Tektronix FG502, trying to get it working

[liaw]

Am trying to get caught up on some old projects, and starting looking at an old FG502 signal generator.  I saw in years past that people had posted about FG502's and other Tek equipment on this forum, so figured was a good place to ask.  It was all messed up and would release magic smoke when hooked up.  I replaced a bunch of the tantalums with higher voltage electrolytics (it was what I had), removed an intermittent short, and got it so it at least runs with no smoke.  The waveforms are not really pretty, and it still has issues.  Had some odd experiences with the LM741's used in the regulator section (at least the schematic lists them as LM741C).  It was not regulating correctly, and seemed like one of the op amps was bad, so I tried swapping in a chinesium LM741CN, and that would immediately explode (literally had two of them shatter upon power up).  Ended up finding a short somewhere, put back the original op amps (the numbers on the IC's do not say 741, though the schematic does) and the regulators worked again.  So not sure if those are a special version of 741 with higher ratings or the schematic is wrong.

The voltages seem ok now, but the frequency range is off.  the max is around 75xx in the lower ranges, and when you jump to the 10x6 range the max instead of being the expected 7.5MHz to be in line with the lower ranges (or the correct 11MHz) is only 5.9MHz.  Also in a few of the waveform settings the oscillation can be unstable.  In the far CCW negative pulse setting, sometimes it works, and then while watching the oscillation will stop.  Flipping to other waveforms and you have oscillation again.

Sorry I am all over the place, am just trying to figure out where to start.  The service manual from what I can see just suggests "setting the max to 11`MHz", but no real suggestions if that does not work.

I guess start at the voltage from the variable frequency control resistor and start poking around to see what I see, but just not sure where to go.  I am definitely NOT an EE, but have dabbled in electronics for many years and actually worked as a technician at a small electronics company when in school.

Thanks,
Bob

[duak]

Bob, are you testing the '502 in a TM either with or without an extender?  Can you probe the various circuit nodes?   When repairing an FG503 It was easier for me to open it up on the bench and  just apply power to the appropriate points in the regulators.  I don't have an FG502 but from the schematics it appears to need only about +/-20 V.

Anyhoo, are you getting the +/-20 V & +/-17 V power voltages?  741 op-amps should be good for +/-18 V with gusts to 22 V.  Do you think there was any chance that the +/-20 V regulators could have put more than that out when the parts failed?  If so, some of the other 741s could have been damaged.

The frequency is controlled on the left side of page 1 of the schematic.  What is the voltage on U135 pin 6 when the frequency setting is X1? When it's X10?  Does the waveform symmetry change when the frequency setting is varied?

Cheers,

[liaw]

Well, I went to basement and took a few readings on this strange device.

I think the supply voltages are working at least good enough...
+19.87/-21.48
+16.94/-17.93

U135 pin 6 varies from 15v at frequency knob set to 1 then to 4.3v at frequency knob 11

The voltage on this IC at pin 7 is +19.8, and pin 4 is -5v (set by a zener)

U100 pin 6 varies from +1.1v at frequency knob 1 to +12v at frequency knob 11.
Voltage to U100 is the +/- 17v rails.

I am not sure what levels I should be seeing.

One other strange thing that may or may not be related - the frequency drops by about a factor of 6 when you switch from the square wave down to the sawtooth waveform.  This one may be a basket case.

Thanks,
Bob

[duak]

I wouldn't give up yet.

Last things first - when changing to ramp or pulse waveform it should reduce the frequency by 10.  This allows the waveform symmetry to be adjusted.  A factor of six instead of ten tells us something but what it means isn't yet clear.

I'm using a manual from 1980.  Page 4-5 section 2 gives the specs on the power supply voltages.  The -20 V looks to be too high.  Did you replace any parts connected to this bus?  Does it change when the frequency or any other control setting is changed?  Do you have a scope to see if there is any ripple or other AC voltage on this bus?

[liaw]

Well, I got it working a little better.  I can't get the supply voltages perfect.  There is about a .6v delta.  So I have the + at 19.7 the negative is -20.3.  That was as close as I could get it.  I am thinking that more of the remaining tantalums might be bad elsewhere, those never had much room for overvoltage.  I will have to dig through it more when I have time.  Not sure why I am even trying to rescue it, I have a working FG504 which is a lot better instrument, but I just hate seeing something sitting there not right.  It seems like they must have drastically changed these things as time went on, as there are parts tacked on in places, and places on the board for parts that are not there, but it looks like the board is for an earlier revision, and the mods on it at least appear almost factory and seem to match what it is supposed to be.

[THDplusN_bad]

Hi Bob,

that FG it is surely worth a repair!

I'd suggest to have a look at the service manual and schematics again. Per the "Theory of operation" on page 3-1, a triangle waveform is the basic signal. All the other waveforms are derived from this.
I'd go one step back, say check the triangle generator first, then have a look at the following stages.

With reference to the schematics diagram 1, probe the input signal at the neg. input pin 3 of U140 (µA741C) and the equivalent pin 3 of U175 (another µA741C).
A voltage within specs (+7 VDC to +17 VDC, and -7 VDC to -17 VDC) shall be present at these inputs. The front panel frequency dial shall result in changes of these voltages.
If this looks good, probe the outputs of each operational amplifier, and check the constant current sources formed by Q140 and Q175.
Then check the triangle waveform at the inputs of the Q200 FET and its output signal.

If everything fails, you might want to join https://groups.io/g/TekScopes and post your question in that group - as it has some very knowledgeable members incl. several former Tektronix engineers.
You will find a lot of enthusiasts familiar with the TM500/TM5000-series equipment there.

Cheers,

M.