6V6 in 18w TMB?

[punkrockpoet]

Has anyone done this?

Is it even possible?


Ideally, a set of noval sockets for EL84 and a set of Octals for 6V6... possible?

[Alexo]

Possible and been done. Sounds good! Use the search function, lots of posts on it.

[LeeMo]

http://www.trinityamps.com/phpbb/viewto ... highlight=

If I copied the link right, it is the Trinity v6.

LeeMo

[phsyconoodler]

i have built a few 6v6 versions of the TMB and they sound great.I just use grid stoppers and screen resistors suited to the tubes.And of course the cathode resistor needs to be 250 ohms of thereabouts.They seem to be voiced a little more towards a vox than a marshall.I like the tone with a strat more,but it lacks the grind of the EL84's for more rocky sounds.Still a very pleasing tone overall.

[zaphod_phil]

I found that you need to change the 18W PI's tail resistor to something around 22k to 27k (from the original 56k or 47k) to get the PI to drive the 6V6s harder, with their lower gain. Same for most other octal power tubes. Like psychonoodler says you also need to change the cathode bias resistor. I used to change the screen resistors to 470 ohms, as in a tweed deluxe, but the Lite IIb's arrangement of a shared 1k with individual 100ohm screen resistors also seems to work well.

[JohnnyCrash]

I built one. I then dropped the Normal channel and cascaded it into the TMB permanently (I originally tried it on a switch).

My 18w 6V6 TMB is now my main practice amp.

[Alexo]

Here is my schematic for my 18 watt 6V6 TMB.

It's not 100% tested. I built the amp, tweaked it a lot and it worked 100%. I decided to rebuild it, just to tidy things up after all the modding and tweaking, so the version in the schematic is the new one that I'm about to finish, the only untested aspects are a couple changes in the resistor values in the power supply rail for the preamp and the bias supply and the addition of a voltage divider to DC reference the heaters. But I am 99% sure it'll be fine. ...famous last words.

I preferred fixed bias in this amp. Some might say that puts it farther away from being an 18 watt, but I think that sonically, it is closer. Don't ask me why, but 6V6's in cathode bias sound a little flabby and floppy to me, they tighten up really well in fixed bias; to me, the 18 sounds more tight than flabby/floppy.

The TMB channel has gain like crazy and needs some shielded wire to avoid oscillations. The normal channel yields a really nice crunchy power tube overdrive, not as much gain as you'd get out of an EL84 normal channel, but it sounds great, especially jumpered!

I originally used a 5U4 and had a higher plate voltage, more like a 6V6 Plexi, but the 5Y3 smoothed out the attack so nicely and gave it a real thick, bluesy sound. The 32 uf cap hasn't killed it yet - DC secondary resistance of the PT is about 140 ohms, hopefully that's enough.

Voltages are approximate.

Hope this helps.

-A

[zaphod_phil]

Looks great!

I preferred fixed bias in this amp. Some might say that puts it farther away from being an 18 watt, but I think that sonically, it is closer. Don't ask me why, but 6V6's in cathode bias sound a little flabby and floppy to me, they tighten up really well in fixed bias; to me, the 18 sounds more tight than flabby/floppy.

I know what you mean. The huge cathode cap trick, that James Marchant at Marshall wised me up to. It stabilizes the cathode bias voltage and helps control the bass.

The normal channel yields a really nice crunchy power tube overdrive, not as much gain as you'd get out of an EL84 normal channel...

Have you tried a 220k load resistor on the Normal preamp. Just a thought...

I see you've also biased the PI warm, Plexi-style with the 470 ohm resistor, for a bit more PI crunch.

[Alexo]

Thanks, Zaphod!

I had thought about raising the plate load resistor in the normal channel, but I played through it a bit more before I tore it down (usually use the TMB channel) and was surprised that I liked the tone quite a bit as-is, so I left it.

Yep, 470 in the pi for a bit more crunch, 15k on the tail, pretty unbalanced, I suppose, but that's the sound I found I liked best. I tried to design it so that the B+ at that stage would be equal to 300 volts + the voltage sitting at the top of the tail, so the ltp shouldn't compromise the tone of the tubes at that stage by low-balling them on plate voltage. Who knows, maybe the reduced plate voltage actually contributed to the good tone.

I had 470K grid leaks on the power tubes to maximize power tube distortion, but I revised them to 330K's after I saw that the maximum recommended value in a fixed bias 6V6 amp was 100K. Hopefully I won't lose too much drive there, (or blow out my tubes) I suppose that is one advantage of a cathode-biased power section, less signal loss going into the power tubes.

Anyway, I finished her up last night. Gave myself a good night's rest before a final check-up, we'll see what happens tonight!

[mashaffer]

This is approximately what I am planning on doing.

mike

[Alexo]

Looks good to me!

Are you using a stock 18 watt power supply?

Should C5 be .1 instead of .01 uf, or is there a reason for .01?

[morcey2]

You need to put C9 on the other side of R34 (master volume). Otherwise you'll have DC on the pot and the PI won't behave correctly. And Alexo is right on C5. It should be much larger than C9.


Randall explains the theory much better than I can:

http://www.aikenamps.com/LongTailPair.htm

So does our own merlinb:

http://www.freewebs.com/valvewizard/acltp.html

Other than that, it looks pretty good.

Matt

[zaphod_phil]

I had 470K grid leaks on the power tubes to maximize power tube distortion, but I revised them to 330K's after I saw that the maximum recommended value in a fixed bias 6V6 amp was 100K. Hopefully I won't lose too much drive there, (or blow out my tubes)..

You do have a good point there, as the spec does say that, although there are plenty of fixed-bias 6V6 amps around with 220k resistors on their grids. Still you don't know how far you can push it, but I suspect the increase drive from having a 15k tail resistor probably compensates nicely for that slight reduction in the grid reference resistors.

[JJGross]

You need to put C9 on the other side of R34 (master volume). Otherwise you'll have DC on the pot and the PI won't behave correctly.
*** snip ***
Matt

You mean on the output of the pot going to the PI and power section, not the pot lug going to ground, correct?

Sorry if that seems Uber-Noob, but I am a noob who's trying to grasp all this stuff on my own with no formal training.
It's not easy since a lot of this is over my head, but I'm starting to 'get it' - at last I think I am.

I know - famous last words ... ZZZZZTTTTTT!!!

Cheers,
- JJ

[mashaffer]

Thanks for the corrections guys. I would much rather get these things sorted out before the solder flows.

I have moved C9 to the wiper of the MV pot and increased the grid cap of the PI to 0.1u.

I have not decided on the PS but I was thinking of something similar to the AX November project. Something similar to the attached digram.

The current source is the PSdesigner simulation of the output tube idle current. I have since realized that for accurate sim I need to split the screen current out so something like 76mA at the input of the choke and 4mA at the output of the coke (screen grid is connected to the right side of the choke via the screen resistors). The PI and final gain stage come off at C4 and the input stages (bright and normal) at C5.

The choke may be over kill but it seems like a clean PS at the screen might be a good thing. The smallish input cap is to limit current stress on the rectifier v.s. having 32u input cap. I may not in fact be using those rectifiers as I might go with 5V.

It is possible that series resistance may be needed to protect the rectifier also. As I said I am in the early stages on the PS design.

I will also probably add buffered (and possibly mixable) Fx loop later but we want to try to get the basic amp ready for a recording date in July so may not get to it in time for that.

mike

[zaphod_phil]

I preferred fixed bias in this amp. Some might say that puts it farther away from being an 18 watt, but I think that sonically, it is closer. Don't ask me why, but 6V6's in cathode bias sound a little flabby and floppy to me, they tighten up really well in fixed bias; to me, the 18 sounds more tight than flabby/floppy.

Sounds like you need the huge cathode cap mod on the power tubes. Just use something like 1000uF, 2200uF or even higher, to nicely tighten up the amp.

[zaphod_phil]

Has anyone done this?
Ideally, a set of noval sockets for EL84 and a set of Octals for 6V6... possible?

It's much easier to just use 6BW6/CV2136 or 6973 power tubes, which are both 6V6s in9-pin bottles. Simple as that!
Note, you could also try 6P1P tubes, which are basically a Soviet version of the 6BW6/6973.