Function Generator repair Q

[cthree]

I picked up an old Philips PM5132 0.1-2Mhz analog function generator recently but I'm having a some problems with it. It sort of works but it has a second noise signal modulated on top of the output signal. It happens on all signal types, all output voltages, all frequency ranges and frequencies. The only thing that has any effect is adding output attenuation which leads me to think it's something near the final output stage after the signal generation but before the attenuation stage.

The noise frequency is always a sine wave of 2.4Mhz ish which sometimes inverts in a glitchy fashion. I've attached a scope capture of the problem on a ~200kHz sine wave @ 3 Vpp. The noise amplitude appears to be proportional to the output voltage, between 300mVpp and 720mVpp. The TTL output doesn't have this problem so I think the main oscillator is fine. This signal is on the DC output as well.

My question is: does anyone know what the likely culprit might be? I'm not sure of what I'm looking for and I'm hoping someone might recognize the likely source of the issue right away.

Thanks.

[sync]

Get the service manual. Check the supply voltages first. Then check the signal from the output backwards to locate the origin of the oscillation (noise).

w2aew had a similar problem.

[cthree]

Thanks!

As soon as I posted I looked at the schematics and noticed a couple of things. First, the stage before the attenuators is an amplifier which was basically a bunch of transistors. There is no obvious source of oscillation. At the input to the output amplifier is the DC switch, just before the DC offset voltage pot. I figured that since the oscillation is on the DC signal and all the generated signals, maybe the oscillation is on one of the DC power rails right from the get go. That led me back to the main uA78GU and uA79GU voltage regulators. I scoped both the input and output of each and I think I've narrowed it to this point. I think the rest of the unit is fine.

I attached two more scope captures. The first is the +/-33V inputs to each regulator (yellow is uA78GU, blue is uA79GU). The second is the output of each regulator. I think my next move is to pull the 3300uF input cap on the uA79GU and test it unless someone else has a better idea. I'll watch that video too. Thanks to the link.

[tom66]

I would check if the op-amp/output driver is oscillating.  Perhaps its compensation is set wrong.

[sync]

The negative supply looks bad. Check the capacitors. Not only the 3300uF also the 1uF and 22uF. You can put a good one in parallel for testing.
Are they blue Philips axial types? I read they will go bad often.

[tom66]

Oscillation like that on the -output- of a regulator suggests more likely a problem with the -output- capacitor.

[cthree]

The negative supply looks bad. Check the capacitors. Not only the 3300uF also the 1uF and 22uF. You can put a good one in parallel for testing.
Are they blue Philips axial types? I read they will go bad often.

Yes, they are blue Philips axial type. The 3300uF input cap in the negative supply "seems" to be ok. My DMM says it's 3347uF. I'll keep checking the others.

[tom66]

ESR more important than actual capacitance. I'd replace the cap anyway as a precaution.

[sync]

A DMM is not a good capacitor tester. I have really bad caps (bulged) that measure it's rated capacitance with a DMM but have very bad values on my LCR bridge.
I don't thing it's the 3300uF one. More likely the 1uF or 22uF.

[cthree]

A DMM is not a good capacitor tester. I have really bad caps (bulged) that measure it's rated capacitance with a DMM but have very bad values on my LCR bridge.
I don't thing it's the 3300uF one. More likely the 1uF or 22uF.

I don't have a 22u. I'll try the 1u and hope I get lucky. I'm attaching the PS part of the schematic just for completeness.

[cthree]

The 1uF didn't change anything but replacing the 22uF with 2 10uF in parallel seemed to fix it! Until, they exploded! Tried again with 2 more, popped them too

Then I remembered I was working on the negative rail and had them in backwards

The fireworks were fun but I ended up with only SMD AL caps left so I had to be creative.

Problem solved! $35 well spent! I'm going to order all new caps from digikey and replace them all with high quality ones and this should be an awesome signal generator for years to come. sync, you were an awesome co-pilot, thanks so much for holding my hand. Thanks also to tom66.

I've attached the wrap-up including the DC output signal, the fixed 3Vpp sine wave and a pic of my DIY 22uF capacitor hack.

[tom66]

Nice one!