Fender Bassman 50 AA371 Silverface 70s Noise

[fenderguy81]

Solved

[pdf64]

What are your other valve amps?
Need to compare like with like, eg both non master volume.
It looks like both input stages have bypass caps on their anode load resistors, are they both still fitted?
https://el34world.com/charts/Schematics/files/Fender/Fender_bassman_aa371_schem.pdf

[fenderguy81]

Solved

[Latole]

This Bassman 50 wiring it's a rat nest ; wiring dress are a must. Picture:

Many wires are too long and and not arranged correctly.

Look at the wiring in you '64 DR and your '70 TR . They act as an antenna capturing the noise. Compare with your other amps, you will understand

[pdf64]

Are the shorting switches on the input jacks working properly?

Does the noise level decrease if the 68k input resistors have a shorting link clipped across them, ie so 68k become 0?

[Latole]

Are the shorting switches on the input jacks working properly?

Does the noise level decrease if the 68k input resistors have a shorting link clipped across them, ie so 68k become 0?

+1000 shorting contacts are often dirty or open.

[fenderguy81]

Thanks, I know these 70s Fenders have a lot of extra cable and poorer lead dressing. I’ve tried moving leads around, crossing at 90 degree angles, etc., but I haven’t ventured to shortening them yet. Perhaps I’ll do that, redo the solder joints, and clean the board with isopropyl alcohol while I’m at it.

The input jack shunts have been cleaned with a burnishing tool and make good contact.

I’ve also lifted one end of each coupling cap and checked for DC leaking. No DC on the board either. I’ve tried freeze spraying individual components with no success. A tap test showed that one of the disc caps on the first channel is microphonic. I’ve swapped it out and ended up putting it back in because it made no difference. Also, I forgot to mention all the resistors in the dog house were replaced with metal oxide.

I’ve read in another post that someone with the same amp and similar problems had a bad output transformer. I wish I had one on hand to substitute in. Does anyone know of an easy way to test the output transformer? I should note that the previous owner had run it with an 8 ohm cab for years (the amp’s rated at 4 ohms). It’s my understanding that this won’t necessarily damage the output transformer—it just will have less power/wattage. Trying to run a lower impedance rating (e.g. 2 ohms) can damage it.

[Latole]

 " I should note that the previous owner had run it with an 8 ohm cab for years (the amp’s rated at 4 ohms). It’s my understanding that this won’t necessarily damage the output transformer—it just will have less power/wattage. "
-fenderguy81

IMO it not very good to run at higher speaker impedance. Is output stage may see like a open circuit ?

______________

In other way, with a bad OT, amp sound will be not good, you may have distortion.
Noisy amp ; what kind of noise ? Hard to say in a forum

[pdf64]

If the amp can get loud and clean, keep up with a drummer, then the OT is fine.
Given the symptoms, the OT can’t feasibly cause the hiss.

[fenderguy81]

I'm using it with a 4 ohm load. The amp gets loud enough to play with a drummer, but there is not a lot of headroom. I'll have to test it again when I'm home, but I remember the signal being a bit distorted.

[bmccowan]

I have one of those. I rewired it to a different Bassman circuit, but do not remember it being overly noisy prior to the rewire. Rabbit hunting is frustrating so when I have this issue I bite the bullet and hook up a listening amp & probe. Its easy with a noise issue because you do not need to generate a signal, the noise is doing that for you. Probe down through the signal path and you at least find the locale where the noise is coming from.

[pdf64]

Are the 220k feedback circuit resistors at the 3rd stage, and the 47k feedback resistor on the power amp, fitted and ok?
How about the 68k input resistor shorting out test?

[fenderguy81]

I have one of those. I rewired it to a different Bassman circuit, but do not remember it being overly noisy prior to the rewire. Rabbit hunting is frustrating so when I have this issue I bite the bullet and hook up a listening amp & probe. Its easy with a noise issue because you do not need to generate a signal, the noise is doing that for you. Probe down through the signal path and you at least find the locale where the noise is coming from.

I have an oscilloscope, but to be honest I do not feel comfortable using it yet until I do a bit more research. I just got an isolation transformer and I might try testing it out on a working Vibro Champ to get a bit more comfortable with its function. How do you go about using the listening amp and probe?

Are the 220k feedback circuit resistors at the 3rd stage, and the 47k feedback resistor on the power amp, fitted and ok?
How about the 68k input resistor shorting out test?

I'll check the fit of those resistors when I get home. I think they are fitted ok, but I'll double check. Also, the input resistor and shorting jacks are good. I used a burnishing tool to clean them and had tested them with a DMM with a 1/4" plug.

[pdf64]

…the input resistor … tested them with a DMM with a 1/4" plug.

The test I’m suggesting is to short them out and review the noise level. What can a DMM tell you about the noise they’re generating?

[fenderguy81]

…the input resistor … tested them with a DMM with a 1/4" plug.

The test I’m suggesting is to short them out and review the noise level. What can a DMM tell you about the noise they’re generating?

It can show the resistance when the jack is open vs close. Sorry if I wasn't clear-- I did also manually open and close the shunts on each input jack and the noise did not change.

[bmccowan]

Listening amp - Doug has good instructions here. https://el34world.com/Hoffman/tools.htm  It is just a probe that plugs into another amp. Very important to have a blocking capacitor in the probe so that if you touch a DC high voltage connection you block that voltage from getting to you or the other amp. For noise you need no input signal. For other issues I use an iPad with a tone generator app to send a signal into the amp. Saves you from reaching over and hitting the strings on your guitar. Or employing a monkey and organ grinder.

[pdf64]

…the input resistor … tested them with a DMM with a 1/4" plug.

The test I’m suggesting is to short them out and review the noise level. What can a DMM tell you about the noise they’re generating?

It can show the resistance when the jack is open vs close. Sorry if I wasn't clear-- I did also manually open and close the shunts on each input jack and the noise did not change.

That’s something different to what I’m suggesting.
Resistors can be noisy yet within their resistance value tolerance.
Does the noise level decrease if the 68k input resistors have a shorting link clipped across them, ie so 68k become 0?

[Latole]

 "  When the amp is on and all of the knobs are off the amp is quiet "

-Federguy81

Which seems to mean that the problem is in the circuitry before the volume pots of each channel.
You must work there.

All the electrolytic capacitors have been replaced; the power supply ones I suppose?
Are they good ? How many volts AC (yes AC, not DC) do you measure on the plates of V1 and V2

[fenderguy81]

Listening amp - Doug has good instructions here. https://el34world.com/Hoffman/tools.htm  It is just a probe that plugs into another amp. Very important to have a blocking capacitor in the probe so that if you touch a DC high voltage connection you block that voltage from getting to you or the other amp. For noise you need no input signal. For other issues I use an iPad with a tone generator app to send a signal into the amp. Saves you from reaching over and hitting the strings on your guitar. Or employing a monkey and organ grinder.

Thanks, I'll try making one when I have some down time. That seems like a really useful tool to have on the bench!

…the input resistor … tested them with a DMM with a 1/4" plug.

The test I’m suggesting is to short them out and review the noise level. What can a DMM tell you about the noise they’re generating?

It can show the resistance when the jack is open vs close. Sorry if I wasn't clear-- I did also manually open and close the shunts on each input jack and the noise did not change.

That’s something different to what I’m suggesting.
Resistors can be noisy yet within their resistance value tolerance.
Does the noise level decrease if the 68k input resistors have a shorting link clipped across them, ie so 68k become 0?

I clipped an alligator clip on the 68k resistors across to the chassis ground and it did not make any audible difference in the noise floor.

[pdf64]

…
I clipped an alligator clip on the 68k resistors across to the chassis ground and it did not make any audible difference in the noise floor.

Which lead of the 68k did you clip to? ie the grid end or the jack tip terminal end?
The end result of the test needs to be to connect the grid to 0V chassis common, as opposed to having the grid stoppers to be in circuit.

[fenderguy81]

"  When the amp is on and all of the knobs are off the amp is quiet "

-Federguy81

Which seems to mean that the problem is in the circuitry before the volume pots of each channel.
You must work there.

All the electrolytic capacitors have been replaced; the power supply ones I suppose?
Are they good ? How many volts AC (yes AC, not DC) do you measure on the plates of V1 and V2

Yes, all electrolytic caps have been replaced (filter caps, bias cap, cathode bypass caps)

Here are the AC Plate voltages

(Channel 1)
V1A .567 VAC
V1B .003VAC

V2A N/A
V2B .006VAC

(Channel 2 Vibrato)
V3A .0025 VAC
V3B .0074 VAC

…
I clipped an alligator clip on the 68k resistors across to the chassis ground and it did not make any audible difference in the noise floor.

Which lead of the 68k did you clip to? ie the grid end or the jack tip terminal end?
The end result of the test needs to be to connect the grid to 0V chassis common, as opposed to having the grid stoppers to be in circuit.

I will redo the test today when I get home to ensure I am shorting the grid end to ground.

[Latole]

AC are low ; filter caps are good

[shooter]

V1A .567 VAC

was your #'s WITHOUT a input signal?
what did you use to measure the AC volts?


.5VAC with no signal is a lot for V1A's plate.  If you used a scope, what is the frequency?

[Grees01]

My understanding is that V1 plate voltage is coming from the last node in the power supply chain and should be pretty clean dc voltage, thru all the filter caps, and resistors. 1/2 volt ac on the plate? With no input signal? Doesn't sound right to me. Don't take anything I say as being accurate. I'm just thinking out loud with my very limited knowledge.
If I'm incorrect in my statement please correct me.
G.

[valvetones]

“ I should note that the previous owner had run it with an 8 ohm cab for years (the amp’s rated at 4 ohms). It’s my understanding that this won’t necessarily damage the output transformer—it just will have less power/wattage. Trying to run a lower impedance rating (e.g. 2 ohms) can damage it.“

You are correct, 8 ohm load would double the OT primary reflected impedance and would not be damaging the amp.

[pdf64]

…8 ohm load would double the OT primary reflected impedance and would not be damaging the amp.

Yes that doubles the OT primary impedance. If using output triodes no problem. But we use pentodes almost universally, and full power into twice the intended impedance will increase their screen grid dissipation. If you check, even with the intended impedance, full power typically results in the screen grids overdissipating. And so twice the intended impedance impedance will exacerbate that.
Note that screen grids are just a fine wire, and shorts there seem to be a common cause of output valve failure, which can be catastrophic.

With beam pentodes (eg 6L6) in the typical guitar amp circuit arrangements, if it’s not possible to comply with the intended load impedance, it’s best to go a step lower rather than higher. Anode impedance will increase but it typically self limits at a non damaging level.

My experience of suppressor grid pentodes (eg EL34) is that they tend to be more prone to damage from overdissipation, so it’s best to avoid using them at high power levels into a different load impedance than intended.

[fenderguy81]

V1A .567 VAC

was your #'s WITHOUT a input signal?
what did you use to measure the AC volts?


.5VAC with no signal is a lot for V1A's plate.  If you used a scope, what is the frequency?

That is without any signal running through the amp and was measured using my fluke dmm with one probe on chassis ground and the other on pin 2 of V1.

I redid the other test of grounding the grid (pin2 of V1) to chassis ground and it did not change the noise floor.

[shooter]

Here are the AC Plate voltages

on pin 2 of V1.

pin2 is G1 not plate.
either way I would expect  < 20mVac at either point with no signal applied.
which fluke?  mine has a frequency button.  If you can measure freq, repeat the test only measure freq instead of volts.

If not move the amp to a different room, garage, shed and repeat.  I'm thinking you have a shielded cable with a broke/bad shield

[pdf64]

…
I redid the other test of grounding the grid (pin2 of V1) to chassis ground and it did not change the noise floor.

That includes the noise with the channel volume turned up?
We don’t expect it to make any difference if the vol is turned down.
But turned up, that seems weird?!

Just trying to get a straight answer, comparing like with like, to the query posed in post #5 

[fenderguy81]

Here are the AC Plate voltages

on pin 2 of V1.

pin2 is G1 not plate.
either way I would expect  < 20mVac at either point with no signal applied.
which fluke?  mine has a frequency button.  If you can measure freq, repeat the test only measure freq instead of volts.

If not move the amp to a different room, garage, shed and repeat.  I'm thinking you have a shielded cable with a broke/bad shield

Apologies! The plate voltages provided in ACV were on Pin 1 and pin 6 (plates of each tube). I accidentally put pin 2 I think because I was thinking about shorting the grid as someone else had suggested. I can try replacing the shielded cable and see if that helps. Thanks for pointing that out!

That includes the noise with the channel volume turned up?
We don’t expect it to make any difference if the vol is turned down.
But turned up, that seems weird?!

Just trying to get a straight answer, comparing like with like, to the query posed in post #5 

Yes, that is with the volume turned up. I used my Fluke 26iii set on ACv, black lead clipped to chassis and red probe on pin 2 of V1A. I turned the volume knob up and down and got the same amount of noise as before. I can try pulling preamp tubes if that helps any. 

Also, I do have a small signal generator if needed. I can run a 1k signal through the input etc. I also have a Tektronix oscilloscope, isolation power transformer, dummy load, but unfortunately I do not have much experience with the scope.

[valvetones]

…8 ohm load would double the OT primary reflected impedance and would not be damaging the amp.

Yes that doubles the OT primary impedance. If using output triodes no problem. But we use pentodes almost universally, and full power into twice the intended impedance will increase their screen grid dissipation. If you check, even with the intended impedance, full power typically results in the screen grids overdissipating. And so twice the intended impedance impedance will exacerbate that.
Note that screen grids are just a fine wire, and shorts there seem to be a common cause of output valve failure, which can be catastrophic.

With beam pentodes (eg 6L6) in the typical guitar amp circuit arrangements, if it’s not possible to comply with the intended load impedance, it’s best to go a step lower rather than higher. Anode impedance will increase but it typically self limits at a non damaging level.

My experience of suppressor grid pentodes (eg EL34) is that they tend to be more prone to damage from overdissipation, so it’s best to avoid using them at high power levels into a different load impedance than intended.

Interesting, so say for the sake of conversation that it’s a 4K OT primary with an 8 ohm secondary, but you only had a 16 ohm speaker and a 4 ohm speaker to work with.
Your saying it would be safer to run 6l6s with the 4ohm speaker/ at 2k primary,, and more dangerous to run the 16 ohm speaker / 8k primary?

I tend to believe you right out of the gate , but I must say that runs contrary to everything if read about it.
Do you have a source where I can read more about that with regard to triode/pentode operation with regard to higher and lower impedance?

I’d like to study up on that.

Thank you
VT

[pdf64]

… so say for the sake of conversation that it’s a 4K OT primary with an 8 ohm secondary, but you only had a 16 ohm speaker and a 4 ohm speaker to work with.
Your saying it would be safer to run 6l6s with the 4ohm speaker/ at 2k primary,, and more dangerous to run the 16 ohm speaker / 8k primary?
…

It’s tricky to find a technical reference covering the topic, sorry, maybe others can help?

It’s perhaps worth noting that vintage Fender combos wired an extension speaker jack in parallel with the main jack, rather than in series. So assuming the combo cab impedance is the intended load value, a lower than intended load is inevitable when an extension cab is used.

It’s also worth noting that JTM45 use 6L6 family valve types loaded with an 8k typical primary OT.

With a typical guitar amp circuit and loaded HT in the range 400 - 450V, I’m happy to overdrive 6L6 family type valves into 3k - 9k loads. If the loaded HT is closer to 400V, then perhaps even down to 2k.
Outside that range may be chancing one’s luck though.

[valvetones]

So are you suggesting it’s low voltage high current that we may start running into issues.

So 4K primary / 8ohm secondary
Used with 16 ohm speaker
Now 8k primary
Is better of at 400v plates 35-40 milliamperes per tube =good

VS. the exact same scenario except 300volts on the plate 80 Milligan’s per tube = Bad

Or are you saying lower voltages with higher current=good

Higher voltage, lower current into that impedance mismatch would be bad?

This one is important knowledge because I use reflective impedance a lot and most of my builds are low voltage , high current .

[PRR]

A too-high impedance load, when the amplifier is OVER-driven, will kick-up ten times the DC voltage on the peaks. This tends to be "OK" for seconds and "dead!" after hours.

One of Gar's now-famous clients was using an un-loaded 5W amp as a fuzz and had to get it fixed weekly, so Gar demanded to know how he was killing it.

As for a mere 2X mis-match: it must depend, and probably on too many factors to ponder.

[fenderguy81]

My reasoning for accepting the higher mismatch came from Dr. Z. Here's a video of him talking about speaker impedance mismatching:

&t=89s

Now, I'm not necessarily saying that all of this information is correct. He is a more credible source than I am, but I do not think he is necessarily the authority on the topic. I thought this video may be helpful though to those interested in the discussion.

[dwinstonwood]

One of Gar's now-famous clients was using an un-loaded 5W amp as a fuzz and had to get it fixed weekly, so Gar demanded to know how he was killing it.

Randy Bachman?

[pdf64]

My reasoning for accepting the higher mismatch came from Dr. Z. Here's a video of him talking about speaker impedance mismatching:

&t=89s
…

I see things differently to that.
It’s easy enough to verify who’s right.
Measure overdriven screen grid dissipation with the intended load impedance and with 2 x the intended load impedance.
Measure the OT temperature when overdriven into the intended load impedance and into half the intended load impedance.
I think the former will be more significant than the latter, ie contradicting what’s being said in that video.

But really, I don’t think it’s a good idea to heavily overdrive a suppressor grid type pentode amp into a different load impedance than intended. Though the cathode bias of the amp he’s using as an example will tend to mitigate / self limit overdissipation.

[Willabe]

One of Gar's now-famous clients was using an un-loaded 5W amp as a fuzz and had to get it fixed weekly, so Gar demanded to know how he was killing it.

Randy Bachman?

Probably. That's why he built him the Herzhog.