57 DELUXE REISSUE W/NOS 5Y3: CAN I SAFELY SUB 6K6GT's FOR 6V6's WITHOUT SERIOUS MODS?

[Mango Moon]

Greetings All,

I tried posting this same thread elsewhere here on MEF (under Mods & Tweeks), but I didn't get much of a response, so I'm going to risk censure by double-posting it here.

There's been a fair bit of discussion here on MEF and elsewhere over the years about taming the power output/volume on a 5E3 by swapping-out the stock (12w) 6V6's for NOS 6K6's (8-10w), and I'd REALLY like to make this change to a '57 Deluxe Reissue equipped with 5Y3 rectifier, if possible, but its not particularly clear to me whether I can safely do so without making other modifications to the amp. In particular, its not clear to me what the plate voltage (B+) is likely to be on this Deluxe Reissue (with 5Y3) or whether I may need to change the PT or other components to reduce that voltage below the maximum specified for a 6K6GT (i.e., <315V).

I've received some limited feedback from knowledgeable others (elsewhere) indicating that it should be no problem whatsoever, but does anyone here concur with that assessment or have anything further to add on the topic? Basically, are 6K6's directly swappable (plug-n-play) with 6V6's in such an amp, without the need for other mods? Or is the plate voltage likely to be too high (>315V), even with NOS 5Y3 rectifier installed?

Many thanks!

[loudthud]

A pair of 6K6s is going to require a cathode resistor of something like 400 to 500 ohms to get the plate dissipation within limits. Then you would really like a higher load impedance of 16 ohms to match the plates better. But people put 6V6s in Twin Reverbs and they don't go up in flames so it might work. (They of course adjust the bias.) Plate voltage is higher than the spec but guitar amps do that all the time.

I liked the idea someone had in the other thread about grounding one of the output tube grids. But, you will get different results depending on which grid you ground. On the plate side of the concertina you are ok but on the cathode side you are turning the concertina into a gain stage. The corner frequency of 56K and 0.1uF is 28Hz. Who knows, maybe you will like it.

What I'm trying to say is go ahead and try it. You might fry a pair of 6K6s and you could harm your amp if things don't go well. Shut the amp off if the tubes red plate or if you see smoke or something smells bad.

[Mango Moon]

Thanks Thud!

What you're saying here is consistent with what others have said (i.e., plug 'em in and see what happens). Barring a complete meltdown which I'm told is highly improbable, it sounds like the worst that could happen is to fry a power tube or two (or rectifier), and I'm good with that, but yes, I'll be watching things very carefully for 5-10 minutes just to see what goes on. And if something does indeed go awry, there's a mainline fuse that I suspect would go on the fritz first pop.

I've been told by a reputable source that the plate to cathode voltage is probably around 330-360vdc, so I should be well within the margin of error for 6K6's (i.e., 315vdc), since I'm running a 5Y3 rectifier, and since the ratings of most tubes (by the manufacturer) are based on 100% duty cycle, not the 65-75% safety margin that most of us use when biasing our amps.

Anyway, thanks again for the great feedback and I'll try to follow-up here in a day or so, after I've actually given this a shot. And in the odd chance you don't hear from me ever again, well, I guess you'll know what happened, eh?? ZZZzzzzzzzzzzzzz . . . . . Bwa-haaaaa!

[tubeswell]

Going back to what loudthud said, it less about the overall plate voltage in this instance, and more about the plate dissipation (which you need to regulate by having the appropriately-sized cathode resistor). Recommended maximum plate dissipation for a 6K6 is 8.5W (vs 12W for a 6V6).

Plate diss. in cathode-biased stages is: plate-to-cathode voltage x plate current.
(and plate current is tube current minus screen current)

[Mango Moon]

Tubeswell,

Thank you for the detailed response and for the technical information it provided, but I'm afraid I'm not a formally trained amp tech, so unfortunately, much of this is over my pay grade. I've received instructions (elsewhere) on how to determine plate and cathode voltages at idle for the original 6V6's in this amp at power tube socket pins ("lugs") #3 and #8 to ground, which will allow me to calculate the actual plate voltage, but I'm afraid its not clear to me how I would go about determining plate current or how I would calculate plate dissipation from those measured values.

[tubeswell]

Plate current + screen current = tube current

Tube current is the current that flows through the cathode (and the cathode resistor in a cathode-biased stage).

If you measure the cathode-to-ground (DC) voltage and divide this by the actual (measured) resistance (in ohms) of the cathode resistor, you will get the tube current. (Ohms Law: Current = voltage/resistance). Say the cathode was measured at 18V and the cathode resistor actually measured 470R, then: 18V/470R = 0.0383A (or 38.3mA)

Similarly, to get the screen current, you can either measure the voltage drop across the screen grid resistor (if you have one) and divide the voltage drop by the actual measured resistance of the screen grid resistor. Say there was 1V drop across a measured 470R screen grid resistor, then: 1V/470R = 0.00021A (2.1mA).

Or you can estimate about 2mA for a typical 6V6 (or 5mA for a typical 6L6).

Ipso facto

38.3-2.1 = 36.2mA plate current.


Or

You can measure plate current more directly by measuring the actual DC resistance across the OT primary, and then dividing the plate-to-platesupply node (DC) voltage by this OT primary DC resistance, in order to get the plate current. Say the DC voltage drop was 2.2V and the OT Primary DC resistance was 60R, then: 2.2V /60R = .0366A (36.6mA). (You need a sensitive R-meter setting to measure a 2.2V drop)

To get the plate dissipation, the formula is Volts x Amps = Watts. (Plate-to-cathode voltage x Plate current = Plate dissipation)

Say we measured 350V plate-to-ground (DC) voltage with the 18V cathode voltage in our earlier example. The plate-to-cathode voltage would be 350-18=332. We then multiply this by our calculated 36.2mA (0.0362A) plate current thus:

332V x .0362A = 12.01W. The recommended maximum plate dissipation from your average 6V6 tube datasheet is 12W. So our example is bang on for a 6V6 SE stage in Class A mode