Ultrasonic cleaner circuitry repair

[docroot]

I have a "Baxter" Ultrasonic cleaner Model ME11. I have seen elsewhere the guts for Metzler ultrasonic cleaners, and they are identical. I bought my cleaner on ebay and was using it to clean a motorcycle carb, and it quit half way thru. Investigation showed two 2A blown fuses which I replaced and they blew again instantly (incandescently) upon turning on the switch. The fuses appear to be on parallel circuitry for each of the two transducers, and I don't know that it was waoring on both when I got it, so the fact that 2 were blown does not mean the failure effected both sides in one event.

I cut off the transducers and tested no continuity in both directions with my analog multitester. Other than that, I test a bit of reverse continuity on one of the DII RL
225 diodes but not on the one in the parallel circuit, but I haven't isolated it so I don't know if the diode is at fault or something else.
I am attaching a photo of the circuitry and my drawing of it in the pdf below the photo (I recorded the shape, color, and markings of each component to avoid a mistake on my part in identification, but I do sort-of know what I am looking at and how it works). So far I have diagramed the part of the circuitry in the box on the photo.


View attachment CircuitDiagram.pdf

I hope to fix this and make it work again. I expect to complete the circuit diagram and then to begin un-soldering one end of each component for testing, but my testing is limited to my multimeter, so I don't know what to do with the transistor. I don't see any obvious burned parts (except the fuses).

Any wisdom would be greatly welcomed!

 

[OLIVE2222]

Hi,

The transducers can be piezo type (nor sure about) so continuity test can not be significant. Are both fuse part of a common circuitry or each transducers driver have is own fuse?

Being "stereo alike" you can maybe try to isolate the drivers to test them separately.

Olivier

 

[docroot]

Thanks Olivier
I have finished diagraming the circuit



View attachment CircuitDiagram2.pdf

I am surprised at the capacitors between the lines-in, but I looked and looked and that's how it's wired. The fuses are not common to anything so there must be two faults.

The only "parallel testing" I have done is on the DII RL 225 diodes with different results, but I haven't gone further (the diagramming has taken quite a bit of time), but that is my plan unless one of you folks on this forum who is more experienced than I (not hard to be more experienced than I) sends me in a better direction.

 

[OLIVE2222]

Hi,

Looks odd that both drivers fail exactly at the same time. Guessing that the fan still cooling them can be a main surge or a dry run.
Anyway you can isolate the diodes (the 225 and the 306's ones) and test them. Same for the transistors, the package should be the collector you can test them as diodes, see http://www9.dw-world.de/rtc/infotheque/semiconamps/semiconductor_amps1.html figure 1.1.2.2
The 2 capacitors // to the main are for filtering purpose (it's allowed when dedicated types).
All other parts looks to be rugged except the 2 grey" TTAM" capacitor who can tend to dry.
Also check that transistors packages are electrically isolated from the frame.

Olivier

 

[docroot]

Thank you! I will isolate the diodes, the TTAMs and the transistors for testing. How can I test a transistor with a multimeter? Do I assume the two wires coming out are the emmitter and collector, and the shell is the base? Should the emmitter amd collector have high or infinite impediance between them?
I do not know that they both failed at the same time. It could have been working on one circuit, and then that one quit... I don't have a baseline to compare its performance.

 

[OLIVE2222]

Hi,

You can test the transistor with the diode check position of a multimeter, like this:

http://www.allaboutcircuits.com/vol_3/chpt_4/3.html

We don't know if its a PNP or a NPN but it's the same method to test them. We can assume that the" shel"l is the collector, the 2 leaving pins are to be determined. Wathever the pinout is you CAN'T have the same indication when you swap the 2 multimeter probes for a working transistor!

Olivier

 

[docroot]

I have isolated the diodes and TTAMs, and they test OK... now for the transistors

 

[docroot]

The Transistors have continuity in all 6 directions
SJ9089A 9309
View attachment CircuitDiagram2.pdf

 

[(*steve*)]

The Transistors have continuity in all 6 directions

That sounds like a catastrophic failure.

You'll need to determine what the transistor was from (hopefully) the markings, or perhaps by investigation of the rest of the circuit.

Then you'll need to find out why they failed.

It is also possible that their failure may have damaged other parts of the device.

Are the markings on these parts readable?

 

[OLIVE2222]

Hi,

According to the OP schematic, transistors are SJ9089A made by Motorola.
i can't locate a datasheet but they are avialable here:

http://www.plccenter.co.uk/en-GB/Buy/MOTOROLA/SJ9089A?redirect=true

But indeed Steve better to find out why they failed however the rest of the the circuit involve a multiple windings (up?) transformer and the transducers. None of them will be easy to source. So replacing the transistors is probably the last chance to fix the item.

Olivier

 

[docroot]

The Transistor is a Motorola #SJ9089A, which is aparently obsolete. I find a recommendation for MJ16018, which is available on Ebay, and I have made an offer.

you'll need to find out why they failed

The transistor is bedded to the heatsink with thermal grease, so I know that heat is an issue, vibration is a fact of life for an ultrasonic cleaner, and the circuit is 20 years old (1993 on the fan). I have tested the 225 and 306 diodes, and the TTAMs, and they pass.
How should I proceed?

 

[docroot]

Hi,
I can't locate a datasheet but they are avialable here:

http://www.plccenter.co.uk/en-GB/Buy/MOTOROLA/SJ9089A?redirect=true

Olivier

I called the PLCCenter and they were gracious but only had one, used, and a $50 minimum. They are selling the same one on Ebay for $35, but I need 2 anyway, so I am more inclined to go for the (maybe) substitute of MJ16018 ($5 ea) and see if it works

 

[OLIVE2222]

Hi,

Yes you can go for an equivalent, be aware that some parts can be plain fake. On the other hand you can buy decent ultrasonic cleaner on Ebay for around 30$

http://www.ebay.com/itm/Mini-Ultras...039?pt=LH_DefaultDomain_0&hash=item4cfef979ef

They probably will not last as long as your but for a time to time usage it can be OK.

If you go for a transistor replacement take great care to clean the heat sink and place all the isolation parts with new thermal conductive grease! and yes they can burn again if the root cause is elsewhere

Olivier

 

[docroot]

Thanks Olivier
I need the size of this one, 11 liter is just big enough for a half carbset for an antique Goldwing GL1000, and that's what I do with this cleaner. I got this without a basket which concerned me until I saw that I wouldn't have been able to fit the carbs into the basket anyway, so I suspend it with sticks and string:


... and the water just comes up to the floor of the plenum:

 

[OLIVE2222]

waouw that size of carb, right small Ebay cleaner will not do it, or you need to bought a Goldwing GL50

 

[docroot]

LOL! Yes it's a lot of carb with yards of tiny passageways.
I got the MJ16018 Transistors but am not sure how to put them into this circuit as the ones being replaced have continuity in all 6 directions. These have continuity with a milivolt analog tester in only one direction

How should I assign the E B C, and what would make sense in this circuit?

I see the pins in the transistors are slightly off center, so in the absence of any other data I could hook them up the same based on mechanical dimensions... is there any risk in doing so or is it just the case of "try it, if it doesn't work, try it the other way?"

 

[(*steve*)]

Check the datasheets for both transistors (this is perhaps not the most useful answer).

Most likely the connections will be exactly the same. Yes, the leads are off-centre and this is used to allow identification of the leads. And as you are clearly aware, the case is one of the connections.

Aah, it's a BJT. I'm almost 100% sure that the connections are standard. If you do wire it up the wrong way, you *could* destroy the transistor.

I'd go with matchy-matchy.

 

[docroot]

OK, I am ready to do that. Please review my intentions and cross check it against my circuit drawing to see if it mekes sense
On the transistor, when the printing is "up", the offset of the pins is "up" too.
My photo of the MJ1618J (posted above) shows continuity when the red testlead is on the right and the black testlead is on the left.
My photo below appears to show the orange wire going to the right, the green wire going to the left, and the blue wire going to the screw/heatsink.

When I try to reconsile this with circuit diagrams I find on the internet like this one (http://donklipstein.com/samschem.htm):

R1 D1
H o------/\/\-------|>|----------+
1, 1/2 W EDA456 |
C1 D2 |
+----||----+----|>|-----+
| .1 uF | EDA456 | 2
| 200 V | +-----+---+ T1 +---+------->>------+
| R2 | _|_ C2 ):: o 4 | | |
+---/\/\---+ --- .8 uF D ):: +----+ | |
| 22K _|_ 200 V ): + |
| 1 W - 1 o ): ):: _|_
+-----------------+---------+ : O ):: L1 _x_ PT1
| R3 | 7 : ):: |
| +---/\/\---+ +-----+ : 5 + |
C \| | 10K, 1 W | F ):: +---+ | |
Q1 |--+-+--------------+ 6 o ):: | | |
E /| | D3 R4 +---+ +----+------->>------+
| +--|<|---/\/\--+ _|_
| 47, 1 W | --- Input: 115 VAC, 50/60 Hz
| | | Output: 460 VAC, pulsed 80 kHz
N o------+-------------------+---+

I endup fairly boggled as I am uncertain translating my "descriptor" circuit diagram (2 posts up) with this symbols diagram so the logic gets lost in the mental gymnastics for me.
Does what I intend make sense electrically? I do have an extra transistor so if I get it wrong I can try again, but all things being equal, I woudl like some guidance to get it right the first time
Thanks

 

[docroot]

IT WORKS! Yahoo!

So, in this curcuit the MJ1618J replaces the SJ9089A, wired in "matchy-matchy", and it has worked thru 6 cycles of 15 min each. I have read elsewhere that this substitution didn't work for more than one cycle, but that may be problems in the other-guy's circuit that I didn't have. I still don't know why mine failed in the first palce... I tried to run it in the sink before failure, and that may have limited cooling, or it may have been age, but they are working now.
I hope my circuit diagraming (such as it is) is a useful resource in the fullness of time for others, and I appreciate your support and encouragement!

 

[docroot]

Oh yes, here's the circuit diagram (descriptor style)