I would like to be able to convert a Threshold S300 to a SA3. From looking at the manuals, it looks like the amps are very similar.
Was looking for input as to if this is possibe? The speakers I'm using seem to work fine with a SA/3 I tried out on them (loves the high current).
Thanks for any help.
Was looking for input as to if this is possibe? The speakers I'm using seem to work fine with a SA/3 I tried out on them (loves the high current).
Thanks for any help.
Completed the conversion of the first of two S300 to SA/3. Also changed out the emitter resistors.
The bias was a bit tricky, but it did stay at the new setting after a couple of tries.
I continue to be amazed at just how different (and better) tha amp sounds runnig in class A. It seems that most speakers are quite happy with 50 watts in class A. The speakers seem to have more bass heft running class A, and the reduced distortion is audible, no question.
This is one my favorite amps from Nelson. It just sounds like music.
The bias was a bit tricky, but it did stay at the new setting after a couple of tries.
I continue to be amazed at just how different (and better) tha amp sounds runnig in class A. It seems that most speakers are quite happy with 50 watts in class A. The speakers seem to have more bass heft running class A, and the reduced distortion is audible, no question.
This is one my favorite amps from Nelson. It just sounds like music.
Attachments
The chassis, plate, and internel layout of the S300 and SA/3 iis identical, so no worries there.
Class A is definately the way to go if the speakes support it.
Mine sadly does not have the extra windings. Might as well look for a replacement transformer.
As the original PS caps are far from great sounding, the reduction in voltage also presents the opportunity to go for lower voltage/higher capacitance/better quality caps should you decide to make the change permanent. If the bridge cannot take it, might as well go for some better sounding rectification too.
Btw, replacing the NFB decoupling caps from the original polycarbonates to Musicaps worked out extremely well.
I replaced the original PS caps with a pair of 68000 uf at 75VDC, along with solen NFB de-coupling caps. Yes, it does help indeed.
Finally got the time to try this out: got a 600VA transformer which would provide 2 x 32v rectified, connected it to my S300 and set the bias at a stable 200mA per device.
Well, that's another audiophile legend down the drain for me Sound suffered in almost every conceivable aspect at even the softest volume settings. Plumpy bass, weak dynamics, sugary coating on everything...maybe i'm deaf.
Yet, why shouldn't it be like that? After all, this bias level doesn't quite transform it into a real class A amp. The 5 class A watts of the S300 go up to 23W, still less than half the Krell KSA50 output. Can doubling the bias at the same dissipation really improve the sound? By keeping the same PS capacitance, stored energy is reduced 4 times and i think it shows. And as far as i can remember NP recommends an even lower bias point at 145mA...
So did the SA-3 use the same PS capacitance as the S300?
Well, that's another audiophile legend down the drain for me
Sound suffered in almost every conceivable aspect at even the softest volume settings. Plumpy bass, weak dynamics, sugary coating on everything...maybe i'm deaf.Yet, why shouldn't it be like that? After all, this bias level doesn't quite transform it into a real class A amp. The 5 class A watts of the S300 go up to 23W, still less than half the Krell KSA50 output. Can doubling the bias at the same dissipation really improve the sound? By keeping the same PS capacitance, stored energy is reduced 4 times and i think it shows. And as far as i can remember NP recommends an even lower bias point at 145mA...
So did the SA-3 use the same PS capacitance as the S300?
Finally got the time to try this out: got a 600VA transformer which would provide 2 x 32v rectified, connected it to my S300 and set the bias at a stable 200mA per device.
Well, that's another audiophile legend down the drain for me
Yet, why shouldn't it be like that? After all, this bias level doesn't quite transform it into a real class A amp. The 5 class A watts of the S300 go up to 23W, still less than half the Krell KSA50 output. Can doubling the bias at the same dissipation really improve the sound? By keeping the same PS capacitance, stored energy is reduced 4 times and i think it shows. And as far as i can remember NP recommends an even lower bias point at 145mA...
So did the SA-3 use the same PS capacitance as the S300?
Well, a couple of points:
1. The transformer with the S300 that sports optical bias is a 700 VA unit, not a 600VA.
2. The driver boards for the optical bias were designed to work with both Class A and AB. I do not think the older driver boards will work as well.
3. The S-300 and SA/3 with optical bias is essentially the same amp. Not true of the older S-300.
When I did the re-configuration. I put in a pair of 68,000 uf caps rated at 75 VDC. Also changed the emitter resistors, wiring, rectifier, binding posts, RCA jacks, etc. I set the bias at 260ma cold start, as per the reference sheet posted here at DIY.
Sorry it did not work in your case.
If you do not have the optical bias S-300, I’m not surprised it did not work out. So, for now, legend still holds up.
Last edited:
Excellent points Redwingnine. It is the reason i mentioned the 600VA - not that i would attribute what i heard solely to the power rating. I may need to examine the circuit of mine (slightly different from the ones on the net) and see if the reduction in PS voltage may lead to current starvation in some stage. I already have nice binding/rca posts and higher capacitance Rifa caps in place. Curious as of what rectifier bridge you chose and why.
Interestingly i only started liking the S300 after a multitude of parts upgrades i did years ago. At the time i intended fitting a second power transformer with slightly higher output and feed the input circuit a regulated voltage. Other amp projects put me off track but i might still try it.
Interestingly i only started liking the S300 after a multitude of parts upgrades i did years ago. At the time i intended fitting a second power transformer with slightly higher output and feed the input circuit a regulated voltage. Other amp projects put me off track but i might still try it.
I used an IXYS HEXFRD bridge rectifier. They have very fast recovery times, and are quiet to boot. Jon S. from Vintage Amp Repair uses them for his Threshold upgrades.
Hope you can sort out the issues. My S-300 conversion sounds outstanding, an provided a noticeable improvement over stock S-300 Class AB operation. The speakers that I’ve used them on are all easy loads, ranging from 88 to 93 db/w.
One thing I did notice was that the amp initially drifted downward in bias, and it sounded a bit like you mentioned. After adjusting a couple of times, the bias circuit locked in at the higher value, and the sound became much better.
Hope you can sort out the issues. My S-300 conversion sounds outstanding, an provided a noticeable improvement over stock S-300 Class AB operation. The speakers that I’ve used them on are all easy loads, ranging from 88 to 93 db/w.
One thing I did notice was that the amp initially drifted downward in bias, and it sounded a bit like you mentioned. After adjusting a couple of times, the bias circuit locked in at the higher value, and the sound became much better.
Thanks. Partsconnexion seems to be carrying these.
Is it possible we simply have a preference for a different type of sound? Bias is not easy to set accurately even with the multiturns i have in place. Takes a couple of hours to get stable, so no, this is not it. Had a quick look at the opto-bias circuit. No reason why it would make any difference. Mystified.
Is it possible we simply have a preference for a different type of sound? Bias is not easy to set accurately even with the multiturns i have in place. Takes a couple of hours to get stable, so no, this is not it. Had a quick look at the opto-bias circuit. No reason why it would make any difference. Mystified.
It’s possible we could like different sound, BUT, we generally want our systems to sound as life like as possible, so all things being equal, we should like the same sound at a macro level.
I reviewed the manuals. One should not write off the importance of the bias setting. The difference between the S-300 and SA/3 is at least 105ma per emitter resistor reading. Also, keep in mind that the optical bias runs a lot hotter (49C vs. 42C). I know from restoring a 400A that it did not sound good until Papa helped me resolve the bias issue.
Lastly, the speakers play a large part into this. What is the db/w of the speakers you are using?
I reviewed the manuals. One should not write off the importance of the bias setting. The difference between the S-300 and SA/3 is at least 105ma per emitter resistor reading. Also, keep in mind that the optical bias runs a lot hotter (49C vs. 42C). I know from restoring a 400A that it did not sound good until Papa helped me resolve the bias issue.
Lastly, the speakers play a large part into this. What is the db/w of the speakers you are using?
Modified ProAc EBS, similar to ATC50. Probably not much better than 85db/w. Did the listening test way after midnight at possibly mW power levels...
Anyway, next stop will be separate, probably regulated PS for the front end.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.