Determining Plate-to-Plate impedance of a tube pair

[Gaz]

I'm specifically trying to find out the plate-to-plate impedance of JJ's Kt77 reissue, but the datasheet does not have the AB1 typical or limiting values.

Is there a formula to figure out the plate-to-plate impedance with only single tube info?

Here's a link to the sheet:

http://lib.store.yahoo.net/lib/thetubestore/JJ-KT77.pdf


Thanks for any help.


Alex

[Chuck H]

I don't have any "formula" that you asked about. But I can say that whatever the tube data sheet says doesn't apply as strictly to guitar amps. A couple of reasons, 1) the tube will be producing most of it's current in a narrower frequency band than the data sheets assume, and 2) the tubes will be intentionally overdriven... Often and repeatedly.

Look at how Fender and Marshall (and many other MFG's) seemed contrary to the tube data in their products.

AFAIK kt77's often used the standard 6.6k primary so popular with most of the other big bottles. I'd assume that for a guitar amp of moderate plate voltage that you could fudge that number the same way you would for, say, el34's or 6L6's. Like a 4k or 8k primary. I can't report on the tonal differences this would make but I can say that for safe operation it's usually better to go with a higher primary resistance as you increase plate volts.

FWIW I saw an old Genalex data sheet that also listed the kt77 single tube plate R as 23k. These data sheets DID show other typical operation modes like AB1. It showed data for 400Vp using a 4.5k primary, 500Vp using a 5.5k primary and 600Vp using a 9k primary. But notice I wrote "showed data for"... A lot of guys think that the operating demonstrations in the tube data is a spec for how the tube should be used, it's not, it's a demonstration of how the tube behaves under certain circumstances.

You don't have to be too anal about it. There's lots of fudge room with guitar amps as primary impedances go.

HTH

Chuck

[J M Fahey]

Hi gaz.
Rather than a "formula", the curves shown allow us to design graphically.
Strictly speaking, most usual datasheets lack some information about tubes, necessary for us guitar amp makers.
What I mean is, as you can easily see, on the right one (I call it gr-rt) they provide full A (plate) current versus G1 (grid) voltage curves, *only* for a single G2 (screen) voltage, that being 250V, a value you know *nobody* uses, the usual value being around 400V.
This means that those curves will be shifted "upwards", providing a lot more current at any grid voltage shown *or* that grid voltages will have to be a lot more negative than shown, for a given plate current, which ends being the practical case.
The graphic on the left (gr-lt) provides some indication of what we can expect with higher screen voltages, although in an incomplete way: they show plate current vs. grid voltage only at a fixed point: Va=Vg2.
From both graphics I can extract some useful data:
gr-rt gives me a reasonable set of currents to work with, although grid voltages suggested will be wrong.
gr-lt tells me that for a Vg2 voltage of 450v , I'll need a "no signal voltage", or "bias" of around -48Vg1 for a 30 mA static current.
There is no 400Vg2 curve, but I can estimate (interpolate) the bias around -43Vg1.
Now on to the A (plate) load:
it should be found graphically, but it can be approximated by the formula: RA (Plate load):
(delta)Va/(delta)Ia ; delta meaning "variation"
There are many pairs of values that can be chosen for particular reasons, we all know that tubes often work happily within a range of impedances, but in this case I choose these, based on these published graphics:
Va no load: 430/450Va
V supply full load: 400V (I assume an excellent power supply, well filtered, SS diodes)
Ia static: 30 mA.
Ia saturated: I choose 240mA ; it means I'll "lose" around 75V, the tube can go no lower at that current .
Delta Va= (400-75)V=325V
Delta Ia= (240-30)mA=210mA
Ra=Plate impedance=325V/0.21A=1550 ohms.
Raa=Plate to Plate impedance = 4x Ra= 6200 ohms.
Of course the 6600 ohms suggested elsewhere will work very well too, it's very close, but this is a calculated value based on the provided datasheet.
I personally would *measure* the actual tubes used, datasheets are always "generic".
No, you don't need an expensive curve tracer (although it would help a lot) ; I measure them on the actual amplifier, with the actual power supply plus an auxiliary one to provide the saturation voltage.
Results can be eye-opening and it's easy to weed out bad or poor quality tubes.

[CitizenCain]

Wow, just the info I was looking for too! Thanks guys

[Gaz]

Thanks guys, those were really the answers I've been searching for!