time span is desoldering. I can't keep up with the suggestions. Takes me forever to get one removed.

I previously posted 0.3nF. I was using my new Uni-T UT61E. Since then, my new Fluke 15b+ arrived. It measures that same mylar as 0.973uF! With that much difference I felt that my first UT61E measurement must have been user error. On retest, the UT61E shows 0.95, so close enough I guess.

My one and only electrolytic test is a rock-steady 789uF for a 2200uF with the Fluke and some bizarre reading on the UT61E. I'll get back to the desoldering struggle. Thanks to everyone.

That 2200 uF cap is shot. If you had a device that measures ESR you would likely see it is way too high. Component testers which can show ESR and test almost any 2 and 3 legged component are very inexpensive on ebay/etc.  There is a very long thread here about these.

Desoldering shouldn't be a pain, especially on a single-sided board. There are some good threads around here about techniques, whether you choose to use braid, a spring click pump, or a full-fledged desoldering iron/gun.

[billbyrd1945]

Enough fingers were pointed to caps that I decided to suffer through taking every single one off for testing. And I actually because pretty good at desoldering. I've got them all hot-glued to a board with Fluke measurements (for what they're worth). A new ESR meter will be here Saturday. And a third check will be with my new oscilloscope in conjunction with signal generator which is due to arrive on Sunday. Obviously, this is for self-teaching, not the most direct route to getting my wine cooler up and running.

[billbyrd1945]

Just noticed I made an error on labeling my cap board. The last two should be pF instead of pK.
And for the record: This noob thinks it's unnecessarily difficult to identify capacitors.
Here's a pretty good one (for the ceramic disks) for anyone else on my level: http://search.alkon.net/cgi-bin/pdf.pl?pdfname=01569.pdf

[Yansi]

I was wrong then...

I was wrong too, sorry.

Reading through the last comments, it is obvious there are some "issues" (cough cough) with some of the electrolytics.

Just change all those suspicious caps and it'll be fine. In case you want to have the supply working and never touch it again, buy some suitable branded caps from Nippon/Nichicon/Rubycon and replace all of the caps with new ones. (for example Nippon SMG/SME series for the generic small stuff, LXZ for the large output ones and KMH for the input mains ones).

[james_s]

I've repaired many dozens of switchmode power supplies and electrolytic capacitors have been bad in probably 80% of them. It's always the first thing I check, often they're so shot that they're bulging or leaking but not always, sometimes the ESR is way high with no visible damage.

[billbyrd1945]

Thanks for the new responses. I'm committed to doing the most thorough job I can and will therefore order best caps I can find. I'll end up with countless hours in this project, but it's not about the wine-cooler. It's about learning. I got in too big a hurry in ordering an ESR and accidentally ordered a mere "capacitor checker". Sent it back the same day and re-ordered. The biggest challenge so far has been in ordering caps. The system could really use some standardizing in marking. (I realize there are markings already in use, but they vary a lot) And it really puzzles me that there can be one 1.0uF/250v cap the size of a pencil eraser while another one is 3/4" long! In that case, I went with the bigger is better philosophy. Almost all caps are coming from Mouser. Had to order the big wide 1.0uF/250v from someone on ebay.

[james_s]

Capacitors are made in a huge number of types and styles, there's no one size fits all, most are optimized for one thing or another, either an electrical characteristic, temperature range or lifespan. Don't get too caught up in trying to get the best of the best, this is a power supply, not an audiophool grade amplifier. Just stick to reputable brands and you'll be fine, I personally like Nichicon, Rubycon and Panasonic electrolytics but there are many others.