how to generate plate curves for higher screen voltages

[pula58]

how to generate plate curves for higher screen voltages than shown in a data sheet?

Like, for example, running a JJ 6V6GT tube with a screen voltage of 420V?

[gingertube]

Shape of the curve will be pretty much the same.

Measure 2 or 3 points, plot them on graph paper and freehand sketch between them. That will be good enough for a +/-10% tolerance device like a vaccuum tube.

Cheers,
Ian

[martin manning]

Like this: https://tubeamparchive.com/viewtopic.ph ... 281#237281

[pdf64]

Ian, just wondering, where did you get the 10% tolerance figure from?
Pete

[pula58]

When all is said and done, I think I am heading in Martin M's direction: All I really care about is the Vg1= 0 curve, and will use the approximation found in RDH4:

current scaled by factor = (new screen voltage/old screen voltage) ^ 1.5

[martin manning]

Don't forget to scale all voltages by Vg2new/Vg2old too.

[pula58]

Don't forget to scale all voltages by Vg2new/Vg2old too.

In the amp I am designing I'll have the quiescent screen voltage be about the same as the plate voltage. So , my reasoning is like this:


I'll be aiming for Vscreen=Vplate = 400V

The Vg1=0 line for a 6V6gt tube (see top plot of attached data sheet), at Vscreen=250V,Vplate=400V shows a plate current of about 114mA

using the equation: new current = old current * (400/250)^1.5
or, multiply y-axis (plate current axis) by a factor of 2.02

so, for the 6V6gt operating with 400V screen and plate voltage 114mA on the y-axis of the attached plot becomes scaled to 228mA, 50mA becomes 100mA...etc. Thsatis, just scale the entire y-axis by a factor of 2.

Does this seem correct to you guys?

P.



http://tdsl.duncanamps.com/dcigna/tubes ... 6gt-4h.gif

[Phil_S]

Build one of these: http://www.youtube.com/watch?v=QEVHcf1iJj0
Thread at AX84: http://www.ax84.com/bbs/dm.php?thread=478112

[pula58]

Build one of these: http://www.youtube.com/watch?v=QEVHcf1iJj0
Thread at AX84: http://www.ax84.com/bbs/dm.php?thread=478112

incredible, love this!

[jazbo8]

The voltage scaling method or the chart used in RDH4 is accurate for scaling <+-60%, so the 2x multiple you used is a bit too high. Here is a worksheet done by someone over at MEF awhile back, I tweaked the graphs a bit so it reads better (or worse) I have not used it much, so please double-check for errors. The idea is to use whichever set of data/graph from the datasheet that is closest to your target voltage.

https://dl.dropboxusercontent.com/u/132 ... 0Table.xls

[gingertube]

Ian, just wondering, where did you get the 10% tolerance figure from?
Pete

Well to be honest, it is an educated "butt pluck".

What? You think I was a bit optimistic? - Maybe I should have said +/- 30%.

With tubes, when it comes to operating points etc. then +/-10% is not worth worrying about. Outside that tolerance you may wish to have a closer look to see if it of concern or not.

Cheers,
Ian

[pdf64]

Ha, 'butt pluck', that's a good phrase!
I was wondering if RDH or any ancient tomes suggest a figure.
Tube info sheets don't mention any tolerance / acceptable variation etc on actual performance characteristics; so how are tube testers calibrated, what the the rationale for sectioning an area of the scale as red/not acceptable?
Pete

[martin manning]

The voltage scaling method or the chart used in RDH4 is accurate for scaling <+-60%, so the 2x multiple you used is a bit too high.

The language in RDH4 says that voltage scaling factors from 0.7 up to 1.5 will be approximately correct. The voltage scalar above is 400/250 or 1.6. However, if I use the GE data sheet for 6L6GC, which has Vg1=0 lines for a range of Vg2, and scale the knee down from 400V to 150V Vg2 (0.38 factor), or up from 150V to 400V (2.67 factor), the error reaches at most about 5% in current, and around 5V.

[jazbo8]

The language in RDH4 says that voltage scaling factors from 0.7 up to 1.5 will be approximately correct. The voltage scalar above is 400/250 or 1.6. However, if I use the GE data sheet for 6L6GC, which has Vg1=0 lines for a range of Vg2, and scale the knee down from 400V to 150V Vg2 (0.38 factor), or up from 150V to 400V (2.67 factor), the error reaches at most about 5% in current, and around 5V.

Good catch on the accuracy range wrt to the RDH4 conversion chart, it should be 0.7 to 1.5. I was thrown off by the 2x scaling figure mentioned above. Should have read the post more carefully...