[Precaud]

I just completed these detailed mods for the SRM-252S amp:

Upgrading the Stax SRM-252S

Comments/questions welcome.

 

[yaluen]

I just completed these detailed mods for the SRM-252S amp:

Upgrading the Stax SRM-252S

Comments/questions welcome.

Thanks for documenting, would appreciate the mentioned link to BOM if have it handy for simpler parts ordering, cheers.

 

[Precaud]

Thanks for documenting, would appreciate the mentioned link to BOM if have it handy for simpler parts ordering, cheers.

Am working on it... may be a couple days yet.

 

[Precaud]

Link to the BOM is posted now, just before the parts list..

 

[yaluen]

Link to the BOM is posted now, just before the parts list..

Awesome, much appreciated

 

[550914]

deleted

 

[Precaud]

I've done the 200k feedback mod today and yes the soundstage is more clear for sure, even with the basic metal film resistors I had in hand.
I'll do your filtering mod later, as I need to order those specific parts.

That's clever! Yes, changing the feedback R is the easiest way to get a taste of what's to come. If you have any green LEDs laying around, change D7 and D8 for the extra current until the other parts arrive.
Even quicker and easily reversible is to solder a 600KΩ across the existing 300k.
The HVPS mods make the biggest difference, though.

This little Stax amp is heavily underrated, I'm pretty sure it can compete with way more expensive gear, especially once modded.

I have not compared it to any other Stax amp but my hunch is the fully-modded one will fare well in comparisons.

 

[550914]

deleted

 

[Precaud]

OK, that's great, you were definitely heading in the right direction. Your HVPS Z curve is similar to the "Tidy" one except higher due to the 2nd 470Ω still being used.
So it will still be very worthwhile to complete the mods.

It turns out, the film bypasses do nothing to reduce the switching noise. They're way too small. Their impedance at 65kHz would need to be less than the 'lytic's ESR (about 0.2Ω) to make any difference. That means they would need to be 24uF to be effective. So you can see why Stax chose the 22uF lytics. It's the minimum necessary.

FYI, C31 may be physically close to the power entry, but in the circuit, it comes after the long, thin traces to and from the power switch. Lots of inductance. (This is primarily why using a "super-regulator" circuit for the -12V makes little or no difference.) But definitely go bigger than 470uF for C31. A 2200uF @ 16V fits nicely. And move its bypass to right across the barrel jack to trap RF.

 

[550914]

deleted

 

[Precaud]

Regarding the input stage, I followed Borzim :
- R1 / R2 Removed
- R3 / R4 Removed (strap placed)
- C1 / C2 Removed

You did not retain those recommendations, is there any drawback of cleaning up the input stage this way?

I didn't concentrate on them, I was looking for big fish, not minnows
But I think the 50k pot is too high, I'd prefer 10 or 20k. R1 lowers it to 28k, so I left it.
Pot + R3/C1 is an input lowpass filter, I live a mile from radio transmitters, so I'd rather have something there than nothing.
EDIT: My parts list changes R3 to 1kΩ and C1 to 100pF, for about 132kHz cutoff. 68pF would be ok too (195kHz). The AM radio band starts around 500kHz, so I want it below that... the original parts had it at 33kHz !!