Author Topic: analog panel meter - repair possible?  (Read 6084 times)

0 Members and 1 Guest are viewing this topic.

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
analog panel meter - repair possible?
« on: December 19, 2015, 03:44:52 am »
Continuing my repair of an HP 6115A power supply, I think I've managed to bolox up the analog meter in this thing. It responds to changes, but it seems to have a non-linear response now, and so if I calibrate it for the full-scale value of 100V, then when I set the Power Supply to 20V, the meter is reading high (about 30V). It was off when I got it, but something I did during the repair made it much worse (possibly the time I powered things on forgetting to plug back in the control board first, leading to the meter seeing an overvoltage condition).

I've looked at the circuitry surrounding the meter, and it just boils down to a resistor in series with it, so I don't see that being the source of the problem. To confirm, I removed the meter from the PS and connected it to a programmable current source. I was able to confirm the non-linear behavior that I see when connected to the PS.

Is it possible to repair this?
« Last Edit: December 19, 2015, 07:13:21 am by motocoder »
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 7645
  • Country: de
Re: analog panel meter - repair possible?
« Reply #1 on: December 19, 2015, 12:48:13 pm »
Sometimes electrostatic change accumulates on the inside of the meter. One indication of such problem would be neadle movement if something conductive is put on or close to the front. One might get rid of such charge by putting the meter in a really humid place (e.g. close container with some water) for quite some time (e.g. 1 day or so), usually followed by getting it dry again.

Otherwise repair on such analog meters are tricky. It could also be some dirt (especially if magnetic).
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: analog panel meter - repair possible?
« Reply #2 on: December 19, 2015, 04:18:59 pm »
I plotted out the response this morning, using an HP 6177C DC current source to drive the meter, and an Agilent 34401A multimeter to measure the current. The response does not look that bad - certainly not the huge errors I was seeing when the meter was in-circuit. So either it was some electrostatic charge as you mentioned, which is dissipating slowly, or some issue with the components (resistors only) surrounding the meter in circuit.

Here is a graph of the measurements I mention above:

 

Offline voltz

  • Frequent Contributor
  • **
  • Posts: 267
  • Country: gb
Re: analog panel meter - repair possible?
« Reply #3 on: December 19, 2015, 06:33:58 pm »
That response curve looks good enough. You can get anti-static aerosol sprays to get rid of static problems. "anti-stat" and others.
If thats not the problem then it would suggest a problem in the electronics driving it. Maybe not enough current?
 

Offline Fank1

  • Regular Contributor
  • *
  • Posts: 159
Re: analog panel meter - repair possible?
« Reply #4 on: December 19, 2015, 06:41:18 pm »
Wipe it down with a fabric softener sheet, works as well as anything I have tried.
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: analog panel meter - repair possible?
« Reply #5 on: December 19, 2015, 06:54:11 pm »
That response curve looks good enough. You can get anti-static aerosol sprays to get rid of static problems. "anti-stat" and others.
If thats not the problem then it would suggest a problem in the electronics driving it. Maybe not enough current?

Yes, I think the current limiting resistor needs to be a slightly lower value. I guess the meter has become a little less sensitive for some reason. It is quite old. Unfortunately this is the hardest resistor on the PCB to reach with the iron - right in-between the two giant caps on the back of the board.

Wipe it down with a fabric softener sheet, works as well as anything I have tried.

I will also give it a wipe down with a dryer sheet soaked in staticide.
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 7645
  • Country: de
Re: analog panel meter - repair possible?
« Reply #6 on: December 19, 2015, 08:17:25 pm »
Usually the nasty static charge is not at the outside, but at the inside of the transparent plastics.

I have an old analog meter that had so much static charge, the needle got nearly stuck at about 25%, making the lower half practically useless.  The charge was stable for more than 2 years - a few drops of  water inside solved the problem.
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: analog panel meter - repair possible?
« Reply #7 on: December 19, 2015, 10:40:16 pm »
Usually the nasty static charge is not at the outside, but at the inside of the transparent plastics.

I have an old analog meter that had so much static charge, the needle got nearly stuck at about 25%, making the lower half practically useless.  The charge was stable for more than 2 years - a few drops of  water inside solved the problem.

You are 100% correct. I removed the front plastic bezel, and sprayed the inside with some distilled water. Shook the water off, reassembled, and retested. Now the meter is behaving 100% correct, and works fine with the original series resistor value (118K).

Thanks for your help.
 

Offline jh15

  • Frequent Contributor
  • **
  • Posts: 485
  • Country: us
Re: analog panel meter - repair possible?
« Reply #8 on: December 21, 2015, 04:23:16 pm »
I had a call at a radio station where the VU meters had suddenly become wonky.
Turned out someone must have recently polished up the board. Just breathed on them and they were OK for the training session.

Think I learned of this problem first hand with my Triplett 630 VOM first time I cleaned it. (1960's, way before ESD solutions were around).

And later working on legacy aircraft d'arsonval movements
tek 575 curve tracer top shape, 535 top shape, 465. 545 hickok clone, Telsa Model S,  Ohio Scientific c24P single board computer, many c-64 from my club days, Giant electric bicycle, Rigol stuff, Heathkit AR-15 receivers 2, Heathkit et 3400a trainer and interface,
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: analog panel meter - repair possible?
« Reply #9 on: December 21, 2015, 06:01:25 pm »
It's amazing how little information there is on the Internet about these devices. I always search the web and try and figure it out for myself before posting here. In this case, I couldn't find anything on the topic.
 

Offline jh15

  • Frequent Contributor
  • **
  • Posts: 485
  • Country: us
Re: analog panel meter - repair possible?
« Reply #10 on: December 22, 2015, 04:04:41 am »
Hi motocoder,

Until I was working on the aircraft instruments, I never really knew what a 'jewel' was in watch movements and the like.
tek 575 curve tracer top shape, 535 top shape, 465. 545 hickok clone, Telsa Model S,  Ohio Scientific c24P single board computer, many c-64 from my club days, Giant electric bicycle, Rigol stuff, Heathkit AR-15 receivers 2, Heathkit et 3400a trainer and interface,
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 10099
  • Country: au
Re: analog panel meter - repair possible?
« Reply #11 on: December 22, 2015, 04:12:00 am »
I had a call at a radio station where the VU meters had suddenly become wonky.
Turned out someone must have recently polished up the board. Just breathed on them and they were OK for the training session.

I so want to have the opportunity to be able to do that one day!  Merlin eat your heart out.   :-DD
 

Offline jh15

  • Frequent Contributor
  • **
  • Posts: 485
  • Country: us
Re: analog panel meter - repair possible?
« Reply #12 on: December 22, 2015, 04:22:00 am »
It's OK, was under contract, not an emergency call.
tek 575 curve tracer top shape, 535 top shape, 465. 545 hickok clone, Telsa Model S,  Ohio Scientific c24P single board computer, many c-64 from my club days, Giant electric bicycle, Rigol stuff, Heathkit AR-15 receivers 2, Heathkit et 3400a trainer and interface,
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: analog panel meter - repair possible?
« Reply #13 on: December 22, 2015, 04:42:34 am »
Hi motocoder,

Until I was working on the aircraft instruments, I never really knew what a 'jewel' was in watch movements and the like.

I guess digital hasn't completely replaced the analog gauges in aircraft yet?
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: analog panel meter - repair possible?
« Reply #14 on: December 22, 2015, 05:00:54 am »
I ran into this with the movement in my RCA WV77E VTVM.  Handily the original manual explains the issue and how to fix it.  In the 50s, the fix was to pour dish soap inside the lens, swish it around to coat everything, dump out the excess, and let it dry.  It worked in 2013 too.
 

Offline jh15

  • Frequent Contributor
  • **
  • Posts: 485
  • Country: us
Re: analog panel meter - repair possible?
« Reply #15 on: December 22, 2015, 05:16:54 am »
motocodr  lots of old stuff up there, and proven many decades of use over the new stuff.

P.S. I like new stuff too, but hope software doesn't cause a crash. (Maybe trying to hide engine emissions)
tek 575 curve tracer top shape, 535 top shape, 465. 545 hickok clone, Telsa Model S,  Ohio Scientific c24P single board computer, many c-64 from my club days, Giant electric bicycle, Rigol stuff, Heathkit AR-15 receivers 2, Heathkit et 3400a trainer and interface,
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf