Clementine 6L6 Build

sweet & juicy SE amp for 1626 Darlings and the 6L6 family

Chubby Clementine on Steroids

Postby EWBrown » Sat Nov 05, 2011 6:19 pm

As an interesting Saturday evening experiment, I pulled the 6L6WGCS out of the Chubby Clementine, and put in a matched pair of "golden S" Svetlana EL34s, and the sound is noticeably nicer than with the great sounding Russki 6L6WGCs. It also delivers about 50% more output power with lower 2HD and 3HD. These toobs have 1.6 amp filaments, vs the 6L6s' 900 mA, and the B+ current for each EL34 is 67 mA each vs 58 mA for each of the 6L6s, so I am running the power tranny at the "edge of the envelope" as far as the 6.3VAC @ 3A and the 120 mA B+ current rating (136 mA vs 118 mA). A PA-774 should be able to handle the extra curernt demands with ease, as it can deliver 5A 0f 6.3V and 180 mA of B+.

After a couple hours operation, I can still put my hand on the power transformer, for about 30 seconds without feeling any thermal induced pain or raising any blisters :'( (try that with an ST35 sometime)

I did not change any resistor values, they are all original for the 6L6GC design.

The tranny ran only "warm to the touch" under the 6L6s and 6SL7 loading, but I will monitor the tranny's temperature during extensive testing. The tranny is a one-off prototype, and I suspect that it was conservatively designed and rated.

I ran a SE-CAD simulation, the 6L6WGC has a mu of 7.5, rP of 1.7K, gm of 4.43 mA/V, IK of 58 mA, PO = 2.46W @ 6.2% 2HD and 0.4% 3HD.

The EL34 has a mu of 7.12, rP of 1.07K, gm of 6.67 mA/V, IK = 67.2 mA, PO = 3.60W 2HD = 4.7 mA, 3HD = 0.2%.

/ed B
Last edited by EWBrown on Wed Jan 11, 2012 10:01 pm, edited 1 time in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby Shannon Parks » Sun Nov 06, 2011 8:36 am

I have done some preliminary testing with some JJ EL34s (all triode strapped and no feedback), and indeed, the results were very nice. The lower Rp and higher mu just really make them shine. I'll probably tweak some in the ULtra circuit and see if I can get close to 10W. I love this power transformer, too!

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby EWBrown » Sun Nov 06, 2011 4:33 pm

I should have added, that i kept the NFB loop (2.2K and 100 ohms) with the EL34s, and that may have actually enhanced them. They seem about the same input sensitivity as with the 6L6s, so that should be compatible with signal sources with more "timid" output voltages. My 6SN7 VA/CF tube linestages can push out over 10V rms, so even the most "deaf" power amp can be driven to full output power [:) With the 23 step attenuator, I generally only haev to go to the "12 o'clock" position to get what I would call an ideal listening SPL (which can be pretty LOUD)

I also SE-CAD modelled KT77s on the Clementine, but I didn't have any available to actually try them out. These tubes gave me the best overall results of power output, and low 2HD and 3HD, at least in the "virtual" world.

I should also try some KT66s, I have a quad of the Russian made ones, yet to be used.

I have a matched quad of the early release J/J KT77s, but these had the under-sized "skinny" pins, and they aren't compatible with most normal octal sockets, because they just don't make good contact. Maybe on my next build, with some "virgin" sockets which haven't been stretched to the normal octal pin diameter. FWIW< these are perfectly good with a new Chinese made octal socket, their receptacles are really a tight squeeze, and I almost need a rubber hammer to drive in a new octal tube with them. Also some of the 40s-50s vintage US made, ceramic octal sockets are also similarly a "tight squeeze".

The later release J/J KT77s have the normal diameter pins, and they are good performers, I tried and really liked them, in an ST70. I just don't want to rob its tubes just for an experiment...

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Tubby Clementine, with 12V6GTs and 12SL7!

Postby EWBrown » Wed Jan 11, 2012 10:07 pm

I finally finished the "Tubby Clementine" with 12V6GTs and 12SL7. It is a "hybrid" between the 6L6GC version, and the 1626 version, the resistor values are all the same as for the 6L6, but I included the Darling's 12VDC filament voltage doubler circuit, to keep everything on the board all DC voltages.

The main difference, is that I used a smaller power transformer, a leftover 285-0-278 @ 120 mA, which I bought from Musical Power Supplies on e-bay several years ago. These trannies are no longer available, but they have "new and improved" versions, which translates to "yeah, they cost more now" :/ $)

Right now, I'm running it in triode-strapped mode, with the same NFB resistor values, as with the 6L6 version, and it sounds very nice!

It has a bit less maximum output power, but I'm going to try the UL conenctions later, though I probably will have to change the NFB resistor from 2.2K to a lower value, perhaps 1K, to get the best combination of power output and low THD. Before doing any of that, I'l put some hopurs on it to make sure everything is "broken in".

Photos of both versions, coming soon....

Next (the third or subsequent one) Clemmy build , I'm going to evaluate if it is reasonably feassible to do some etch "surgery" on V2 and V3 in order to allow the board accomodate 12AV5GAs, as I really like the 6AV5GA and 12AV5GA souns and power capabilities. I could always just to the jumper wires to a standard octal socket, but I prefer to keep the sockets soldered directly to the PCB, if possible.

/ed B in NC
Last edited by EWBrown on Sun Jan 15, 2012 2:26 pm, edited 2 times in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby Shannon Parks » Fri Jan 13, 2012 7:13 am

Sounds cool, Ed! I *do* plan to test some 6AV5s eventually. ;)

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby EWBrown » Sat Jan 14, 2012 10:17 pm

I did the requisite voltage and current measurements, I've got 325VDC P-K across the 12V6GTs, at 43 mA so that's maxing them out at the 14 watt PD rating. The 12SL7 is running at about 1.44 mA per triode, which is where I expected it to be, pretty well puts the plate voltages at half of the driver stage B+ supply, where I prefer it to be.

It may be worthwhile to pull out one of the two 1K. 2W cathode resistors for each V2 and V3, and replace it with 1.2K / 2W just to bring up RK from 500 ohms to approx 545 ohms just to lighten up the cathode current slightly, which would also give me a couple more volts of maximum control grid swing voltage.

I'm definitely happy with the sound from the 12V6GTs, much better than I was really expecting.

Nice thing about having only DC power voltages on the PC board, is that there is absolutely NO hum, buzz or ripple, the background is "black" quiet as can be. [:)

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby Shannon Parks » Sun Jan 15, 2012 11:48 am

EWBrown wrote:It may be worthwhile to pull out one of the two 1K. 2W cathode resistors for each V2 and V3, and replace it with 1.2K / 2W just to bring up RK from 500 ohms to approx 545 ohms just to lighten up the cathode current slightly, which would also give me a couple more volts of maximum control grid swing voltage.


Yep, those dual 1K resistors have been fixed in the latest schematic and parts list. I like a 1.2K//1.8K combo for max grid swing. Linearity loss is minimal even with the big power tubes like EL34s and KT88s.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby EWBrown » Sun Jan 15, 2012 2:25 pm

IT all depends on the output tubes and plate voltages, I found 1K + 1K in parallel were perfect for the 6L6HCs and EL34s with my particular B+ voltages. - lower voltage, miore current approach. The 12V6GTs are a tad more "dainty", so they shouldn't be overstressed too much, even though they're relatively inexpensive.

I'll run the amp for a while before trying any cathode resistor value changes. Maybe they are like 6AV5GTs, in that perhaps they can take a lot more "punishment" than their rather "shy" PD ratings.

Right now I'm listening to a Procol Harum CD through it. Sounds pretty good, though the CD's recording quality isn't the absolute best, it's the old AAD variety, and not digitally mastered. Little rough around the edges, and the SE tube amps bring out the recording quality imperfections.

I'll be buying some more Clemmy boards, for future experiments, including modding the circuit to take two, and make each one a floating paraphase / PP monoblock.

Pretty simple, jsut a couple of minor changes around the 12SL7, and using one PP OPT rather than two SE OPTs. 1940s "old school" technology, with a modern twist [:) O:) ;) (lol)

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Previous

Return to clementine

Who is online

Users browsing this forum: No registered users and 17 guests