The ultimate DIY? A Stax SRM-T2!

[Craig Sawyers]

Not quite sure where to post this.

However, I've just been in touch with Linear Systems regarding their replacement to the 2SJ109. They've come back and said "We have plans for this year. Check our website for updates."

So it looks at last like the 2SK389/2SJ109 lives again.

[kevin gilmore]

they lie. they continue to lie. In january they promised me samples. All i get is excuses.

They continue to say the j109 does not meet specs. They refuse to tell me which spec

it does not meet. There is another reason...

[Craig Sawyers]

they lie. they continue to lie. In january they promised me samples. All i get is excuses.

They continue to say the j109 does not meet specs. They refuse to tell me which spec

it does not meet. There is another reason...

I'll not hold my breath then

[luvdunhill]

Paul Houlden hand delivered my transformers on Friday. Awesome - as a supplier to high end audio,

cool! I got mine is as well. Your story reads like mine... I think mine are nearly twice as thick as the SumR transformers (62mm each) due to the multiple taping layers...

[luvdunhill]

ok, this is probably a dumb question. Are you guys splitting up the heater wires in one umbilical and the DC wires in the other? This approach sorta makes more sense to me than trying to be dual mono for the sake of being dual mono...

[spritzer]

I've never had any hum issues with tightly twisted filament wires bundled in with the DC but separating them can't hurt. Just remember to use high voltage wire for the filaments.

[luvdunhill]

this pic should do everything for you

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

looking at Kevin's pics, he goes the dual mono approach, notice the splitting on the PSU DC output block. I think I'd rather split on the amp side, with a dedicated AC umbilical... far less wires between the chassis'. What do the other builders plan on doing?

[kevin gilmore]

The wiring diagram is on the schematic. Each of the channels is seperate and has all the wires to run that particular channel.

For the one extreme lunatic out there that wanted dual mono power supplies.

Ac and dc all on the one connector. Interchangable. Also easier if there is a problem to troubleshoot one channel at

at time.

[Craig Sawyers]

The wiring diagram is on the schematic. Each of the channels is seperate and has all the wires to run that particular channel.

For the one extreme lunatic out there that wanted dual mono power supplies.

Ac and dc all on the one connector. Interchangable. Also easier if there is a problem to troubleshoot one channel at

at time.

FWIW I twisted my heater wires and then ran them through a high temperature glass fibre braid sleeve (why? I just happened to have a load from a project about 10 years ago). That sub-bundle runs with the other wires inside the usual woven net sleeve. Times two of course as per KG's scheme and chassis mechanics.

[Inu]

I�m struggling with a 680-700Hz ripple noise that originally coming from the -500V line.

Both, -560 and -260V lines are also noisy too because of stacking from the bottom rail of the -500V PS.

And, I noticed that one of the two HV transformers getting very hot. I measured current at the R33 15ohm, about 92mA (R=46.5+L=45.5) from the -500V line. (+500V line is about 56mA)

Kevin,

If you have a chance to open the top cover of your PS next time, would you check the current on yours?

[kevin gilmore]

The -500 volt supply should be the current of the +500 supply plus the current of Q26,Q27,Q34 which is about 25ma.

On my amp i measured 50ma on the +500 line, and 75 ma on the -500 line. Ultimately the current on the +500 line

is controlled by the led's that do the current sources.

If the supply is noisy or oscillating, a small cap across R41 (say .1uf) should shut it up. Or you have a bad opamp

or a noisy zener on the power lines.

Now why the transformer is getting hot, i don't know. The transormers were spec'd at 140ma i think. (have to look when i get home)

[Craig Sawyers]

Now why the transformer is getting hot, i don't know. The transormers were spec'd at 140ma i think. (have to look when i get home)

Looking at the basic rectifier circuit - there is a secondary feeding a bridge feeding at 430uF cap (two 680u in series). Now it depends on the effective secondary resistance, but I calculate:

For 50mA delivered, transformer current = 130mA

For 75mA delivered, transformer current = 184mA

So possibly the -500V rating is a bit close - but it depends on the detail of the transformer design.

With that in mind, might it be worth using the 60V winding on the +500V supply rather than the -500V one? That way the dissipation would be shifted from the more heavily loaded -500V transformer to the +500V one.

Inu, bear in mind that most mains transformers, when operating at design ratings run hot - typically 40C surface (and hotter inside) - and too hot to comfortably touch. The only way to avoid this is to massively overrate to keep ohmic and core losses low.

[kevin gilmore]

The specs on the transformer was to deliver 130ma on the high voltage line DC. SumR calculates however they calculate from there.

With all the problems i had with plitron, i made sure that he understood the current requirements.

60ma on the 285 volt winding and 60ma on the 75 volt winding.

I did actually use the minus transformer for the 60 volts, but switching to the other transformer is definitely an OK idea.

As people can attest to that visited CJ, the thing was on all day, and was definitely HOT, but not untouchably hot.

[spritzer]

There were some people whining about burnt fingers on the volume knob but they should have tried the stock Stax unit...

[Craig Sawyers]

Revised transformer currents. The effective secondary resistance is much higher than I though, and that limits the secondary current spikes going into the reservoir caps. On my transformers (remember these are not SumR), the 485V secondary is 71.7 ohms at room temperature, and the primary is 33.5 ohms. The effective 2:1 step up ratio to 485V (from a 240V primary) gives a four times reflected primary resistance into the secondary. So the effective secondary resistance is 71.7 + 4 x 33.5 = 206 ohms. The SumR transformers will give different resistances, but not by a huge percentage I suspect.

Running the analysis with that resistance gives

50mA, rectified voltage 647V and secondary current of 108mA at 52VA

75mA, rectified voltage 629V and secondary current of 150mA at 72VA

With hot windings and loaded other secondaries the rectified voltages and VA will come down a tad for the 485V winding. So a totally loaded secondary spec for the 485V winding of 130mA is about right.

Those numbers are consistent with Kevin's figure of around 630V on the packages of the current sources - hence the need for the ceramic insulators.

[Inu]

Small cap wouldn’t help to stop the “switching” 2SC3675.

In my case, I changed R14, R29, R41 and R53 to 33Kohm to stop the "switching". Now all lines are stable and very quiet.

My OP27 is PMI “OP27GZ”.

[kevin gilmore]

Schematic updated with note from Inu.

I'm using vintage parts, because with the amount of stuff i have in stock, i don't actually see the reason to buy more.

Inu's mod increases the open loop gain of the power supply feedback. So the new versions of the op27's must be

a different process or gain from the old ones. Completely makes sense.

Now i'll spend the rest of the day synthesizing it, trying to find a better op27 model.

[Inu]

Small cap wouldn

[luvdunhill]

ok, getting closer. Any reason firing one side of my PSU up with no load and just the devices properly coupled to the angle bracket would be a bad idea? sorta a quick sniff test to see what the voltages look like on one side. Then I'd bolt it to the heat sink (haven't drilled it yet) and then do the same test under load? Seems like there wouldn't be a whole lot of heat on it and I could measure it rather quickly if I had a meter setup and what not...

[kevin gilmore]

you can certainly power it up without a load, but make sure the angle brackets are attached otherwise the current source that drives the zener to power

the opamp gets very hot very fast.

[luvdunhill]

yeah, I'll have things attached to the bracket and will make it as quick as possible.

For dual primary transformers, J3 is 1+3 (120VAC), J1 is 1+4 and 3+6 (120VAC), then for J4-J6 one winding goes across 1 and 3 and the other across 4 and 6?

[kevin gilmore]

I think that is right, but if you have the 100 volt windings, don't hook them to anything.

[luvdunhill]

didn't quite get to it last night, it will be waiting for me when I get to it tonite or tomorrow...

[kevin gilmore]

Its hard to tell from the picture, but you did use the ceramic washers right??

at least on the non-isolated parts.

[luvdunhill]

yup, everything is isolated. The no load test went fine. I had no idea how long the delay would be, so that added a bit of anticipation, but other than that the rails came up to 252V and 498V. I had +12V and -15V, but I'm assuming that's because there is no load on that -15V at all... before you ask, yeah at first glance, the part number was correct.. heh. tomorrow I'll see about mounting the heat sinks and getting the loads ready.