i'm on a roll... the kgsshv

[Lil Knight]

Very nice revision of the boards. Being able to source all the parts from a single vendor is a big plus.

I've just done with huge schoolwork and will be pretty relaxed for a couple of months to do another run of group buy, if that really happens. Been planning for ordering a huge stock to keep the price as low as possible but haven't got chances yet.

I also found a local source for custom made EI-core or toroidal transformers. They are very well-made and all built by hands, not machine, so utmost care is guaranteed. I would just bring the psu board and try with every single transformer to make sure it works perfectly. A group buy for transformers+boards and ship in a single package would save a lot of time and cost.

[kevin gilmore]

it is amazing how fast electrons travel the globe

 

http://gilmore.chem.northwestern.edu/kgsshvamp11mini.jpg

board is 7.26 x 4.84 inch

 

http://gilmore.chem.northwestern.edu/kgsshvps8g-mini.jpg

board is 7.3 x 7.35 inch

 

updated power supply schematic

http://gilmore.chem.northwestern.edu/kgsshvps8g.pdf

 

you can use ksa1156 in place of 2sa1486 if you put in the 270v zeners

[spritzer]

A group buy on transformers sounds like a good idea and the PSU board will be even smaller. 

[RiStaR]

It's late over here, but just finally got my KGSSHV onboard version up and running

 

Looking forward to this newer version and probably going to try the offboard version, so this is just in time!

 

Edit: Too tired to pull pics out of the cam... tomorrow. In the meantime, thanks to everyone who helped! (Tran; Geoff; Birgir; Victor) for parts KG for the design (of course) and Victor for helping me check my stuff before I powered everything up each time!

 

Edit 2: Updating the post with pics

 

Some notes for myself, or hopefully helpful for anyone building:

• Had to change the resistors in series with the DC offset pot to 1.2k (couldn't get the offset up from one of the boards)
• I used Op07s (predecessor to the Op27) instead because the local shop didn't have any on hand... after an hour the balance and dc offset is within +-1V so I guess it works?
• I initially tried to pull the 15V from one of the amp boards instead of the 15V source - there was a burning smell, so I disconnected that cable (I'm not too sure if it's the 180ohm resistor in the heatshrink that was smelling funky or what else... will redo this when I get the right switch.)
• Will need to shorten some of the cables - it's not exactly fitting in the "lines" I had in my mind when I did the cabling!
• The bottom-right-most 4686as on both the amp boards (above the C2705) seem cool to touch; its as if the warmth of the heatsink comes from the surrounding sinks?
• It's been random and it's only happened once, but there was static sound when the volume pot was set to minimum - it went away when moved just a touch up [haven't been able to replicate it but will check my grounding again to make sure]
• I confused myself between the Alpha pot; Birgir's circuit boards + help and my DMM. Figured it all out when I realized I was looking at pin1 with the legs facing me while Birgir was referring to pin1 with it pointing away. Either way easy to figure out.
• My Mytek DAC actually has a pretty "Hot" (I think it goes all the way to 12Vrms) balance output but the KGSSHV is handling it perfectly
• I bought two types of sinks and I understand why some folks are having difficulty with them. The legs on the ohmite sinks are smaller whereas the ones most of the builds here seem to be using is about 0.5mm thicker. I used a philips head screwdriver (the ones you use for glasses) to widen the wholes so they fit much easier.

 

First power up:

 

View from the top:

 

At it's temporary location:

 

[chinsettawong]

Will there be a new revision for off board PCB too?

[livewire]

Congratz!

Another HV sees the light of day.

or is it the light of night? 

 

Impressions? What cans do ya got?

Edited April 1, 2013 by livewire

[spritzer]

Very nice build, what case is that? 

 

Will there be a new revision for off board PCB too?

 

It's next on the list and I'll start shrinking it today.  I had done a "mafia version" earlier but I'll see if I can make this one better. 

Edited April 1, 2013 by spritzer

[RiStaR]

Congratz!

Another HV sees the light of day.

or is it the light of night? 

 

Impressions? What cans do ya got?

 

Hey Livewire! Like you when you first built the KGSSHV, I'm just riding the SR202 for now (I made a deal with myself that I'd only get the O2Mk1 if I could build a working KGSSHV - so yay when I get to Japan in August ). With the 202s, I think you're right... it's pretty much at its limit. I do notice a difference in terms of detail and depth in comparison with the 252A that I have though. I haven't actually done an AB test, yet. The 202s have been plugged into the KGSSHV the whole day

 

Very nice build, what case is that? 

 

 

It's next on the list and I'll start shrinking it today.  I had done a "mafia version" earlier but I'll see if I can make this one better. 

 

 

Thanks!

 

Case was bought through a local shop (he probably got it made from China for me). It's aluminum and pretty nice (they screwed up cutting up the backplate for me - was lucky the PS fit by flipping it around.) The size is about 423 x 410 x 130 usable space inside is really 390 x 399 x 120 (you can probably push it by another 10mm in width but you'll have to do some creative drilling.)

 

 

Did another adjustment in the middle of the day earlier. The balance and dc offset are all 0.4v or less... I'm gonna stop now - so addictive to open it up to tweak... guess i'm trying to milk my money's worth with the pot tool lol.

[kevin gilmore]

quit messing with it and listen for a few hours.

[RiStaR]

quit messing with it and listen for a few hours.

 

 

I worked from home today so it was >8 hrs listening hehe

 

Thanks again for the design!

[chinsettawong]

Nice work RiStaR!

 

Wachara C.

[chinsettawong]

 

It's next on the list and I'll start shrinking it today.  I had done a "mafia version" earlier but I'll see if I can make this one better. 

 

 

Looking forward to seeing that soon. 

Edited April 1, 2013 by chinsettawong

[spritzer]

So the PSU is pretty much finished:

 

 

The old version was 9" by 7.5" exactly.  This one is 6.92" by 6.56".  I call that a result...

 

Changed a few things, isolated diodes for the HV section, 270V zener on each side to be able to use KSA1156's instead of the 1486's.  The NPN is the Fairchild part which was always a sub.  LV supply main filter caps are now 15mm in diameter, 7.5mm pitch 3300uf/25V.  There is a Panasonic FC which fits this.  The LV output caps are 330uf/25V, 10mm in diameter and a 3.5mm pitch. 

Edited April 1, 2013 by spritzer

[Remolon]

A minor observation. My experience with the pro bias Zener is that the actual value is a little lower than the nominal value. I am using three 200 Volt (Mouser 625-Z4KE200A-E3/F3) to obtain 582 Volt.

[spritzer]

Yeah but if we spec 10% or even 5% parts then it should be close enough.  Still 560V is still a lot closer than what Stax get with their unregulated PSU's and a simple voltage divider for the bias. 

[RiStaR]

Nice work RiStaR!

 

Wachara C.

 

Thanks and cheers Wachara!

[eggil]

Congratulations RiStar ! Good to see more builds.

Look Mah! No heat sinks!     

Edited April 2, 2013 by eggil

[palchiu]

Nice work! RiStaR

[RiStaR]

Congratulations RiStar ! Good to see more builds.

Look Mah! No heat sinks!     

 

 

Nice work! RiStaR

 

 

Thanks for the kind words guys!

 

Unfortunately, disaster struck earlier... It's really my fault I left it on the whole night and I wanted to listen to it during the day (My wife calls it the "men with new toys syndrome" )

 

After some analysis, I blew two resistors and an LED (which I found out later). The 2nd 25 ohm resistor between the 4686A and the IXYS part was really charred (I saw arcing); The 50k below the C2240 was slightly brown so something wasn't right there and I changed it. The LED at the top right hand corner was also blown.

 

Had only changed the two resistors and powered it on again, the LED didn't turn on, but the 4686A parts above the IXYS parts started to turn hot really fast. Changed the LED and all is fine again. 

 

I have to say though; even after it blew the KGSSHV was still kicking out tunes (I didn't hear anything through my headphones and only smelt something funny)

 

Hope it all works out from here on out! (And I'll remember to switch it off every now and then )

[deadlylover]

Hope it all works out from here on out! (And I'll remember to switch it off every now and then )

 

Nicely done with the build, and good luck! I didn't switch off my on-board KGSSHV for a month as a torture test (safely), and it didn't break a sweat.

 

Anyway, in regards to the potential group buy, I'm up for re-distributing the boards for the Aussies if needed, along with some sand (I've still got ~50 spare 4686A if anyone needs some).

 

And on that note, are there any notable performance differences between the 4686A's and the 01N100D's as the output devices?

[spritzer]

The output capacitance of the IXYS is much higher, the 4686A is almost as good as tubes in this regard.  High output capacitance is in some ways similar to high output impedance in dynamic amps and is best avoided.  That said, we can only use what parts there are. 

[deadlylover]

Thanks spritzer!

 

I'm not too sure what's going on here, rs-online says they have some 4686a coming in a couple of weeks despite the part being discontinued. I might clean them out if that's not an error.

 

Failing that, I'll clean out satodenki, but it might be expensive at ~$1.50 a pop. 

[eggil]

RiStaR, I wonder why your parts blew. I have left mine for long hours before and never had any problems.

[RiStaR]

Nicely done with the build, and good luck! I didn't switch off my on-board KGSSHV for a month as a torture test (safely), and it didn't break a sweat.

 

Anyway, in regards to the potential group buy, I'm up for re-distributing the boards for the Aussies if needed, along with some sand (I've still got ~50 spare 4686A if anyone needs some).

 

And on that note, are there any notable performance differences between the 4686A's and the 01N100D's as the output devices?

 

 

 

RiStaR, I wonder why your parts blew. I have left mine for long hours before and never had any problems.

 

 

Deadlylover and eggil, I'm not sure why those parts blew as well. It's essentially been on (except for one hour) since I got it up the other day (so more than 24 hours). The weird thing really is how it continued to work (music was coming out fine) after those parts blew. After the repair I had it on for another 6 hours and it all seemed fine. Hopefully nothing major there!

[jwzhan]

I really wouldn't hope for the best in your situation...

I had one channel that was doing similar things (after some initial damage by accident,) but I didn't try if it were able to produce sound after the damage.

I reaplaced the blown resistors, powered on without any problem, and it ran some 4 hours before I turned it off.

Meanwhile, I tested the current draw across the damaged resistors against the other working board, and the resistors that were damaged (replaced) were drawing some 30%~40% more current (forgot the specific values.) I also wasn't able to get the offset to 0V with the default resistor values.

Right now, I threw that damaged board out because I couldn't identify the cause... and the newly populated board is running flawlessly. (left on for some 10 days during the longest run.)

Good luck on the build, but just as back up, I would get some spare boards and parts just in case if you decide to toss the current one out.

By the way... trying to fix the damaged board wasted almost a week of my time while populating the new board only took me one afternoon.... a much easier and quicker fix.

Edited April 3, 2013 by jwzhan