The ultimate DIY? A Stax SRM-T2!

[Horio]

No thanks to you! I'll get over it if you bring your KGSSHV to run the triple threat.

Sounds like a plan Al! We'll have to make sure to do some serious and careful comparisons, and then post a minimally helpful 3 sentence review for everyone.

Edited December 10, 2011 by Horio

[morphsci]

The music list needs to be three pages, followed by the three sentence comparison.

[wink]

How about DIYT2<BHSE<KGSSHV...? Says it all....

[bozebuttons]

congrats AL!

Looking Good

[livewire]

Some boyz got all da toyz...

[Voltron]

I now have the source going into the BHse and then looping out to the T2. Both amps have NOS Blackburn xf2 EL34s. Listening to hi-res files through Amarra's cached playlist. The SR-Omega and 009 are charged up and I hope to spend enough time to be able to compare them on both amps a bit. The SR-Omega sounds remarkable on the T2, and the amazing bass is the first thing to stand out. Nice.

[deepak]

I now have the source going into the BHse and then looping out to the T2. Both amps have NOS Blackburn xf2 EL34s. Listening to hi-res files through Amarra's cached playlist. The SR-Omega and 009 are charged up and I hope to spend enough time to be able to compare them on both amps a bit. The SR-Omega sounds remarkable on the T2, and the amazing bass is the first thing to stand out. Nice.

Damnit Al you are supposed to say the SR-Omega sucks on the T2!

Waaant

[Voltron]

Well, I can say the 009 sounds even better.

[morphsci]

Well, I can say the 009 sounds even better.

Bastard! That is from my wallet

Edited December 12, 2011 by morphsci

[shellylh]

Just noticed the T2, very nice Al! Also congrat on the DIY work.

[pedefede]

Well, I can say the 009 sounds even better.

You are evil Voltron ... as if the T2 wasn't expensive on its own ...

(Otherwise i need to 1. get broke 2. send my sister in law to fetch me a "cheap" pair from HongKong, )

[bozebuttons]

I now have the source going into the BHse and then looping out to the T2. Both amps have NOS Blackburn xf2 EL34s. Listening to hi-res files through Amarra's cached playlist. The SR-Omega and 009 are charged up and I hope to spend enough time to be able to compare them on both amps a bit. The SR-Omega sounds remarkable on the T2, and the amazing bass is the first thing to stand out. Nice.

Al are you using the amperex 6922s for the input tubes on the T2?I would suggest getting some NOS mullards to replace them a little more organic sounding.

[kevin gilmore]

Messing with the input tubes can cause significant DC output which requires a pair of DVM's and

a strong heart to adjust. The T2 is not like other tube amps. Best left alone.

The wood is for more knobs.

At some point i'm going to try and make a knob with the display inside it.

Similar to the one on my Wolf range. And it will work the same way.

Or maybe like that new thermostat from the ex-apple people, except

made out of ebony.

Edited December 24, 2011 by kevin gilmore

[cobra_kai]

Does anyone have a recommendation for a good, reasonably priced input tube, something on the order of the winged c output tubes?

[ujamerstand]

+ Reliable in the long run would be nice.

[luvdunhill]

Ideal would be 8 inch-pounds (NOT foot-pounds!), assuming a nutted 6-32 steel screw. (12 in-lbs max)

If the #6 screw is going into a threaded aluminum heatsink, 6 inch pounds is safe.

PEEK (plastic ) screws have a much lower torque rating. ~1.33 in-lbs for the M3 or #6 screw sizes.

Here's an interesting read on this topic:

http://www.vishay.com/docs/72674/72674.pdf

[Craig Sawyers]

Does anyone have a working link to Craigs excelsheet from post 1830, about RV1 and RV2 ?

Sorry I've been away folks - looks like around 7 months - and I've been scanning the back mails (boy does that take some time).

Anone interested in the pot adjustment spreadsheet, drop me a PM. This was created using a Spice simulation of the battery (using simple DC spice models I derived from the semiconductor data sheets), But I used these graphs as the method for setting the pots to give the needed 740V.

Basically RV2 is a coarse setting at 34V increase per clockwise turn (550V to 790V with RV1 mid range). RV1 is fine setting at 3.4V increase per ANTIclockwise turn (500V to 790V with RV1 set to 740V nominal). I set them at mid travel, tweaked RV2 until it was close, and then used RV1 to fine set and tweak as the beast warmed up.

Variation with FET Idss is weak - only two turns of RV2 covers the yellow, green and blue varieties, Green is a good compromise, with less than one turn of RV1 covering the Idss band.

Probably more as I catch up.....

Craig

[Craig Sawyers]

Keratherm film might be too thin.

The problem with thin film materials is this. Although the material itself has excellent high voltage properties, it you draw a sketch you will see that the weak point is where the insultating bush goes through the hole. At that point, the distance betwen the device tab and the heatsink is only the thickness of the pad - a few thou. Actually, I think Kevin zapped something in his first build of the power supply as a result of this, somewhere in the bowels of this thread, and it was that event that led to the use of ceramic insulators.

The very best solution is to use not only the ceramic insulators (the weak spot is still the hole - but the thickness of the insulator is amply good enough for the voltages present. Then use the Aavid long PPS bushes - I think it is the 7721-3PPS with 1/8" long shaft.

In my (not so humble!) opinion, PEEK and ceramic screws are not capable of handling enough torque for optimum heat transfer. The datasheet optimum for a TO220 package is 1.1Nm (8.9 in lb). So I used stainless screws - there is ample insulation length in the Aavid bushes.

Craig

Edited December 31, 2011 by Craig Sawyers

[luvdunhill]

I wasn't aware of the fact that the longer bushing insulators would fit through the ceramic pads or the devices themselves. If this is the case, this really is the best solution. The parts I have don't fit.

[Craig Sawyers]

Wow - what exciting times on the list!

Lots of new T2's out there, looking real great. Bloody well done all. And new and exciting problems overcome. Who would have thought that non-Xicon resistors would give rise to a noise problem when used in the batteries? That takes some serious detective work. I just used Xicon's throughout, as per KG's suggestion. I was considering upgrading to Dale RN60's, but I don't think I'll take the risk now.

And lots of exciting knobs - mallee burl (I used brown mallee burl forknobs of the chest of drawers I built), and now gaboon ebony!

Mine just keeps rolling, and sounding astonishing. The only upgrade was removing the filtered mains socket, adding an adaptor plate, and using a Furutech rhodium plated unswitched, unfused and unfiltered IEC. Now this is probably a contentious statement, but the T2 sounds a whole lot better without the mains filter (even Mrs S heard the improvement, and she has zero interest in high end audio).

Next is to add a switch to the front panel of the power supply. Spritzer - where did you get the sexy switch for the front of yours? And what is its rating given the level of inrush current that this animal takes?

Craig

[ujamerstand]

Welcome back!

[Craig Sawyers]

I wasn't aware of the fact that the longer bushing insulators would fit through the ceramic pads or the devices themselves. If this is the case, this really is the best solution. The parts I have don't fit.

Well there is some tolerance in the holes in TO220's. They just slipped into some of my devices nice as ninepence, and others I had to press into place - nothing worse than an engineering interference fit, a thou or so. But they fit the ceramic insulators just fine. It wasn't my idea to use the Aavid bushes by the way - Inu found those during his build.

If you use the regular short bushes, you definitely need insulating screws. As far as I know the 7721-3PPS is the only bush with a really long tube, and that allowed me to use steel screws and the correct high torque. I think I got mine from Future - yes, they have stock at around 15 cents each and 11,500 in stock.

Check the part number from searching this list just to make sure I've got the right number.

Craig

[cobra_kai]

I actually tried to use the aavid things on the KGSSHV on the 10m90s, and ixtp01n100D, but they didn't fit either of them. I only tried a couple of them though and then gave up and just used the PEEK with the ceramic insulator by itself, so maybe I should have tried a few more (I got 100 I think). This has been working fine so far though.

[Pars]

Kerry indicated in a previous post () that he had to drill the 10m90s and some of the FETs with a #28 bit to get those to fit. The part number appears to be correct Craig. Nice to have you back!

[Craig Sawyers]

I actually tried to use the aavid things on the KGSSHV on the 10m90s, and ixtp01n100D, but they didn't fit either of them. I only tried a couple of them though and then gave up and just used the PEEK with the ceramic insulator by itself, so maybe I should have tried a few more (I got 100 I think). This has been working fine so far though.

Yeah - I just persuaded them through with a small vice and a ceramic insulator to capture the bush as it came through. But it is certainly surprising when they just slip into some devices and not into others. Wear in the press tools that are used to form the tab I guess, since they must start out at nominal dimension with a new press tool.

Craig