The ultimate DIY? A Stax SRM-T2!

[Lil Knight]

So this is the spreadsheet with common parts and price. In Yen, of course.

https://docs.google.com/spreadsheet/ccc?key=0AoNVeoQi85GgdHZPMlJPVnFKZ1A0d3dxRnpNR0xPSFE#gid=0

[luvdunhill]

Interestingly, as I went through adjusting the pots, the LEDs did all sorts of weird things - some unlit ones lit up while others turned off, adjusting one channel seemed to affect some of the LEDs on the other channel. 

 

George, did you tie together GND and EARTH inside the PSU chassis? Still thinking about your channel dependance...

[GeorgeP]

Going by memory:

 

Circuit ground (pin P) from amp to psu board. Then as directed by Kevin (and difficult to see in one of his photos, but you can see it in mine), a short wire from the psu board ground to the nearest screw. Continuity-wise, this is connected to earth ground on IEC via the chassis.

 

Chassis ground (pin K) from amp to earth ground on IEC.

 

I know a few others had slightly different chassis grounding schemes. Hope I didn't miss something.

Edited September 3, 2013 by GeorgeP

[GeorgeP]

this pic should do everything for you

http://gilmore.chem.northwestern.edu/t2power3.jpg

R50 sets the output voltage. so you probably want to go to the standard value of 287k which should be

a tad under 500 volts, but 300k would certainly work.

All of the .6 inch resistors are supposed to be .5 watt.

will update the schematic when i get home if i remember.

pick load resistors out of your junk bin, and test at say 60% of max power.

make sure they can handle the power, or dump in a plastic pail filled with

water and stay away....

The 47 ohm resistors are just to ground the input filament. Would be the same as

having a center tap on the transformer...

In the middle you can see the bare wire connecting circuit board ground to the

chassis nut close to it. Don't forget this.

 

Here is Kevin's post I was referring to - took me a bit to find it.

[GeorgeP]

George, did you tie together GND and EARTH inside the PSU chassis? Still thinking about your channel dependance...

 

I meant to ask, how did you end up wiring yours, and how did the LEDs respond when you first turned it on? Thinking back I am pretty sure I wasn't imagining things, but it was about 4am and anxiety was running high so who knows - I certainly would not go about trying to recreate the effect.

[chiguy]

 

I am ordering from satodenki soon, so anyone needs anything from there, just let me know via PM. I will finalize things on Thursday.

 

To make it easier, here are the required parts for the T2 that are available from them:

 

34  2sa1486  8  2sk216  4  2sk246gr  4  2sj79 

 

 

I would like a few things from Satodenki if that'd be okay.  I'll send a PM with a list soon.

Edited September 3, 2013 by chiguy

[luvdunhill]

Very similar. I don't use the ground strap on the PCB and connect the earth that runs through both umbilicals at this point. The chassis is connected through the XLR connector itself. I do not remember how the LEDs interacted at power on.

Thinking about this more, has a ground loop been created? I realize both umbilical cables need to have a chassis connections so they can be used separately. Maybe it would be useful to someone accomodate for this with a jumper or switch?

Maybe a ground loop breaker is better than direct ground to earth in the PSU chassis? Can antiparallel diodes act fast enough in case of a fault?

[GeorgeP]

So the effect would be the same in how you have done it. The two ground wires are connected in one spot in the amp box then split between the two umbilicals and then reconnected in the psu box at the IEC (my XLRs are also connected as yours are). Not sure if that in and of itself creates a ground loop, but maybe where it is grounded as well in the psu chassis which in turn connects to the amp ground - not sure. But I can say it is dead silent at all volume levels and with volume turned down all the way (max resistance). If it is a ground loop, it makes my head hurt trying to conceptualize it - sort of like the "Inception" of ground loops - a ground loop within a ground loop.

 

Maybe Kevin will weigh in on his design considerations.

[nattonrice]

Depending on component tolerances and how you set the pots when you installed them the first power on can have: all led lighting up like magic, nothing lighting up, some combo in between.

[Headampbro]

Does anyone knows where can I purchase the PCB boards for T2, from Justin?? 

[Victor Chew]

Contact Lil Knight. Amongst other boards, the group buy for the T2 is on at this moment.

 

[Lil Knight]

Just got back from a short vacation... sorry for any delay.

 

The boards arrived yesterday. The extra thickness is just awesome, bullet-proof is all I could say. Will try to take some pics later on

A very helpful delivery man did manage to drop the whole box and damage the corners of the boards a bit :-/

 

I will bring a set of them to the machine shop and begin to sketch up the chassis this weekend.

[Lil Knight]

Some updates...

 

 

The board's thickness will need more time to fully solder the thing. You don't want to de-solder anything though...

 

I've come up with what I think final design for the top and bottom plates.

 

 

 

Bottom and top plate of the PSU will be the same. I wanted to add more venting holes but afraid it will not be strong enough to hold everything.

 

Small parts such as tube rings and volume knobs are already in production yesterday.

 

I am thinking of making a small badge made of stainless steel, and then a layer of leather with laser engraved logo on top of the amp.

Lots of options for leather, will get back with this later.

Edited September 30, 2013 by Lil Knight

[s_r]

Shaping up to be quite nice there. If only my wallet would have allowed a T2 build...

[shipsupt]

Thanks for the update LK.  Looking good.

[Lil Knight]

Any suggestion for the front panel's design?

[kevin gilmore]

looking very nice

[luvdunhill]

If the top and bottom are the same as above, where are the PCB mounting holes?

Also I would recommend engineering a place in the PSU case to accomodate some bolt down 50W resistors. I did this and found it very handy. Plus it encourages people to test the PSU.

[GeorgeP]

If the top and bottom are the same as above, where are the PCB mounting holes?

Also I would recommend engineering a place in the PSU case to accomodate some bolt down 50W resistors. I did this and found it very handy. Plus it encourages people to test the PSU.

 

Yes, definitely holes for the standoffs on the bottom covers and also (obviously) holes for the transformers (tthough guessing they were just not in your drawings) - and I am assuming the traffos from KG's original case will fit here as well? Also any sense of the clearances from the board to the top and bottom covers for both the amp and psu?

 

Regarding the comment "You don't want to de-solder anything", would it be advisable for people to use higher quality trimmer pots to reduce the risk of them going noisy (however unlikely that may be)? Bourns makes some higher quality wire-wound ones (instead of cermet - they say its better), but they are about $26/pc instead of $2.60/pc.

 

Regarding the volume knob, were you going to do it like Kevin did where the knob had about 1 inch of shaft that extended into the chassis (you probably don't want the delrin rod extending through the panel bearing)?

[Lil Knight]

Those are only top panels. Mounting holes will be there.

 

Yes, traffos from SumR will fit. New PSU board is a bit shorter too.

 

I got some more expensive trimmers from Bourns already. Just haven't got time to try them... More than $200 just for the trimmers doesn't sound good though. Measure the resistance and replace with fixed resistors is another option.

 

No, the volume knobs will be 'standard'. The extended rod will be through the panel bearing.

[palchiu]

Received some 2SC3675 but they looks not like normal. I can't make sure it's real or fake.

 

 

I've test voltage breakdown, all 1050V+. And hFE is 35+

 

Should I find another source?

 

Thanks!

 

Pal

 

[justin]

they look like ones that i have. which KG also put on his curve tracer. only thing left is to crack one open and make sure the die size is the same as a real one, if you have one to compare. at worst, it seems there's a part with very similar if not identical specs that is being labeled 2sc3675, not a typical fake

[palchiu]

Justin, Thanks!

 

Here it is. But I don't have other 3675.

 

Edited October 31, 2013 by palchiu

[justin]

cant really tell from that photo

 

 

the one on the left is the one i got that looks like yours, the one on the right is a 'real' one

[palchiu]

Justin, thanks for your pics.

 

I'm thinking another source for 2SC3675.