[heatsink]

Hi Jim,

thank you for chiming in. For the 007t I'm using a pair of heatsinks like the SK 92/50 AL from fischer elektronik (if I recall correctly). Each is about 100x40x50mm @ 1.9°C/W. Each heatsink has two insulated 10M90S on it. Should be something like 5,6W per heatsink, right?

Unfortunately, I'm not able to tell which fet "goes south" first, as you phrased it.

What other influence other than thermals could lead to the gate (of either one or both fets) getting punctured, so to speak?

 

[JimL11]

That should be adequate heat sinking. I am assuming you are using ceramic high voltage insulators and heat sink grease to attach the 10M90S. No other obvious thoughts.

 

[heatsink]

Yeah, I thought so too, thanks for the confirmation. I'm indeed using these white ceramic insulators and thermal grease as you did. And bushings to insulate the screw from the fet, of course.

Exceeding the Vgs limits could harm a fet and limit its capabilities. Could that may be the cause, either during power-up or switching the amp off?

Spritzer used some caps on his CCS board, apparently between GND and B+. Are they meant to protect the fets?

That's so annoying, I never had any issues with the CCS in my 006t.

 

[heatsink]

Any other idea what to look into? I don't know what else to try...


Do you guys think it's reasonable to buy some new 10M90S (originating from another lot, hopefully) and give them a try?

 

[heatsink]

Found this link https://db.audioasylum.com/mhtml/m.html?forum=kandk&n=9229&highlight=oscillation&search_url=/cgi/search.mpl?forum=kandk

Could that help?


Apparently, finding 10M90S is very hard these days...

 

[kevin gilmore]

the 1k gate resistor was added to prevent oscillations.
the oscillation is way beyond 20khz, so you need a scope to find it.

 

[heatsink]

Thanks Kevin, too bad I don't have a scope with said capabilites. Do you think the suggestion would be beneficial?

Could osscilation harm the 10M90S permanently?


People were having problems like mine earlier, as stated e.g. here https://www.audioasylum.com/cgi/vt.mpl?f=tubediy&m=252144&VT=T

 

[kevin gilmore]

this is a different application with different currents, don't know if it makes any difference.
however i have never seen or heard of a problem with my current source board, or any of my boards
with the ccs circuit on them. (kgst,bh,kggg, t2 with this ccs, grhv) etc.

maybe the 1k resistors you are using are spiral cut film resistors, that certainly could cause trouble.
the dale resistors definitely work fine.

 

[heatsink]

Thank you for sharing your experience, Kevin, I appreciate that a lot!

I'll certainly look into dale resistors instead of whatever I'm currently using.

 

[Firschi]

Has anybody ever seen getter (?) material chipping off 6S4A tubes for no apparent reason before?

Any idea what the root cause is? Balance of the affected channel is about 275V now, even after replacing the tube, yikes. Offset can be adjusted to about 0V.

 

[KDS315]

Has anybody ever seen getter material chipping off 6S4A tubes for no apparent reason before?



Any idea what the root cause is? Balance of the affected channel is about 275V now, even after replacing the tube, yikes. Offset for L+ and L- can be adjusted to about 0V.

Nope, not seen that. Overheating, mechanical stress like a hard bump etc (?).

Did it short circuit anything inside? Even massive imbalance AFTER tube replacement indicates an issue in the surrounding electronic circuit in my opinion.

 

[kevin gilmore]

that is way to thick to be getter material. also looks to be the wrong color.
likely a piece of mica insulator.

 

[Firschi]

I guess you're right, Kevin. Should have looked at the tube more closely before posting. The chip is very, very thin, though, and also very shiny on one side.

That's why I confused it for getter material. One piece of it even made it to the lower deck.

 

[kevin gilmore]

definitely a piece of mica some of which got getter on it when the getter was flashed.
its possible to shake the other piece down to the other mica where it won't be able to short out anything.
so the tube might still be usable. but i would never use that tube again.

as it is looks like the plate is shorted to the cathode.

 

[Firschi]

I certainly won't use this tube again, that's for sure.

Since the amp was moved and I shaked around the tube itself a bit, I can't tell what exactly was shorted against what. The chip is not attached to anything, in moves freely, more or less. In your experience, what short would explain high balance und close to 0 offset voltage?