i'm on a roll... the kgsshv

[Remolon]

It is D1, output stage

[livewire]

The resistor selection thing has been discussed before in this thread as well as the T2 thread.

No time to research all this again for you, so i'm speaking in general.

The Xicon though rated for 350 volts tend to work well,

although in one case somebody posted a pic of one that had shorted to the ground plane of the pcb.

If you have doubts, stand them off ~1/16" so the resistor body is not touching the pcb.

Again, if you have doubts 1/2 watt wont hurt. They fit and the brown Dale ones (CM60?) work well.

KG also employs the method you mention about stringing resistors in series to divide down

the voltage they see when a practical one component alternative is not available.

Looks like your pcb is already populated so all this is prolly moot.

 

One concern that I have is which plastic shoulder washers have you used on the heatsinks.

Please list the part number that you ordered / used.

They had better be the deeper ones or you'll be seeing fireworks when you power your amp on.

The correct plastic shoulder washers are a bit oversize and require drilling out the metal transistor tab

hole a few thousands of an inch (I used a dremel rasp) or pressing it in under pressure if you are lucky.

This too has been previously discussed in depth.

[insanity]

I used the washers that extend into the ceramic plate. I had to drill the holes on all TO220 packages. Part No. 7721-3PPSG.

 

I'll be trying the XICON 1/2W then, maybe elevate it a bit. 

FYI: I did a search on "kgsshv and 787k" and did not find the asked information. I did not know, that the T2 has the same layout in this section. 

 

The resistor selection thing has been discussed before in this thread as well as the T2 thread.

No time to research all this again for you, so i'm speaking in general.

The Xicon though rated for 350 volts tend to work well,

although in one case somebody posted a pic of one that had shorted to the ground plane of the pcb.

If you have doubts, stand them off ~1/16" so the resistor body is not touching the pcb.

Again, if you have doubts 1/2 watt wont hurt. They fit and the brown Dale ones (CM60?) work well.

KG also employs the method you mention about stringing resistors in series to divide down

the voltage they see when a practical one component alternative is not available.

Looks like your pcb is already populated so all this is prolly moot.

 

One concern that I have is which plastic shoulder washers have you used on the heatsinks.

Please list the part number that you ordered / used.

They had better be the deeper ones or you'll be seeing fireworks when you power your amp on.

The correct plastic shoulder washers are a bit oversize and require drilling out the metal transistor tab

hole a few thousands of an inch (I used a dremel rasp) or pressing it in under pressure if you are lucky.

This too has been previously discussed in depth.

[livewire]

@ insanity - Since the conversation was IXYS "in general",

 I was referring to the LED in my amp where IXYS was only an alternative in the third stage.

I just checked, this pcb revision is much different than my prototype.

Do what these guys say on the fourth stage.

[livewire]

The T2 does not have the same layout as the KGSSHV although much of the parts selection wisdom does apply here too.

(they are both HV stat amps released by KG around the same time)

[livewire]

I used the washers that extend into the ceramic plate. I had to drill the holes on all TO220 packages. Part No. 7721-3PPSG.

 

Good! Them are the correct ones.

[spritzer]

No, scrap that.  I was working from the second oldest version of the PCB and it didn't have these changes.  Now I see what Kevin did... 

 

All the parts labeled A should not be used with IXYS parts. 

[insanity]

Ok thank you very much. Do you happen to have the schematic on hand for my version? I don't have it. If you give me your address Birgir, I'll send you some chocolate

No, scrap that.  I was working from the second oldest version of the PCB and it didn't have these changes.  Now I see what Kevin did... 

 

All the parts labeled A should not be used with IXYS parts. 

[livewire]

Now that's participatin!

[kevin gilmore]

this matches the current board

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

 

send me the chocolate

[kevin gilmore]

for all those people stuck with a very over priced chassis...

in progress.  stereo single board about 8.75 x 7.5

designed for +/-400v. But of course you can run it at +/-450v or +/-500v

and it needs the +/-15v power supply

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

 

there will be an alternate plug board that's the srm323 version

that runs only on +/-400v without the +/-15v power supply

[Emooze]

Someone in charge of putting parts in bags didn't eat their wheaties:

 

 

I needed more of the 10m90s (right) and what I got is on the left. Customer Service isn't open until Monday either...

[kevin gilmore]

1) those are mosfets and you really should not be putting them on material like that!

 

2) what are those things on the left anyway? the right thing in the wrong package?

[insanity]

I got the same IR receivers as you, when I ordered the 10m90s. But my bag only contained those, not a mix.  But I have the correct replacement already.

Someone in charge of putting parts in bags didn't eat their wheaties:

 

 

I needed more of the 10m90s (right) and what I got is on the left. Customer Service isn't open until Monday either...

[kevin gilmore]

that's a pretty good screwup on mousers part. if you buy them 25 at a time like birgir and I do, you

get a tube of them, pretty hard for them to screw that up.

[insanity]

BTW: Why would anyone want to change the feedback with the amp powered on? I would not dare to stick pincers in there when it is on.

[kevin gilmore]

One person wanted to wire in a toggle switch!

[spritzer]

This shit is like tuberolling, people want to do that on the fly as well. 

Edited September 14, 2013 by spritzer

[Remolon]

Thanks Kevin for the new schematic.

 

I do not have any 2SA1968. So, I am considering the option of using 2SC4686 for the current source (Q5, Q6). Do you think that it is a good alternative or it is better to go with the optional current source (IXTP01N100D for Q7, Q8)?

[kevin gilmore]

You cannot use a npn to replace a pnp as the current source as q5,q6

It's either the 2sa1968 or the ixys part.

[Remolon]

You are right. The current source need a PNP transistor but the 2SA1968 is a NPN one, and I was looking for an alternative to the 2SA1968 and not for a high voltage PNP, according to the schematic. Sorry.

[kevin gilmore]

The 2sa1968 is a pnp

It is incorrectly listed in a few places as a npn

[Remolon]

This is the Data Sheet that I have

 

http://www.sm0vpo.com/_pdf/2SA/2SA1968.PDF

[RiStaR]

I've looked into this in the past and I think the PNP part is: http://www.sm0vpo.com/_pdf/2SA/2SA1968LS.pdf

 

Only reason why I found those is that I chanced upon some 1968s that were marked 1968LS but they are non-isolated - they test fine in the transistor tester so I assume they are remarked 1967s? The ones that I got from Sato Denki don't have the LS label but they are isolated like you expect.

 

At this point I'm testing the non-isolated ones in the Megatron that I'm building... just waiting on the transformers and my LT1021s to show up before I can power everything up.

[Kerry]

As a rule of thumb, anything marked 2SA.... is a PNP and anything market 2SC.... is an NPN.