SissySIT R.3

Member
Joined 2018
Paid Member
Great!
I'm about to build those PSU too, but I'm afraid that some semis won't be available this year

:confused:
I had to order from both Mouser and Digikey to obtain matching hFE suffix for the pass BJTs. I think I was able to get everything else off of Mouser a few weeks ago, tho.

While it took some time (and $$$) to pull together, it appears to be a solid choice for Class A amps with lower PSRR.
The SLB needs to have extra bulk storage capacitance on its final output. From 10,000 uF to 24,000 uF per channel will do the trick.
The SLB is good at reducing ripple, but it still needs energy storage on its output.
I've seen some here add extra bulk storage even with the standard FW CRC. What's the benefit in doing so, out of curiosity?
 
Oh man... This is promising!

I went ahead and swapped in the SLB PSU and Neurochrome goodies, and from a quick test tonight I could not tell if the amp was on or off when pressing my ear to the woofer.

Wondering what you used from Neurochrome?

I was planning to use a Prasi LT4320 active rectifier, now considering changing to a SLB.

Randy
 
Member
Joined 2018
Paid Member
Wondering what you used from Neurochrome?
The Guardian-86 speaker protection & Intelligent Soft Start. Compact, high quality, and the soft start has the added benefit of additional remote start & power button/LED options. Not inexpensive, though. You get what you pay for, I spose.

I was planning to use a Prasi LT4320 active rectifier, now considering changing to a SLB.
Prasi's is a great PSU, but, I think it's probably best suited for a nice class AB build -- or, could maybe try adding a ripple eater to its output for the SissySIT. I'm thinking of selling the two boards I built (and xformers) on here since I don't suspect I'll be using them anytime soon (Sissy is keeping me busy and I need to buy better speakers now lol).
 
  • Like
Reactions: 1 user
Member
Joined 2019
Paid Member
Prasi's is a great PSU, but, I think it's probably best suited for a nice class AB build -- or, could maybe try adding a ripple eater to its output ........

Much better would be to add simple Cap Mx instead of resistor in the CRC chain. Prasi posted project with gerbers long ago:
https://www.diyaudio.com/community/...sy-capacitance-multiplier.297921/post-6031533

Using this ripple eater, brings a little les performance in ripple suppression than with SLB, which uses CFP (Siziklai pair) instead of Darlington transistor, but still very good.
 
  • Like
Reactions: 2 users
Much better would be to add simple Cap Mx instead of resistor in the CRC chain. Prasi posted project with gerbers long ago:
https://www.diyaudio.com/community/...sy-capacitance-multiplier.297921/post-6031533

Using this ripple eater, brings a little les performance in ripple suppression than with SLB, which uses CFP (Siziklai pair) instead of Darlington transistor, but still very good.
Thanks for the link. Since I already have the Prasi PWB's, this looks like a better way to go for me.
This will let me build as planned, and then add the cap mx and see if I like it.

For anyone following, I found this link to use to order the cap mx PWB. There is a bom and schematic there too.

Randy
 
  • Like
Reactions: 1 user
Member
Joined 2019
Paid Member
The SLB needs to have extra bulk storage capacitance on its final output. From 10,000 uF to 24,000 uF per channel will do the trick.
The SLB is good at reducing ripple, but it still needs energy storage on its output.

I've seen some here add extra bulk storage even with the standard FW CRC. What's the benefit in doing so, out of curiosity?
I made some measurements with explanation. As lengthy answer would be off topic here, answer is at the appropriate forum section:

https://www.diyaudio.com/community/...-at-the-capacitance-multiplier-output.389938/
 
  • Like
Reactions: 1 users
Official Court Jester
Joined 2003
Paid Member
tnx for that, far from being unnecessary

though, for me - I did those tests simply observing CRO and without any data save, issue long ago resolved

and I did wrote several times that cap multiplier is sort of replacement for series choke, while not replacement neither for choke energy storage nor for capacitance energy storage

in short, big cap is still necessary
 
Official Court Jester
Joined 2003
Paid Member
Naaah, not necessary with Class-A amps that draw constant current from the PSU.

and what with amplifier having non-equal rails sag, dynamically ?

say exactly as with amp of this thread - mosfet in negative rail is working as hell, while SIT is working against it, and speaker getting not sum but difference of their currents?

same applies to Schaded mosfet included in game

yeah, still all A Class amps, but as Tombo did measure - if nothing else there is difference in ripple with vs. without fat cap after cap multiplier ..... which is sorta important when you have SIT acting as variable resistor of not much effective ohms

though, ZM is stubborn, I did tried eons ago several sorts of series regs for amp main rails, same as several options of cap multiplier

nothing of these is sounding for my by Elephant treated ears, as simple fat CLC ...... and I'm happy camper - set of 4 proper EI chokes for dual mono amp cost me around 50 greenies

while any satisfactory regulator is easily complicated as amp itself, even without taking in account logic that more parts is equal to greater possibility of Gray Smoke

capmulti - brings politeness, steals vividity
 
Great opinions, boys.
I was about to build a cap Mx PSU, but it may be a better idea to build a beautiful CLC supply and call it done.

Am I wrong if I say that the choke specs have to be very carefully selected?
If so, it's not possible to build a versatile PSU, to be used with various class A amps 24V 1.3~2A ?
 
It is certainly possible to build a CLC supply that will work well for various First Watt and related amps. Dual-mono is the way to go. For ease of reuse, the supply will probably be best in a separate chassis, with a carefully built umbilical power cord to supply the amp.

The chokes do need to be chosen carefully, and will end up being some of the larger components in the supply. Look for inductance around 2.5 mH or larger, and DC current capacity of 2A or greater. Chokes that are rated for DC filtration will be Ok running close to their rated amperage, and may exhibit slightly reduced inductance when run at higher current.

I have included a picture of the recent supply that I built for my M2x. This was remarkably quiet with speakers of average efficiency, but showed some slight hum pickup when I switched to higher efficiency speakers. The hum was due to coupling between the power transformers and the Edcor signal transformers.
 

Attachments

  • M2x Revised.JPG
    M2x Revised.JPG
    883.1 KB · Views: 156
  • Like
Reactions: 1 users
Thank you guys.

@TungstenAudio, nice build. I have standalone synchronous rectifiers aswell, and will reuse them.
Could you tell me the purpose of your two 3W resistor? I imagine one of them being a bleeder, but the other one?

Regarding the iron, I don't have Trafomatic guys next door, but I may break the bank with our friend Alexander, the crazy winder, for some fancy chokes that give better sleep :cool: