[SOLVED] Help with Ultrasonic Cleaner repair.

[solarbot]

Hi all,

Wondering if anyone can give me a few pointers on how to track down the fault on a nearly new ultrasonic cleaner?  The failure was caused when the transducer suddenly detached from the basin - the spot weld had failed and left a small hole through which water got onto the PCB.  There was a pop and unit display went out.  I found the PCB fuse had gone (marked red) and 5 diodes (marked cyan).  I've replaced these and everything including the heater works again but when I switch on the ultrasonic side the fuse and diodes 'pop' again.  I'm not sure where to look next but keen to repair as it is basically new but the seller is of course unresponsive.

Any thoughts much appreciated.

[dacman]

Check U1, Q1, and Q2 and make sure the transformers are dried out.

[SilverSolder]

Quite apart from replacing the broken parts - was the water that spilled on the board clean?  (detergents, anything at all in it)?

Even if plain tap water has splashed on a PCB,  it is a good idea to clean it off with distilled water, to get any impurities that may have been in the water out of the way.  On this particular board, it looks reasonable to clean it with IPA after the distilled water, to aid drying.

Finally a treatment with a heat gun on low, to properly dry the board out, is a good idea.

[solarbot]

@dacman - thanks, I guess an easy way to test would be to take them out of circuit and put them in a component tester?
@SilverSolder - thanks, I will do that as the 'water' had degreaser in it.

[solarbot]

Hi,

I've now tested the 5V regulator U1 as good, the NPN Power Transistor Q1 when measured with a multimeter in dioide mode returns a voltage of 0.0015V between the collector (BLACK) and base (RED) - same for the Emitter.  Resistance between the pins is about 0.1 ohms.  I guess this confirms they are blown?

How could I go about testing the transformers?

Thanks.

[Armadillo]

For your consideration;
Off resonant, may goes into avalanche [high enough current]. Furthermore the transducer must be loaded and cannot be run empty [without load]
So for electronic troubleshooting, hook a large enough resistor to stimulate for troubleshooting purpose until you are able to weld back the transducer and loaded it with water for tuning/commissioning.

[Armadillo]

As for the transformer, you can ring test it [Use a oscilloscope, pulse a short fast edge current through, and counts the rings.... 10 or more].

The Xfrm is vanished, so unlikely the problem. Norm air drying will do. But you can try passing a small current through it overnight for faster drying if you want.

[solarbot]

Thanks Armadillo, I've replaced the transducer and bolted it through the basin so hopefully fixed - means I can test with a 'real' load.

I will try a ring test for the transformer, never done that before so will learn something new - many thanks.  The failure was a couple of months ago now so I expect the transformer to be dry.

[Armadillo]

Hi, it will be wise to cater for "No oscillation" situation and transistor "HARD ON" across the transformer.

I would first test it on variac and in series with a light bulb or isolated xfrm -> variac -> series light bulb so that I can oscilloscope it to make sure oscillation and frequency is about correct.

Just my 2 cents.

[do read the instruction manual on the correct level of water to fill as optimum load.]

[solarbot]

Hi Folks, I'm still trying to get this working!..... my current worry is that the RED power supply wire is connected to the PCB pin labelled N and the black one to L which is backwards to my logic.  The Black wire traces through to 3 pins on the transformer and the 2 relays and the Red wire traces through to the other 3 wires on the transformer.  May I confirm that the Black wire, connected to the relays, is therefore the LIVE and the RED is NEUTRAL?  Thanks.

[solarbot]

Hello again,

May I ask if my previous question was a bit daft :-)   ?

As we are dealing with AC does it matter which way round Live and Neutral are wired to the PCB?

Thanks.

[LateLesley]

It's not as daft a question as you might think. The neutral is tied to earth potential, whereas live carries the AC voltage in respect to earth. If you have a circuit, with a fuse in the live side, if the mains is wired back to front, and the fuse is blown, on what would now be the neutral side, then potentially all the rest of the circuit could be a live potential, and in a very dangerous state.

Of course, it can be argued that any part of the circuit would be dangerous anyway when working with these circuits, so it may be a moot point, but it could give rise to a situation, as a service technician, that there's live voltages where you don't expect them.

But then, you would be up on safe practices of working on power supplies like this, and how to turn off, and discharge everything, and test as much as possible with the power off, and how not to touch anything metal when probing with the power on.

[solarbot]

Thanks Lesley,

Good to know I'm not 100% daft :-)  Now I just need to figure out if the PCB label is wrong and reverse the wiring so Live runs to the 13A plug fuse (UK) - any idea how I can confirm how this?  Currently the RED mains wire has continuity with the PCB fuse but the PCB wire label says Neutral.

[Nusa]

Ignore labels and colors to determine which is line and which is neutral. You never know if they're actually right. Test it yourself from the prongs on the wall plug (I presume you know which is which for your country) to see which goes where.

[solarbot]

Thanks Nusa, I know which is Live until it gets to the PCB, just trying to confirm how I can check which terminal is Live on the PCB if I disregard the labels?  The currenlty labelled LIVE doesn't go to the fuse or relays, just the transformer which seems wrong to me?

[LateLesley]

I think if the live is on the fuse side, then leave it as it is, no matter how it's marked. There's been a lot of cases in Chinese design with the fuse on the neutral, and they've maybe later swapped the connections. The reality is it doesn't matter hugely, many pieces of equipment run off ac, which swings positive AND negative in respect to earth, and many have dual rails inside once rectified and smoothed which are +ve or -ve with respect to gnd.

It's just because we reference things to ground that we think it matters, what really matters is the potential difference. At high voltages, your body wont care if its touching +240V or -240V, it just knows that there's a potential difference causing current flow through you that will hurt and possibly kill. We reference everything to ground, because invariably we are standing on it when we touch something electrical, and we want to know if it'll let current flow through us.

It's why some of the cheap chargers can be so dangerous. A lot of times, the output isn't referenced to earth, and even through the two contacts inside might only have 5V difference (the potential difference), they may be sitting at 100V and 105V when measured between each contact and ground. So you can measure a small difference between two points in a circuit, but that does not mean it has a small voltage compared to ground, and isn't dangerous.

As always, when working with equipment with live mains, ALWAYS treat EVERY part of the circuit as if it is live, and could kill you. It'll keep you safe. It's also why isolation transformers are recommended when working on this stuff. It can stop you getting a full on zap. (unless you still go across contacts of both sides of the transformer.)

[solarbot]

Many thanks Lesley et al for all your input, I can now confirm that the PCB is labelled incorectly and the fixes were successful so back up and running :-)  FAB - have a good weekend.