So this started a few weeks ago when someone found a picture of my 100 watt night light. Its an 833A tube. (new picture below) And he wanted a megatron made from 8 of these. (10 vac at 10 amps each tube) so 800 watts filament power just for the output tubes. By the time you make the +/- 1kv power supplies running at 40ma, and the power for the 300b driver tubes etc, way past the limit for usa 120v 20 amp service which is 1850 watts. And with the top of the tube at silly and dangerous high voltage...  bad idea.

So i started looking for slightly less silly alternatives.  I think the 211 tube fits the bill. running with +/-750v power supplies at 35ma. So a megatron with 211 output tubes, 300b drivers and a high gain input tube.

unless someone has a better idea.

This might be too crazy for me but why not go further.  I happen to have a couple of 30kWh, 400V battery packs so a DC-DC converter and make it "portable"?   

March madness, redux.

Was the original March madness in 2013?

2 hours ago, spritzer said:

This might be too crazy for me but why not go further.  I happen to have a couple of 30kWh, 400V battery packs so a DC-DC converter and make it "portable"?   

portable if you are godzilla or king kong....

 

I like the idea with the 211. It also helps keeping the noise from the filament supply at bay with the common cathode configuration. If matching tubes for the CCS sounds too much, we can always consider the 1700V depletion mode SiC JFET. Thermal management is going to be a challenge though at 35ma.

Can we call it juggernaut?

i like that.  i also like "david and goliath"

there will be 4 output boards. 12 x 6 inch each.  filament boards will stack on them, each filament board will have 5 x 5v 4a switchers. 2 switchers in series for the 211

so chassis is likely to be a bit over 24 inches wide and 20 inches deep.

 

updated picture

 

 

Edited March 30, 20242 yr by kevin gilmore

Sign me up, probably be cheaper than paying for heat in the winter.

@kevin gilmoreI think 833A tubes with 1Kv supplies would work with 230V mains 10A.

Nice Omron relay 3D model, don't you think? Zoom in for a good look.

Edited September 11, 2025Sep 11 by JoaMat

Very nice! A new JoaMat "interesting twist"-project in the works I believe ? 

Edited September 12, 2025Sep 12 by MLA

Have made schematic out of Kevin’s published gerbers of David and Goliath filament. From schematics a PCB layout following the original gerbers is made and the above 3D model is the result.

I’m addicted to make my own twist of others work. So, now can the process of shrinking Kevin's board start.

Just to be clear, David and Goliath project is insane. I like that.

Edited September 12, 2025Sep 12 by JoaMat

no real reason to shrink the filament board, but go ahead and try.  the output tube spacing is already at the minimum and that results in a specific board size. no reason for the filament board to be any smaller than that because all the holes need to line up. filament board stacks on the output board.

Yep, I’ve noticed output and filament boards match up to each other. 

Here is how it looks like on my screen, all output tubes side by side. Size 24 in x 12 in – huge.

Question regarding jumbo socket – what socket do you have in mind? This from a French company might fit.?

was planning on using the teflon versions on ebay. but i think that ceramic one is exactly the same size.

Thanks.

I avoid ebay. Goods tend to be stuck or disappear in Swedish customs.

Suggestions about +/-750V?

there are a couple of places that sell the same thing that are not ebay. 

circlotron power supply works fine for 700v, 800v etc.

Lovely, finally some use for circlotron psu. 

Most things settled - only finance issue… no prediction.

so here is the latest version of the kgsshv-carbon  with the c3m series parts including the current source. Good for up to +/-600v power supplies.

board file updated.

 

 

kgsshvcarbon600v.PDF

kgsshvcarbonv6.1c3cs - CADCAM.ZIP

Edited October 29, 2025Oct 29 by kevin gilmore

and the power supply that goes with it.  standard off the shelf antek transformers are $44. need 2

schematic fixed.

 

 

kgsshv600vpower - CADCAM.ZIP

circlotronpsupdated.pdf

Edited October 29, 2025Oct 29 by kevin gilmore

Does it have enough power to driver the Jackling Float?

Yes, it should have no problem with that.  The only issue might be gain and bias but easy to solve.  

definitely has enough power for that.  2400 volts peak to peak stator to stator at 25ma bias.

only otl dc coupled solid state amplifier design that can do this.

Q12 in the PSU seems full-on and not doing much if anything.

I noticed that the 2N3904-based current limiter is not there. Many people say it is too slow, not realizing that it is protecting the passing SiCFET at every power up, before the opto coupler soft start switch is released. On the other hand, if the SiCFET decides to give up, nothing can hold it back anyways. The 2N3904-based current limiter should not be the one to blame.

The amp section still uses PZTA42 despite the linearity issue we found earlier and better options are available. The STN9360 although is specified with a 600V VCEO, its curve shows onset of breakdown at a lower voltage. I didn’t take a picture with the STN9360, but the one below is similar. Technically the breakdown voltage is specified at point A, but you wouldn’t want to operate the part at point B, even though that’s still “within spec”.