Kerry

Thanks. @Cee Tee - I was always thinking about doing a curved face. I always thought to mill it out, but printing might provide alternate options. I looked at the site and went to the materials section to look up alumide. I'm not clear that this material would be acceptable for a face plate. What were your thoughts?

KGSShv milling is proceeding I had to buy a new controller for the mill. I had one a cheap ebay one from china and it just messed up too often. I ended up with the gecko g540. This is the bottom plate. It is 8.25" W x 14.25" D x 3/16" H. I decided to go a 1/4" wider just incase I want to put a larger transformer in.

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

I didn't use the 3900pF cap in my version. A lower capacitance would certainly work. I didn't do a comparison of the difference under load. Under no load and with either version I was unable to measure the noise level since it is so small. In LTSpice it was slightly lower noise without that cap. Kevin probably had a reason for including it but I'm not clear on it though I'm guess it would be to prohibit oscillation.

Sorry for the delayed response, I'm just seeing this now. This project is still sitting on my bench on breadboard. I never made up any boards, but do want to finish this up after a couple more projects.

Congrats!

I got a chance to put the amp on the scope. It looks great! Response is totally flat from 10Hz up through 40KHz + (I didn't look much beyond this). I also looked at the FFT function on my scope and that looks great as well. Here's some shots of the scope with the FFT window. I'm showing 1KHz, 10KHz and 20KHz. I also looked at noise with the volume all the way down and I was seeing about 50mV. Not bad for my house. I'm going to keep the chassis width at 8" since I don't see that contributing to the noise floor of the amp.

I will PM you.

I don't have a comparison from the lower powered to the higher powered KGSShv, but I did spend 4 hours listening and A/B'ing the Blue Hawaii and the KGSShv last night. I was listening to the Stax O2/Mk I and switching them between the two amps. Without going crazy loud, it was hard to tell much of a difference. I occasionally turned up the volume and the KGSShv did seem to hold its own. All the things that make me smile with the BH were there - wide sound stage, deep and clear bass, remarkable separation of voices and instruments with incredible detail. Makes me happy just thinking about it.

Hi Birgir - Any progress? I was inspired and took the original version of the board and modified it so the servo is the same as yours. I used a 2uF cap instead of the 10uF. Also lowed the resistor coming from the opamp to 20k. works well. now looking at an SMD version. Thinking about the sot23 version. At 15 mA per should support it. Could use the sot223 version instead.

I milled the bottom panel out of wood to make sure everything fits and there is no hum. so far so good. Still need to get it on the scope. here's some shots... http://img189.imageshack.us/img189/5108/t782.jpg http://img824.imageshack.us/img824/2716/nocl.jpg http://img855.imageshack.us/img855/1337/mdfk.jpg the back plates will be powder coated to match the heat sinks.

Yeah I saw that as well. Believe it or not, Radio Shack sells them on line.

That's really nice. Looking forward to seeing the results

I haven't turned the current down on mine yet, but glad to hear that it isn't substantial.

You're right. Luck was with me I've been there too. Good luck.

Back up Here's what I broke: Glad it wasn't more extensive. I'm going to keep the current where it is for not with the 100 ohm set resistors. If the heat in the chassis comes down to 120 deg F or less I'll probably keep it here. I've not put it up on the scope yet, but will get there at some point.

I should be more careful when measuring the amp. Just shorted the output stage

I forget too much... My amp is running at 3 mA (per leg) on the 3rd stage and uses IXCP10m90s with a 1200 ohm current set resistor. This will only work up to 900V (+/- 450V) on the power supply. Output - Silk error #2 I'm using a 100 ohm set resistor (must have been playing with this value in spice and forgot to set it back when I made the boards) to set the current - about 10.7 mA per leg - I was confused why it was running so hot - now I know. Good news - nothing went boom and the amp seems to run stable. I have a version 3 jpeg of the amp board that had a 120 ohm set resistor for the current which puts it at about 9 mA per leg. Version 12 of the amp uses a 180 ohm set resistor which puts the current at about 6 mA. In a conversation with Kevin he said he was targeting about 7 mA (depends on the LEDs you use as well). This is what I though I was setting the amp up for (default / standard current). EDIT: I also saw another post in this thread that had Kevin putting the output stage at 5.5 mA

yes. I set it up with the default current. I would want to check the temperatures once it is fully cased up before thinking about this. I'm pretty happy so far with the way it sounds. ... I never thought about issues with bugs inside the amps. Good luck.

Just a quick update on my amp... I wired up both channels this morning and spent an hour listing (I finally pulled out the O2s). I haven't had a chance to A/B with the Blue Hawaii, but from memory it sounds every bit as good.

So far it is dead quiet. I'm curious how the noise will be once it's cased up too.

I have music I had one error on the silk. Had a 2k resistor instead of a 1k coming out of the 1k trim pot. I couldn't adjust to zero. Changed it out and it works perfect. I scratched the solder mask on one of the boards and wanted to cover it. I haven't had to do that yet, but I'm thinking some clear nail polish should do (could someone please confirm). I've tested each board but not both together. I've got to make the second power connector to the board and wire up the input section. Here's a picture of the amp running... I tried it first on some electrets and then on my 404s. Very nice so far. I let it heat up. The sinks got to 129 deg. F. once I case it up I think it will do a bit better. Sitting on the table doesn't help and I will have another 20 lbs of aluminum to help cool. The STN9360 transistors only got to about 100 deg. F, so we're good there.

I agree. I was trying to keep the front of the chassis free of any noise. For the control board, I could mount it to the face of the amp and shield it from the amp boards. I've got .75" there so I could fit it. I was going to use .5" aluminum plate for the face and could mill the display into that so the electronics would be very flat. I just bought a small 12V 1A (70mm x 35mm x 30mm) switched supply that could sit under or above the Bias board. Don't know how I feel about that. It would not be part of the audio circuit (only control), but I'm worried about interference being induced into other components.

Thanks. I want to use the digital attenuators, but I have to see if I can fit another power supply for them (didn't want to tap into the existing 15V supply). I might be able to mount another transformer on top of the existing one, but also need to see if I can fit the 12V / 5V supply. I think I can, but it will take some thinking. I can make a digital controller that could easily mount to the front with a digital display. If I'm doing that I would add and IR remote control as well. @Victor - Yes, the power supplies are split + one side - the other. Had to do everything possible to get this down in size.

Here's some progress. I'm shooting for a chassis size of 8" x 14.5" x 3.33". Here's a few pics...