Fender Bassman 70 buzz

[JAelec]

I'm trying to cure a Silverface Fender Bassman 70 from severe buzzing at 100hz

Schematic is here: https://schematicheaven.net/fenderam..._bassman70.pdf

Gut shot:



Filter caps:



I've tested paralleling the filter caps with a 100uF capacitor one at a time. When connecting the capacitance as in the pic above the buzz level drops from very loud to loud (from 65 mVpp to 50 mVpp), indicating that the cap in the green circle is bad.


The dog house on the Bassman amp has a total of 7 capacitors in the doghouse compared to the Twin Reverb. The two extra capacitors are 63uF 100V and 100uF 100V. I'm guessing these are the negative bias caps but they are moved to the doghouse compared to the Twin reverb, is that correct?


The negative bias caps (if I am correct in my assumption) seem to have different values... could that be a problem?

What else could be causing the 100hz except for the bad filter cap discovered?

Adjusting the hum balance would only effect 50hz filament hum if I understand correctly... so that won't be a solution.

[Jon Snell]

Check the output valves are actually balanced, if not you will get hum.
You will need to either put a set of balaced valves in the output stage or fing out what is wrong with the biasing and fix itm if that is the problem.

[JAelec]

Will balanced tubes and biasing fix 50mV 100hz buzz that is independent of volume control pot? Since the source is the power supply filter, shouldn’t the potential fix be somwhere in the power section?

[Helmholtz]

Will balanced tubes and biasing fix 50mV 100hz buzz that is independent of volume control pot? Since the source is the power supply filter, shouldn’t the potential fix be somwhere in the power section?

Even with good and sufficient capacitance filter caps there will still be some ripple on B+.
As a consequence of push-pull operation power supply ripple entering the LTPI or the power tubes will not show at the output if the amp is well balanced.
This amp has an "Output tubes Matching" pot. Use it to minimize the hum.

[g1]

This amp has an "Output tubes Matching" pot. Use it to minimize the hum.

Something has been messed with in this amp. It looks like one of the rear panel pots is going to a single yellowish wire. Where it goes through the chassis, there are 2 white wires cut. I'm guessing maybe an 'output matching' pot has been changed to regular bias?

[52 Bill]

I can't figure out what is going on on the back panel ...3 fuse holders where the ac and standby switches normally are. No rotary voltage selector switch and there seems to be a fuse on the back of the pilot lamp holder.

[The Dude]

Maybe some of those things that look like fuse holders are some kind of odd replacement switches? The amp has definitely been messed with.

[JAelec]

Something has been messed with in this amp. It looks like one of the rear panel pots is going to a single yellowish wire. Where it goes through the chassis, there are 2 white wires cut. I'm guessing maybe an 'output matching' pot has been changed to regular bias?

From what I can see the tube matching circuit is just like in the Bassman 70 schematic:



The pot has four lugs:



The middle lug goes to ground through a resistor. Supposed to be a 33K but in my amp it looks like 22K:
Is that OK or should it be the original value?



The two outer lugs go to two resistors, one 47K and one 68K resistor, like in the schematic. I'm guessing these resistors set the bias for the grids of the output tubes? How come they are not equal in resistance?


The "extra lug" on the back has a single yellow wire that goes to one of the bias caps located in the doghouse just like you pointed out.




The two wires that are cut come from the doghouse at this point:



They seem to not be part of the schematic, from what I can tell. So my guess is that they are a modification that someone has restored to how it is supposed to be in the original schematic.

[g1]

That's a better look, I couldn't see the other wires on the balance pot before so wasn't sure. The 4th lug 'tap' is perhaps not exactly in the middle, thus the difference in the 47K and 68K.
I would think someone changed from 33K to 22K to run the power tubes hotter. For myself, I would put it back to the 33K stock value.
Looking again at where the cut wires go, I see they are going to the filter caps as you say. Probably an alternate routing was used in the filter cap wiring. Without a layout diagram it is tough to say whether those wires were the stock version or not. Lots of stuff has been changed in this amp.
Is there now a standby switch, or not?

[JAelec]

That's a better look, I couldn't see the other wires on the balance pot before so wasn't sure. The 4th lug 'tap' is perhaps not exactly in the middle, thus the difference in the 47K and 68K.
I would think someone changed from 33K to 22K to run the power tubes hotter. For myself, I would put it back to the 33K stock value.
Looking again at where the cut wires go, I see they are going to the filter caps as you say. Probably an alternate routing was used in the filter cap wiring. Without a layout diagram it is tough to say whether those wires were the stock version or not. Lots of stuff has been changed in this amp.
Is there now a standby switch, or not?

There is no standby switch on this amp.

I found an alternative answer to the cut wires... someone posted at the "Guitar Amp Repair Techs" Facebook page concerning a Bassman 100 with a similar snipped wire from the doghouse.

Some quotes from that thread:

"That snipped wire is factory. I’ve seen it in several of these.
I can only assume the cap boards were assembled and put in a pile to be used for several different models, and this model didn’t need that particular wire, so they just snipped it clean at the grommet.

"I’ve seen this cut off wire many times in 100 watt Fender amps that did not have vibrato and reverb circuits. Node B would have powered these circuits. Same pre-wired power supply board. The wire was not needed so it was efficiently sniped off in production. No mystery no mistakes. Although I’ve seen many mistakes in Fender amps this isn’t one of them. Have a great weekend."

"I recently had a Bassman 100 on the bench. I encountered the same snipped wire. After looking at other chassis pics, I realized that it was factory."

[g1]

So the standby switch has been removed and several of those fuses are non-stock. There should be one mains fuse and that is it.

[JAelec]

New matched pair of tubes + Tube matching pot made the buzz go away...

Although paralleling the filter caps with my 100uF capacitance did away with 1/3 of the noise amplitude. Switching to matched tubes and adjusting the tube matching pot were enough to make the noise go away completely, without needing to replace the cap. Not sure what this means and if I should replace the cap anyway?

The old tubes were impossible to match with the tube matching pot. Is this because they were too far off from each other perhaps?

The hum balance pot does nothing, no difference in pot position. However there is no 50Hz hum in this amp to begin with... I was expecting some hum to start when the pot was turned as an experiment, but no. So all good!

[Helmholtz]

.
Although paralleling the filter caps with my 100uF capacitance did away with 1/3 of the noise amplitude. Switching to matched tubes and adjusting the tube matching pot were enough to make the noise go away completely, without needing to replace the cap. Not sure what this means and if I should replace the cap anyway?

Increasing filter capacitance lowers ripple riding on B+.
As balanced push-pull operation suppresses PS ripple, filter capacitance can be lower than with single-ended amps.
So the effect looks normal.

[JAelec]

Increasing filter capacitance lowers ripple riding on B+.
As balanced push-pull operation suppresses PS ripple, filter capacitance can be lower than with single-ended amps.
So the effect looks normal.

Ok, thanks for the clarification and info!

I'm thinking perhaps I should check the bias of this amp. From what I've read between 60-70% plate dissipation is ideal. Any more might lower tube life and colder will decrease headroom.

I was thinking of adding a small bias trimpot in addition to the bias balance present in this amp. Similar to picture below (10k trimpot and a 6.8k resistor in series).

[g1]

From what I've read between 60-70% plate dissipation is ideal. Any more might lower tube life and colder will decrease headroom.

This is not really true. The 70% is 'do not exceed', not a target. Many fender amps are biased in the 50 to 60% range from the factory. Recommended setting on some modern fender models is sometimes 45%. Some as low as 33%.