Mixing power tubes

[lowell]

So here's my amp I've been building and tweaking now for 6 months. It sounds great, however I'm curious. The 6V6 (cathode biased 32v) grids are clipping before the 6L6 (fixed bias -47v) are. I'm not TOO concerned, however it's caused some new brainstorming. I've thought about 2 separate PIs for the pairs. Maybe a LTP for the 6L6 and keep cathodyne for 6V6. I've also considered changing the cathodyne to clip later by increasing the supply voltage and the load resistors (Ra and Rk), however I don't really want to change this as far as the clean channel is concerned when the 6L6 tube-cancel switch is engaged and only the 6V6 are running. I've also considered cathode biasing the 6L6 and getting them to run at a Vgk that is closer to the 32v on the 6V6, and considered biasing the 6V6 colder up towards the -47v that the 6L6 are at. Hope you're still w/ me here. I guess essentially I don't want to change anything except for the fact that the 6L6 aren't seeing a big enough drive voltage. Is there another way I could up the drive voltage to just the 6L6 grids aside from adding another tube and PI? Or maybe someone has a better way to arrive at my solution?

[mooreamps]

Well, my first thought would be to go with the additional PI. For this reason, I do not mix the smaller power tubes with the larger power tubes. The bias and drive requirement are quite different.


-g

[lowell]

ok thanks for your input Gary.

[pdf64]

How about using larger grid stopper resistors, say 10-22k, to each tube? Then when the 6V6 grids conduct, they won't short out the signal to the 6L6 grids.
I couldn't open your schematic?

[MWJB]

"Or maybe someone has a better way to arrive at my solution?" I think that you are overcomplicating the issue. You have 4 sockets, if you want higher power use 4x6V6, or 2x6L6, fit a fixed/cathode bias switch & set the power tube up as a regular parallelled PP quad. When you want 1/2 power just pull/use 2x 6V6.

You may be lucky to find a pair of 6V6 that run acceptable current and a pair of cooler running 6L6 that can be subbed without a bias adjustment, I have found this posssible in a couple of amps that run JJ6V6 @ 20mA & Sovtek/Reflector/EH 6L6/5881 @ 35mA.

Getting 6V6 & 6L6 to clip simultaneously in the same amp is not going to be easy.

[mooreamps]

Well, I was thinking about this. I suppose if I were to mix a 6L6 with a 6V6, one could feed the signal into a pair of resistors forming a voltage divider. The top of the stack driving the 6L6, and at the tap driving the 6V6. Also, if one used a pot ; say 500K ; you could drive the 6V6 from the tap, and then adjust the drive level to "dial in" just the right amount of signal into the 6V6.




-g

[lowell]

Gary that's a good point. I'd also thought about that but the only issue is then that the 6V6 aren't going to clip as much or as early. Reducing the 6V6 drive is basically the opposite of increasing the 6L6 drive, which will "fix" my issue, however not sure I want lessen 6V6 clipping.

[lowell]

ok so I increased all grid stoppers to 100k in the PA. I've also reloaded the schems so you can seen them now. The grid drive now has about 4 more volts of headroom. that helps!

[mooreamps]

So ; then drive the 6L6's from the tap ??

[lowell]

but that'd be lessening the ratio of 6l6/6v6 grid drive, I want the opposite. More 6L6 grid drive. Amp sounds good w/ 100k grid stoppers. I'm gonna leave it at that.

[black_labb]

why not a voltage divider after the PI that feeds the 6v6's, and keep the 6l6's at maximum potential. using a dual ganged pot you can have the 6v6 run clean or dirty and the same with the 6l6's.

[lowell]

that would work. i guess putting a pot or switch in that enables/disables the voltage divider. I had another thought too. I'm using 5881wxt, not 6l6, so if I change to 6L6 I can juice the quiescent current - at least getting them closer.

[MWJB]

The only way I can see this working at a practical level is a stereo amp, one side feeding 6V6 one side feeding 6L6, a volume for each preamp and an input for "both". Otherwise, when you have a signal level that is working the 6L6, the amp will be battering the 6V6s...if the 6V6s are working fine the 6L6 won't be up & running.

Otherwise use 6V6 or 6L6, it only takes a couple of minutes to change tubes.

"I'm using 5881wxt, not 6l6, so if I change to 6L6 I can juice the quiescent current" I'm not sure of the relevance of this statement, the 5881WXT is/was often dual labelled as 6L6, it only has the 5881 designation because it's clean & has a military origin (5881 is a hi fidelity, mil spec 6L6). Current requirements are the same as for 6L6.

[lowell]

The 5881wxt is 23 watts max diss. The JJ 6L6 is 30.

[MWJB]

Not true. 23W was the max dissipation for NOS 5881, the Russian made WXT is one of the most rugged 6L6 style tubes around...along with the JJ.

Remmber those Electar Tube 10 amps? They ran the WXT at 28W all day long.