question about cathode bias on 6L6's with increased B+

[terminalgs]

I've got a Valco built Airline 62-9052A that I found sans tubes and transformers. Luckily, Classictone still makes the 5Kohm:8ohm OT that Triwec built for all those various 2x6L6 Valco amps from 1961-68,  and TAD 6L6GC-STR's fit in the small clearance area of the cab.  The only original-spec part that I struggled to source was the PT.

The original supplied two pairs of output transformers, each with two 6L6's plus four 12AX7's and a 6973 reverb driver. 5.5A @ 6.3V and a 320V secondary.  I simply could never find iron to fit that bill *and* physically fit the cabinet/chassis. So I found something I pulled from an old amp that I figured would work.  I also decided to built out only one of the two power sections (i.e. 1 6L6/pair/OT instead of two), which if you look at the schematic is pretty easy since they share a PI.

So I got it all together and full of tubes and everything worked! However.. there evidently was an oversight..  ahem.  Instead of 360V at the plates, I have 483V. cough...cough...  Obviously my carefully labelled tag on the PT was based on some not so careful measurements?  but, hey it sounds great!!  I also had to replace several electrolytics that were over the 450V rating, and in the end I had to re-work the power supply a bit (Valco used a bunch of 20-10-10 cap-cans that they'd double up (either two 20 to make 40uf or two 10's to make 20uf), so I found places inside for some F&T 500V axial caps.

Question:  I'm a little on the fence about what to bias the 6L6's at.  I looked a bunch of Valco's with the same output stage configuration and found plate voltages ranging from 360V to 400V, and current per 6L6 from 46ma to 83ma. With a 250ohm (measures 246) resistor, I get 36V @ K ~73ma, with a 330ohm (measures 336) resistor I get 40V ~ 59.5ma.  Based on what valco did with these amps, I'm leaning toward the 250ohm resistor, but is something I'm not considering since I've increased plate voltage by 35% ?

• B+ 488V (center tap of 5K:8 OT)
• plate 483V
• screen 470V
• cathode 40V (with 336ohm resistor).

Schematics:

• original: http://i.imgur.com/5SQy4wD.jpg
• modified part-1: http://i.imgur.com/rC410tt.jpg
• modified part-2 (PS): http://i.imgur.com/6e3SX4U.jpg
• tubes I'm using: http://www.tubeampdoctor.com/images/File/TAD-Datasheet_6L6WGC-STR%20Version1.1-08.07.11.pdf

Here are some other 2x6L6 Valcos that are cathode-biased with the same OT:



By the way, this is one of those Valcos where a stereo jack was used for the output and the two OT's each fed either the tip or the ring.  there was no marking on the back of the amp to say "use a special cable", and the burned tube sockets one the side of the "ring" output suggest the use of a mono cable is what caused its earlier demise that probably led it to get gutted.

Another mod was that replaced the pricey 6973  reverb OT driver with a 12AT7  ala Fender.  it's not drawn on the schematics above, but the power supply's mod to lower the 12AT7's plate to 406V is in there for the "F" node.

I originally posted about this in 2014 here: http://el34world.com/Forum/index.php?topic=17320.msg173420#msg173420

[John]

IF I'm figuring right, it all gets down to plate dissipation of watts. You have 40 volts being dropped by 336 ohms, which is .12 amperes. Take the .12 x 483 = about 58 watts pdiss, divided by 2 tubes is 24 watts per tube. Should be fine. I'm not taking into account the screen current, but you're not idling right at the 30W max per tube, either.


If you put in a 250 ohm K resistor, most likely you'll notice the plate voltage drops a little too. Conversely, if you put in a larger value K resistor, your plate voltage would rise a little.

[PRR]

> Instead of 360V at the plates, I have 483V
> Obviously my carefully labelled tag on the PT was based on some not so careful measurements?

Utility AC is Sine, measured RMS. Our rectifiers tend to "peak catch". The peak of a 360Vrms sine is 1.414 times higher, 509V DC.

So 483 was to be expected.

> what to bias the 6L6's at

"Bias" includes plate voltage. Cathode bias amps must run HOT. If Vp is increased, Ip must DEcrease so the tubes don't melt. Then by R=V/I, the Load Impedance must increase to properly absorb the power. From 360-400 to 480V, the load should be 1.4 to 1.8 times higher, 7K to 9K. But these are odd values and is not the OT in hand (chassis).

So go to the STOCK bias. I'm amazed at the spread you found in your research. I figger almost 400V is a fine target. If it comes out 380V or 360V, that seems to be good-as Valco intended.

> Schematics:      original

Gah!! Who designed that cap-ghetto??

Without mucking much with the Vintage Design, _I_ would insert a big resistor and another cap. Being cathode-bias the current won't change much, a resistor will give a semi-steady B+. Take the 480V on a new 500V-600V cap, insert ~~700 Ohms, and feed your "400V" point, going to OT B+ lead and all those other R-C filters. A side-effect is much lower ripple, I hope that is not a problem. If you really need the B-flat-flatt overtones in OVERdrive, this may need more thought.

[PRR]

> replaced the pricey 6973  reverb OT driver with a 12AT7

6573 "is" a high-rated 6V6, a suitable sub for the 6F6 in Fender's classic stand-alone reverb. If you need THAT much booiiing, you need that much tube, and 6V6 will work with these 6573 values. Of course many people think the 12AT7 reverb is all the surf they need.

[terminalgs]

> replaced the pricey 6973  reverb OT driver with a 12AT7

6573 "is" a high-rated 6V6, a suitable sub for the 6F6 in Fender's classic stand-alone reverb. If you need THAT much booiiing, you need that much tube, and 6V6 will work with these 6573 values. Of course many people think the 12AT7 reverb is all the surf they need.

When Valco went bell up in 68/69, someone bought the amp division and continued supplying Montgomery-Ward with these amps for another year. they are nearly identical except they used a 6V6 instead of a 6973. regarding the 6973, when they started using it as a reverb driver (61/62) they were using 6973's as power tubes so they had some buying power.

[terminalgs]

Without mucking much with the Vintage Design, _I_ would insert a big resistor and another cap. Being cathode-bias the current won't change much, a resistor will give a semi-steady B+. Take the 480V on a new 500V-600V cap, insert ~~700 Ohms, and feed your "400V" point, going to OT B+ lead and all those other R-C filters. A side-effect is much lower ripple, I hope that is not a problem. If you really need the B-flat-flatt overtones in OVERdrive, this may need more thought.

PRR, thank you for your help!


I was afraid there would be a gotcha with this parts-bin-special.  as-is I'm out of room inside the chassis, for either a cap or 3 big resistors. I'll have to move to a couple of can style caps (which isn't a problem)  it might be the weekend before I can try it out.


and I suppose using a 16ohm sec. load would make the load on the primary side too high for 6L6's cathode bias...


I'd like to get it right because the amp sounds really good.  The two preamps are different and nice, and the reverb/vibrato is nice too.

[terminalgs]

I finally got back to this amp.


I installed this PT: http://www.classictone.net/40-18103.pdf


Which is an upright 300V@200ma.  I had to lay it flat to fit in the chassis (stole some end bells off a spare tranny that happened to have the right lay-down dimensions).


Now, the plates are 410V, the screens are 400V and the cathodes are 30V with a 220ohm 10W resistor. 


 

[terminalgs]

Valco put these two 6L6 bottles pretty close together.  clearance to the top of the cabinet is about 1/2".  After the amp was powered up for 30 minutes, I checked the temp of the roof of the cabinet: 204deg.F.