Author Topic: 8116A Code 42  (Read 32535 times)

0 Members and 1 Guest are viewing this topic.

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
8116A Code 42
« on: January 18, 2024, 09:05:42 pm »
Hello all, Quick catch-up. This was given to me because a screwdriver was dropped into it, it was working before this. It had multiple issues & killed the -24V rail. I replaced the Shaper output amp transistors Q505 ~ Q513 (except Q507) and a couple of emitter resistors. R559 & R560. The display came up now. and so did CODE-42 I started on the troubleshooter at chapter 10.4.1 as per the manual.
the first test, Input pins 18-19 I have a signal, but they are NOT 180° OUT of phase, but ARE IN PHASE. TP4 looks ok. At this point
I got sidetracked by Q502 being too hot to touch. (only this transistor) the others aren't even warm. Testing output transistors
Q501 ~ Q513 (except Q507) when testing PNPs, Q511 & Q513 in diode mode red lead on the emitter black in the collector to shows a cap charging and then open circuit. OK. When testing Q510 & Q512 red lead on the collector black on the emitter I see a cap charging up but never gets to open circuit.? i pulled 10 &12 they test fine caps charge up and goes to open circuit. I'M LOST.
« Last Edit: January 31, 2024, 08:37:13 pm by Mick B »
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: [HP] 8116A Code 42
« Reply #1 on: January 30, 2024, 12:03:59 am »
(So for other readers, this would be an HP 8116A function/pulse generator...)

Hello all, Quick catch-up. This was given to me because a screwdriver was dropped into it, it was working before this. It had multiple issues & killed the -24V rail. I replaced the Shaper output amp transistors Q505 ~ Q513 (except Q507) and a couple of emitter resistors. R559 & R560. The display came up now. and so did CODE-42 I started on the troubleshooter at chapter 10.4.1 as per the manual.
That's an awful lot of replacements.  Did you verify they were all faulty after being removed?

Quote
the first test, Input pins 18-19 I have a signal, but they are NOT in phase. TP4 looks ok.
So, when you say TP4 looks ok, you mean the test configuration is producing the waveform in figure 10.5-6?  And when you perform the test at amplitude 9.99V and 16.0V, TP4 is following those voltages too?  Do you have any screen shots or scope photos of what TP4 looks like?  Does the unit produce any output?

You might have to come back and address the pin 18 & 19 issue, but if TP4 is operating normally (which is the input to the output amplifier), I'd recommend working on the output amplifier section first.

Quote
At this point
I got sidetracked by Q502 being too hot to touch. (only this transistor) the others aren't even warm. Testing output transistors
Q501 ~ Q513 (except Q507) when testing PNPs, Q511 & Q513 in diode mode red lead on the emitter black in the collector to shows a cap charging and then open circuit. OK. When testing Q510 & Q512 red lead on the collector black on the emitter I see a cap charging up but never gets to open circuit.? i pulled 10 &12 they test fine caps charge up and goes to open circuit. I'M LOST.
When in diode test, your multimeter is applying a test voltage to the probes.  On a good PNP transistor, it should display a voltage of 0.6V to 0.7V, which is one diode drop, when measuring red (+) to Emitter and black (-) to Base.  You should also get a diode drop with the black (-) on the Base again, and red (+) to the Collector.  Reverse polarity in both of these positions should show open.  On NPN, the polarity for the four tests is reversed.  Try it on a known good transistor out of circuit if this isn't clear.

Note that it's not always possible to do the diode test in circuit because other components can interfere, and often do.  If there's a capacitor, as you seem to be seeing, it might take a while for the test voltage to settle to the final value.  Or there may be a lower impedance path that will prevent the diode inside the transistor from turning on and your meter will show an incorrectly low voltage.  Or the transistor could be open, but another component is conducting the test current and creating a "good" reading on the meter.  It's just not 100% reliable.  In one case however, if you get a reading of 0.0V that's usually a good sign it's shorted (but again not *always*).

So, unsoldering a transistor, testing it, and then putting it back is one way to do it.  But that is time consuming, and risks damage to the board.  A better way, if the board can be safely powered up, is to look at the various voltages around the transistor leads.  If the transistor is in an active state, you should be able to measure the 0.6V to 0.7V diode drop across the B-E pins with your voltmeter (called "bias").  Although there's a lot else that can go wrong, looking for the bias voltage is usually a good test for a first pass.

If you don't see a bias voltage, a closer look at the schematic is needed to determine if the bias voltage should be there.  A transistor might be normally off (for example, when used as a current limiter) and would not have a B-E bias voltage of 0.6V to 0.7V.

So, if Q502 and other components are not smoking hot or turning odd colors, I would power up the board for short intervals and check the voltages described on page 10.4-14.  This section is checking voltages around some of the transistors that you replaced.  Feel free to post your readings.  There might be some clues where to look next.

You can also use your meter's diode test to check the various diodes in this circuit.  Although it has the same caveats described above for in-circuit testing of transistors, if a diode is bad it's usually shorted and that will show up as 0V.

If you remove a component that you suspect was bad as a result of in-circuit testing, I'd recommend always testing it again out of circuit to confirm your diagnosis.
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #2 on: January 31, 2024, 11:23:47 pm »
Hi Mark, Thanks for showing up. I put the Q510 & Q512 transistors back in They tested good out of circuit. Once again Q502 is to hot to touch. I checked TP4 as you can see at 999mV its just a blond one high and just gets nasty at 9.99V & 16V I corrected an error in my post above about pins 18 & 19 it now says pins 18 & 19 are in phase but supposed to be 180° out
I have not retested them again.
« Last Edit: April 11, 2024, 09:02:18 pm by Mick B »
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #3 on: January 31, 2024, 11:39:45 pm »
Here is the output @ 1K 1V this is sad. The output files are named output S and T and SQ = sine, tringle, square   Mark what time zone are you in?
« Last Edit: January 31, 2024, 11:41:56 pm by Mick B »
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #4 on: February 01, 2024, 04:14:56 pm »
Ok, so as you've noted, when the unit is set on the 999mV range TP4 looks normal, and when set higher it starts to distort.  This is because less attenuation is switched in between TP4 and the output amplifier input on the higher ranges (relays K501-K504).  As you set the unit's output range higher, there's less isolation between TP4 and the output amp, and misbehaving voltages from the output amp input are making it back to TP4 (probably).

And looking at the unit's final output, clearly it's stuck above 0V when the waveform should be more or less symmetric around 0V.  So something is wrongly dragging the output high.  Q502 is one of the intermediate drivers trying to pull the output low, so it's not a surprise it's hot.  It's probably fighting against something that's keeping the output high.

The next thing I would check is voltages around Q501 and associated circuitry.  Q501 is likely the component Q502 is fighting, since Q501 is the high side of the Q501/Q502 pair.

The suggestion of looking at the voltages described on page 10.4-14 is still valid, although perhaps we only need steps 1 to 3 at the moment.  I would further look at the voltages on EBC of both Q501 and Q502 with respect to ground.  If you post your findings, it will probably provide some clues where to look next.  If needed, we can also compare with a working 8116A I have here for voltages not listed or not apparent from the schematic.

(I'm on the east cost of the US, so EST.)
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #5 on: February 01, 2024, 11:11:38 pm »
HP8116A
Hello Mark, I am on the east cost also NC.
Initial Diagnose: All display LEDs on and stops. Found no -24V rail.
Transistors NPN Q512 is bad but not a dead short, PNP Q513 is ok,
NPN Q510 is open--- PNP Q511 is shorted.---Q508 & Q509 are ok. R559 & R560 were burnt and replaced along with all the power transistors from Q505 ~ Q513 except bias transistor Q507 was not. It tested ok. (in circuit)
The 24v rail problem is gone !
Next:
CODE 42: Diagnostics with MarkL settings @ f~1k @ 999v output on.
Testing Transistors Q501 & Q502 results Q501 =
E.- testing at R518=  0.67mV
B.- testing at CR501= 0.670V.     
C.- testing at VR501= 23.58V
results Q502 =
E.- testing at R519= -111.3mV
B.- testing at CR502= - 0.746mV
C.- testing at VR502= - 8.888V
Testing CR501 & CR502
CR501 Anode to Gnd.= 0.66V            Cathode to Gnd.=047.5mV   DROP=0.7V
CR502 “…………………” = -047.3mV        "…………………… “=-0.73V        Drop=0.69V 

Well, after doing this I retested Q503 and Q504 the diode drops are way off, as are CR503 & CR504 I’m hoping it's just because there in circuit.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #6 on: February 02, 2024, 02:31:07 am »
From your measurements, it appears that Q501 is completely off.  It has near +24V on its collector, so is therefore drawing little to no current.

So, my first question is, since Q501 appears to be off, where is Q502 getting its current that's making it so hot?  And should it be on to this extent in the first place when the system output is supposed to be 0V?  I'll bet R527 is hot too.  What voltage are you seeing on R527?

What voltage are you measuring at the junction of R518 and R519?  This is the input to the output amplifier.  It should be near 0V with the unit set to the 999mV range.

It could also be helpful if you could measure the voltages across zeners VR501 and VR502.  They should be limiting the drop across the associated 2k resistor to 12.1V.  It might be on in VR502's case.

Q503 and Q504 are part of the next stage, which is a gain stage, but it may help to know what they are seeing from the Q501/Q502 pair.  It might be helpful if you could do the EBC measurements on them too.

Opamps U501 and U502 both play role in setting quiescent current of the output amp, so it might be interesting to take a look at their respective outputs (pin 6), and in the case of U501, the junction of the emitters of Q514 and Q515, which is the output from the offset current stage.

Also, note that the designers have provided a nice hint:  The heavy line in the schematic is showing you the waveform signal path from input to output, which can be helpful.

In the meantime, I will extract my 8116A so I can open it and take at some voltages too.  It's currently at the bottom of a stack of other instruments with many GPIB cables, so it's a little involved.

Also, a clarification:
...
Testing CR501 & CR502
CR501 Anode to Gnd.= 0.66V            Cathode to Gnd.=047.5mV   DROP=0.7V
CR502 “…………………” = -047.3mV        "…………………… “=-0.73V        Drop=0.69V 
Did you mean "Cathode to Gnd.=047.5mV" is -047.5mV (minus) ?  It should be the same as CR502 anode.
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #7 on: February 02, 2024, 10:30:40 pm »

2/2/24
Hello Mark, day 2.
This is as far as I got. Some emergency personal stuff came up, I don’t think I’ll be back in town until Mon the 5th. Thanks for your help, it's just getting interesting. I also noticed U33 and U37 getting hot there on A2 with option 001. On page 10.6-5 OR 236 in acrobat. Maybe there supposed to hot.
PS: Yes, CR502 anode is the same my mistake
I can’t tell if R527 is getting hot It’s between a cap and Q502. The voltage on R527 = -21.2V to ground and 2.8v across it.
The voltage at the junction of R518 & R519 is -026.5mV
The voltage across VR501 is 0.24V and at VR502 its 12.3V
Thanks, Mick.
 
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #8 on: February 07, 2024, 11:28:02 pm »
Not sure when you might get back to this, but I had a chance to extract and open my 8116A and I made some measurements on the points from my previous posts so you can compare with a working unit.

First, in text form:
  Settings
  --------
  No Waveform selected
  Amplitude 900mV
  Offset 0mV
  Output enabled
  Freq 1kHz (does not matter)
  Duty 50% (does not matter)
  30 minute warmup
  All voltages with respect to GND (using GND testpoint near the
   large electrolytic capacitor in the power supply section)

  +24V: +24.0729
  -24V: -24.0089
  +15V: +14.9824
  -15V: -15.1512
  U501.7: +14.9222
 
  Q501: E+0.100942  B+0.74338 C+10.6051
  Q502: E-0.105894  B-0.75663 C-10.7324
 
  R518.R519: -0.0022828
 
  CR501: A+0.74359    C-0.0047704
  CR502: A-0.0047704  C-0.75784
 
  U501.6: -0.46839
  Q514E.Q515E: -0.165501
 
  U502.6: +0.0049809
 
  VR501: A+10.6072  C+22.6236
  VR502: A-22.5623  C-10.7285
 
  Q503: E+21.9350  B+22.6268  C+23.2524
  Q504: E-21.8193  B-22.5627  C-23.1768


The above is a lot to probe and digest, and some of the things I suggested in my previous post are duplicates of the same node.  With this many nodes to probe, a better way is to put the measurements right on the schematic, which also makes it easier for troubleshooting.  Attached is the schematic of the output amplifier annotated with the above measurements.

Because the ground I'm using is physically far from the output circuit, there is a consistent small drop of around 2mV for most of the readings.  This ok as long the same ground is used for all measurements.  See photo for suggested ground point.

U33 and U37 should be warm, but shouldn't be so hot you can't keep your finger on them.  I measure them around 42C.  The four chips in that row are ECL, so they tend to use a lot of power and run hot.

My Q501 and Q502 are also quite hot, around 65C.  There are other components that are running even hotter, like R530, R531, and R532 which all seem to be running around 82C.  U401, the shaper IC, is running at a balmy 73C.  I never looked at the insides of a 8116A with a thermal camera before.  Wow, it's a really toasty unit.

I'm starting to wonder if the collector of your Q501 is open.  Q501 has a reasonable bias but doesn't seem to be doing anything.  You could take it out and test it.  You could also try a test in-circuit and try to force more bias current into the base by temporary holding a 10k in parallel with R516 (an 11k) and see if there's any voltage change on Q501's collector or emitter.

As per my previous post, it would be interesting to know what the voltage is at the base of Q503 (which is also the cathode of VR501).  VR501 is also a suspect, but not proven innocent or guilty yet.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #9 on: February 08, 2024, 08:20:12 pm »
Here's another thing if you could please clarify:
...
results Q502 =
E.- testing at R519= -111.3mV
B.- testing at CR502= - 0.746mV
C.- testing at VR502= - 8.888V

...

CR501 Anode to Gnd.= 0.66V            Cathode to Gnd.=047.5mV   DROP=0.7V
CR502 “…………………” = -047.3mV        "…………………… “=-0.73V        Drop=0.69V 
...

Did you mean that Q502 B is at -0.746mV or -0.73V?  It would make more sense if it was the latter.  Please take care to be accurate in recording voltages.  It adds a layer of difficulty having to guess what was meant.

Because this is an inverting amplifier with feedback, we should also consider what's coming back towards the input on the feedback path since there might be something faulty downstream of the stages currently under scrutiny.  A good place to measure the feedback node would be the junction of R528 and R529.  Can you check the voltage there too, please?  Mine reads +0.0026147V (same settings and ground reference point).
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #10 on: February 09, 2024, 03:29:52 pm »
Analyzing this a little more closely, I want to correct something I said:
...
And looking at the unit's final output, clearly it's stuck above 0V when the waveform should be more or less symmetric around 0V.  So something is wrongly dragging the output high.  Q502 is one of the intermediate drivers trying to pull the output low, so it's not a surprise it's hot.  It's probably fighting against something that's keeping the output high.
...
First, we've determined Q501 and Q502 just run hot, when comparing to a working unit.  So the elevated temperature of Q502 may not be a useful clue.

Second, I was not taking into account the inverting stage in this amplifier (which happens at Q505 and Q506).  So, when the collector of Q501 goes high, the output is forced low.  Therefore, the collector of Q501 might actually be doing the right thing by being pinned high, as instructed by the feedback.  It could be trying to correct for whatever is dragging the output high.

I'd like to add Q507 to the list of measurements.  It sits conveniently between the inverting gain stage (Q503/Q505 and Q504/Q506) and the output driver (emitter follower) stage consisting of Q508/Q510/Q512 and Q509/Q511/Q513.  It may at least help divide the schematic in half for your troubleshooting focus.

Let's also take a look at voltages around Q508 and Q509, since that's where you had to replace them and their associated collector resistors (R559 and R560).  Maybe there's something else dead in the output driver section that you didn't catch.

Note that Q510/Q512 are essentially in parallel with each other, as is the pair Q511/Q513.  They should be sharing currently equally and a quick check should show the emitters in each pair to be at the same voltage, within a mV or so.

I should have some time later today to update my diagram with measurements for these nodes and will repost.
« Last Edit: February 09, 2024, 03:32:18 pm by MarkL »
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #11 on: February 09, 2024, 07:10:01 pm »
I didn't read the above post yet. I'll be Back.

2/9/24

Hello again Mark, these are the newest measurements. Your schematic idea is brilliant. I was going to edit yours, but I have Adobe Acrobat Pro X, and it doesn’t have the fonts (LibreOffice) has.
I’m having a major issue with my Saab someone stole the ECU. And now I have to learn how to work a Tech 2 Scanner, and the software it uses THAT ONLY WORKS IN XP! Had to bone up on VMware. The updates need to be downloaded (I don’t think XP will access the internet, so I must find the files on the internet and figure out how to get them to load in the software in XP. You can just buy an ECM and reprogram it USING your OLD ECM. See the problem?
 This part is supper complicated, and I’ve only scratched the surface. 
The ECM must be married to the ECM apparently through the security system I have been reading into all hours of the night. For the moment until a few more hours are added into the day is it OK if I forgo the schematic until I get caught up and just post the results.

I used the settings & Gnd. locations noted.

+24v:    +24.0140
-24v:     -24.0123
+15v:    +15.0319
-15v:     -15.0273

U501.7: +14.04263

Q501:    E +066.501mv     B +0.6592v     C +023.8454v
Q502:    E -069.3mv          B -0.7245v     C -08.868v NOTE: these are fluctuating up & down.

R518.R519:  -016.9463mv

CR501:   A +0.65984v      C -035.46mv
CR502:   A -0.2474v         C -034.67mv

U501 . 6:  +013.5805v NOTE: yours is neg.
Q514

Q515: E +012.8370v NOTE: yours is neg.

U502. 6:   -21.2336mv NOTE: yours is pos., and this starts out at -100 and drops to 21. And then starts climbing.

VR501:   A +023.1866v     C +023.9821v
VR502:   A -021.3346        C -08.8716v

Q503:   E +02.23366v       B +23.9812v       C +02.6464v
Q504:   E -020.5156v        B -021.2258v      C -023.6773v

Tempts FREAKEN HOT!! Using UNI-T:  uti260B
U33:  78c
U37:  80c
Q501:  32c
Q502:  53c
R530,31,32:   76c
U401:  74.5c

NOTE: a lot of the voltages were fluctuating,
I also tried jumping R516 w/10K no changes,
Removed Q501 Using a small transistor tester, it tests good. This a 2N2369A
I have NOS   Q501 and 2 transistors coming- it couldn’t hurt.
Thanks Mark
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #12 on: February 09, 2024, 11:46:58 pm »
Attached is the schematic with added measurements on the output driver side.  If it would be helpful, I can post the LibreOffice Draw file instead of the PDF.  I can also use a different font if it's picking some weird substitute font.

Let me take a look at your latest measurements and I'll post back.  The pinned output on U501 immediately stands out.  As long as you have waveforms off, I don't see how anything should be fluctuating.

Are you referring to the U33 and U37 on the A3 board where the processor sits, or on the A2 control board?  HP duplicated the U designations on these two boards (how nice of them).  I was referring to the A2 board in my previous post about them being ECL.  But in either case, your U33 and U37 are still too hot, ECL or not.  At the moment I don't think it has anything to do with the distorted output, but it's certainly is something to keep in mind.  Let's focus on the output amp for now and come back to that.


Good luck with your ECM.  I experienced a similar thing with the module security code stuff trying to fix my Toyota Prius about 15 years ago.  After many, many wasted hours, I finally got it to work only to find out the problem was something internal to the motor generator in the engine anyway.  I just take my cars to the dealer now.  I don't give up easily, but they win.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #13 on: February 10, 2024, 12:08:19 am »
And to answer your question about forgoing the schematic, feel free to use text if that's easier.  I put the measurements in the schematic because it's easier to visualize what's going on, but don't let fancy delivery be an impediment to getting the data out there.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #14 on: February 10, 2024, 03:54:31 pm »
Ok, based on your latest measurements there's some issues with and around Q503.

Note that the Q503 B-E junction is nowhere near 0.7V, and the collector should be around +23.3V or more.  It should be +24V - one diode drop from CR503, or R533 could pull the collector voltage even closer to +24V.

So, Q503 appears dead (open B-E junction, >0.7V), CR503 is also dead (open, >0.7V forward voltage, A to C), and R533 too (open, since at those voltages it would be dissipating about 23W and glowing).

I would double check your voltage readings on Q503, and if confirmed pull it out and verify it out of circuit, as well as CR503 and R533.

When double checking voltages, I would also make sure you can see the +24V rail on the upper side of R533 and the anode of CR503.  It's not beyond possibility that there's a fried trace.

If Q503 is indeed bad, it also wouldn't be a bad idea to check R534 and Q505.  The current that killed it had to go somewhere.
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #15 on: February 18, 2024, 06:30:24 pm »
Hello Mark, down for parts on the Saab. Well, I was a lot more careful with the measurement's a lot of them not directly connected to the power rails had fluctuating readings as if caps charging. I tried to get the first reading. Where it says mv, that is right off my meter. HP34401A its readings match my Keithley 2015 to within microvolts.
I replaced Q501 because I had a new one and it was already out. I have the TO-93pkg heatsinks so at some point I will replace Q502 also. Looking at the resisters around Q501 and Q503 readings made with 4wire, R516. 10.962k.  R524. 100.437Ω   R525. 2.0012K.  R533. 19.635Ω   R534. 19.599Ω those numbers on the schematic sure look like an 8 but definitely a 6.
I'm going to pull CR503. right now.  Oh ya. notice the polarity difference around U501 & U502?
Thanks, MarkL. 
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #16 on: February 18, 2024, 06:33:33 pm »
For some reason it didn't post my schematic here it is.

UPDATE: CR503 checks fine out of circuit!
AND: found a bad solder joint AT the topside of the MAIN board where Q510 Collector meets the board making a sketchy +24v connection. It was soldered on the bottom but not the top where is gets the +24v Makes me want to test output transistors with heatsinks where you can't see the legs from the bottom of the board at the legs what a pain that will be. perhaps Mark you see some transistors I might want to recheck looking at new voltages? 
« Last Edit: February 18, 2024, 08:37:16 pm by Mick B »
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #17 on: February 18, 2024, 09:40:00 pm »
The voltages around Q503 that you're showing on the schematic are very different than those you posted before.  I can no longer conclude there's anything wrong with Q503, CR503, or R533.

The polarity difference wrt. U501 is not a major concern.  U501 is providing an offset current to the amplifier input to keep the amplifier's virtual ground at 0V to compensate for imbalance due to component variability.  So, it could come to rest at any value ideally close to 0V.  But because the amplifier is currently way out of balance due to whatever failure (note mid-point between Q501.E and Q502.E), U501 is pinned to one side as a side effect.  I've learned to never rule anything out completely, but I don't see anything at the moment that leads me to believe there is  a problem with the way U501 is behaving.

I'm seeing some voltages around Q512 that don't make sense: How can it have +0.36152 on its emitter when the base, collector, and other side of the emitter resistor are all higher than that?

Before we go any further, can you describe the "fluctuations" you're seeing?  Or maybe put your scope on a couple of the nodes where you're seeing a lot of fluctuation?  It's not out of the question that in its current failure mode the whole amplifier is oscillating.  Obviously it shouldn't be, but that would make it impossible to obtain stable voltage measurements.

Are the power supply rails stable according to your scope?
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #18 on: February 18, 2024, 10:19:04 pm »
In addition to the above, fluctuations notwithstanding, the voltage on the collector of Q506 is not making a lot of sense.  I'd like to know what kind of current is flowing through the emitter.  Can you take a look at the emitter voltage on Q505 and Q506?  We can determine the current from the emitter resistors R534 and R537.  Thanks.
« Last Edit: February 18, 2024, 10:39:04 pm by MarkL »
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #19 on: February 20, 2024, 10:23:56 pm »
In addition to the above, fluctuations notwithstanding, the voltage on the collector of Q506 is not making a lot of sense.  I'd like to know what kind of current is flowing through the emitter.  Can you take a look at the emitter voltage on Q505 and Q506?  We can determine the current from the emitter resistors R534 and R537.  Thanks.

Hi Mark, Ok I spent all day yesterday looking over voltages I am dyslexic and i have to make multiple measurements and read them over and over to make sure I have not transposed numbers and since my first set of numbers, I have been diligent in this. and trying to make since of these voltages. I left question marks at some of the fluctuating numbers. I circled just 3 the numbers were any ware from 0.32??? to what you see in the circles Just an hour or so apart, and it had been warming up well over an hour. This is what I mean by fluctuating the last 3 digits do not stop sometimes changing 2nd digit. the Power supply is rock solid. I looked at a few nodes with my scope really nothing to see just DC voltage, I saw no AC until I got down in the weeds @ 50mv = noise. (I had that thing running all day the 2 ICs on the A2 board haven't caught on fire yet). :-+
When it rains, went to post this yesterday and I internet went down. the same network I had up 5yrs or so no issues, I saw ATT workers at one of their junctions where a lot of their stuff meets. I'm sure they did it. My whole house is wired with Ethernet, using a Cisco 48 port switch & an ATT, fiber router/modem with typical default IP address, long story short. all of a sudden my switch has the same IP add. as the router. That took a min to get to the bottom of and fix. 
« Last Edit: February 20, 2024, 10:27:23 pm by Mick B »
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #20 on: February 21, 2024, 11:02:21 pm »
Regarding your description of "fluctuations", it appears you are getting readings that are unstable in the mV in most cases.  It could be noise or possibly a low level oscillation, but you are not seeing anything on your scope.  I'm going to assume for the moment the digits you've shown are stable and valid, and ignore any digits with a "?".

One thing to try is to set your DMM on AC and see what kind of readings you get on some of these nodes that are unstable.  This will block the DC and show just the varying part of the voltage on that node (within the frequency limits of your DMM).

As far as the latest updated measurements, there are several items of note:

1) Q506 B-E junction is significantly more than 0.7V (approx. 3.2V).  This would indicate Q506 B-E is open.  I would remove Q506 and check it out of circuit.

2) Q505 is off, as indicated by the B-E voltage of only about 0.4V.  The collector is therefore floating, and free to wander around (within the bounds imposed by other circuit components attached to it, such as other transistor junctions).  Even connecting your probe to the collector could be influencing the reading.  This could be the reason for the unstable readings.

3) Both Q510 and Q512 also have B-E junctions > 0.7V (approx. 0.9V to 1.0V).  Recall that they are in parallel, and this would indicate neither is working.  I would remove them both and test out of circuit.

I would try to correct #1 first and leave #3 alone for the moment.  Let's find out why Q506 is not reading correctly, fix it, and then move on.  Taking a shot-gun approach and changing multiple things at the same time creates too many variables.

You stated in the beginning that you replaced a large number of transistors.  When checking Q506 out of circuit, make sure it is the right type, PNP or NPN.  You might also want to purchase a transistor tester, as it will identify the transistor type, the pinout, and other important parameters like gain (hFE) and leakage.

For reference, my Q505.E is +22.6483V and Q506.E is -22.5548V.
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #21 on: February 24, 2024, 12:01:42 pm »
I pulled Q506 and sure enough it was toast! Q506 & Q505 have the heatsinks on but not the cross bar they got very hot, can you tell me what your temps are for these transistor's before I move on. and the good news NO CODE!
Note: the settings were as you recommended, also all the pots on the main board are centered.
Thanks
« Last Edit: February 24, 2024, 12:09:55 pm by Mick B »
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #22 on: February 24, 2024, 03:06:16 pm »
I pulled Q506 and sure enough it was toast! Q506 & Q505 have the heatsinks on but not the cross bar they got very hot, can you tell me what your temps are for these transistor's before I move on. and the good news NO CODE!
Note: the settings were as you recommended, also all the pots on the main board are centered.
Thanks
So, have you replaced Q506 and it now has no error code, or do you mean there is no error code with it out?

My calculations for my Q506 show that it's dissipating about 0.6W when quiescent.  Without a heatsink on such a small part I could see why it's getting quite hot.  I wouldn't run it that way for longer than what's needed to get a very quick check done, maybe less than a minute or so.  The Q505/Q506 heatsink temperature on my unit after a 15 minute warm-up is 60C.  With the case open there's not much airflow directed across it.

How did all the pots get centered?  Were they that way or did you do that?
 

Offline Mick BTopic starter

  • Regular Contributor
  • *
  • Posts: 94
  • Country: us
Re: 8116A Code 42
« Reply #23 on: February 24, 2024, 05:17:28 pm »
Ok I replaced Q506 fired it up and error code 42 is gone. I put the heatsinks on. It still has some issues, I'm doing a rough calibration on it so far it is doing ok, those 2 IC are still blazing hot. The pots were centered by the guy I got it from, He was going to calibrate it that's when he dropped the screwdriver in it, and why I ended up with it. Just going to try and get it in the ballpark before the next big dive. Unless you (MARKL) think I should start else ware in this endeavor. Thanks   
« Last Edit: February 24, 2024, 05:22:39 pm by Mick B »
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2152
  • Country: us
Re: 8116A Code 42
« Reply #24 on: February 24, 2024, 10:22:23 pm »
Centering of all the pots by the former owner was not a great move.  I can say from experience these units can be a real bear to align.  Some of the pots are way too sensitive, i.e. a small movement can radically change parameters, and there is interplay among some that can be difficult to optimize.  It's much easier to start where it's almost right, and then tweak whatever is necessary.  Oh well.

At this point, if something isn't working right, it would be more difficult to say if there was another bad component or an adjustment that's way out.  I agree you should try to get through the adjustment procedure.

On the hot chips, you mentioned they were on the option 001 A2 board.  On page 8-20 of the service manual, there is a short verification procedure for option 001.  See if it passes.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf