I'd check the bottom just to be sure.  Then you can start replacing parts. 

I'd check the bottom just to be sure.  Then you can start replacing parts. 

 

Problem solved.

 

I started testing resistors and found 2 bad ones R35 and R37 both 100R, measured 140K and 450K.  Considering these 2 probably saw 450V, I also pulled the adjacent Q13 and Q14.  One tested bad; replaced them both.  This is in the vicinity of the 1000V film cap that shorted to ground.

 

Both balance and offset now set near zero.

 

Thanks for all the tips guys!

Great news!! 

Movin' on to the other amp board now.

you forgot to tell us that you let out the magic smoke

  Magic smoke indeed.

 

 

there is a thread over at diyaudio saying that since wima has moved production to china,

quality has turned to absolute shit.

 

time to buy caps from vishay

Cant find that thread. Where are Vishay made

Sent from my iPhone using Tapatalk

no idea. probably china too.

 

what evidently happened to wima is that when they switched to china

they changed the sizes of most products, making them smaller with

different films that are more tightly wound.

Good thing I stick with my surplus of Russian and German NOS caps for high voltage applications. Or modern Epcos and CDE, Wimas and Siemen mkts are good enough for low voltage stuff.

Pretty sure the Vishays are Taiwan or Malaysia.

Anyway I'm doing a 2 box solution for my kgsshv with the trafos and psu board in a separate enclosure whats a good connector for the umbilical connection?

the same connectors used in the diy-t2 but less pins.

ugly but rated for the voltage and not plastic crap

 

amphenol industrial circular connectors

Cant find that thread. Where are Vishay made

 

Needing a few 1KV caps and not wanting to trouble Mouser for a small order from England, I got

EPCOS - B32653A104J - CAP, FILM, PP, 100NF, 1KV,  made in Spain

 

VISHAY ROEDERSTEIN - MKP1840422104M - CAP, FILM, PP, 220NF, 1KV,   made in Portugal

 

from Farnell who do sell in the USA.  as Farnell Element 14

I decided to try something new for the next round of HV's:

 

 

I like the "black line" look and it will fit with the next batch of amps using some exciting transistors. 

Exciting as in easily available and cheap?

Yup and it's a mosfet that actually doesn't suck for this role. 

Even as a notorious Silverist, I will concede that that looks pretty fucking trick.

Turned on my kgsshv today, heard some louder than normal lamination hum along with some crackling, opened things up and nothing seems to be amiss except for the transformer being extremely hot, and a particularly acrid smell. I used the specs in ujammerstad's BOM so I don't think the specs are at fault...

 

450VAC output for +/-500VDC rails Sumr part #: RC0100 050 3 100VA / 117V primary 2 x 450V (@125mA output ea) secondaries 1 x 30VCT (15V-0-15V) 10VA secondary Core band Static shields

 

I didn't realize these transformers retain heat for so long too...makes me worried about the SumR units I had potted. Is this retention normal, or is something happening inside?

Edited June 11, 201412 yr by nopants

I shall maintain the Silver bulwark.  Silver forever, semper fi.

The specs are just fine as I have SumR units here which worked just fine with that spec.  Richard just did something wrong somewhere and we are paying the price.  The units I have that actually work were after I got rather cross with him...  The BHSE units I was having trouble with got very warm on the Blue Hawaii so we'll see soon how they are on the KGST. 

 

The transformers I had made in Hong Kong were designed from the point of keeping max temp rise low so they run cool and very few have any noise at all.  Not cheap but as if I care. 

What's extremely hot? It is sitting on a thick rubber mat? A short to ground may explain the sudden heat...?

I've measured my SumR in my BHSE at 70C. I've since re-housed it in a well ventilated metal box, temps are much better after a few hours (40 - 45C). They're rated max 105C, so I'm not too worried, and they are relatively noise free.

However...

The two SumRs in my FX5 buzz if I push the current up to the Amps optimum range.

I've had to dial it down because the piss poor quality of the transformers.

I usually go for no more than 35°C above ambient.  That's at max load which I'd never run a transformer at. 

the SumR traffos in my two KGSSHVs (when I had them) ran just slightly warmer than ambient. My SumR traffos in my T2 are at about 25-30C above ambient, but they have a 1 inch clearance from the top of the case (whereas the spec calls for something in the neighbourhood of 4-5 inches - going by memory).

My transformer also ran slightly warmer than ambient under normal conditions, which is why there was some alarm on my part when I touched the case underside. I didn't have something on hand to measure the heat, and I didn't want to leave it on long enough to check. During that period, the amplifier LEDs were all light as normal. I used the insulating mats that Richard provided. The real red flag was that smell...I still can't figure out what exactly was burning. I'll find out once I pull the transformer tonight.

 

I understand that ambient temperature may affect this as well, as it has been ~90s at my place recently. However, I've been pointing a fan in its general direction (to cool my SDR2000, whose feet melted during normal operation)

 

Richard has implied that he can build a replacement, and has offered to install a reset fuse in case of overheating. Any thoughts on this? Spritzer, in the past you've mentioned some sort of construction fix to pass to Richard (larger diameter wire)? 

 

I'm a bit concerned about using another SumR as a replacement, but if that falls through I can scrape together two hammond 269GX's as a substitute.

I replaced a SumR transformer on my first build and no problems after that. Free of charge.

Richard mentioned something about problems with the first unit so I bought the Megatron transformer from him. It works well without load. Will see how it works when completed.

One of my KGSSHV transformers melted so Richard replaced it.  That replacement cost me about 300$ in customs fees as it was declared as if I had just bought them... 

 

The replacements seem to work just fine but I've only done brief tests.