I have seen bigger caps than the 470uf caps mounted horizontally, so I'm sure its ok.

 

a singles version of the 900v supply would be 5.95 x 4.5 inches

 

http://gilmore.chem.northwestern.edu/circlotronhvpowersingle.jpg

 

so each side of the heatsink with everything mounted vertically from the front would be

 

driver board

 

output board

 

900v power supply

 

900v power supply

 

hv power supply

 

so minimum height of the heatsink needs to be 5.95 inches

and the boards stick out a maximum of 4.5 inches

 

definitely fits in the 5U box ??

transformers on the back panel??

 

2 inch spacing between the driver and output board

 

then 3 inch spacing on the power supply boards (or whatever the cap height is)

Edited July 4, 201511 yr by kevin gilmore

How about mounting the driver/output boards vertically & the PSU boards horizontally? Should put less stress on the caps & leave enough space for toroids on the bottom plate (this is assuming 500mm depth). I can include a rudimentary sketch if needed.

if it fits, it fits.

 

not a bad idea to do the hv power supplies horizontal and stacked

depth might be an issue

Good to see this one moving forward........

 

RE: Cases and enclosures? Who needs em, have em built Don Garber style, open frame and easy to touch. 

Edited July 5, 201511 yr by DefQon

This is an amp you'd really want a 6' stick just to turn it on.... 

that was also part of my reasoning for using a preamp... something with a remote so I can listen comfortably behind the blast shield

its really no different than the kgsshv, its the same +/-450v output swing.

 

and the 900v supplies are the same as the original supply, they just float.

 

no big deal, nothing to see here, move along.

 

made the single power supply a bit smaller.

 

5.95 x 4.35 inches

 

by the way, the pdf of the kemet capacitors specifically says they

can be mounted in any position.

Edited July 5, 201511 yr by kevin gilmore

this is the new power supply schematic. i like it much better than the previous version.

 

new board soon

 

http://gilmore.chem.northwestern.edu/goldenreferencehv900new.pdf

 

runs on much less voltage, so for the dht amp running at 50ma, less heat

also less noise than the old version, and no pile of zeners

impossible to keep up with your progress, not that that's a bad thing

 

When's the DHT amplifier coming out? Will there be a kickstarter?

"350V/us", does that mean you plan to bump the current in the output stage to 42mA (assuming 120pF load)?

 

Would it be safe to use the original PSU board for the floating supplies if one lowers the voltage from 900V to 700V  then the original 450V caps can be used I assume

should you actually find music that will drive the amp at the speeds necessary, the

output stage will drop out of class A and go into AB.

/

Edited November 11, 201510 yr by JoaMat

kevin gilmore said:

"this is the new power supply schematic. i like it much better than the previous version."

 

And I assume if you ground the input side and the output side at the junction of the two 470 uf caps you can use this PS for the KGST, KGSS, KGSSHV Carbon, BH, etc - with appropriate adjustment for the output voltage?  BTW, shouldn't R8 be 890k nominal?  Will need to find a 1 kV rated resistor for that one, or else three 350v resistors in series.

yeah, there is about 5 volts of offset, no biggie

 

board posted there are 3 resistors in series for the feedback

and its 890k total on the board.

 

no, you definitely should not try to use this as a dual voltage supply

because only one side is regulated.

 

like this

 

http://gilmore.chem.northwestern.edu/circlotronhvpowersinglenew.jpg

Edited July 7, 201511 yr by kevin gilmore

apart from a lower voltage swing I assume the circlotron can be doneusing battries at 700V rather than 900V?

and then you have to change the driver power supply to +/-350

Thnx, can you share the part number for the 1kV terminals you are using? (cannot seem to find any at Mouser)

http://www.mouser.com/ProductDetail/Phoenix-Contact/1902547/?qs=sGAEpiMZZMvZTcaMAxB2AKJ8wpDgQJg5WwkpmYrVFB4%3d

has anyone used/traced this http://gilmore.chem.northwestern.edu/kgsshvcarboncirclotron.jpg....? seems like the upper left opto hasn't been connected at pin4/5 (assuming no ground plane) ?

updated picture

thnx...looks ok to me, thnx for the link to Mouser 

Oh, what a great project! So It will be 8 boards+ 3 transformers to build the circlotron?

10 boards plus 3 transformers plus 5U case

 

per side

1 x hv ground reference

2 x hv900

1 x driver

1 x output

For some reason thoughts of flies and sledgehammers comes to mind.

Edited July 20, 201511 yr by JimL

 

i also don't know where to get sheets of that silicone shown above

 

 

Bergquist do SIL PAD which Farnell/Element 14  have and also similar transistor retaining springs

 

http://uk.farnell.com/fischer-elektronik/thfu-3/clip/dp/1892352?ost=transistor+spring+retainers

 

Good luck with this project!

Edited August 1, 201511 yr by headinclouds