Tweedle-Dee Build (lengthy wrap-up)

[Raoul Duke]

Hi All,
Been in “lurking and learning” mode for awhile and recently decided to put together a Tweedle-Dee. First step was reading the 22 page thread originated by Charlie - which is a gold mine of info; then on to mhartman’s documentation in “Files”. All very informative and interesting! Many thanks to both those guys as well as all the other contributors! Charlie’s practical approach and narrative (to include the pictures) throughout were my guides and I learned quite a bit - so thanks again Charlie!

I didn’t do a build thread because I wanted to take everything I’ve learned from this fine community in the past year and try to apply it without assistance unless ABSOLUTELY necessary. Anytime I got confused or didn’t understand something - I searched this site as well as EL34.com, Aiken’s info pages, and of course Blencowe’s info pages to review and get a handle on what I was doing. I know this amp isn’t terribly complicated compared to my last three (and I probably should have started with this one) but it really connected a lot of the principles I’ve picked-up over the past 12 or so months in a way that was easy for a novice to follow - so thanks to everyone who’s imparted knowledge to me as I've started this hobby.

Here’s the wrap-up for any beginners like me who want to see the “journey”:

- Used the AES stainless chassis. Thought about saving some dough and using an aluminum blank and doing my own cutting - which I WILL do next time. The AES is nice, but IMO was tricky to work with. Drilling holes was very hard - even with the correct type of bits for stainless; and the stainless reacted differently while chassis grounding with my 100w iron. It all worked out fine in the end - it’s just frustrating to me that two things that are relatively simple for me became somewhat complicated due to my choice of material. Looks purty though.

- Was tempted to add some mods, but stuck to the TAG files with three small exceptions: added a filter cap bleeder resistor, back-up diodes to the rectifier socket HV in, and the fourth input. Otherwise built as spec’d (with 10M PI grid leak).

- Used the “TM 6/14” board plan with an added 33k grid stopper for the 4th input. This board didn’t line up with the stock mounting holes in the AES chassis, and you can’t just use one and extend the other due to the layout; both need re-locating. This also caused the OT holes to need to move apx. 1/4” c-to-c because the new board hole location put the screw slightly under the OT. Tried to find different solutions to these two fairly minor issues and this is how I solved them. I’m sure there are other solutions had I realized these things earlier in the process - but “live and learn” I guess.

- Based on a thread highlighted by sluckey, I opted for the Hammond 290BX PT and used a Mojotone LENCO OT to round-up to free shipping on my tweed order, lol.

- Speaker is a lightly used MOD 12-35 I’ve had in a box for about 15 years. Sounds pretty good - might stay.

- Cab looks close to a TD, but I took a half inch off the depth because I had a 1x10 in the garage. I measured everything 17 times just to be sure the speaker and transformers all had some room to breathe and it worked out fine. Did the floating baffle, but didn’t cover it - opting for the grill cloth frame/velcro approach. Front panels are 3/4” thickness vs the thinner Tweed Deluxe type. First time working with tweed and it worked out great. Experimented with Mapei 711 flooring adhesive to stick it and it worked very well (I already had it, it was cheap to begin with, and only needs one coat on the wood and nothing on the fabric to stick). As with most adhesives - the key is letting it dry and using heat to reactivate as/if needed.

-Experimenting with different rectifiers (5Y3 vs 5V4) and currently have a mid 60’s Sylvania 5Y3 in. The 5V4s seem to sound as though they give more headroom, but the lowest I could get with them (with the current 250R cathode resistor) was 15.6w @ 111.4% per tube on a set of matched JJ 6V6S. I tried four 5Y3s and 5V4s each and the one that sounds best right now actually had the lowest voltages and dissipation. Here’s the final voltages as the amp begins to break-in:

B+1: 365
B+2: 330
B+3: 286
B+4: 285
PT Plates: 354
Volts across cathode: 20.7
V1a: 153.1/1.0
V1b: 157.3/1.0
V2a: 239.2/40.5
V2b: 189.2/1.4
I get 13w per tube @ 92.9%
DISCLAIMER: This is my first cathode bias build - so if my math is off, please correct me. Always ready to learn. I’m fairly confident because I used the Rob Robinette bias calculator page - but if I mis-applied any of that data - it’s my inexperience showing.

I have JJ 12AX7 in both pre and PI for now. I’m sure I’ll try some others. Also thinking about trying some TAD 6V6GT-STRs. Got a set in another amp sitting here, and I believe they are a 14w tube as well.

Overall real happy with this amp. Sounds good and I’m proud to say I didn’t have to bug anyone to get it going, lol. Thanks again everyone; on to the next.
Here’s some pics:

[martin manning]

Nice work! One thing that I recall there was some debate is the 10 Meg grid resistor the PI, which IIRC was revised to 1M.

For the plate dissipation I get:
Va-k = 365 - 20.7 = 344.3 V
Ik = 20.7/250/2 = 0.0414 A
Pa = 344.3 * 0.414 = 14.25 W

[Raoul Duke]

Thanks Martin, I appreciate the compliment and showing the work! I actually understand it, lol.

I have some 10R/10 watt resistors I can put in series as well as 10w 270, 300, and 330 to sub in if needed.

In your opinion, is this something I should be contemplating or do you think at 14.25w it should be safe to let it break in some and see where I land after awhile?

Also, I went back and forth with the 10M vs 1M in the PI given all the info exchanged by those who know better than I - but opted for the 10M based on Charlie’s final thoughts on the matter. Like I said - I have a bag of 1Ms to try eventually and may do so just to compare at some point.

All very interesting for such a basic amp. Good stuff!

[martin manning]

The JJ's are rated at 14W so you are just about 2% over. If they aren't red plating you can run them like that, but I would be inclined to get them to run closer to 13W.

[Raoul Duke]

Understood. Quick follow-up regarding terminology:

I measured what I thought was B+1 at the junction of rectifier pin 8 and the 4k7 dropping resistor/33uf filter cap - and measured plate voltage at pin 3 on the power tubes. The pin 3 measurement is how I calculated the bias on the calculator using both methods, but I see that you used what I may have mistakenly called the B+1. Am I using the wrong terminology and/or using the wrong voltage measurement; or possibly both?

Using the layout voltages as reference, I surmised the B+ nodes as:

B+1: 398.1
B+2: 359.4
B+3: 311.7
B+4: 310.9

Just want to make sure I’m using the right terms. Thought I had it figured out based on the flow but hoping to gain the correct perspective (and not run my tubes so close to the edge next time, lol).

Thanks again!

[martin manning]

I missed your PT Plate voltage and grabbed the B+1 as “close enough.” The voltage at the plate pin is the one to use.
Va-k = 354 - 20.7 = 333.3 V
Ik = 20.7/250/2 = 0.0414 A
Pa = 333.3 * 0.414 = 13.8 W
%Pa Max = 100*13.8/14 = 98.6 %

[Raoul Duke]

Awesome, thanks for the info as always Martin!

I feel good about 98.6… Something about it feels familiarly “just right”, lol.

I’ll probably pop the 270R in today and see where it takes me.

[Guy77]

Congrats on a great looking build! Beautiful cab too.
I have all the parts for one of these in my basement just have to find the time to build it !

Its fun to experiment with tubes.
I built a 5e3 many years ago and liked how the Groove tubes 6v6GT (sovtek) sounded.
They also handled running at close to 100 % with no problems.

Cheers
Guy

[Raoul Duke]

Thanks Guy, good info and much appreciated!

I’m starting to enjoy the fine tuning more than the building lately - but it seems to be a cyclical thing for me.

Trying different tubes never gets old though. I have an old box of a dozen various-branded NOS 6V6GTs from the ‘60s sitting here (bought at a “going out of business” from a local TV repairman about 30 years ago) and I’m tempted to cool the bias way down to try those; but I’ll probably just stick with the 14w tubes in this application for now and let it break in.

Those tubes intrigue me and scare me at the same time, lol.

[Raoul Duke]

Ok, put the 270R (measured at 272.3 when warm) in with the same Sylvania 5Y3GT; here’s what I came up with:

B+1: 374.5
B+2: 340.4
B+3: 296.3
B+4: 296.8
Plates: 363
Voltage across cathode: 22.3
Plate to cathode voltage: 340.7
Plate current: 38.7ma
Dissipation: 13.2w @ 94.3%

Pre and PI just for info:
V1A: 160.2/1.1
V1B: 162.3/1.1
V2A: 248.6/42.3
V2B: 196.6/1.4

I think I’ll stay here as I play it in a bit more unless anyone sees something I might be missing in the voltages.

[dorrisant]

Very nice build, Marc. You should be proud!

The JJ's are rated at 14W so you are just about 2% over. If they aren't red plating you can run them like that, but I would be inclined to get them to run closer to 13W.

Not to derail but those JJ6V6S can handle a lot. Out of curiosity, have you run any tests to check past the said limit for those JJs? It may be good for another topic. If you were in on that, I'd pull up a chair and get some popcorn. I understand that what you have stated works for this particular circuit and am not suggesting any major deviation... I would just state that in my opinion, I wouldn't be worried at all about being a slight bit hotter on the bias just for the tube's sake. That said, I think you'd be hard pressed to get them to redplate.

Disclaimer: Relative to JJ6V6S only!!

[ijedouglas]

Those JJ 6V6s are damn near indestructible!

I ran a test for a year on my Tweedle Dee (voltages were almost spot on with layout) after building it, running it on a dummy load for 8 hours a day 7 days a week. The only thing that failed was one of the Ruby filter caps. I still have the amp and its still on the same set of tubes 7 years later

[martin manning]

Very nice build, Marc. You should be proud!

The JJ's are rated at 14W so you are just about 2% over. If they aren't red plating you can run them like that, but I would be inclined to get them to run closer to 13W.

Not to derail but those JJ6V6S can handle a lot. Out of curiosity, have you run any tests to check past the said limit for those JJs? It may be good for another topic. If you were in on that, I'd pull up a chair and get some popcorn. I understand that what you have stated works for this particular circuit and am not suggesting any major deviation... I would just state that in my opinion, I wouldn't be worried at all about being a slight bit hotter on the bias just for the tube's sake. That said, I think you'd be hard pressed to get them to redplate.

Disclaimer: Relative to JJ6V6S only!!

I looked at the JJ 6V6S using my newly finished uTracer 3 (10 years ago now), and found them to be similar but not exactly like 6V6GT. I haven't traced them at the voltages the OP is running, but that is possible now with uTracer 3+. They seem to be ok at the stated 14W Pa. It would be interesting to see where the typical 8k load line would fall on the curves at Va = Vg2 = 400V. I just don't happen to have any new ones around at the moment.

[Raoul Duke]

Thanks dorrisant, I appreciate the compliment and thank all the great teachers I’ve had on this site!

I also have read a lot about the JJs being pretty durable as well and may take you up on that experiment; however, there’s no way I could be more thorough than Ian’s test - which I think speaks volumes for those tubes.

Speaking of great teachers, thanks Ian and Martin! You guys have been very patient and generous and I am always grateful for your expert tutelage, sage advice, and value-added commentary throughout this site. I continue to learn from you both (and many others) just researching past threads randomly as I “connect the learning dots”.

My wife asks me why I keep building these because my 102 was originally supposed to be a “one and done” kind of thing; but like I tell her - I learn at least a hundred new things every time I build one and it’s incredibly rewarding to hear the end result. It’s actually inspired me to play differently and stretch my musical learning more.

Haven’t found a downside to this hobby yet - other than space - but I’m probably going to start moving some stuff after the holidays to make room and raise capital… FOR MORE BUILDS, lol!

Thanks again guys!

[Colossal]

Marc,

In response to the Mrs.' inquiry about how many amps do you need, and, one-and-done, I will politely inquire as to the number of pairs of shoes she has and if more than six linear feet are required to house them.

One pair of shoes should be enough, right?

Well, that's settled. When are you building your next amp then?