Fender Bassman 100 Head Silverface 1974 - Blows Fuse

[Enzo]

Sure, but what matters regarding shock is the voltage difference between the two points you touch and not the difference to 0V. If you touch -200V on one side and -200V on the other, you won't get shocked. So this example doesn't prove anything.

But it only becomes -200v when it is with respect to ground. Otherwise it is just a floating 200v source.

If you are grounded and the supply is grounded, the -200v point will shock you more than the -20v point. SO it seems counter intuitive to say we reduce the -20 to -200.

[Helmholtz]

I think the only thing we are going back and forth over is how to label the voltage change.

It all seems to be a matter of definition and maybe convention.
I am using the international definitions of physics and engineering literature. But if this is not helpful here, I will shut up.

[tubeswell]

Yeah but when taking voltages in a guitar amp that is plugged into the main supply in your workshop, you are not going to be at -200v, and if you were, the Q in the 80uF 73 bassman 100W bias supply cap would probably kill you (since it is only at about -60V) if you were stupid enough to touch it in such a way as to provide a return current pathway through your heart).

[soundmasterg]

I wasn't aware of e-caps eventully shorting. Might be a good reason for adding a B+ fuse as the failure doesn't seem to be predictable. If even new caps may fail after only 2 years I prefer to stay with the old ones as long as they measure O.K. and show no signs of physical leakage - at least in my own amps.

In my former company we continuously tested hundreds of e-caps of several brands for over 50,000 operating hours at 85°C. They were measured before and after testing. Never heard of a shorted one. But we only used SMPS grade 105°C types.

I think one thing to keep aware of in regards to e-caps and audio and MI tube amps is that they are often not in continuous use, and sometimes they will not be in use for decades at a time. E-caps are known to go bad faster with non-use than when used continuously, and in your prior testing it sounds like the caps were in constant use.

It is surprising that some of the vintage caps last as long as they do. Take for instance the "Planet" brand caps that were in vintage Silvertones. It said right on them 'guaranteed for one year' yet many last 50 years. Then some new caps (Illinois caps come to mind) often fail before the warranty is up. We hear about how new e-caps are so much better than the old ones.....they are certainly smaller today, and they are probably better too, but will they last as long? Who knows. That said, I will tend to err on the side of replacing the e-caps in older amps if there is any question or if one cap in there has the same vintage as the rest and it is bad. I could run my 50 year old Chevy with original tires, but new ones will perform so much better there isn't much point in sticking with the old ones. Now the signal caps are a completely different story....those only get replaced when they fail.

Greg

[soundmasterg]

In regards to the Bassman 100's, I have seen a lot of bad power tube sockets on these in the ones I have seen. You think the amp is ok, put new tubes in, seems to be fine on a bench for hours, and first gig the socket will arc and there go your tubes. I'm suspicious of the sockets on these as a result and check very closely. Haven't had an issue with any other amps which makes me wonder if CBS was using cheap stuff in these amps in that era.

Greg

[misterc57]

Bias is cold

I replaced the following; both bias caps with F&T 100uF 100V caps, the single diode in the bias circuit and its 470 ohm resistor (was reading around 700 ohms).

I set the bias balance pot in the middle. All readings below are for V4, V5, V6, V7 in that order;

pin 5 VDC; -57, -58, -56, -56

current mA; 16, 47, 9, 16

So I have 3 cold tubes and one hot tube. I swapped tubes 5 and 7 and now read

current mA; 16, 14, 9, 51

So the hot tube followed the valve swap. I swapped V7 with another old 6L6 I had available. Now I read

current mA; 16, 14, 9, 13

Bias is cold.

I monitored V4 plate voltage during all the above and it was 420 VDC.

The two 68K resistors following the bias pot both read 71K.

Bias readings were done with a bias tester that sits between amp tube socket and the tube, and read VDC on your DMM, one tube at a time (very tedious).

Thoughts? Thank you! MC

[Leo_Gnardo]

Thoughts? Thank you! MC

Good so far. Especially getting that hot tube out of the mix. Now to dial down the bias voltage so you can get a bit more current. I think you'll find a 33K resistor from the bias pot wiper to ground. Clip or tack-solder a resistor in parallel to reduce the overall bias voltage. Start with say 220K, see if those bias currents head up to 25-35 milliamps. Remember, with all 4 tubes in place they will be sharing current from the high voltage supply, the result being lower bias current per tube than you measure when testing individually.

At some point it would be nice to have a matched set of output tubes for your amp. But you're proving that it works first while learning how to deal with setting bias.

FWIW I replace that 33K with a lower value say 15K, with a 20K trim pot in series. This allows a wide variation of bias voltage, an advantage these days when we find tubes that are at the far ends of the emissivity bell curve being marketed.

[g1]

From the schematic in the first post, the resistor Leo mentioned would be 15K in this version (coming off the bias pot wiper).

[Chuck H]

So maybe parallel 100k to start. 220k won't give much gratification. If you change to a pot I would use 10k (wired as a variable resistor) in series with a 10k resistor.

[Leo_Gnardo]

From the schematic in the first post, the resistor Leo mentioned would be 15K in this version (coming off the bias pot wiper).

Dang, I should'a looked. OK then Chuck's advice looks right. Thanks for pitching in guys.

[misterc57]

Yes it is a 15K on the wiper to ground. I worked my way down to a 15K in parallel to ground which yielded 38 mA current on the one tube (V7) I was monitoring (note that all 4 output tubes are inserted in all prior and current tests).

From what I read about Fender amps, I thought 38 mA was a bit high. I bumped up to 19.6K (a 15K + 4.6K in series) in parallel to the wipers 15K and then read 32 mA.

I still need to play with the values more. I think the ideal number would be 30 mA?

Thank you! MC

[g1]

You should be close enough. Your plate voltage should now have dropped a bit so you need to factor that in to the equation.
How does it sound? Any audible crossover distortion on decay of notes?

[misterc57]

The amp is sounding great! Many thanks to all of you. I love learning more and I learned a lot about biasing on this one.

If anything else comes up I will be back to this thread but I think all is good now.

MC