Ultrasonic cleaner repair

[drummerdimitri]

I have a cheap 50W ultrasonic cleaner that worked fine at first, then it would only start working after I smacked the stainless steel tank and now it wont even start anymore.

Not sure what could be the issue here and I have no clue if it is a power supply issue or a faulty transducer as it is a ceramic type and not an "industrial" one.

While it was working, I was able to get a 40 KHz, 320V sinewave output on my DMM but now it is dead and I have no way of generating such a signal with an audio amplifier or a variac.

How does one test such a transducer?

I am not sure how to proceed with troubleshooting the device so any help would be much appreciated.

[drummerdimitri]

I have replaced the transducer with a brand new one and it still isn't working.

I got it to work once when I banged on the unit to make is start oscillating but wasn't able to do it once more.

What else could be the problem? Any specific components I need to check?

[Alex_Baker]

Probably a long shot but I would check the power supply input. A while ago I got a 25W unit at a junk shop for cheap, but it didn't work. Turns out that the common-mode suppression choke (if you are familiar with SMPSs) was open on both coils.

I think the high frequency can damage components or cause solder joints to crack. Judging by the fact that you have to smack it I would put money on there being questionable solder joints in the power supply or the driver circuit.

[drummerdimitri]

Probably a long shot but I would check the power supply input. A while ago I got a 25W unit at a junk shop for cheap, but it didn't work. Turns out that the common-mode suppression choke (if you are familiar with SMPSs) was open on both coils.

I think the high frequency can damage components or cause solder joints to crack. Judging by the fact that you have to smack it I would put money on there being questionable solder joints in the power supply or the driver circuit.

Good point thanks for that.

Will first go over each solder joint with my soldering iron in case one or more of them have developed a high resistance contact with the PCB and if that doesn't work out, will try to probe around the power supply circuit to see if something is acting up.

[james_s]

If smacking it affected the operation then it is almost certainly a bad connection somewhere, probably a solder joint. Inspect the PCB closely and resolder anything suspect.

[drummerdimitri]

If smacking it affected the operation then it is almost certainly a bad connection somewhere, probably a solder joint. Inspect the PCB closely and resolder anything suspect.

It's definitely a mechanical issue however not a solder joint related issue and I went over each one and ever directly soldered the transducer's wires to the main PCB instead of having a plug type connection and the problem persists.

Since we've isolated that it is most probably a mechanical issue, the only moving parts I can see are two relays. My guess is at least one of them is toast so will be investigating that next.

[coppercone2]

it sounds like a crimp, these ultrasonic cleaners often get water spilled in them, then what happens is something corrodes because its high voltage electrolysis.

test every wire for continuity from the beginning to end (i.e. test header  on PCB 1 to header on PCB2)

also check for shorts. always test on the solder side, not the cable side

[james_s]

The relays are possible. I would also check any connectors, I've seen breaks inside the insulation and bad crimps that weren't making a good contact.

[drummerdimitri]

I've taken out the relays and tested them and both are working just fine.

Actually, when I try starting the machine, the relays click but most of the time nothing happens.

I measured a 20V potential across the coil side of the relays and I'm wondering if this is a bit low (should be 24V?) and maybe that's why the unit is not switching on?

Maybe it needs a "kick start" but it is not receiving either enough voltage or current.

[drummerdimitri]

So after further testing, it turns out that one of two 47 K  resistors was burnt and the other out of spec (55 k  ).

Didn't have any 47 k  resistors on hand so decided to replace both with 50 k resistors just to see if it solves the issue and guess what, problem solved.

Now the machine works again however after taking out my thermal imager, it was apparent why these resistors went bad after some time: After a couple of minutes of on time, the two 2.2  resistors adjacent to them were reaching into the 140 C 

Not sure why that's the case and also the power seemed to drop drastically as the power draw from the wall went down from 26W to around 17W in the same time span.

My guess is since these resistors are getting so hot, their resistance values is increasing causing a large voltage drop which in turn drops the output power down.

What is strange however is that I never noticed this kind of behavior prior to the unit failing. Should I just replace these with 5W models to reduce the heat output and power sag or is there an underlying problem that I am missing here? 

[Alex_Baker]

Do you have a schematic for the ultrasonic cleaner?

[drummerdimitri]

Do you have a schematic for the ultrasonic cleaner?

I do not unfortunately.

[Alex_Baker]

Hmmm, I might have some ideas if you had a schematic but without one I have no idea. The other ultrasonic cleaner I worked on had a driver board where the AC input is similar to a SMPS but after that they are a bit different...

sorry, 
what brand is your unit? mine is a "SharperTek", I may take mine apart and see if they are similar. Could you post a picture or two of the circuit board?

[coppercone2]

wire wound resistors can run really hot, when I first started doing thermal imager repairs I was suprised by whats in your average linear lab supply, usually a very hot resistor.