need help with a tetrode screen supply / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

need help with a tetrode screen supply

previous 1 2 3 4 next

Move Thread

LAN_403

Steve Conner

Mon Apr 13 2009, 08:58PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

The resistor approach works fine, it just wastes a good deal of power for most transmitting tubes.

Another possibility that sometimes works is to connect the two grids together and drive both as a control grid. This can overdissipate the control grid sometimes. Or, connect the control grid to the cathode and use the screen grid as the control grid. You should be able to tell from the datasheet curves which method would work best. If you're not sure how to read the curves, maybe the tetrode should go back in the cupboard for now!

The self-rectifying approach is nasty and dirty, to be sure. The valve designers would probably cry if they knew what you were doing. But it's the easiest and safest route to a VTTC, as there are no big smoothing capacitors, and it also gives nearly the longest spark length per unit input power. (The longest probably comes from level-shifted power.)

Or maybe it's not so dirty. If the tube is specified for Class-C telegraphy use with high-level AM, then running in a Tesla coil off level-shifted AC is no different from AM with 100% modulation. (And as with 100% AM, the PEP is four times the average, which explains the more efficient spark growth.)

Harry, you should tell us some of these gruesome stories of vacuum tube meltdown. And how they happened, if you ever found out

Proud Mary

Mon Apr 13 2009, 09:46PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

I can't begin to imagine how this valve would perform in self-rectifying Class C - presumably it would hop into 'dynatron' mode (and perhaps mischeviously change frequency since there so few frequency determining elements) when it hit the 'negative resistance' zone of the tetrode kink, and then move on again. I have no idea.

As for the spectral purity and stability of the oscillation, well, I'm only an old codger with a shed at the bottom of the garden, and no longer have to concern myself with such things.... but I shall be interested to see the final results.

teravolt

Tue Apr 14 2009, 03:32AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

for the moast part boath of you are right Harry and Steve. A VTTC is not a Class C amp accept in it's basic configuration, But a VTTC is a oscilator. the conduction angle of the tube will be anyware from 15 to 40% an each half cycle so HF harmoncis will exist hopfully the low pass RL filter on the plate will fix that.

in regard to Dr. Sparcs resistor from the plate to the screen I found this scematic for a 4164a VTTC

witch might work. Hanging a resitive divider from the plate might wast to much power from the plate. I have concidered a shunt type power supply that uses 2 4cx300a that will track the plate for the screen

harry is right about protection of the circuit. because I may end up using tens of kilowatts some current monitor and tube arcing will have to be watched but it may have to wate till more of it gets built.

some of the problems I have is understanding the plots for the gu-39b-1 and understand how this will apply to A VTTC to keep the system in a SOA and then disigninign a proper circuit so any help is welcome thanks

Proud Mary

Tue Apr 14 2009, 02:06PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

teravolt wrote ...
for the moast part boath of you are right Harry and Steve. A VTTC is not a Class C amp accept in it's basic configuration, But a VTTC is a oscilator. the conduction angle of the tube will be anyware from 15 to 40%

ahem, perhaps you meant to write 'up to, but less than, 180 deg'

At the risk of causing further contention, and assuming that you are running the valve from reasonably smooth DC, I don't think these very simple screen supplies as suggested in places above are at all suitable for a multi-kilowatt valve. Simply strapping the screen to HT+ with a by-pass to earth is a set-up you'd use for a watt or two. The screen supply must be able to both source and sink current. The very simplest credible arrangement would be the potential divider with a by-pass capacitor in the bottom leg.

This has the advantage that if HT+ sags, as it will

the screen voltage will sag in something like proportion, so avoiding the possible calamity of the screen voltage exceeding the anode voltage, which will cause prompt failure.

I'm not recommending this approach, which is really inadequate for costly kilowatt installations, but it will, in principle, work, (big, fat, hot resisitors!)

However, at the kilowatt level, I feel that you must have proper screen regulation which can track the HT voltage, and trigger a disconnect safety relay in the kathode line if the screen current should deviate beyond safe limits.

Proud Mary

Wed Apr 15 2009, 04:25AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Something like this might be a good start for a hybrid design:

Dr. Drone

Wed Apr 15 2009, 05:58PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

teravolt

Wed Apr 15 2009, 08:01PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

thanks for your reply Dr. Spark sorry you had to divulge your project but it looks nice.How many watts? I would like to modulate the tesla at some date. right now I am in the setup phase. to Harry thanks for the scematic I have the artical and I also have a site to get boards for tetrode supplies. unforchinitly the pcbs are limited to 550Vdc. Dr. Spark has a point about a 1.4 kv screen voltage and using semiconductors for the shunt. I am planing to use tubes for the shunt. This way it should be flexible. I don't have a scematic yet but I am toying with idias. The circuit that you have desplayed could be made to work.

I have been doing a little reeding from the eimac book care of power tubes and the class C amplifier that they use for demonstration is biased along a load line, then decupled to a pie network for impedance matching. In a classic VTTC the plate is connected direcly to the tank and operated in a mode where the tube is oscilating only in the positive half cycle of the AC. I am thinking that one way of doing this is to biased the tube then decuple the ac into the primary then a stable sceen supply can be used. another option is to build a screen supply that varies with the plate. any coments wecome

Dr. Drone

Wed Apr 15 2009, 08:40PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

teravolt

Thu Apr 16 2009, 12:57AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

nice picture of your kitty, is she water cooled to

Dr. Drone

Thu Apr 16 2009, 01:08AM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

previous 1 2 3 4 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob