Author Topic: 833A Maximum Dissipation Ratings (Read 3670 times)

kyledellaquila · « **on:** October 25, 2020, 11:16:17 PM »

Hello! Advice time

Here is another 833A post... and somewhat in line with what @MRMILSTAR was concerned about in the other thread about excessive red-plating of the 833A vs 833C (whos' forum post I have just came across today).

For my purposes, I will not be using the classical tickler coil for feedback. The grid of the 833A in my application will be driven by a Solid State (designed by Ernst van den Bergh) generator that produces +150-0-150 square waves.

So in order to keep the valve operating in the MFGs recommended specifications (to avoid excessive Red-Plating), I will need to bias the grid down. To what amount? I could use some help with determining that.

(http://www.ur4ll.net/833a.pdf)

On 4 of the datasheet, last section. It says 4KV and 450mA,
that would be 1800W (see DC Plate Input) but the plate dissipation is
only 350W.
Ernst indicated to me that the resistance calculated from 4KV @ 450mA is impedance and not resistance... hence why the tube does not simply set ablaze when 1.8kW is being ran through it... So what is this 350W plate dissipation number in respect to? How do I stay below this danger point?

I plan on putting a variable DC supply (0-220VDC up to 480W) that can handle the grid current needs as shown in my sketch below:

T3sl4co1l · « **Reply #1 on:** October 26, 2020, 02:49:25 PM »

To calculate plate dissipation, you need the load characteristic, and some estimation of how the tube behaves between saturation to cutoff. Procedures can be found in e.g. RC-26, and probably the equivalent transmitter tube manual. If you're running a resonant primary, and sinusoidal grid drive, the load is probably mostly resistive, and you won't have to model an elliptical load line (or anything higher order due to harmonics).

Still not very straightforward, as plate load and tuning varies with amplitude (breakout length).

Tim

Mads Barnkob · « **Reply #2 on:** October 26, 2020, 08:19:37 PM »

You can roughly estimate your plate load impedance and with the primary specifications you can match that with your tank capacitor, there is a little math on it here: http://kaizerpowerelectronics.dk/tesla-coils/kaizer-vttc-i/

kyledellaquila · « **Reply #3 on:** October 28, 2020, 06:30:05 AM »

Here is a schematic for what Ernst and I have in mind:

The driver supply will have a +150v -150v bipolar supply. The output will be crisp square waves sync'd with the resonant frequency of the secondary (feedback via current transformer at the ground connection)
My HV supply will be limited either to 500W or 1000W at 2kV.
In order for me to keep the tube consuming more than my supply can dish out... I figure the smart move here is to have an adjustable bias supply. (adjustable DC ranging up to 200V)

This is how I figure based off of the chart (amplitude modulating the driver supply will be less sufficient in modulating than bias control.. is this sound?):

And for Plate dissipation, I found this quote from that RCA Receiving manual that Tim suggested.

So for Class-C activities, it would appear that the tube is ~90% efficient right?

T3sl4co1l · « **Reply #4 on:** October 28, 2020, 11:57:11 PM »

Heh... class C is typically 60-75%, with an upper limit of I forget what, maybe 80%?

Tubes in class D can do a bit over 80%, but may require quite high grid/screen power dissipation, or careful drive and load matching, to achieve.

Tim

kyledellaquila · « **Reply #5 on:** January 22, 2021, 01:26:13 AM »

ok, so now I have a fully functioning 3kV DC power supply.
Has anyone had any successes with cathode modulation of an 833a?

I came across this forum page and got some ideas for a simple modulation technique:
http://amfone.net/Amforum/index.php?topic=43982.0