gslikker
Super Member
Indeed, but you have to measure for differences between the off board and the card if wanting to include the tracks going off board and connections as suspects.
Last edited:
Calculating Current Through a Transistor
- Thread starter LBPete
- Start date
it would require a bit more insight into the connections setup for the board for me to make an informed assessment, but my first thought is that if any external connections get cancelled out as suspects if you swap the boards and the issue stays with the board....
I just solved a similar issue...I had a differential pair (j-fet) composed out of two separate transistors. Sporadically, the DC would erratically go up or down into the hundreds of millivolts...then back to normal (+/- 1-2mv). The issue followed the board and it was the differential pair in case.
I replaced the two j-fets with a single dual-fet and the problem went away immediately. Having a differential pair on the same substrate is naturally better for reasons of thermal stability.
The problem does follow the board when switched left to right but that doesn't rule out a problem with the connections to the connector pins or how those pins mate together. I'll take a very close look at this today. I have previously touched up the solder joints for the connector on the problem board.
Thanks for suggestions, I appreciate the help.
- Pete
Thanks for suggestions, I appreciate the help.
- Pete
Maybe a picture of the insides would be helpful so we can see the physical layout and connections.
I didn't have much time yesterday to work on it but did do some checks.
Here's a picture of the boards. In general, this appears to be a very well designed receiver. The connectors are different than I've seen before but they have very long pins that seem to engage well.
- Pete
Looking at my notes, I believe the voltage is higher on the output side of R612 but I didn't write it down when I checked the base side. I'll re-check this.
I did a continuity check across all the pins on the edge connectors. They were all the same, an ohm or two. I checked from R612 to the output emitter resistors and nothing there either. I was not able to check between the driver transistor emitters and the bases of the output transistors. Can't get access to the output connector pins without a lot of disassembly.
Checked the continuity of all the pins and moved the board around to see if it's position in the connector changed anything.
Can't get to the base connections on the output transistors
I moved the board in the connector to see if it changed anything.
Here's a picture of the boards. In general, this appears to be a very well designed receiver. The connectors are different than I've seen before but they have very long pins that seem to engage well.
- Pete
There is a discrepancy in the voltages on the driver transistor emitters.
Bad board TR608 the emitter voltage started up around 1V and dropped to about 0.8V as it warmed up. The good board quickly stabilized to a little more that 0.5V.
Bad board TR609 had an emitter voltage wound up around 240mV after several on/off power cycles. The good board quickly stabilized to a little more than 0.5V.
These measurements were taken at zero volume, no load. After taking these measurements I swapped the boards left to right and got similar results. So I don't think it's an issue with the circuit on the board the output transistors are on.
- Pete
Bad board TR608 the emitter voltage started up around 1V and dropped to about 0.8V as it warmed up. The good board quickly stabilized to a little more that 0.5V.
Bad board TR609 had an emitter voltage wound up around 240mV after several on/off power cycles. The good board quickly stabilized to a little more than 0.5V.
These measurements were taken at zero volume, no load. After taking these measurements I swapped the boards left to right and got similar results. So I don't think it's an issue with the circuit on the board the output transistors are on.
- Pete
gslikker
Super Member
Ok so this means total difference is about/slightly over 1 Volts on both channels, so the emitters to output transistors base is correct (about 0.5V each output transistor base to emitter).
Now the input side, what is the measurement ground reference? Do the incoming shields go to ground or to some other board? Read: Is the incoming ground (E on left side on schematic) zero when situation is wrong?
( As incoming ground looks soldered, that connection does go with the card.
further, I guess I am out of ideas....
Now the input side, what is the measurement ground reference? Do the incoming shields go to ground or to some other board? Read: Is the incoming ground (E on left side on schematic) zero when situation is wrong?
( As incoming ground looks soldered, that connection does go with the card.
further, I guess I am out of ideas....
gslikker
Super Member
One more thing: I see R617 and R618 called F 33. Would this be fusible resistors and are they measuring stable when open circuit and cold or warm? (read: no power, board unplugged and measuring Ohms.)
Maybe also check the driver transistors collector voltage presence when cold and powered up, to make sure the collector resistors are conducting when cold.
Maybe also check the driver transistors collector voltage presence when cold and powered up, to make sure the collector resistors are conducting when cold.
Last edited:
Turns out that all the resistors that are raised off the board have that F designation. I've changed them all with no change in the symptoms.
Ground level on both boards is zero volts. While measuring the ground level at the ground connection to R602, I also measured the junction between R602 and R601. There are 600mV at that junction on the bad board and only 70mV on the good board. That's a huge difference. I double checked R602 thinking it would have excessive resistance but it's at 100k which is what is called for in the schematic and what it's mate on the good board measures.
It just occurred to me that if that resistor is good and the voltage at the R601/2 junction is high, that supports your theory of a high ground level but it measured zero volts. I'm going to try supplying a separate ground to see if that changes anything.
- Pete
Ground level on both boards is zero volts. While measuring the ground level at the ground connection to R602, I also measured the junction between R602 and R601. There are 600mV at that junction on the bad board and only 70mV on the good board. That's a huge difference. I double checked R602 thinking it would have excessive resistance but it's at 100k which is what is called for in the schematic and what it's mate on the good board measures.
It just occurred to me that if that resistor is good and the voltage at the R601/2 junction is high, that supports your theory of a high ground level but it measured zero volts. I'm going to try supplying a separate ground to see if that changes anything.
The voltage levels on the driver transistors do add up to around a volt but there is a big discrepancy between the good and bad boards. The good board they are both around 0.5V. That seems balanced. The bad board the positive rail emitter is about 4x the negative rail emitter. That doesn't seem right to me, why so great a difference?
- Pete
gslikker
Super Member
Out of ideas again, my ideas were just not to find the culprit but to exclude possibilities hoping there is something left.....
Especially, since everything is normal after some minutes, which excludes any hard fault.
BTW the very REASON I asked to measure incoming groung also, was that according the picture, it looks soldered to a wire, which would mean that the fault does not 100% move with the boads swap, as the incoming stuff stays connected.
Especially, since everything is normal after some minutes, which excludes any hard fault.
BTW the very REASON I asked to measure incoming groung also, was that according the picture, it looks soldered to a wire, which would mean that the fault does not 100% move with the boads swap, as the incoming stuff stays connected.
Last edited:
petehall347
the brandy coffee man
are the bases equal voltages ?
elnaldo
Lunatic Member
I'd disconnect the signal input and short the input to ground to discard anything from a previous stage (confirm with a schematic, but shorting the input is usually OK)
I've seen weird behaviors like this caused by: a transistor backwards, a PNP instead of NPN, a leaky capacitor or diode, conductive glue between components, wrong value replacement resistors.
I've seen weird behaviors like this caused by: a transistor backwards, a PNP instead of NPN, a leaky capacitor or diode, conductive glue between components, wrong value replacement resistors.
The issue occurs on the bad board regardless of which input (left or right) is connected. The good board is good regardless of which input is connected.
This issue is heat sensitive. The wrong transistor or resistor would be hard faults that wouldn't change as the board heats up. The capacitors and diodes have been replaced and there is no glue.
The base voltages correspond to the emitter voltages. TR609 has the high emitter voltage and it also has a high base voltage. TR608 has the low emitter voltage and the base voltage is low as well.
I appreciate your help. Often just putting a response in words helps me better understand what's happening
I haven't worked on it for the past couple days. The symptoms have actually changed a bit. The last work involved supplying an auxiliary ground to the junction of R607 and R608 to ensure the board was getting a proper ground. That appear to drop the offset voltage to under 10mV. I replaced R607 and the offset stayed low for as long as I played it. The next day it was back to it's old tricks, high offset. I didn't touch it yesterday and today when I turned it on for the first time, the offset quickly dropped to under 10mV. It's never done that before on first start of the day. I turned it off and let it sit for awhile then turned it on again and the offset was still low. The third time I tried it, the offset stayed above 200mV or so.
Temperatures for the past couple of days have been in the upper 90s (35C) which is about 10 degrees warmer than the last time I worked on it. That may have something to do with the change in symptom. It may not be cooling off enough to cause the issue.
- Pete
In an act of desperation I've replaced every component on the board. All the semiconductors, all the electrolytics all the ceramic capacitors and all the resistors. Still the same problem, high DC offset when cold. Normal operation when warm. Part of the process of replacing the remaining parts was to ensure good solder joints on those components. Today I removed the solder from the connector pegs and resoldered them. Still does the same thing. Every solder joint has now been re-done and every component replaced. I can't see anything wrong with this board. Maybe I need to make a new one.
- Pete
- Pete
gslikker
Super Member
The only thing you did not replace, is the board indeed.
I had a problem with an old single opamp board once, it turned to be hygroscopic and although normal resistance values were used, the opamp started acting weird after a few months. Replacing the opamp apparently heated up the environment by soldering and the problem was gone for months.
I was about to suggest put SK10 solderable laquer on the pcb, but the hygroscopic problem was within months not minutes.
I could not figure out but some guy at work did the suggestion to seal that board (using professional stuff) and it worked.
Still, you could have a bad trace on the board, not apparent when measuring.
gslikker
Super Member
Suggest solder two wires on a track to check, run low voltage high current (1 Amp or so) through the traces, and measure remaining voltage over the tracks, I know it is time consuming doing all tracks.
Maybe even bend the board a little and see if voltage goes up (or the trace burns
).
One day I found a problem such way. Of course, best to be done with an empty circuit card for components
(of course, only to be done if desparate....)
Maybe even bend the board a little and see if voltage goes up (or the trace burns
).One day I found a problem such way. Of course, best to be done with an empty circuit card for components
(of course, only to be done if desparate....)
Last edited: