Power Supply - Intermittent

[raphaeltube]

Hi

I have 14 fairly old (maybe 10 years) domestic electronic sewing machines for students to loan out & use (over and above the industrial machines they normally use) at a Fashion Design Academy.
A few times in the past, they report a machine dead. I check it, it's dead. I pull out the PSU for testing: AC goes in, there it mains DC after the rectifier, fuses are good, but no DC out.
Then before I get much further, it works. I put it back and everything is good (for a month or 2 anyway.

Of the 14 machines, I now have (had) 6 playing dead. -the 1st one I opened up has just come alive while checking voltages .
So I thought I'd get some opinions before prodding the other 5 to life, making them un-testable.

My only suspicion is dry joints? -do the symptoms fit?
They seem to go dead after sitting unused for a while (months).

I don't see any resetable fuses or anything. My guess is it's the mechanical movement that get's them going, but could be electrical I suppose.

I can post pics if necessary, but there doesn't seem to be much to see.

 -R

[raphaeltube]

My guess is it's the mechanical movement that get's them going, but could be electrical I suppose.

Nope. 3rd one today (working now ), I had hooked up to the machine so I could see exactly when it came on (makes a noise & light comes on).
I gave it a few good knocks and nothing happened, then took a break from probing, and it just came on after 5-10 minutes.
 -so maybe thermal?

I then switched it off & left is without touching anything, but still worked when I tried it an hour or so later.

 -R

[George Edmonds]

Hi

Without a picture or more details it is impossible to help

George G6HIG

[andy3055]

Agree with the comment above.

One thing you could do anyway is to check the motor brushes and the commutators. If they are bad and full of carbon residue, they will not make proper contact. The brushes may be getting stuck inside the brass housings. May be the springs have got hot and weak. May be you need new brushes altogether!

Check the foot pedals. If they are not pure electronic, (in the older machines), if people are learning to sew, they tend to keep the foot on the pedal lightly, making it arc before the motor starts to spin. This in turn will make poor contact.

Check and make sure they are mechanically not stuck with lint/thread under the bobbins (or any other issue) as that can severely load the motors.

Just a few suggestions.

[raphaeltube]

Without a picture or more details it is impossible to help

Ok, I'll post some pics...

[raphaeltube]

...check the motor brushes and the commutators...
...May be the springs have got hot and weak...
...Check the foot pedals...

No, no. The problem is definitely with the PSU board.
It's a switch mode supply. It has 32V, 12V & 5V outputs for the motor, the light & servos, and the electronics.
When is it's 'dead' condition, there is no AC across the transformer, but there is mains DC after the initial rectifier, so the issue seems to be somewhere around the PWM chip, oscilator, MOSFET or feedback couplers? -one of them came alive while probing around the optocouplers, but could have been coincidence.
I'm not much good with switch mode PSUs, so difficult for me to understand what to probe for.

The symptoms are very predictable though. I have the luxury of lots of units with the same issue.
Basically, after standing for a long time, they play dead. Then after poking around a bit they work again, quite consistently.
So my question really is, are there obvious components with that kind of failure mode? Do those symptoms fit dry solder joints?
I have re-wetted all the joints on one of the boards, but I have no way to tell if it fixed the underlying issue or if it's just decided to work for a few months.

 -R

[raphaeltube]

Pics:

[George Edmonds]

Hi

First working on mains powered SMPSU's is very dangers as there is some 360VDC at high current around the switching device and the bulk storage capacitor.

The resistor that you have marked "this resistor powers up ...." tend to go high in value due to being over voltaged. Also check the electrolytic capacitor it supplies.

Failure of the opto-couplers used to be a very common problem with SMPSU's

Any and all of the electrolytic capacitors can cause the problem, most used are not up to the job that they are being asked to do.

George G6HIG Dover UK

[Audiorepair]

I would change the orange Electrolytic Capacitor next to the chip Power supply resistor. 

If this is leaking/high ESR  the chip may not always start.

[Gyro]

I would change the orange Electrolytic Capacitor next to the chip Power supply resistor. 

If this is leaking/high ESR  the chip may not always start.

Yes, that would be my guess for intermittent startup too. It's a small cap that dries out easily (especially next to a biggish dropper resistor). If the UC3842 doesn't have a stable supply then it can fail to start switching. Sometimes playing a hairdryer over it can confirm the diagnosis.

[raphaeltube]

I would change the orange Electrolytic Capacitor next to the chip Power supply resistor. 

If this is leaking/high ESR  the chip may not always start.

Wow!
Thank you very much. That seems to be it.

I had 3 remaining that were still playing dead. The 1st one came right immediately when I replaced that cap. 2nd one, same.
The last one, I 1st tried just re-wetting/heating the joints on the old cap 1st, and nothing. Then replaced & bingo.

So now, I'll get to work replacing all of the 'working' ones too.

I'll post again if one of them dies again, but I suspect you have solved it for me.

Thanks again 

 -R

[perieanuo]

hi, change first the orange capacitor ('05' on bottom pcb??) who is placed not far from 3842 chip