My first build: Ceriatone 36W EF86-questions!

[RockinDoc]

Once a tube goes microphonic it's more prone to picking up mechanical vibrations like that. Making EF86 tubes less than ideal for use in combos with their increased vibration from the speaker(s). If it's so bad that it's picking up its own internal vibrations then it can self oscilate and squeal.

FWIW, I found a sleeve (5) of these NOS 6267s locally for $50. They're the Amperex Orange Globe branded tubes and so far all are very quiet and very strong. I also have several Sylvania branded 6267s that are also very good. I've had a pretty good experience with the RI Tung Sol EF806... maybe I've just been lucky?!? I've also read in various places that the EH EF86 is just as "good" as the TS, but while the EH tube that came with my Ceriatone was not uPhonic... it was a little on the dull side. Regardless of what tube is in the floating socket, the silicone rings help immensely in keeping them quiet.

Tarzanalog,

When you say "dull" do you mean like lacking in high-end clarity, or lacking in gain, or exactly what?

I'll have to check out those 6267's if I can find them for a decent price.

The EF 806's are just a supposedly high-end EF86, aren't they?

Thaks again for all your help.

[tarzanalog]

Dull = not as much sparkle as the Amperex or Sylvanias I have.

The Sylvania branded tubes are marked "Made in Germany" so I'm thinking they are from Valvo perhaps... they sound almost as good as the Dutch Amperex tubes and I can recommend them as well.

It's a shame these are getting so expensive. But I guess it's no big deal to spend $100 on a single pentode when it's the core of your sound in a $3500 Neumann U67 or something. I just got really lucky on mine.

FWIW - I have one Genalex Z729 and while they're supposed to sound amazing, mine tests at around 85% and has lost its luster and is my least favorite of the EF86 tubes I own.

[Holmes]

RDoc - In case you are interested this is my current schematic.

Holmes

Holmes, very helpful as well. Just getting OK w/ reading schematics, and I can see the similarities and differences in our amps.

Questions, if you don't mind:

What is the alternate V1 you have and what's that for? Different tone mostly?

Do you ever use NFB?

The top cut, is that like on the old AC-30's? How do you use that?

Do you have to mod the amp to use a reverb unit (free standing)

thanks again,

RDoc,

The 6SJ7 is an old pentode design that was used extensively in early PAs and amps including some Fenders. It doesn't quite have the quality chime that a great EF86 has but still is a good sounding tube. I was having problems with microphonic EF86s and had some known good 6SJ7s laying around so I just popped a socket into the chassis and wired it up. Both tubes work OK with same plate, screen & cathode components so I just paralleled the 2 sockets. Only run one tube at a time however.

NFB - Yes, use it a lot. Offers a little more treble, a little less gain and seems to offer more definition when you get distortion going strong.

Top Cut - Yes, similar to the feature on the AC-30. If you find your amp way too dark then it will be useless. Mine with appropriate speaker is reasonably bright and with single coils can sometimes benefit by having a bit of the bite trimmed off, which is what the top cut mod does real well.

Reverb - No, you don't if we are talking something like the Fender 63 RI Reverb or similar.

This is something different. I was fortunate to have a very kind and gifted contributor on AX84.com nurse me through building a solid state driver and recovery that was suitable for marrying a long delay Accutronics tank (like the Twin Reverb tank) to the 36 watt. This is definitely an extra-credit sort of project and is not guaranteed to work to your satisfaction. It uses amp B+ voltage to drive the circuit and is not very efficient, dissipating around 5 watts to do almost no useful amplification. In my amp I have an oversized PT providing extra power and put a little doghouse enclosure on top of my chassis to house the board and components.

The reverb only works on the TMB channel unless you jumper channels. I love it and use it almost all the time. It hums a little but is pretty innocuous.

Have fun,

Holmes

[Holmes]

RDoc,

How did your biasing come out on the EL84s?

From other posts it sounds like you are happy with the amp. I think the 36TMB + EF86 is just a killer design (if we can just find good EF86 tubes)!!

Holmes

[RockinDoc]

Hi Holmes,

Thanks for asking. Work's been heavy-duty lately, so I didn't have a chance to try the dissipation calculations until this AM before work.
I didn't have time to really pore over your instructions, but I wasn't sure about the cathode resistor voltage drop.

This is embarrassing, but I got it powered up, full power, pots zero, speaker in, and got stuck on the first damn reading!
So... to measure voltage drop across the 68ohm resistor, I tried to measure just probing the leads on either side, which gave me essentially the same voltage as the catode voltage, ie, 11V (the cathode pathway runs up through the half-power switch, and then over to the resistor, or at least I think that's the set-up. I tried from chassis ground to either side and got the same reading.
What am I dong wrong here?

Checking plate voltage, no problem. BTW, on half power the Rk will just be the sum of the two resistor values, correct? What about the voltage drop(s) with BOTH?

Gotta go, more info later. thanks a lot!

[Holmes]

R'Doc,

You are doing nothing wrong. All sounds OK. My directions may be a little terse.

One end of the cathode resistor is grounded. In full power mode the switch connects only the 68 ohm resistor to both pairs of EL84s. Measure voltage across the 68 ohm resistor and you have Vk for full power mode.

In half power mode you measure voltage across both the 68 and 62 ohm resistors. The switch connects to only one set of EL84s.

Rk in half power is 68 ohm + 62 ohm = 130 ohm.

Take two asprins and call me in the mornin (har, har).

RM

[RockinDoc]

R'Doc,

You are doing nothing wrong. All sounds OK. My directions may be a little terse.

One end of the cathode resistor is grounded. In full power mode the switch connects only the 68 ohm resistor to both pairs of EL84s. Measure voltage across the 68 ohm resistor and you have Vk for full power mode.

In half power mode you measure voltage across both the 68 and 62 ohm resistors. The switch connects to only one set of EL84s.

Rk in half power is 68 ohm + 62 ohm = 130 ohm.

Take two asprins and call me in the mornin (har, har).



Hi Holmes,

Har har indeed Doc jokes, gotta love 'em!

Sorry for not responding sooner, I was consumed by work for a bit.

Anyway, I got through the cathode voltage/current part, and the plate voltage (Vk was about 11V on full, 6.somethingV on half, Vp about 340V on full, 360V on half), but I could NOT get the readings on the screen grids to make sense.
Not sure what I was doing wrong. I tried reading the voltage on either side of the 82 ohm screen resistors, in reference to ground, and taking the difference, but there WASN'T any difference, and then taking voltage in reference to each side of the resistor, but the readings didn't make sense to me (IOW, they got me no where NEAR the usual range of dissipation that one would see, ie, 10 to 12 watts).

Frustration! Then, tonight I was at practice with my bandmates, and the first power tube after the preamp was red plating big time! I know it had to be that, because I was getting **** for sound when I turned up my guitar volume knob, and then I looked and the JJ's plate was glowing like a cherry popsicle! Aaaarrrrggghh! I flipped off the power pronto, and borrowed one of my buddy's (many) combo amps to finish the rehearsal.
I have been having great tone and no problems up until now. 10 to 20 hours of hard playing, usually on half power, through an Air Brake attenuator (Dr. Z's).

Any ideas? I am bummed out. The only thing out of the ordinary, is when I was in the midst of getting those readings of which I spoke, my probe slipped off the plate or screen pin of one of the tubes, I forget which component, but I don't even think it was this tube (the red plating one), and the probe hit the chassis very briefly (pop crack!)
The fuse didn't blow and neither did my circuit breaker switch off.

Help...

RM

[Holmes]

RDoc,

Don't worry about the screens for the time being. If you assume 3 mA per screen (each tube) and subtract from the cathode current Ik and calculate dissipation you should be real close.

You have seen red plating now and know what it looks like. Note that playing can push tubes that are biased on the hot side over the edge into red plating. The goal is to increase cathode resistance in increments until you are in the safe operating zone. The extra load on the PT with 4 tubes operating usually results in lower plate voltage and lower power dissipation on full power versus half power. Full power operation is usually safest on this design.

The probe slip, snap crackle pop thing happens to all of us sooner or later. If no fuses blow you are probably home free. Be most careful not to electrocute yourself, everything else is "small change".

Kind regards,

Holmes

[RockinDoc]

RDoc,


You have seen red plating now and know what it looks like. Note that playing can push tubes that are biased on the hot side over the edge into red plating. The goal is to increase cathode resistance in increments until you are in the safe operating zone. The extra load on the PT with 4 tubes operating usually results in lower plate voltage and lower power dissipation on full power versus half power. Full power operation is usually safest on this design.

Kind regards,

Holmes

Holmes,

Thanks for the prompt response, I do feel much better now after reading it.
I just used your 3mA per screen and my dissipation value is 12.2 watts, just a hair above the manufacturer's upper limit (I think it's 10 to 12 W).

Is that cool? is that the cause of the red-plating? and why only one tube?

Also, if I'm increasing my cathode resistance, what increments would you recommend to start with? And do I add or swap out or use wire coil resistors like those on the board or what? Again, assume I know nothing here Parallel or series arrangement?

One last thing: did I damage the tube that was red-plating in any significant way? Could I have damaged my transformer?

Thanks again, look forward to getting this resolved and getting back to playing. Don't think I could handle another rehearsal with my bandmate's Peavey combo...

[Holmes]

R'Doc,

My own experience has been that I don't notice a lot of difference between biasing at 12 watts vs 10 watts.

I have had a single tube in a push pull amp red plate before. It could be a tube that is defective, or is just not well matched to your other tubes. There may be other wiring, connection or component problems causing this as well but I am not knowledgeable enough to comment. If you switch tubes and have the same problem in the same position it leads us to believe it is not the tube and more likely something wrong with the build.

I would recommend you source the following 10 watt power resistors: 75, 82 & 91 ohm. Typically I order Xicon 5% cement power resistors from Mouser (P/N like: 280-CR10-VALUE-RC).

I would recommend you install the 91 ohm resistor in the half power position (probably where the 68 ohm is currently installed) and the 82 ohm resistor in the other location. Check dissipation with this arrangement. If below 11 watts see what it sounds like. Lower idle dissipation, if you like the sound will give you longer tube life. If dissipation is low (below 10 watts) you may want to replace the 91 ohm with the 82 and the 82 with the 75. That is what I currently run.

If you played long enough you could have destroyed the tube that red plated. They are generally tougher than you would think. Try the tube biased a little colder and see whether it is still functional.

Keep on trucking.....

Holmes

[PeterS]

Digressing way back to the start of the thread, you mentioned the voltages on your pentode, and that the screen was higher than the plate--- try increasing the value of the resistor that feeds the screen.

[RockinDoc]

Digressing way back to the start of the thread, you mentioned the voltages on your pentode, and that the screen was higher than the plate--- try increasing the value of the resistor that feeds the screen.

Peter,

Thanks, I will. Any particular value of increase that you might recommend?