Oscilloscope Repair

[shimniok]

You probably were going to do this anyway but ... as cheap as the 2SC458D equivalents are, and since there's several in this scope, might as well get a few just in case another one is bad or you want/need to replace the 'twin' of Q11 with one too... anyway, let us know when you're back

 

[pierreretief]

Hey guys, Im back. I ordered the transistors and now Im waiting for them. Cant wait to get them here and fix my scope! Will keep you posted!

 

[shimniok]

Hey guys, Im back. I ordered the transistors and now Im waiting for them. Cant wait to get them here and fix my scope! Will keep you posted!

Cool deal, eager to hear what happens!

 

[pierreretief]

Hey guys, I am back, got the transistors and replaced Q11, but still wonky... what now? :-(

 

[Rolf]

Worth Checking.........

Hey guys, I am back, got the transistors and replaced Q11, but still wonky... what now? :-(

Many years ago when doing oscilloscope calibration and repair I did on at least one occasion find a broken (intermittent) trace in one or the center layers of their multilayered PCB.

 

[shimniok]

That's a drag, but don't worry we'll get this figured out.

*** Are the measurements the same on Q11 as before?

We'll get back to circuit troubleshooting ... I will take a look as soon as I get more free time. I recall we were seeing odd behavior on the one side of the circuit... it was either totally on or totally off.... will stare at the circuit diagram some more... we thought one of the voltages across Q11 looked weird (was it Vbe? Usually should be 0.7V and it was nowhere close)

If Q11 itself isn't the cause... we'll have to figure out what is... bad component, bad upstream or downstream transistor? Stay tuned...

Michael

 

[shimniok]

Haven't had a chance to look into this much... been super busy... intend to look at it soon

 

[pierreretief]

Great, thank you so much! Thought you guys have forgotten about me. Yeah, it was Vbe, it was something like 2V but now it measures correctly, so we did fix that at least, so it wasnt for nothing. All the measurements and info is still on this forum somewhere, just have to page back a bit

Thanks alot

 

[shimniok]

I've been working on repairing an old Harman Kardon 680i receiver and have been doing some transistor testing. I found this worksheet to make it easier to go through the tedium of BJT transistor testing.

First, gotta know if it is pnp or npn and use the right chart. I start with red on 'base' then measure using diode test, with black on C then E and fill in the corresponding cells. Then move red to C, measure black on B and E and fill in. And then red on E and measure black on B and C. Then compare the readings with the pre-populated numbers to make sure it is aok.

Of course I was doing it by hand, in pencil, but the nice thing is that these worksheets also show what the readings "should" be.

Michael

 

[shimniok]

EDITED a couple times.

As for the o-scope... the only way we're going to get this done is to be thorough and methodical, testing components patiently until we find and fix the problem for good. There may be multiple bad components.

If it makes you feel better, I spent about a month or two chasing down the intermittent horizontal amp problems on my Hitachi V1050F. It takes time and patience. I've been chasing down problems on this HK 680i for quite awhile too, and had to take a break from it for awhile. But in both cases, I found the problem. It can be done.

Btw, I have learned more about BJTs and their associated circuits in the last few months so hopefully that will help us. I've also thought about this a lot and had some light bulbs go off in my head.

We found that Q11 Vbe was wrong when the trace was in the "up" position and so it is replaced, now. Where we left off -- we thought that Q13 was not getting the correct drive current into its base with the trace up.
Vbe for Q13 was 0V in the "up" position.

But our wonky trace problem is occurring with the trace in the middle and lower positions, not the up position!

We WERE seeing both Q13 AND Q15 (looking at Vce) either fully on or fully off, instead of being partly on with the trace in the middle. What should happen is that their Vce values should linearly vary (inversely proportional to each other) with the trace position knob. We were also seeing AC voltage across Vce for Q13/15 which should NOT be happening when displaying a DC (or 0V) signal trace! More on this in a minute.

It's quite possible Q11 had nothing to do with the wonky trace. We need to make sure swapping Q11 actually fixed or changed something. Either way we'll gain valuable information about what is (still) going wrong in the circuit. But if the behavior changed and we don't know about it, we'll be chasing ghosts so we have to know how things are behaving now. You can use this spreadsheet on google docs to record your measurements:

https://spreadsheets.google.com/ccc?key=0AjFa-cL8YjH_dHpVaWM1MHhZQnBwNWxSUjFYSHNKNkE&hl=en

1. Measure Vbe for Q11 with trace up, wonky, and down. Hopefully it is normal now.

2. Measure Vbe for Q13, Q15 with trace Up, Wonky, and Down.

3. Measure Vce DC Volts for Q13, Q15 with trace Up, Wonky, and Down.

4. Measure Vce AC volts for Q13, Q15 with trace Up, Wonky, and Down.

Since you are still getting the wonky trace, I suspect that whatever was wrong is still wrong.

If you think about it, something was (is?) causing AC voltage and incorrect DC voltage across Q13/15, but that same something wasn't affecting Q14/16 despite the fact that they appear to me to share the same source for base drive current (despite errors on the schematic)

It seems logical that whatever is causing the problem is related to Q13/15 but unrelated to Q14/16. Like, Q10, Q17, Q19, maybe even Q5. There are other possibilities like bad resistors or capacitors, but we'll get to that.

Let's not get too distracted. Let's focus on doing the following (use the spreadsheet above)

1. in-circuit BJT diode test on Q9 (PNP) Q10 (PNP) Q17 (PNP) and Q18 (PNP) using spreadsheet (numbers may not exactly match due to being in-circuit) -- use the "diode test" tab on the worksheet

2. Test Vbe DC voltage on Q9, Q10, Q17 and Q18 - up, wonky, down.

3. Test Vbe AC voltage on Q9, Q10, Q17 and Q18 - up, wonky, down.

Let's see if that turns up more suspect components. We're slowly tracing our way through the circuit until we find the root cause of all this.

 

[shimniok]

Hello? Didja give up?

 

[pierreretief]

Hey, sorry I havent replied in so long. I have been so incredibly busy, and we are busy wih something we call engineering week. Big tests every day. I am finishing tomorrow, and I am going to do work on my scope on saturday, and do eveythging you suggested.

I really thought we had it with that transistor. Oh well, guess well just try again...

Thank you so much for all the info you have given so far. We'll see what I can find on saturday

Thanks again

 

[pierreretief]

Had a look at the spreadsheets, Im gona do all thoese tests and surely now we should find something amiss, this thing is so sneaky

 

[shimniok]

Cool deal. Back in college, Engineering Week was a sort of festive event with contests and demos going on or something. I never really paid a lot of attention. Anyway best of luck with the tests.

We'll nail this thing down!

Meanwhile I am closing in on fixing that receiver I blew up, with help from a guru over on an audio forum I'm on. It feels really good to fix electronics.

 

[unclejed613]

Had a look at the spreadsheets, Im gona do all thoese tests and surely now we should find something amiss, this thing is so sneaky

i don't know where this quote came from, but it fits.... "It's just a machine, it can't outsmart you much longer"......

i thought we had it too, but i'm sure we can find the cause.

 

[pierreretief]

Right, lets do this...

 

[pierreretief]

mmm, ok, starting with everyhting, but already something is fishy, I could have sworn it was 0.6V Vbe for Q11 but testing it now, I got 2.68V...

But I will continue with the tests.... Maybe something else is causing it. I am going to do a lot of tests now!

 

[pierreretief]

Since looking at that excell sheet really confuses me, and is quite daunting to say the least, I have decided to approach it like this. Starting with transistor Q9 and moving up. Finishing one after the other and having a seperate post for each transistor, to avoid confusion also...

Next follows Q9

 

[pierreretief]

Q9:

DC - Vbe - Wonky - 0.64
DC - Vbe - Middle - 0.65
DC - Vbe - Up - 0.66

DC - Vce - Wonky - 11.36
DC - Vce - Middle - 9.4
DC - Vce - Up - 7.24

AC - Vbe - Wonky - 0
AC - Vbe - Middle - 0
AC - Vbe - Up - 0

AC - Vce - Wonky - 0
AC - Vce - Middle - 0
AC - Vce - Up - 0

 

[pierreretief]

Q10:

DC - Vbe - Wonky - 0.66
DC - Vbe - Middle - 0.65
DC - Vbe - Up - 0.64

DC - Vce - Wonky - 7.27
DC - Vce - Middle - 9.08
DC - Vce - Up - 11.35

AC - Vbe - Wonky - 0
AC - Vbe - Middle - 0
AC - Vbe - Up - 0

AC - Vce - Wonky - 0
AC - Vce - Middle - 0
AC - Vce - Up - 0