EEVblog Electronics Community Forum

Electronics => Beginners => Topic started by: outofstepper on June 04, 2012, 07:04:26 pm

Title: Know what'd be nice? . . .
Post by: outofstepper on June 04, 2012, 07:04:26 pm
a video tutorial on debugging / fixing ... anything (broken  ;D ).
Using some/all the basic equipment: scope, DMM and function generator.

Maybe I'm too green to even realize what I'm asking / proposing.. but it would
be great, as a beginner, to get a glimpse of the logic / reasoning a pro goes through
to sort out broken kit they are likely not familiar with.

For example is see a lot of youtube vids on "how to use a function generator" and
all they really go through is the operators manual.  "I pushed triangle.. lets hook it up
to the scope.. look at that its a triangle wave.. this knob changes the frequency!"

and I'm not sure if this is a question or a suggestion.  :)

Thanks!
Tony
Title: Re: Know what'd be nice? . . .
Post by: Architect_1077 on June 04, 2012, 07:49:25 pm
Well, from beginner to beginner... you first need to learn the basics of electronics to understand the steps the "pros" go through to discover what the problem is and solve it. IMHO...
Title: Re: Know what'd be nice? . . .
Post by: SirDan on June 04, 2012, 07:53:31 pm
Dave did a really nice tutorial on debugging a circuit board from a helicopter or something. In the end it turned out to be a bent pin on a connector, but the debugging steps were very useful. When I watched the video I was thinking more of this kind of stuff would teach how to debug/fix your electronics. I cannot find the video now. Maybe someone can post a link to it.  :)
Title: Re: Know what'd be nice? . . .
Post by: Neilm on June 04, 2012, 08:13:40 pm
Dave did a really nice tutorial on debugging a circuit board from a helicopter or something. In the end it turned out to be a bent pin on a connector, but the debugging steps were very useful. When I watched the video I was thinking more of this kind of stuff would teach how to debug/fix your electronics. I cannot find the video now. Maybe someone can post a link to it.  :)


I believe that you are after episode 230 http://www.eevblog.com/2011/12/31/eevblog-230-arducopter-ardupilot-troubleshooting/ (http://www.eevblog.com/2011/12/31/eevblog-230-arducopter-ardupilot-troubleshooting/)

Neil
Title: Re: Know what'd be nice? . . .
Post by: HLA-27b on June 04, 2012, 09:07:18 pm
You are in luck. It turns out today Shahriar (aka Hugoneus on this forum) posted exactly what you asked for.

Teardown and Repair of an Agilent E3634A Power Supply (https://www.youtube.com/watch?v=3EjmCWmarwE#ws)

 
Title: Re: Know what'd be nice? . . .
Post by: w2aew on June 05, 2012, 12:48:42 am
...and, I posted a debug and repair video a few weeks ago that dealt with a problem I was having with my function generator...
Function Generator debug and repair - Leader LG-1311 (https://www.youtube.com/watch?v=x3z2QSVtjSA#)
Title: Re: Know what'd be nice? . . .
Post by: vk6zgo on June 05, 2012, 03:10:49 am
Nothing happened with either video!
I clicked on the links below them, & got the "Trained monkeys"message! ;D*

To the OP:
It is a bit harder to demonstrate than to do some of these things,as they are a combination of theoretical knowledge,experience & educated guesses.
Without the theory,you can run out of ideas sometimes!

* Thought I'd better put the "n" in before some smart-alec asks what a "mokey" is!
Title: Re: Know what'd be nice? . . .
Post by: vk6zgo on June 05, 2012, 04:30:22 am
Well,I tried again---No "mokeys" this time! ;D

Both videos are excellent demonstrations of faultfinding techniques.
Title: Re: Know what'd be nice? . . .
Post by: codeboy2k on June 05, 2012, 05:23:40 am
...and, I posted a debug and repair video a few weeks ago that dealt with a problem I was having with my function generator...
Function Generator debug and repair - Leader LG-1311 (https://www.youtube.com/watch?v=x3z2QSVtjSA#)

Thanks for that video. Very interesting repair.  I'm curious, have you tried looking at the output filter caps on those 17v rails?  As you said, the adjust cap is there to slow the feedback a little and help with ripple, but with the cap installed it's just too slow and can't react fast enough.  Again, as you said, obviously it worked when the engineers designed it.  I was thinking perhaps the output caps on the 17V rails have gone bad, thus developing a high ESR and are not providing any big load for the rails anymore.  This would increase the dV/dt of the 17V rails and thus it might shoot up too fast, beyond what the adjust cap can handle (as ripple), so it's just oscillating.

Are the schematics online?  I'd love to read them too.  I love a good analog read.  I did a quick google for it but I didn't see anything that stood out.

Cheers!


Title: Re: Know what'd be nice? . . .
Post by: david77 on June 05, 2012, 07:33:37 am
I've watched Shahriar's video although it was quite good, it doesn't really teach you much about repairing things.
It's highly unlikely you will find exactly the same fault anywhere else.

In my experience the only things that help you fault finding are experience, a structured, methodical approach and a manual/circuit diagram.
It also always depends on what you are trying to fix. A dishwasher is quite different from some lab gear - I know I've fixed both ;).
Also good knowledge of some commonly used circuits is usefull so you can identify what's going on in a circuit.

It used to be normal to get a proper manual with explanation of the circuit for expensive lab gear, sadly that's no longer the case apart form maybe Agilent.
Title: Re: Know what'd be nice? . . .
Post by: johnwa on June 05, 2012, 09:09:03 am
Have a read of the Repair FAQ: http://www.repairfaq.org/ (http://www.repairfaq.org/) It will probably take you about a month to get through it all, but it is well worth it.
Title: Re: Know what'd be nice? . . .
Post by: outofstepper on June 05, 2012, 09:42:24 am
holy smokes, this forum is something else!

wait wait, let me try this again:

"Know what'd be nice ... a million dollars"

 ;D

Thanks all, I've got a lot to take in.

Tony
Title: Re: Know what'd be nice? . . .
Post by: Architect_1077 on June 05, 2012, 11:00:14 am
Did it work???  ;D
Title: Re: Know what'd be nice? . . .
Post by: w2aew on June 05, 2012, 03:33:09 pm

Thanks for that video. Very interesting repair.  I'm curious, have you tried looking at the output filter caps on those 17v rails?  As you said, the adjust cap is there to slow the feedback a little and help with ripple, but with the cap installed it's just too slow and can't react fast enough.  Again, as you said, obviously it worked when the engineers designed it.  I was thinking perhaps the output caps on the 17V rails have gone bad, thus developing a high ESR and are not providing any big load for the rails anymore.  This would increase the dV/dt of the 17V rails and thus it might shoot up too fast, beyond what the adjust cap can handle (as ripple), so it's just oscillating.

Are the schematics online?  I'd love to read them too.  I love a good analog read.  I did a quick google for it but I didn't see anything that stood out.


I'll have to circle back and re-check them.  I did not notice any significant ripple on the 17V supply rails, so I didn't suspect any problems with the bulk filtering caps.  Of course, the main filter caps after the rectifiers were OK.  I'll have to look through the schematic again to see what bulk filtering is being used after the regulator, and check those caps out with my homebrew ESR meter. 

Unfortunately the schematics are not online.  I had a contact at Leader that emailed the documentation to me.
Title: Re: Know what'd be nice? . . .
Post by: jucole on June 08, 2012, 11:21:13 pm
You are in luck. It turns out today Shahriar (aka Hugoneus on this forum) posted exactly what you asked for.

Great video!