All Triode Express - Standard Wreck Sized Chassis

[johnnyreece]

Are there any oddball tubes you DON'T have hanging around?

[vibratoking]

... lots of mellow and not so much chime or edge. This is typical of my experience with triode outputs.

That's been mine too. Very uninteresting. I've put it on a switch a few times...what a waste of a switch.

[Regan]

Another one to try would be 12a6's, probably get 7-10 watts out of it,,,, super cheap tubes.

[RJ Guitars]

Are there any oddball tubes you DON'T have hanging around?

I have a tube problem and my name is Bob...

[rp]

I have a tube problem and my name is Bob...

Many here have a tube problem. I always said we need a withdrawal subforum, at minimum a twelve step sticky. The good part is we could drink and smoke all we want at our support group meetings.

[BLT]

Hi Bob!

[RJ Guitars]

The most interesting of the tubes I have here is the 815 - aka R2D2, although it won't be part of this amp project.

I modified the heater wiring by running directly from the 12V DC heater supply to the preamp tubes then running a 2nd set of wires in series through the power tubes so I could try more options for output tubes... although I am thinking I will buy a few of the 12a6 tubes and see how they sound. That is an interesting tube and I appreciate the posting of that option.

As I sorted through my tubes I found almost a dozen options of uncommon guitar amp tubes in the 6_6 style. The 6G6-G tube was overly bright with no bottom end and I didn't like the sound of the distortion it produced. The best sounding tube I've tried so far in this experiment was a pair of smokey gray coke bottle 6Y6 tubes. Many of my options sounded really similar to the 12V6gt tubes I tried earlier (6W6 - 6U6 - 6K6) which is to say they sounded good but were really louder than I planned for this amp to be. One surprise entry was an afterthought on my part - I would give good marks to the 6F6 metal jacket tube (no photos yet) although again it was pretty loud.

One of the up sides to this whole effort is that I really dug in and learned some of the fundamental elements of understanding and using load lines. Because of the extreme differences in each type of tube I tried, I got creative on the bias supply configuration and added a trimmer pot to allow me an easy way to get a wider range of bias voltages.

More results to be posted but for now I hold to my thought that the 1626 triode doesn't give me the sound I wanted while discovering that the 6Y6 tube sounds really good!

[johnnyreece]

Excellent update! So, what do you think of the other 6Y6 varieties (I mean, the ones that weren't the smoky-glass coke bottles)? I may give a 6Y6 Express a run. Is the OT suitable for all those different octals? I'd not really reviewed the data sheets on these (and the other octals) yet.

[rp]

I think it's time to see if that R2D2 can fly a space ship or not. You must have something in mind for it.

[RJ Guitars]

Does anybody want to give a tutorial on how this can be done... the readers digest version first please if it can be done?

I think Merlin has a talent for explaining these sorta things:

http://www.valvewizard.co.uk/pp.html

Mike

I only had to read this about 25 times to start to get the concepts... lots more to be absorbed!

Excellent update! So, what do you think of the other 6Y6 varieties (I mean, the ones that weren't the smoky-glass coke bottles)? I may give a 6Y6 Express a run. Is the OT suitable for all those different octals? I'd not really reviewed the data sheets on these (and the other octals) yet.

I like all the 6Y6 variants I've tried... they all sound similar and all of them are quite decent sounding. I am hitting the plates with 259 VDC and you can see from the spec sheet that this is beyond the specs. I did read where the HiFi guys use them at voltages up to 300 on some occasions, but I'm not sure they are in push pull class AB. So I am still wrestling with the load lines to sort out the particulars but I might be tempted to VVR the plate voltage down a bit.

In all of the load line stuff I've found it's still hard for me to tell where I want to be as well as where I am at, but I did find with the 6Y6 tubes my load lines drawings started to look a lot like the "ideal" load line for the EL84 in the Valve Wizard document linked in the quote above, using my 259 B+ and the 8K PP output tranny...

Is there some optimum place for the load line to cross on the left side of the graph... I read something about below the knee for guitar amps?? If somebody is really sharp on these things, did I do things correctly and what would I do to optimize for this tube?

[David Root]

The data sheet for a single tube in pentode mode shows Ra=2600 ohms. Typically at that point a p-p pair would be good around twice that, or Ra-a=5K.

Your operating point is different, not as well centered on the load line, so maybe a little different. Being a bit off center gives a bit more harmonic distortion which is part of what we want. Plus your plate voltage is higher too. Maybe 6K or so at your voltage.

I did notice that the next page in that Tung-Sol datasheet shows that for a single tube in pentode mode the output power is virtually flat from 1K to 3K so it should be pretty tolerant of a wide range of primary impedance in p-p pentode too. Although that graph is at 135V plate not 259, or even 200V.

[RJ Guitars]

David, I knew you had a good understanding of this stuff and I was hoping you would post. I've tried to find a graph with a different center point but so far there doesn't seem to be any others... Do I assume the results are skewed if not irrelevant using the 135V center point?

In my study of load lines for use of the 6Y6 tube in this amp, what I found was that I was really studying amp biasing and the significance of transformer impedance on the bias point. The bias point I arrived at through the load line calculations I made didn't really give me the optimal amp sound according to my ears... I tweaked the bias current above the 70% Dissipation point of 33mA up to 44mA (with no correction for screen current so it's actually a bit less) and as I approached that 44mA value the amp really started to sound better and better... I stopped once I couldn't hear any improvement and the tubes don't seem to be complaining - not too hot to the touch and no sign of redplating. Am I deceived?

I have tons of questions, but maybe I'll ask if anybody knows of a primer for reading and understanding datasheets so I don't have to ask about each item individually.

In the meantime, one thing I'd like to know is the code to understanding the lingo so that I could intelligently fill in the blanks on the "Class AB push-pull poweramp calculator" © Lars Soderquist 2007... some of it is sorta obvious I know, but the rest of it I need some help with.

What is:
Gain PI
Ua
Iaq/tube
Us
Rs
RL a-a (ac),
RL a-a (dc)
R transformer secondary
R1 NFB
R2 NFB


Ugq
Ua line 1, Ug=min
Ua crosspoint line 1 & 2
Isq/tube
Is/tube at line 1 Ug=min
Ugq + Ugmin line 1

[David Root]

Gain PI is just that-the gain of the phase inverter (PI) you are using. 25 to 30 is a fair approximation for most LTP PI. Or you can calculate it see Merlin's book on long tailed pair PI. If you have TubeCad that makes it quicker.
Ua & Us is V plate (Vp) and V screen (Vs) respectively. Note he specifies Ua at max power output, so maybe 90% of Vp at idle.
Iaq I believe is anode current, quiescent ie at idle.
Rs would be the screen resistor size.
RL a-a (ac) and (dc) would be the primary impedance (ac) and DCR (dc) respectively I think. I could be wrong here, I don't get "DC impedance"-DC cancels in a p-p primary.
R transformer secondary has to be the DCR of the OT 2ry.
R1 NFB and R2 NFB are the ratioing NFB resistors. I would assume R1 is the one connected to the OT 2ry tap and R2 is the one on the base of the PI below the tail resistor.

The others are a bit more diffuse, I'd have to look it over a bit more closely but you can infer a lot from what I gave above.

Never seen this calc. before, will check it out.

In the case of the 6Y6G, I get the impression that 135V is the max screen voltage allowed at 200 Vp. If you pull up the Sylvania 6Y6GA datasheet off tubedata.org it has a sheet on page 4 that indicates max screen voltage allowed as a % of the max allowed screen input voltage. I guess they are pointing to the use of some hefty screen resistor values. If you look at the screen current curves on the plate curve sheet you can see why.
None of the 6Y6 sheets I pulled up show any other screen voltage than 135V so I think that is Absolute Max.

[RJ Guitars]

Still working on calculations and lots more to work out on understanding this amp... but it is sounding awesome. I decided to put it into a cabinet and work it a little bit. If these 6Y6gt tubes can handle what I am hitting them with, this could be one of the best amps I've put together. It has better clean sounds than any of my other Express builds, is super quiet (probablty due to the DC heater supply), and has a very sweet and rich distortion. It doesn't hit you in the face as hard as the full sized Express - I think because I really knocked the signal down going into the PI and it all seemed to hit a great sweet spot. It's about half as loud as my 35W Express builds and weights half as much as well... will the tubes last?

[johnnyreece]

The only thing I noticed with these is that the filament current is brutal. 1.25A each...Yowch! I may see if I can get one of these done sometime in the near-ish future. I think wifey may insist I sell something first...