Heathkit IG-72 Repair

Discussion in 'General Electronics Chat' started by Georacer, May 10, 2014.

  1. Georacer

    Thread Starter Moderator

    Nov 25, 2009
    5,142
    1,266
    Hello everyone,

    While cleaning the electronics bench at the university lab the other day, I found this little fella beneath the dust and junk. It's a Heathkit IG-72 Audio Waveform generator. Basically a sine wave generator over audio frequencies, with variable amplitude.

    Heathkit IG-72 002 (Copy).jpg

    It's pretty vintage: the model must be around 1962, but I don't know how old is this particular unit. I couldn't find any date codes on any components.

    The thing is, it doesn't work too well, as you can see in the following pictures.

    Here, I have selected the 10Hz @ 3Vrms (for which amplitude I calibrated the oscillator with the internal pot). You can see the output is badly distorted.

    [​IMG]

    The same distortion follows through all the frequency range, as seen in this capture.

    [​IMG]

    Finally, the voltage noise is pretty bad as well. In the lower voltage scales, the waveform is indistinguishable from the noise. This waveform is generated with 100mVrms and the scale of the instrument goes as low as 0.003mVrms, with the same noise power.

    [​IMG]

    (Continued in the next post)
     
  2. Georacer

    Thread Starter Moderator

    Nov 25, 2009
    5,142
    1,266
    Naturally, I have opened the case. This is what I found inside:

    Heathkit IG-72 003 (Copy).jpg
    Heathkit IG-72 004 (Copy).jpg
    Heathkit IG-72 005 (Copy).jpg

    I have blown the dust off, cleaned most of the atmospheric pollution, poured dry contact cleaner to all moving parts and sanded the tube contacts.
    Nothing really helped.

    I have found an assembly manual, the best documentation I have found on this so far. I couldn't upload it, since it's a bit large (17MB), but here's a dropbox link from my account: https://dl.dropboxusercontent.com/u/8286935/IG-72%20Audio%20Generator.pdf
    I'm not sure if I'm legally allowed to re-post it, tell me if you think otherwise.

    I noticed that all tubes glow when the thing is on, but the incadescent lamp on the bottom doesn't. I'm not sure if this is normal or not.

    Any suggestions or thoughts? Larger quality (and size) images are availabe on request.

    Thanks in advance.
     
  3. n9103h

    New Member

    May 10, 2014
    2
    0
    I am restoring a HW-100 and SB-104... as soon as I get my feet on the ground I would be happy to help you find the schematic.
    George
     
  4. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    The lamp is used as a PTC thermistor in the negative feedback network that stabilises the oscillator amplitude - IIRC, it should glow a sort of orange/yellow maximum.

    If there's any waxed paper capacitors, they're fairly notorious for going leaky - especially high voltage ones.

    Electrolytics drying out in proximity to the heat from valves is a distinct possibility.
     
    Georacer likes this.
  5. #12

    Expert

    Nov 30, 2010
    16,246
    6,733
    Oh yeah. Those carbon resistors are old enough to drift high. Anything 100k and up is likely to be out of tolerance.

    Check your transformer voltage to see if the power line voltage increases across the last 30 years are going to make the 420 volts high enough to blow a 450 volt capacitor.
     
    Georacer likes this.
  6. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Definitely test the filament on that incandescent bulb.

    I'd replace the two orange electro caps, and test the big aluminium can dual electro the best you can. Are it's values marked?

    Then maybe swap out some tubes? Those old kits used common tubes if I remember right, like 12AX7, 12AU7 etc. They are all still available new (due to the surge in popularity of tube guitar amps).

    (edit) re the age it could be 1950's, it looks to use a tube rectifier and not semiconductor diodes.
     
    Georacer likes this.
  7. #12

    Expert

    Nov 30, 2010
    16,246
    6,733
    The schematic is in post #2.
     
  8. MrChips

    Moderator

    Oct 2, 2009
    12,405
    3,352
    If you have time to spare I guess you can attempt to restore it just for the personal satisfaction. Otherwise, what would be the incentive?
     
  9. #12

    Expert

    Nov 30, 2010
    16,246
    6,733
    Antiquing.

    I have some great old stuff that still works like new.
    A 1973 Radio Shack AM-FM radio with 7 watt outputs.
    A pair of 1977 Cerwin Vega H-15 speakers.
    A spring reverb in a wood cabinet.
    A 1976 Randall 12-2 mixer (for music on stage or recording).
    A 1938 Walker-Turner drill press. Cast iron!
     
  10. JohnInTX

    Moderator

    Jun 26, 2012
    2,340
    1,015
    Can't help much beyond what the others have said but here's a reduced-size version of the manual.
    Cheers.
     
    Georacer likes this.
  11. #12

    Expert

    Nov 30, 2010
    16,246
    6,733
    That looks just like the schematic in post #2, including the pencil marks somebody made on the assembly steps but it uses up a lot less megs of bytes.
     
  12. Georacer

    Thread Starter Moderator

    Nov 25, 2009
    5,142
    1,266
    It could actually be useful! There have been about half a dozen occasions in the last 3 years where I needed a signal generator to test audio circuits and I had to power up my PC and fire up a software one, paired with the sound card.

    Other than that, yes I find it pleasant to look at. And it could possibly be the oldest thing in my house right now.

    I'll go to the shops tomorrow and see what I can find.
     
  13. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,976
    3,677
    I think the distortion is because one of the two filter caps is bad (I didn't look at the schematic yet) the waveform just looks like the issue I had when I rebuilt a heathkit IG-102 signal generator (100k-100MHz).

    Also, the copper contacts on switches get "carboned up". It is not really carbon, it is copper oxide (Cu2O). The multi-position switches are especially painful unless you find a spray - I forget the name. Don't waste your time trying to do it mechanically - it doesn't really work unless you are willing to invest 10x time I was. The sprays work really well.
     
    Georacer likes this.
  14. Georacer

    Thread Starter Moderator

    Nov 25, 2009
    5,142
    1,266
    From what you guys say, it seems I have found myself a nice archeology project...
     
  15. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,976
    3,677
    Yes, it is. I learned a lot rebuilding mine. The troubleshooting section in the manual was great help to diagnose shorts/opens and align the dial. Not so helpful, obviously, for finding bad caps or bad switches caused by 60-years of aging.

    I found a heathkit audio tone generator but the wiring inside was completely crispy. All the cloth insulation was friable and falling off of the wires. The caps were all pealing open. The unit was stored in a chemical lab near a running vacuum pump. Ozone damage - I dumped it in the bin after pulling the tubes.
     
  16. Ragwire

    Member

    Dec 9, 2013
    36
    1
    I wouldn't even run it until those electrolytics have been replaced. Other than that, pretty much what the others said. 90% of what I do is restore vintage tube radios and test equipment as a hobby, and the basic procedures as follows fixes the vast majority of the units I work on:

    Step 1: Don't try to run it or even plug it in.

    Step 2: Remove the tubes and gently clean. I use isopropyl alchohol mostly.
    Deoxit on the switch contacts and tube sockets and pins...and general electronic cleaner on the potentiometers. Let everything dry out well.

    Step 3: Replace all electrolytic capacitors and wax/paper capacitors.

    Step 4: Replace any resistors that have drifted out of tolerance, or wiring with bad insulation.

    Step 5: Replace the power cord with a safe (usually grounding type) cord and check any fuses for proper ratings.

    Step 6: Then I replace the tubes, plug it in--with hands off--and test for unsafe leakage or voltage to ground.

    Step 7: If it's OK, I run through the pin voltage tests and attempt any calibration procedure. Any needed troubleshooting can be done at this point.

    This is just how I do it, and it may omit certain steps. Be especially careful with tube equipment because of the high voltages involved.
     
    Georacer likes this.
  17. MrChips

    Moderator

    Oct 2, 2009
    12,405
    3,352
    Too late, he's already done that.:D
     
  18. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    Not sure it applies so much to Heathkit gear, but Elektor does a Retronics page - in one issue the featured a famous HP audio generator, and stated; "if you have the model that came before this one - check your insurance cover now!".
     
  19. JohnInTX

    Moderator

    Jun 26, 2012
    2,340
    1,015
    Its Georacer's file. The full (paid) version of Acrobat will squeeze the .PDFs at the expense of compatibility with early versions of Reader. This one's good with 9.0+. Handy.
     
  20. ian field

    Distinguished Member

    Oct 27, 2012
    4,412
    782
    A quick glance through the assembly manual: the introductory part on the resistors mentions 5%, 10% and no band for 20% - its probably safe to assume nothing in the kit is tighter than 5%.

    In the parts list and assembly directions, there's no mention of a tolerance band - they might all have been 20% to start with.
     
Loading...