Ok, I'm back at it. I replaced Q112 (with the KSC3503DS) along with D105 in both channels. When I switch on the amp, the DBT works as it has been, lighting and then dimming within a few seconds. Then, I hear a relay click on and power is applied to the power amp boards. Both overload LEDs are still lit. Measuring at TP12, I get:
TP12R: 78.9v
TP12L: 77.6v
So, more or less the same condition. Ideas of what the check next?
TP12R: 78.9v
TP12L: 77.6v
So, more or less the same condition. Ideas of what the check next?
Well bummer, I was hoping you'd at least see some difference in your symptoms.
Measure the collector of Q110 and collector of Q112 in ref to Gnd. According to the schematics this should be +2V and -2V respectively.
You know now that I think about it what you should do is to take a print out of the front end of the circuit and take a mass of voltage readings and write down what you're measuring next to the value that Nak has in the schematic. Then post a decent pic of that drawing with your readings or use MS paint and add text to a screen grab of the front end circuitry if your familiar with using that to edit photos. I'd at least do this for the second diff amp, consisting of Q108,109,110,111 & 112. This may help to more quickly identify where the problem may be and it may be easier than going back and forth to take the readings.
Likewise you may only need to do it for one channel because they're both exhibiting the same symptoms its likely once we figure out what failed in one channel the other will fall into place.
It would sure be nice to disconnect the signal to the output section of the circuit from the input stage but it looks like the board would not allow that. Unless there's a set of jumpers somewhere that take the signal from Q114/115 over to Q116/117. I see that there are a few jumpers on the board.
Measure the collector of Q110 and collector of Q112 in ref to Gnd. According to the schematics this should be +2V and -2V respectively.
You know now that I think about it what you should do is to take a print out of the front end of the circuit and take a mass of voltage readings and write down what you're measuring next to the value that Nak has in the schematic. Then post a decent pic of that drawing with your readings or use MS paint and add text to a screen grab of the front end circuitry if your familiar with using that to edit photos. I'd at least do this for the second diff amp, consisting of Q108,109,110,111 & 112. This may help to more quickly identify where the problem may be and it may be easier than going back and forth to take the readings.
Likewise you may only need to do it for one channel because they're both exhibiting the same symptoms its likely once we figure out what failed in one channel the other will fall into place.
It would sure be nice to disconnect the signal to the output section of the circuit from the input stage but it looks like the board would not allow that. Unless there's a set of jumpers somewhere that take the signal from Q114/115 over to Q116/117. I see that there are a few jumpers on the board.
I made the measurements on the right channel and noted the voltages I measured next to the voltages called out in the schematic. One thing that was very strange to me is that when I attempted to measure the voltage on the pin of Q101R near R102R, it trigger the DBT. So naturally I thought I shorted something. But then I repeated the measurement at C101R and R102R and it triggered the DBT every time. Ground on the meter was connected to chassis with an alligator clip, and I was just measuring DC voltage.
Annotated right channel schematic here:
Nakamichi PA-7AII - Google Photos
I couldn't find any obvious jumpers to remove between Q114/115 and Q116/177. There may be some, but I'd probably have to remove the PCB from the heat sink again.
Annotated right channel schematic here:
Nakamichi PA-7AII - Google Photos
I couldn't find any obvious jumpers to remove between Q114/115 and Q116/177. There may be some, but I'd probably have to remove the PCB from the heat sink again.
So a few things stand out to me.
The emitter of Q110 should match the emitter of Q111. The bases of both of them are both referenced off the same divider network of R109, ZD12, & R111. You also only have a 0.1V drop across R109 (75r) that should be closer to .5 - .6V so very little current is flowing through there or the resistance shifted down drastically. I'd check that one. There is 3v across Zener ZD12 like there should be and you checked R111 before and found it to be in spec. Q110 could also be bad.
Also the 79.2V at the collector of the current source Q106 is not right. The base voltage on that one should be about 1.2 V closer to zero than the full neg rail so about -78.5V. Check Q106 and R105.
The emitter of Q110 should match the emitter of Q111. The bases of both of them are both referenced off the same divider network of R109, ZD12, & R111. You also only have a 0.1V drop across R109 (75r) that should be closer to .5 - .6V so very little current is flowing through there or the resistance shifted down drastically. I'd check that one. There is 3v across Zener ZD12 like there should be and you checked R111 before and found it to be in spec. Q110 could also be bad.
Also the 79.2V at the collector of the current source Q106 is not right. The base voltage on that one should be about 1.2 V closer to zero than the full neg rail so about -78.5V. Check Q106 and R105.
I looked over your picture of measurements again while drinking coffee this morning and you should definitely look at that Q106. The fact you have -79.2 from E to C on that transistor sure looks like its dead. This assumes that you have the required 1.2V drop across D103/D104. You could replace this Q106 current source with another C3503 if you bought a few spares.
I measured R109 in-circuit, and it's 76 ohms vs a spec of 75. I just remeasured and got 78.9 at the rail and 78.9 on the other side of the resistor. So, no drop, but it measures ok, so I guess as you say there's no current flowing through there.
Sounds like I should probably pull both Q110 and Q106 to measure them. I have 2 spare C3503. Are they suitable replacements for Q106?
I get about 0.6v across each of D103 and D104, from -79.0v on one side of the pair to -77.8v on the other.
I'll pull those two transistors early this week. Wiped out from yard work today.
Sounds like I should probably pull both Q110 and Q106 to measure them. I have 2 spare C3503. Are they suitable replacements for Q106?
I get about 0.6v across each of D103 and D104, from -79.0v on one side of the pair to -77.8v on the other.
I'll pull those two transistors early this week. Wiped out from yard work today.
I'd concentrate on Q106 and yes the C3503 will work in that position. Something has got the front end of the circuit shut down based on the voltage drops across R103/R104 and it looks like the Q106 current source is probably the culprit. The second diff stage may come up once the input LTP starts working properly.
I pulled and tested Q106 from both channels and unfortunately they both test out ok:
Q106L (NPN):
Resistance Check
B to C: 3.5M
B to E: 6.1M
C to E: OL
Diode Check
B to E: 0.685v ok
B to C: 0.675v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
Q106R (NPN):
Resistance Check
B to C: 3.3M
B to E: 6.4M
C to E: OL
Diode Check
B to E: 0.675v ok
B to C: 0.662v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
Should I pull and check Q110?
Q106L (NPN):
Resistance Check
B to C: 3.5M
B to E: 6.1M
C to E: OL
Diode Check
B to E: 0.685v ok
B to C: 0.675v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
Q106R (NPN):
Resistance Check
B to C: 3.3M
B to E: 6.4M
C to E: OL
Diode Check
B to E: 0.675v ok
B to C: 0.662v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
Should I pull and check Q110?
I just took another look at the schematic and realized I’ve been making a basic mistake in my test setup. This whole time, I’ve been leaving the input connectors disconnected from the PCBs. As a result, the part of the circuit near Q101 has no ground. I guess I was assuming that that part of the circuit would have been connected to the ground that’s used elsewhere on the PCB. Maybe they were worried about ground loops? Sorry for wasting your time, that’s a stupid mistake on my part.
I really wish Nakamichi had make it easier to install and reinstall the PCBs from the chassis. As it is, I have to solder and desoldering the smaller chassis wire and the input connector wires, which are really short and limit the kind of testing you can do.
I’ll give a shot at reinstalling the boards tomorrow, this time with the input wires connected.
I really wish Nakamichi had make it easier to install and reinstall the PCBs from the chassis. As it is, I have to solder and desoldering the smaller chassis wire and the input connector wires, which are really short and limit the kind of testing you can do.
I’ll give a shot at reinstalling the boards tomorrow, this time with the input wires connected.
Its not a waste of time if it corrects the problem that would be great. However looking at the schematic if its the RCA ground connection that you're talking about that's not going to do anything for you from what I can tell.
The whole front end of this circuit is run from pos to neg rail, there is no path to ground in operating the front end LTP. I might be wrong so check it and see, but I don't think so.
I don't know if you could splice some temporary longer wires in to allow you to leave the channel module sitting out. It might save you some time in troubleshooting the issue.
The whole front end of this circuit is run from pos to neg rail, there is no path to ground in operating the front end LTP. I might be wrong so check it and see, but I don't think so.
I don't know if you could splice some temporary longer wires in to allow you to leave the channel module sitting out. It might save you some time in troubleshooting the issue.
I was thinking that, with the ground on the RCA cable disconnected, the gate of that JFET would be floating. With the ground connected, R101 and R102 would act as a weak (51.3k) pulldown.
Recall that when I tried to measure that 0v point on the gate of Q101, it triggered the DBT. I think that's because the gate was not longer floating once I probed it. But, by the logic, I guess that connecting the RCA ground might not have the effect I hope it will.
I should try splicing in a temporary wire like you say. I'll give it a shot tonight.
Recall that when I tried to measure that 0v point on the gate of Q101, it triggered the DBT. I think that's because the gate was not longer floating once I probed it. But, by the logic, I guess that connecting the RCA ground might not have the effect I hope it will.
I should try splicing in a temporary wire like you say. I'll give it a shot tonight.
R106L: 0.999k
R106R: 1.002k
Looks ok
For the JFETs, I went by this guide Meter Check of a Transistor (JFET) | Junction Field-effect Transistors | Electronics Textbook
Measurements are with the parts removed.
Q101L (n-channel):
Pins 1-3:
G to D: 7.6M not ok?
D to G: OL ok?
G to S: 7.6M not ok?
S to G: OL ok?
Pins 4-6:
G to D: 7.7M not ok?
D to G: OL ok?
G to S: 7.6M not ok?
S to G: OL ok?
Q101R (n-channel):
Pins 1-3:
G to D: 7.1M not ok?
D to G: OL ok?
G to S: 7.5M not ok?
S to G: OL ok?
Pins 4-6:
G to D: 7.5M not ok?
D to G: OL ok?
G to S: 7.4M not ok?
S to G: OL ok?
Please double check my interpretation of the JFET test!
R106R: 1.002k
Looks ok
For the JFETs, I went by this guide Meter Check of a Transistor (JFET) | Junction Field-effect Transistors | Electronics Textbook
Measurements are with the parts removed.
Q101L (n-channel):
Pins 1-3:
G to D: 7.6M not ok?
D to G: OL ok?
G to S: 7.6M not ok?
S to G: OL ok?
Pins 4-6:
G to D: 7.7M not ok?
D to G: OL ok?
G to S: 7.6M not ok?
S to G: OL ok?
Q101R (n-channel):
Pins 1-3:
G to D: 7.1M not ok?
D to G: OL ok?
G to S: 7.5M not ok?
S to G: OL ok?
Pins 4-6:
G to D: 7.5M not ok?
D to G: OL ok?
G to S: 7.4M not ok?
S to G: OL ok?
Please double check my interpretation of the JFET test!
Alright your SK240 contains a pair of matched 2SK170's so I took one of my 2SK170 Jfet's and did the same measurements. I also checked my Jfet on my transistor tester to make sure it showed up as a functional N-type Jfet which it did.
In effect these readings look the same as yours so I do not think that Q101 is a problem.
G to D 4.5M
D to G OL
G to S 4.5M
S to G OL
I think that the full neg rail coming in on the feedback side of Q101 is whats really causing havoc with all of the front end measurements.
I know it'll be a hassle but I'd remove Q116,117,118,119. This will eliminate the drivers as the source of the full neg rail. If the full neg rail goes away then Q117 or 119 is bad. If not then we'll look upstream.
In effect these readings look the same as yours so I do not think that Q101 is a problem.
G to D 4.5M
D to G OL
G to S 4.5M
S to G OL
I think that the full neg rail coming in on the feedback side of Q101 is whats really causing havoc with all of the front end measurements.
I know it'll be a hassle but I'd remove Q116,117,118,119. This will eliminate the drivers as the source of the full neg rail. If the full neg rail goes away then Q117 or 119 is bad. If not then we'll look upstream.
You know what, if there's one thing I have right now, it's time! I'll pull Q116-119 and report back.
Thanks for testing your 2SK170 and comparing to my results, that makes me feel much better. I ordered one of those transistor testers you linked, received it later today, and tested my 2SK240s, and all four transistors (2/package) appear to be recognized as N-type JFETs, so it seems as if they're good.
Thanks for testing your 2SK170 and comparing to my results, that makes me feel much better. I ordered one of those transistor testers you linked, received it later today, and tested my 2SK240s, and all four transistors (2/package) appear to be recognized as N-type JFETs, so it seems as if they're good.
I removed Q116-199 from both channels and all looks nominal.
Q116L: 2SC3856 (NPN)
Resistance Check
B to C: 2.5M
B to E: 2.6M
C to E: OL
Diode Check
B to E: 0.513v ok
B to C: 0.508v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q117L: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.524v ok
C to B: 0.521v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
Q118L: 2SC3856 (NPN)
Resistance Check
B to C: 2.5M
B to E: 2.6M
C to E: OL
Diode Check
B to E: 0.518v ok
B to C: 0.505v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q119L: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.515v ok
C to B: 0.509v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
----------------------------------
Q116R: 2SC3856 (NPN)
Resistance Check
B to C: 2.5M
B to E: 2.7M
C to E: OL
Diode Check
B to E: 0.520v ok
B to C: 0.514v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q117R: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.523v ok
C to B: 0.518v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
Q118R: 2SC3856 (NPN)
Resistance Check
B to C: 2.4M
B to E: 2.5M
C to E: OL
Diode Check
B to E: 0.510v ok
B to C: 0.506v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q119R: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.509v ok
C to B: 0.506v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
After I reinstalled the transistors, I measured pin to pin on the transistors to verify I didn't short anything. On the right channel, Q119 B to C and Q117 B to C were 1 ohm, and were around 2M on the left channel. I also measured R125R in-circuit and found 1 ohm rather than the expected 3 ohms.
With Q117R and Q119R removed, the PCB pads for both locations measured 2 ohms B to C. I haven't found the source of the short yet, at least there's nothing obvious. I'll keep looking.
Q116L: 2SC3856 (NPN)
Resistance Check
B to C: 2.5M
B to E: 2.6M
C to E: OL
Diode Check
B to E: 0.513v ok
B to C: 0.508v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q117L: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.524v ok
C to B: 0.521v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
Q118L: 2SC3856 (NPN)
Resistance Check
B to C: 2.5M
B to E: 2.6M
C to E: OL
Diode Check
B to E: 0.518v ok
B to C: 0.505v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q119L: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.515v ok
C to B: 0.509v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
----------------------------------
Q116R: 2SC3856 (NPN)
Resistance Check
B to C: 2.5M
B to E: 2.7M
C to E: OL
Diode Check
B to E: 0.520v ok
B to C: 0.514v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q117R: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.523v ok
C to B: 0.518v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
Q118R: 2SC3856 (NPN)
Resistance Check
B to C: 2.4M
B to E: 2.5M
C to E: OL
Diode Check
B to E: 0.510v ok
B to C: 0.506v ok
E to B: OL ok
C to B: OL ok
C to E: OL ok
E to C: OL ok
transistor tester says BJT-NPN
Q119R: 2SA1492 (PNP)
Resistance Check
B to C: OL
B to E: OL
C to E: OL
Diode Check
B to E: OL ok
B to C: OL ok
E to B: 0.509v ok
C to B: 0.506v ok
C to E: OL both directions ok
transistor tester says BJT-PNP
After I reinstalled the transistors, I measured pin to pin on the transistors to verify I didn't short anything. On the right channel, Q119 B to C and Q117 B to C were 1 ohm, and were around 2M on the left channel. I also measured R125R in-circuit and found 1 ohm rather than the expected 3 ohms.
With Q117R and Q119R removed, the PCB pads for both locations measured 2 ohms B to C. I haven't found the source of the short yet, at least there's nothing obvious. I'll keep looking.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.