Project "scrap salvage"

[stw307]

Hello!

I have a bunch of 120w Bugeras lying around. Plan is to take out the guts and replace them = use only the trannies and the chassis with a turret board type of an amplifier. In my eyes these transformers are beefy units. In my tests the OT went easily down to 30Hz. PT has secondaries of 360, 18-0-18/3A and 6.3/8A. Heater voltages in power amp are low, I measured 5.3-5.5 VAC depending on load. Even w/o load only 5.9 volts. This is way out of specs, and every tranny is alike. One could always use a diode bridge 'n stuff to fix this. Pre tubes are heated thru rectified 18VAC = 24VDC = two tubes in series. And these really are ~12.1V across one tube.

First question is, what are the pros and cons of different layouts? Mainly Marshall type long board vs. Fender type multiple boards + cap housing?

If you come up with any other ideas or opinions, please post here. Very likely "Don't use chinese crap" crosses one's mind, but there is no need to post THAT opinion here 

I'll keep this thread updated as the project goes on.

[stw307]

I have built a fully working proto unit.
Right now I don't have a chance to take a photo of the layout. For the "prototyping" stage all modules have their own turret board. No hum or any other problems so far.
Boards:
1a Power supply section with diodes and most of the capacitors
1b Connected to this board, bolted on chassis: biggest caps, switches and the coil
2a Preamp heaters & bias voltage supply board
2b Attached to the chassis, there is a bias master pot, ranging -30 -70V. From there, every power tube has an individual bias balance pot. All of these have their own test points.
3 Power amp board, which includes 5W screen resistors
4 Preamp board

Preamp board is of 5F6A type, with some modifications to suit 12DW7 in V1. I really do like the results, when playing the bass. From clean to Lemmy.

[mresistor]

I think we would like to see some documentation and some photos of the project when you have time.  Sounds interesting.

[stw307]

I think we would like to see some documentation and some photos of the project when you have time.  Sounds interesting.

There you go.
Schemas are not exactly up to date, but close.

[stw307]

I think we would like to see some documentation and some photos of the project when you have time.  Sounds interesting.

There you go.
Schemas are not exactly up to date, but close.

Preamp does not have a schematic of its own, as this is a temporary preamp. We're just testing with a friend, how a 5-string bass sounds with a tube amp. Very good, actually (just my 2c).
I took a Fender schema, parallel ch layout from Robrob, made some calculations for the 12DW7, added cascade mod w/ switch (for guitar) and drew that uploaded layout "on-the-fly". Also there is a K bypass switch 1-0-1 to test different C values. So, it has many recalculated component values on the V1 side. I suppose it can take any 12A_7, too. Haven't tested any other tubes yet.

Preamp layout is up to date, I think only input resistors were drawn "wrong". 1M should not lift off when jack is inserted. 56K might be actually 68K or even higher to dampen active bass' high output.

I also found the other layouts. Bias board is "final", the other one is far away from a completed drawing. Component values are up to date, but the bias voltage section is a mess. It went in as a separate board. Now it is also redrawn. That black block is a piece of aluminium bolted to bridge. This small board delivers bias voltage and 24VDC voltages for heaters and does a good job now. Bigger layout is uploaded only to show, how the individual bias pots work. They are drawn in the wrong place, as is everything else.

Drawing layouts is hard for me. I'll learn it some day. Almost every DIYLC file name ends in _not, not ready. Actual layouts inside the chassis are cleaner. I spend much more time with LTspice to simulate the power caps and stuff. When designing layouts, I have DIYLC, schemas, papers, pens, boards and components at hand. I play around with all these to find the "right" way to assemble. Then I get excited, forget finishing the drawings, just solder it together and toss the board in. So far this has not been a problem in scratch builds. My methods are way worse than results...

[stw307]

A schematic of power amp, including the 5-pot biasing system. Plus preamp layout, one switch is for parallel / serial channels, second for cathode bypass caps, and the last one switches between coupling caps (bass/guitar mode). Channels are marked as 12AX7 and 12AY7 as V1 is a 12DW7. Tone pot switch bypasses the TMB stack via a "raw" potentiometer. This mod was quite useless in this application, but it is very useful in some other amps.

[stw307]

When looking at these photos, remember that what you see is not a ready product, but my test bench. After testing, the preamp will be replaced and so on...

[stw307]

This 5-pot bias is a must in a test bench in my opinion. Yes, you can do a perfect balancing, but as soon you throw a signal at it, the balance is gone. What this system actually does is great hum canceling and it can also balance mismatched tubes. Heck, I even threw in a mixed set of really worn out 6L6GC's and I managed to balance these. Our bass player played with these tubes ~3 hours with ~50% power. No problems. Extreme testing, I'd say. When biasing, I turned all balancing pots to center-click position. Then the base level was set cold with the master pot. Power on, and after that I had to check every tubes mA reading at least three times to achieve the 32mA goal. The tubes interact with each other, so this is a bit like tuning a guitar with a Floyd Rose bridge.

Biasing one tube intentionally out of balance caused a hum out of hell. This pretty much explains, why some amps tend to hum more when reaching the output tubes' end-of-life. But it also tells me, that all of the tubes are not necessarily dying, only the weakest one(s). Nevertheless I wouldn't recommend replacing power tubes one at a time. Instead, I save the old tubes to tinkering projects like this...

Now it has a quite fresh set of JJ EL34's in it. Readings are 454V anode, -41V at bias test point. Remember, that test point does NOT show you actual readings at tube sockets, it just points to the right direction. Tubes were biased @ 42mA, which gives ~19 watts total dissipation. If I did my math right, this is about 15-16 watts anode dissipation and the rest is grid current. Anyhow, I was forced to quit testing after biasing. My old bass cabinet could not handle the power. Sounded good with lower volume settings, though. I'll visit a friend of mine to test with a proper 6x12" cabinet.

[stw307]

I'm not happy with my soldering, I've had trouble finding the "good stuff". Yesterday one guy came by to drop some rolls of 40/60 tin. Waiting for way better results.

By the way, feel free to tell your opinions, even the negative ones.
Please ask questions.

This thread is like a blog, yes, but it feels stupid when talking alone at a discussion forum 

[dogburn]

This is a rather complex build compared to what I've done. I see how you've set it up to change the power tube bias when using different tubes like 6L6 or EL34 - is there anything else you need to change when switching tube types, or is everything else in the circuit compatible with the different types?

Also, what are you plans for the other amps?

[Willabe]

I see how you've set it up to change the power tube bias when using different tubes like 6L6 or EL34 - is there anything else you need to change when switching tube types, or is everything else in the circuit compatible with the different types?

If your going to use different power tubes in an amp, 6L6GB/C, EL34, 6550, KT 66, KT88, etc., you have to make sure the socket is wired up to use any/all of the tubes. You can't run all those tubes in a socket wired up like Fender amp is wired, grid stopper R and screen grid R on the socket. 

[stw307]

This is a rather complex build compared to what I've done. I see how you've set it up to change the power tube bias when using different tubes like 6L6 or EL34 - is there anything else you need to change when switching tube types, or is everything else in the circuit compatible with the different types?

Also, what are you plans for the other amps?

This is only my second scratch build. It is "beginners work" and looks more complex than it actually is. For example, look at the preamp: replace parallel input with four jacks, throw away those switches and you're left with a fairly standard Bassman. This amp is designed piece-by-piece, then every piece was fine-tuned to fit each other. Except power section was designed to feed everything I might throw at it, so it is pretty stiff. And I dont like sagging...
I take that note of complexity as a compliment, thank you!

When I find the proper layout, sound and so on, I'll take the next two pieces from the scrap pile. Then I'll just make myself an amplifier and an identical spare amp. I suppose I'll opt for 3-board layout. DC feed & caps, preamp and power amp. This one will stay at the soldering station, serving as a versatile work horse much like its smaller brother already does.

If your going to use different power tubes in an amp, 6L6GB/C, EL34, 6550, KT 66, KT88, etc., you have to make sure the socket is wired up to use any/all of the tubes. You can't run all those tubes in a socket wired up like Fender amp is wired, grid stopper R and screen grid R on the socket.

Tube rolling: As Willabe said (and I think he really knows his stuff), there are things to consider, wiring being #1. This is NOT wired up like Fenders. It can accept any of those types mentioned on the back panel (two out of four types tested as of today). Or any other type, which can withstand ~450 volts A/G2 and has a proper pinout. For example, 7027 which is identical to 6L6GC, would require rerouting of pin #1.
If I were trying any other types, I'd plan carefully, draw the load lines etc. Especially with the big bottles it is possible to literally blow up something. Hopefully the fuse...

Thank you all for sharing your thoughts!

EDIT. One more thing about different tubes. When bias master was rolled all the way down, it strangled those EL34s to death @2mA. So in my opinion it is safe to try almost anything with a proper pinout. Also those 10K balancing pots have quite a wide sweep. They need an extra careful touch when biasing. It might be possible (I think it is) to use EL34s and 6L6s simultaneously. Or some other types? These voltages combined with coil fed 680ohm/5W screen resistors can accept a wide range of power tubes, for what I've studied from countless schemas.