Hameg HM208 Oscilloscope Timebase fault

[Philips210]

Hi Roger.

I've retaken some of the pictures to help clarify things.

Pic (k) is the 1kHz Calibration applied to Ch1 and Ch2, timebase set to 0.2ms/cm. Both Dual and Add/Chop buttons in. I can't explain why I quoted the timebase setting as 2us/cm in Pic (i), post #17, sorry for any confusion.

Pic (l) is the 1MHz Calibration signal applied to Ch1 and Ch2, timebase set to 0.2us/cm. Both Dual and Add/Chop buttons in.

Pic (m) is the 1MHz Calibration signal applied to Ch1 and Ch2, timebase set to 0.5us/cm. This time only the Dual button in.

Pin (n) as Pic (m) but with increased brightness. Notice how the rest of the trace starts to appear.

Pic (o) Again with the 1MHz Calibration signal applied to CH1 and Ch2, timebase set to X-Y mode. Both Dual and Add/Chop buttons in.

I carefully tried to adjust VR901 and return it to it's original position but it hasn't changed things. I also reseated the opto isolator IC901 but with no change. I suppose it's feasible this could be defective and worth trying a replacement

I did some dc checks on the X plates when the timebase was set to the X-Y mode with no signal inputs to Ch1 and Ch2. There was a swing of approximately 0 to 50V wrt to chassis on one X plate according to the position of the X shift control. The opposite voltage conditions applied to the other X plate. With the dot in the centre of the screen, the voltage wrt to chassis was about 33V on both X plates. It therefore seems this part of the CRT is actually OK.

In Pic (n) it's interesting to see more of the trace appearing as the brightness is increased which seems to point to a blanking problem.
I hope the pics help to give a few more clues as to the source of the problem.

Regards
Symon.

[Philips210]

Hi.

I unplugged the opto coupler IC901 (CQY80N) from the Z Board. This optocoupler is a little different to the types I've previously encountered in that it has a base connection to the opto coupled transistor. Checking the forward volt drop of the LED reveals a reading of 0.995V (on diode test) as indicated on my Fluke 23 Multimeter. Checking the transistor, there appears to an open circuit of the base/emitter and base/collector junctions. I tried biasing the LED and checking with another meter across the collector/emitter but there doesn't seem to be any perceivable conduction across the ce junction. It certainly seems suspect and I will order a replacement. Attached is the data sheet for the CQY80N.

Regards
Symon.

EDIT: I wonder if a 4N35 would be a suitable replacement to the CQY80N?

[RogerEvans]

Symon,

Congratulations, it looks like you have the culprit. The schematics in the manual show C905, C906 and C907 couple the high frequency components of the blanking waveform into the high voltage circuitry but the low frequency and DC rely on coupling through the opto-isolator, hence the unblanking works for a few microseconds and then fades out. CQY80N are available on eBay for a couple of pounds with one day delivery, you can pay a little more and get what look like genuine Vishay parts. On the other hand the 4N35 looks like it would be a perfectly good substitute.

There are fake parts around so it is worth taking a little trouble to buy from someone who should be reiable.

Roger

[Philips210]

Hi Roger.

Your explanation regarding HF coupling via C905, C906 and C907 and the LF coupling via the opto-coupler certainly fits the logic of this fault.

I used to be in the repair business and I had a hunch the fault was going to be on the Z board where there are all those high voltages present. With CRT TVs, a substantial number of faults were usually in the line output stages and power supply due to the HV stresses. I had done some cold transistor/diode junction tests and couldn't find anything amiss. I had also checked for any increased value resistors but couldn't find anything wrong.
I must admit to overlooking the opto-coupler, and your suggestion of checking it's connections via it's socket pointed me in the right direction.

I agree about the worry of fake components and in the case of an opto-coupler, a critical safety component, then it's best to buy from a dependable supplier. I'm going to buy some 4N35s from CPC and when the replacement is fitted, I'll hopefully be in a position to report back with some good news on this fault.

Thanks again for all your help, I've definitely learnt a fair bit more on the blanking circuits in oscilloscopes.

Regards
Symon.

[Philips210]

Hi.

Well I am extremely pleased to report that I've sorted out the oscilloscope which was due to the opto-coupler IC901, type CQY80N. I was quite baffled by this fault and appreciate Roger's assistance on the matter. I was going to give up and place it on one side but now I am really happy with the outcome.

Incidentally, the Hameg HM604 uses the same opto-coupler arrangement. I discovered this as I have a HM604 to look at as 'another project'. I borrowed the opto-coupler from there to test it in the HM208. I first did some cold tests on it proving the LED and transistor junctions were OK. I have also just received some replacements from ebay and tested those as well. They all work well, and the trace is now back to normal.

I had wondered about changing some of the small electrolytic capacitors while I have the scope apart or may be not for now.

Attached is part of the circuit diagram of the Z Board together with the culprit!

Thanks again for all members' help.

Regards
Symon.