5c3 Proluxe help

[Jason C]

This is where it stands now. It does indeed have A LOT more signal now with the new and improved cathode resistor. The tone and volume controls work on the signal just not the hum. Maybe it was that the signal was too low to hear it before but the signal now sounds like it's talking into a fan. I guess that's oscillation of some sort? Now I'm just tempted to rebuild the whole front end to the stock 5C3P circuit and see if I can get that to work and worry about the GA-75 preamp later....dunno. Yes, those symptoms were correct, PI kills the hum but vol. control has no effect on the hum and the hum fades away when the standby gets flipped.

[loudthud]

I looked carefully at your photos and have one question. The colors don't appear crystal clear on my PC but there is one 25K resistor (R16 on the Weber schematic) that looks like it might be 2.5K. In fact the 25K and the 2.5K (R9) look the same to me.

Edit: See next post before you try this stuff.

Just as an experiment, solder a wire from the input to the PI to the ground side of R9. You should have no hum.

Another thing to try is to use wires for all the grounds in the volume/tone controls and ground them to the ground side of R9.

Weber Schematic: http://music-electronics-forum.com/t14986/

[loudthud]

One thing I should mention. The problem might be ripple on the output stage bias supply. When the PI tube is pulled, the ripple looks like a common mode signal to the 6L6s but when the PI tube is installed, the bottom side PI tube sees it as an input. If you don't have a low voltage cap to try, you can use a 500V cap just to see if it kills the hum. Remember, + side of that cap goes to ground.

[Jason C]

OK status report. I took the bias supply out at "D" and unsoldered my 1 Ohm resistors off of my 6L6 cathodes and put in a 200 Ohm resistor (because I had one handy). No more hum. I do, however get a wierd tremolo kind of effect on some notes some of the time and that resistor was getting HOT but that's probably because it should be more like 250-300 Ohms right? So how do I fix the bias supply because I would really like this amp to be fixed bias.

[loudthud]

The cathode resistor for a pair of 6L6 needs to be 270 or 300 ohms 10 watts. You are probably getting "ghost noting" because the 6L6s are running hot and there is excess ripple on the B+. That should clear up somewhat with proper bias.

Looking back over all the Fender Tweed schematics, I'm not finding any examples where the 5C5 or 5D5 type PIs were fixed bias was used. And for good reason, it's hard to make it work! Ampeg did it slightly differently in amps like the SB-12 which is more like the 5D5 aka See-Saw inverter.

What Weber did is almost a combination as far as obtaining a signal for the second half of the PI. The 25K is on the big side, Fender used something like 6.8K in the 5C5. But the 25K is in the correct position to unbalance the resistors like the 5D5 if you ignore the impedance of the bias supply. Some of the siganl is going to get past the bias supply from the bottom 6L6s grid drive and it's going to depend on where the bias is set. I noticed that Weber's layout and schematic don't match in the way the bias supply is wired. You always want the pot on the shunt to ground side of the bias voltage divider so if the pot fails you have max negative bias, not zero bias.

My suggested solution is to increase the filtering in the bias supply per the attached schematic. I'm not sure how much AC is coming off that transformer tap so to get the bias in range, the 10K or 27K resistors might have to be changed. Along with this you should change the 25K resistor where the bottom side of the PI gets it's signal. It probably needs to be something between 5K and 10K and it will change how the amp sounds. I would solder a trim pot in the circuit and adjust by ear if I didn't have an oscilloscope.

Edit: Don't ground the bias circuit to the filter caps. Run a separate wire to the brass plate or to the ground side of the PI cathode resistor.

[Jason C]

Well I might just put a 300 Ohm resistor in this thing and call it good for awhile. Also, do the cathode resistors typically get so hot that you don't want to touch it? I put a 120 Ohm in series with the 200 Ohm I had on there and they still got pretty hot (not as hot as before but still pretty hot) and the 6L6's sure seemed a lot hotter too. I still had the weird notes though. Well, on second thought a 300 Ohm won't do it any way will it? Will the bypass cap help with any of this?

[loudthud]

Any resistor will get pretty hot when running near it's power rating. Calculate the power as Volts squared divided by resistance. A good safety margin is to run resistors at about half their power rating. Mount the resistor up off the eyelet board for good air flow and away form any electrolytic caps (heat is their enemy) as much as possible.

Plenty of threads on this forum about calculating bias of the output tubes. Basically you calculate the current for both tubes my measuring the cathode voltage and divide by the cathode resistor resistance, then calculate the effective plate voltage, plate volts minus cathode volts, multiply volts times current and you get the plate dissipation for both tubes. Divide by two to get the number for each tube. You would like this number to be about 70 to 80 percent of the 6L6GC rating of 30 watts per tube. Opinions vary on what the best number to use is.

Ghost noting is described as sounding like an out of tune note that's kind of in the background behind the note you are playing. It can sometimes be caused by a speaker (called tone cry). It is sometimes confused with other sounds caused by cabinet rattles and tubes that are rattled because they are too close to the speaker. Try playing through a speaker in a different cabinet some distance from the amp. It is usually present to some degree in most amps. B+ and screen node filtering, and some say rectifier choice influence ghost noting.

[Jason C]

Hey Loudthud, many many thanks for helping me out on this one. I'll just go with cathode bias for awhile until I get up the nerve to rework it. Again thanks for the help. You're aces, man.
Jason

[loudthud]

I modified the schematic to something that should work with fixed bias. It's not true to the old Fender Pro amps, but that went away with the GA75 front end. There still could be some hum injected by the bias supply because the bottom tube of the PI has no common mode rejection, so the mod to the bias supply might be required.

[Jason C]

Would it just be better to go with a different PI all together? I did think about just using the PI and power section from the 5e3P just keeping the 6SL7's. I know, not really anything left of the 5c3P save the octals but I'm really just looking for fat cleans.
Jason

[loudthud]

IMHO the things that contribute to the fat tone you are looking for (given a push pull output stage) are: Grid leak bias input stage, Octal preamp tubes, Non-scooped eq network, Less than perfect phase inverter, No feedback output stage, Cathode bias output stage, Lowish B+, Tube rectifier, Poorly regulated screen supply, Open back speaker cabinet, AlNiCo magnet speaker. There may be some disagreement here, and there are exceptions. You can go too far and you don't need every one of the above items to make a good sounding amp. It is the control and artfull combination of these things that makes a good sounding amp.

That being said, I think a Concertina PI like the 5E3 would be worth consideration. You'll retain most of the early tweed flavor, and fixed or cathode bias shouldn't be a problem. I'm somewhat disappointed you don't want to try my suggested fixes but, it's your amp. I've got a similar amp with 9 pin preamp tubes I could tear up and install that phase inverter.

[Jason C]

I will certainly try your fixes first. I just didn't know if I was asking for something this PI isn't really equipped to do. Much like asking a Pinto to perform like a Ferrari etc.
Jason