I have a weird issue going on with this amp.
1. The clean channel plays fine, the distortion channel cuts in/out. If I hit a note it comes back, and then at some point it cuts out, instead of fading out.
2. The 15V rail is low, only 11V. This may or may not be related to the first issue.
Issue 1 seems to be happening around Q5. OK before Q5 but cuts in/out at the Q5 Source. Q5 Gate is around 0.5V and then at 8.5V when distortion channel is switched on.
Issue 2, I have the proper 21V at TP17, but only 11V at TP19 (should be 14V). Something is pulling it down. Maybe C47 is leaking, or D18 zener is bad. All op-amps should have 14V at pin 8 but are actually 10.5V.
I would like to focus on the rail issue first, and once corrected see if issue 1 is resolved also.
Suggestions please!!!
Schematic link is https://www.audioservicemanuals.com/c/crate/crate-bt/7294892-crate-bt-50-pwa-bass-amplifier-pcb-schematics-848sch-a
I am not sure how to attach a PDF to this thread.
Also, anyone know if Music Electronics Forum is ever coming back online?
Thank you, Gonflable
You would have 45.45mA going through R42 to drop 10 volts between TP17 & TP19.
Where is it being drained too ?
Measure the voltage drop across R45 and R82 to see how much current is going through them to find out. (amps = volts / resistance)
If the current through R45 and R82 adds up to 45.45mA then the circuit is using it somewhere and you have to follow the path it takes.
If the current does not add up to 45.45mA then I would suspect the zener diode D18 is faulty.
I can't see anything wrong with those readings on Q5 if your 15v has dropped to 11v, TP15 will be low if the supply rail has dropped.
I can't find any pictures of that amp on the net, any chance of taking some photos of the insides while you have it open ?
There is apparently a noise gate circuit in the distortion channel that needs the clean channels Level set to more then 0 to operate according to the user manual.
I use a very strong magnifying lens to look for bad solder joints on older amps, look for tiny cracks on every solder joint and then I re-flow any bad or suspicious one.
Intermittent faults are often caused by bad solder joints on older amps.
Voltage drops
R82 0.55 V drop, 47 ohms = .011 A
R45 1.3 V drop, 47 ohms = .027 A
R42 8.6 V drop, 220 ohms = .039 A
So R82 and R45 add up to .038 A or 38 mA?
Trying to figure out how to attach pics.
Thank you
Did you verify the resistance of R42 and check it's solder joints?
try this link for photos https://photos.app.goo.gl/VEjDH76nUFq6TubS7
Quote from: g1 on June 08, 2023, 12:31:58 PMDid you verify the resistance of R42 and check it's solder joints?
Yes, all resistors in that area were checked and are in spec. Solder joints look good. I added pics of the underside at https://photos.app.goo.gl/VEjDH76nUFq6TubS7
I had replaced the input jack that was broken, cleaned all pots, and reflowed the 3 distortion channel pots.
Maybe it is Q4 that is responsible for Q5 source drop-out.
Looks like a limiter circuit consisting of Q4,Q5, U8 and U12.
Also, U8 and U12 are the only things running off 15VX (via R82), so maybe one of those IC's could also be responsible for the +15V rail issue. Do either feel hot?
(edit: that would be the noise gate circuit that tassieviking mentioned, not a limiter)
The noise gate is shown on page 7 of the user manual in the block diagram, I'm not smart enough to know how that works. =)
Good photos Gonflable, I like being able to see what it looks like and sometimes a photo shows bad solder joints that we miss and others notice.
Something is not adding up with the voltages, TP17=21v, Vdrop on R42=8.6v, TP19=11v.
We seem to have lost 1.4v in those measurements.
Could you please measure TP18, TP20, Vdrop on R43, R46, R85 just for reference.
It would be interesting to compare the positive rail and the negative rail current drain.
I just noticed that the tuner and foot-switch circuit are all run of the +15v rail as well, this would add further load on the +15v rail compared to the -15v rail.
Just the distortion red LED would add more voltage drop to the +15v compared to the -15v rail.
Led = approx 8.3mA, 0.0083mA x 220R = 1.826v in R42, 8.3mA 47R = 0.39v in R45.
Surely the engineers who designed the amp would have realized this and made slight differences to the design.
Maybe I'm on the wrong path there but I would be wondering about tweaking the values of R42 to maybe 150R, but I would do it in small steps.
D18 is a 15v 1W zener diode, but it can't do anything if the voltage drop across R42 is less then 6v, it needs more then 15v to start conducting as far as I know.
Measure the -15v rail first and we will start to work it out, after all I could be completely wrong.
What does the X designate in "15VX"?
The tuner is not in circuit during bench testing.
TP17 20.7
TP18 -20.6
TP19 11.7
TP20 -14.8
I see that D18 and R42 are operating much hotter than D19 and R43
Temperature
D18 100F
D19 85F
Is 100F too hot?
I have new zeners arriving tonight.
Thank you
What does the X designate in "15VX"?
The tuner is not in circuit during bench testing.
Zeners arrived and swapped out. No changes. Updated readings...
Same temp readings with D18 hotter than D19
D18 100 Faren
D19 85 Faren
Negative Rail...
TP18 -20.5 V
TP20 -15.4 V
R43 5 V drop
R85 126 mV drop
R46 0.78 V drop
Positive Rail...
TP17 20.5 V
TP19 12 V
R42 8.8 V drop
R82 0.53 V drop
R45 1.37 V drop
Zener D19 is up alongside R45, while D18 is beside R82.
R45 runs hotter, so that may be why D19 is hotter.
See post #7 about 15VX supplies.
Is there much difference in voltage across R45 when ch.2 is selected compared to ch.1?
What is the voltage across R86 (to tuner) ?
layout diagram attached
PWA Bass Amplifier PCB Layout (848PWA_A).pdf
The 1N4744 diodes are ok at up to 175C (347F) temperature, but the power handling will be severely reduced.
At 100C (212F) the power rating drops in half to 0.5 watts (boiling water hot).
At 100F they are still working at 100% capacity, that's not even warm for a semiconductor.
The X in VX means nothing as far as I know, maybe the designer picked X because it feeds the effects (FX in shorthand) circuit. (Noisegate)
I'm guessing the CD4013 flip flop chip might be noisy when operating so they made a separate supply for it to reduce noise in the rest of the amp.
I think when the signal into U8 drops below a certain level Q4 and Q7 turn on, Q4 shorts the signal going into the octave unit and Q7 drops the gain in U6:B, ;that drops the signal right off.
Play with the clean channel gain when using the distortion channel and see if it affects the cut out level.
If I am wrong about that then I hope someone will jump in and correct me.
(I'm old school, I learn better if I get a clip over the ears so don't hold back).
I am wondering where you are picking up the test points, I can't see them on the layout.
Test point TP17 to TP19 voltage don't exactly match R42 voltage, but it should.
If you are measuring TP17 far away from R42 then I guess you could loose some voltage in the tracks but with only 34mA it seems strange to me.
Could there be some bad solder joints on JW7 or JW32 ? Maybe it is just the track loss but I would want to know if it was my amp. Measure voltage drop from D14 to R42 leads to find out.
G1 will lead you in the right path as I think he knows a lot more then me.
Cheers
Mick
Quote from: g1 on June 09, 2023, 10:24:56 PMIs there much difference in voltage across R45 when ch.2 is selected compared to ch.1?
What is the voltage across R86 (to tuner) ?
across R86 1 mV. note that the tuner is not connected at this time.
across R45 V dist. ch. selected = 1.36V
across R45 V clean ch. selected = 1.2V
across R42 V dist. ch. selected = 8.9V
across R42 V clean ch. selected = 8.4V
TP17 and TP19 measured on either side of R42
TP18 and TP20 measured on either side of R43
Play with the clean channel gain when using the distortion channel and see if it affects the cut out level.
[/quote]
Yes it does effect cut out level. As I raise the clean channel gain, it takes longer to cut out. Even with clean channel gain dimed, it still cuts out, but takes longer to happen.
Also, I realize now that I can cause a note cut out by lowering the clean channel gain knob, and make the note cut back in by raising the gain knob.
And, it could be my imagination but, it seems more likely to cut out on higher freq. notes versus lower notes.
As you have both indicated, it looks like the issue is in this circuit. I feel that we are getting closer!
Quote from: g1 on June 08, 2023, 06:45:42 PMMaybe it is Q4 that is responsible for Q5 source drop-out.
Looks like a limiter circuit consisting of Q4,Q5, U8 and U12.
Also, U8 and U12 are the only things running off 15VX (via R82), so maybe one of those IC's could also be responsible for the +15V rail issue. Do either feel hot?
(edit: that would be the noise gate circuit that tassieviking mentioned, not a limiter)
Q4,Q5, U8 and U12 are all running cool. 65 F
Trying to understand the circuit and figure out where I could run a jumper to bypass the "limiter" feature without losing the "octave" feature.
I am thinking in the Q4 area?
Thoughts? Thank you
I removed Q4 from the circuit. Amp works fine now and I do not care to fix the limiter feature.
I bench tested Q4 and found it to be open (not turning on).
My positive rail is now at 12.5 V, still below spec but the amp seems to work fine and sounds great.
Thank you for all the help.
Gonflable
Q4 was the correct part to remove in order to disable the noise gate.
Glad you got it sounding better.
Q4 would not test like a usual transistor as it is a FET. It must have been turning on as the noise-gate was doing something.
Did you ever try the octave feature to see if it is working? That is the only other thing I can think of that might be dragging down the +14V line. U8 and U12 are for the octave circuit, I was mistaken when I earlier said they were for the noise gate/limiter.
Yes the octave circuit is working great. Nice feature on this amp.
Thank you
I tested Q4 as a FET switch. Drain and source were always open, even after charging the gate. But I would imagine it was partially working in circuit as you stated.