Hi,
I'm new here, I've been researching a specific old switching power supply, it's from an old audio amplifier.
I think this would be the correct subforum for posting this, since even though it's an amplifier, I've tracked the issue down to the power supply.

This is the schematic for the particular power supply, I've just removed all the upper portion of it, since it's not present on my particular piece of gear.
The supply should be able to provide 42.5v at the output, however currently it kind of "creeps up to" that voltage, only if i leave the output floating. On any kind of load it starts to go down. In particular what's happening is that I can use the amplifier all the way up to about 60W, when I apply more load (turn the volume up even higher) the voltage goes so low that the protection circuitry of the amp kicks in and shuts it down until the voltage can creep up again to a sufficient level.

From what I understand, Q203 is there to "start" the switching, and Q201 and Q202 are the actual switching transistors, this is such an old design that Q201 and Q202 are 2SC2335, as in, they are not mosfets.
I tried testing most of the components in the board, to see if the meet specifications, all capacitors have the correct capacitance, the electrolytic ones that had high ESR were changed along with the pair if it was required. Resistors were also tested and came up with correct values. And I've changed the 2 switching transistors with NOS pieces, and Q203 for a new KSC1815. However I've seen no changes.

The only thing I find that is "incorrect" is that if I probe Q201 for a frequency (between emitter and collector) I see double what I should be seeing (as in, instead of 45uS I see 22uS).

I'll also leave the explanation of how this circuit is supposed to work that is present on the service manual.
Any kind of help is really appreciated, I'm honestly at a loss for how to proceed here, from what I read in the service manual the frequency is established by the windings in the small transformer (T203), and I also don't see any kind of "feedback", so I wouldn't know where or how to affect the voltage/current that the power supply generates.

Thanks in advance!
Juan

 

What is the supply voltage for the oscillator, where do pins 32&33,34 go.

What is the model number of the amplifier, ?

 

This appears to be some sort of blocking oscillator, which were much in vogue prior to widespread SMPS control ICs.

Do you have access to a scope to see the collector waveforms? You may require a diff probe if the raw DC voltage comes straight from the mains.

 

Hi,
check one more time C218 , 0.27uF, low value resistors R215 (off board ) ,R 214,213,212 (on board), winding continuity T203, T204 ( on board) and trigger diode D205.
Incoming voltage 320VDC is OK?

 

Hi! thanks for the quick replies!!

What is the supply voltage for the oscillator, where do pins 32&33,34 go.

What is the model number of the amplifier, ?

The amplifier in question is a Technics SE-C01, the service manual is this (https://elektrotanya.com/technics_se-c01_sh-c01_sm.pdf/download.html), although my version is a US specific version, where both the power supply (SH-C01) and the amplifier live on the same casing, so there's some differences between both, namely, mine is a 120v only amplifier.
The supply voltage is 160 +/- VDC and comes through 35/36/37, as for 32,33 and 34, that's one difference between both versions as mentioned, they are to support several voltages, in my scenario T204 doesn't have a center tap on the second coil, so you can imagine 33 and 32 as being bridged to 34.

This appears to be some sort of blocking oscillator, which were much in vogue prior to widespread SMPS control ICs.

Do you have access to a scope to see the collector waveforms? You may require a diff probe if the raw DC voltage comes straight from the mains.

I do have an oscilloscope, but I don't have any way to isolate mains from the amp, since the input (pins 35,36 and 37 are not isolated from mains), at least that's what I think, what's happening (and sorry for my ignorance) is that when I touch the ground of the oscilloscope to the collector I see a spark, as if current were flowing, not sure if shorting something or if the ground probe of the oscilloscope is sending the current from the collector over to my house's ground.

Hi,
check one more time C218 , 0.27uF, low value resistors R215 (off board ) ,R 214,213,212 (on board), winding continuity T203, T204 ( on board) and trigger diode D205.
Incoming voltage 320VDC is OK?

Will do, T203 and T204 I do remember I checked them off the board.
As for incoming voltage I see 336v between V+ and V- (pin 35 and 37), If I measure them separately between 36 and either 35 and 37 I see about 168v each. I also measured the mains I'm connecting to the amplifier and it's about 124v AC, so I'm guessing the 8v difference I'm seeing is related to that, would that be correct?

 

Hi,
check one more time C218 , 0.27uF, low value resistors R215 (off board ) ,R 214,213,212 (on board), winding continuity T203, T204 ( on board) and trigger diode D205.
Incoming voltage 320VDC is OK?

Alright, just measured these:
C218 measures at 265nF (so 0.265uF)
R215 measures at 2ohms (rather low)
R213 and R214 measure at 1ohm
R212 measures at 10.15ohm
Transformers have continuity between their coils correctly, as in, almost 0ohm between both legs of a coil.
As far a trigger diode D205, forward voltage drop looks to be 2.7v according to my MM, which sounds pretty high, although I might be testing it wrong, since it's not a normal diode but something that I gather is used as a trigger.

 

Could it be that the trigger diode has issues?

Most likely the device is a Diac, which unfortunately cannot be easily tested, at least not with a multimeter.

 

Apparently this strange power supply has already given grief. Here is another thread from a few years ago, and how he solved the problem.

https://forum.allaboutcircuits.com/...ansistor-be-substituted-for-a-2sc2335.173010/

Unfortunately, the SVDN413 is pretty much unobtanium. If you are feeling adventurous, you could try to replace it with a DB3 Diac which is still available.

 

Thanks for the link! I had seen that one as someone had previously changed the 2SC2335 transistors for MJE13005 ones, initially I started looking for original 2SC2335 ones, since I thought maybe the MJE13005 ones were too "quick" in comparison. But it seems it's something else.
I ordered some DB3 Diacs, to replace the one in the board and see if it makes any difference.
Thanks for the replies!

 

What a weird component SVDN413 ? I can't even found a datasheet of it .

As about the DB3- you can take it from a faulty CFL lamp

 

Alright, thanks Pyrex and schmitt trigger for the diac suggestion, tested it with a variable power supply, and it was acting very high, about 38v, today got some DB3 diodes, tested that one and it was triggering at about 30v, much better.
Soldered it but no dice, same result unfortunately
After reading about Q203 on the other post, tried the power supply without it being in place and same result.
I'm running out of ideas, besides the transformers itself being somehow bad.

BTW, interesting thing, the SVDN413 diode is a black DB3 diode... or it was already swapped before lol

 

Could the high frequency I'm seeing (45khz instead of the ~22khz the service manual mentions) be the reason for the lack of response to a load the transformer has?

 

T203 determines switching frequency. The ferrite core is not cracked?
And a note about low value resistors checking. I had some repairs when resistors (below 50 Ohm) shows good values when checked with my DMM, unfortunatelly, when sufficiently loaded, increased in value a dozen times. Those resistors were in series to the Mosfet gates. The MOsfets were found burned out, shorted internally, so the resistors were affected by high voltage. After that I checked them with my Laboratory power supply, the appropriate voltage was applied and the current was measured. Several "good" resistors were found defective

 

Just tried with a power supply testing the following resistors at about 30v (in series with another big one to limit current)
R215 - 1.7 ohm (should be 2.7)
R213 - 1.0 ohm (should be 1.0)
R214 - 1.0 ohm (should be 1.0)
R212 - 10.2 ohm (should be 10)

I checked another couple things, first checked T203, I don't see it cracked, but not 100% sure since it has the plastic casing.
Also on T203 I checked the direction of the windings and it looks ok.

Talking about transformers, I did disassemble T204 in the past, since it was showing a coil open, turned out it was a capacitor layered on the transformer... thanks Technics for doing weird stuff! The cap joined the "input" and the "output" of the transformer, as in the coil on the left to the one on the right of the diagram.
The transformer is a ferrite core one, I know that by disassembling it and assembling it I probably lost some efficiency. Could this be what's causing the issue? (if it is, I don't think it can be fixed though)

As R215 is a bit low, tried placing some 0.3ohm 5w resistors I had in series, to get it to 2.7ohms, but still, same issue.
Totally baffled!

Thanks for all the answers I think this thing is jinxed haha

 

Just learned that I can use the second channel of my oscilloscope to see waveforms using math (ch1 - ch2 or viceversa)
Looked at what schmitt trigger had asked, and it looks like the waveform across Q201 is as shown except for the frequency (seeing the square wave at about 320v pk-pk)

 

 

So, resistors are OK.
I don't think transformer disassembling and assembling impacts transformer's efficiency very much.

T204 was disassembled in the past because of an broken winding... I suspect there's more problems with T203 or T204- an interwinding fault.
It's not easy to check this type of fault. If you have an electronic transformer for halogen lamps, 230VAC/12VAC, you can power up T204 to the secondary winding. So, T204 is to be tested as a step-up transformer. Put a big power resistor to the primary and see output voltage and form with oscilloscope. In case of an interwinding fault you will see short needles on the secondary winding instead of pulses of same shape as on the primary.
Those electronic transformers do not work if not loaded properly, so put a 12V lamp or a big power resistor to the output. Usually a 20W lamp is enough for 70W transformer

 

Your post/ picture #16.
Is the transformer loaded?

 

So, resistors are OK.
I don't think transformer disassembling and assembling impacts transformer's efficiency very much.

T204 was disassembled in the past because of an broken winding... I suspect there's more problems with T203 or T204- an interwinding fault.
It's not easy to check this type of fault. If you have an electronic transformer for halogen lamps, 230VAC/12VAC, you can power up T204 to the secondary winding. So, T204 is to be tested as a step-up transformer. Put a big power resistor to the primary and see output voltage and form with oscilloscope. In case of an interwinding fault you will see short needles on the secondary winding instead of pulses of same shape as on the primary.
Those electronic transformers do not work if not loaded properly, so put a 12V lamp or a big power resistor to the output. Usually a 20W lamp is enough for 70W transformer

How much load is needed? I have a 12v AC transformer handy, but I'm almost certain it's not good for a lot of power.

Your post/ picture #16.
Is the transformer loaded?

Yes, it's hooked up to the amp (even though it's not hooked up to speakers and input is floating, i'd expect some load)

Thanks for the answers Pyrex!

 

One more picture needed . T204 primary winding, secondary and feedback (related to R215 ) windings.
More load needed- speakers and steady input, say 1kHz at the amplifier input