The ultimate DIY? A Stax SRM-T2!

[Craig Sawyers]

First things first. Then i will read upward later.

First things first - welcome back!

The power supply problem is fixed. Verified by Inu. Took about 5 tries

I've done the mod, board bolted to heasinks, just wiring in the transformers. Hopefully later today this puppy will get fired up.

[nattonrice]

Welcome back =)

From my "2 second before bed" scan of the schematic it looks like just the 4 x .1uf caps?

Nice!~

[Craig Sawyers]

Welcome back =)

From my "2 second before bed" scan of the schematic it looks like just the 4 x .1uf caps?

Nice!~

Plus the 10k resistor between the voltage reference and pin 2 of the OP27.

[luvdunhill]

Is this change applicable to all four supplies?

[Craig Sawyers]

Is this change applicable to all four supplies?

So I understand from KG - so that is what I've done. I think he's updated the PSU schematic to include all this stuff - usual link.

Took more than half an hour to find my 1/4" crimp receptacles. Aagh.

[nattonrice]

Plus the 10k resistor between the voltage reference and pin 2 of the OP27.

Oops dunno how I missed that.

[wink]

Oops dunno how I missed that.

.

Too early in the day here in Aus.

[kevin gilmore]

The resistor may not be necessary as the output impedance of the reference is about 1k. In which case the cap would need to be 1uf.

I have not tried this. Also instead of cutting a trace, you can put the reference up on a pair of sockets stacked together with the resistor

in the middle. Picture on this at some point.

[kevin gilmore]

I have finally figured out the discrepancy in my spice simulator on the power supply vs ltspice.

Although they still don't entirely agree. The explanation would take 3 or 4 pages, so i'll omit

that. Back to the 20 year old software.

[Craig Sawyers]

Whohoa! Power supply lives. Just went for broke, screwed the whole shebang into the casework, wired it up and turned it on. Unloaded volts are just about bang on. Haven't checked for ripple yet

Now the question is - do I dummy load it, or just go for broke and build the amp up?

[Craig Sawyers]

Geez - I've just spent five hours (a) fitting the DACT and ( fitting half the transistors to the heatsinks (just the 20-off c3675's).

Gotta get it more or less finished tomorrow - wife and daughter have gone to Rhodes (I hate hot holidays; I'm a cool climate guy) so I've been using the week as play time. They come back Wednesday, and will expect to find the house not looking like an electronics production shop floor.

[nattonrice]

I say wire it up and go nuts.

Dummy load shmummy load =P

[luvdunhill]

here Craig, use this:

[Craig Sawyers]

here Craig, use this:

??

On other topics, I was merrily torquing down the devices on the amp board, with the torque screwdriver set to the correct value of 1.1Nm for TO220 packages. I then moved on to the 2SA1486's - and promptly split the first one in two.

Then I checked - the maximum torque for a TO126 package is 0.4Nm, and use of a washer under the screw head is recommended to further spread the load.

Taking those remedial actions all is OK.

By the way, I have gone back to using steel screws, but with the Aavid 7721-3PPSG bushes. These have a 3.18mm (1/8") shaft, which goes pretty much all the way through the 4171G ceramic insulators. The steel screws mean that you can get the torque set for maximum heat transfer (with heatsink grease too of course). The 7721-3PPSG is a slip-fit in some device holes and a tight interference fit in others - tolerance stack up I guess. Also, they cannot be used for metric screws because the hole is not quite big enough - so for that insulating bush those of us in metric countries need to get hold of the imperial 4/40 screws and nuts specified in the BOM.

What I am doing with the 2SA1468's is to put a 4.7mm length of heatshrink over the threads of the 4/40 screws. This makes a nice fit in the device hole. 4.7mm goes most of the way through the ceramic insulator.

[luvdunhill]

??

http://www.head-case.org/forums/do-yourself/6925-ultimate-diy-stax-srm-t2-57.html#post369734

[luvdunhill]

What I am doing with the 2SA1468's is to put a 4.7mm length of heatshrink over the threads of the 4/40 screws. This makes a nice fit in the device hole. 4.7mm goes most of the way through the ceramic insulator.

this seems totally unnecessary to me. It's a plastic package after all, why are you using ceramic insulators again?

[luvdunhill]

It's a plastic package after all

oops! nm

[spritzer]

Here I was thinking Marc had finally lost it, worrying about how fuses sound etc...

[kevin gilmore]

this seems totally unnecessary to me. It's a plastic package after all, why are you using ceramic insulators again?

It may be a plastic package on that part, but its not insulated, collector is on the bottom.

[Craig Sawyers]

http://www.head-case.org/forums/do-yourself/6925-ultimate-diy-stax-srm-t2-57.html#post369734

Goddit!

[kevin gilmore]

power supply fix pictures courtesy of Inu

http://gilmore.chem.northwestern.edu/noise%20fix1.jpg

http://gilmore.chem.northwestern.edu/noise%20fix2.jpg

But if you don't want to cut the land, you can put the ref chip up on a pair of sockets with

a resistor in the middle. Picture on that later.

[Craig Sawyers]

Just the Amphenols to wire up now

http://www.tech-enterprise.com/tekstuff/P9070512.JPG

http://www.tech-enterprise.com/tekstuff/P9070513.JPG

[nattonrice]

Hehe I take it work on the bh is on hold?

[Craig Sawyers]

Hehe I take it work on the bh is on hold?

Ah - you spotted that lurking in the background

[kevin gilmore]

For those interested here are the prices. It really only makes sense if i do another 20 piece board run.

power supply board quantity 5 $157 each

quantity 10 $97 each

quantity 20 $61 each

amp board quantity 5 $242 each

quantity 10 $166 each

quantity 20 $102 each

so prices did go up ever so slightly due to increase in sales tax.

which makes a set of boards quantity 5 $399

quantity 10 $263

quantity 20 $163