exxos's Atari, Amiga & retro forum

Hi there,

I'm trying to troubleshoot an SR98 PSU.
It failed at some point with no other symptom than not to power up the Atari.
When measuring its output it only gives 3V/1V instead of 12V/5V.
I checked all the caps, resistors, transistors, transformer and diodes. When measuring the secondary transformer's voltage on my oscilloscope I get a very ugly signal.


I'm almost sure the regulation circuit on the secondary side is okay : I tested it by disconnecting mains, and applying 5V to the output and measured the voltage on the optocoupler's led. The current draw is 100mA at 5V.

Fortunately I also own a working device, and when I do the same measurement I get a nice output.


At this point I think the problem lies in the oscillator circuit, highlighted in light orange on the diagram.


However, that's also the only part of the circuit that I don't understand :-)
So if anyone here has an idea what to try next, I'd be delighted. I already swapped out many components, replaced all EL caps, the full bridge rectifier, …

Pretty desperate actually ha ha

Hi again,

I managed to measure the frequency in the lower transistor coil at the primary side on the working device : 130kHz (37% duty cycle).
On the faulty device, I only get between 10 and 20 kHz.

My question is : what elements in the circuit create that high frequency, and what could be causing its low value ?

The frequency isn't likely going to be consistent. It is the feedback loop via the optocoupler which sets the max frequency. This will depend much on how much current is being drawn. Cheap optos are generally slow. I forget the maximum frequency in general, but I wouldn't really expect a frequency to be higher than about 20KHz on these old PSU's.

Also looking at your scope it is stamped at 10MHz. So I don't see how you could even measure high frequencies. I would assume it would struggle with anything above like 1MHz anyway.

Hi,

Thanks for your quick reply.
As I said the good PSU seems to be generating 130kHz as measured by my DMM.

faxmodem wrote: 14 Nov 2023 19:03 As I said the good PSU seems to be generating 130kHz as measured by my DMM.

Is that with or without a load ? Without a load the PSU will enter overvoltage protection and turn off / on very fast. Without a load on a working PSU I would expect to see a frequency drop.

Well, there's R205 and R204 acting as load so I didn't bother adding a power resistor in parallel to those

faxmodem wrote: 14 Nov 2023 20:10 Well, there's R205 and R204 acting as load so I didn't bother adding a power resistor in parallel to those

Those resistors are there to stop the power supply from exploding. They do not give a adequate load by any means. You need something like 1amp load on the 5V to test things properly. Otherwise the power supply will be constantly turning on and off due to overvoltage protection.

I just tried with a 6 Ohm power resistor on the 5V rail on the good PSU I get a frequency reading of 84kHz, 40% duty cycle, 0.4V (AC)