As for PSU for any SOTA SS amps.
This is my suggestion:
1. CLC RF Fitter -> R Core or E-I Core power transformer -> Fast Switching Soft Recovery Diodes -> CLC RF Filter -> CRC Mains Frequency Filter.
All in separate enclosure.
2. In the amp's enclosure, C -> SuperTedyyReg -> Salas Shunt regulator.
The above (1) and (2) would better be double, one for each channel.
In addition, a capacitance multiplier for each stage.
This is my suggestion:
1. CLC RF Fitter -> R Core or E-I Core power transformer -> Fast Switching Soft Recovery Diodes -> CLC RF Filter -> CRC Mains Frequency Filter.
All in separate enclosure.
2. In the amp's enclosure, C -> SuperTedyyReg -> Salas Shunt regulator.
The above (1) and (2) would better be double, one for each channel.
In addition, a capacitance multiplier for each stage.
My goals, from the beginning, included finding low cost, available, parts for the whole project. Only the regulator portion of the Kubota is needed when the r-core, rectifiers and caps are located externally, hence the Kubota pcb only (available from same seller as JC-80). My pet peeve is BOMs that do not align (missing, inconsistent, not matched to silk screen, version mismatch) with other parts. IP keeps many of these alive.
My cache of SERIOUSly NOS Dale 1% resistors arrived today. They were declared surplus 7/18/88! A single value, 36.5R, can be used with all the lm334s (paralleled in the higher current case). Later I can add trimmers to achieve tempco=0 should drift rear its ugly head. I really enjoy the matching activities and it makes me feel SERIOUSly fanatical
The only real compromise to date wuz the Caddocks. John would likely view this entire enterprise as compromised but, as he has pointed out, what's a fanatic to do?
I dig you completely!
I do match all my components manually. Most kits I have bought in the past, I have always changed resistors, mkp, capacitors. Kept the board and transistors since it was still a value... I have a stack of those unknown resistors that I don't use...
Having the rectifiers on a separate location should help in the noise since you're keeping the AC away. Good thinking!
Keep on measuring!
Do
Hi Joshua_G,
I don't see how piggy backing two regulators could help in #2. This is most likely to cause more issues than help, no? Speed might get affected...
Do
When iz enuff ENUFFF?
Sorry, Do, the lm334 is listed as an alternate to the crds in the BOM, but the resistor values are inconsistent with the specified crd currents. In my quest to simplify, minimize different parts and shipping I _MAY_ have stepped across that fine line between fanacism and obsession
The crds were spec'd @1.5 and 3.5ma but the resistor+lm334 came in @2 and 4.5ma. Using 36.5r I can get to 1.85 and 3.71ma w/o a trimmer (R1). The lm334 datasheet suggests "For ±10% adjustment, select Rset 10% high and make R1 ~ 3 Rset" Zero tempco would require more but even this fanatic / obsessive hopes to avoid that.
I guess I missed a part somewhere...
Sorry, Do, the lm334 is listed as an alternate to the crds in the BOM, but the resistor values are inconsistent with the specified crd currents. In my quest to simplify, minimize different parts and shipping I _MAY_ have stepped across that fine line between fanacism and obsession
The crds were spec'd @1.5 and 3.5ma but the resistor+lm334 came in @2 and 4.5ma. Using 36.5r I can get to 1.85 and 3.71ma w/o a trimmer (R1). The lm334 datasheet suggests "For ±10% adjustment, select Rset 10% high and make R1 ~ 3 Rset" Zero tempco would require more but even this fanatic / obsessive hopes to avoid that.
Totally John Curl + ticknpop at beginning of thread. 'twas indeed good thinking!
Sorry, Do, the lm334 is listed as an alternate to the crds in the BOM, but the resistor values are inconsistent with the specified crd currents. In my quest to simplify, minimize different parts and shipping I _MAY_ have stepped across that fine line between fanacism and obsession
The crds were spec'd @1.5 and 3.5ma but the resistor+lm334 came in @2 and 4.5ma. Using 36.5r I can get to 1.85 and 3.71ma w/o a trimmer (R1). The lm334 datasheet suggests "For ±10% adjustment, select Rset 10% high and make R1 ~ 3 Rset" Zero tempco would require more but even this fanatic / obsessive hopes to avoid that.
Yeah... I just looked at the BOM... I actually never looked at it in detail before since I had bought kits... They were to power some test equipment at the time. Now they're resting in my drawer until the next project! But they're really good so I might use them for a head amp.
Do
The SuperTedyyReg serves to clean the supply lines to a great degree, however it has a relatively high Z Out.
The Salas Shunt Regulators clean the supply lines further and the have very low Z Out, up to very high frequencies.
The SuperTedyyReg can be replaced by a CRC Filter in the amp's enclosure. However, for cleanest possible supply lines, I'd retain them.
In any case, there is co conflict and no possible problem by having those 2 regulators in cascade.
Whatever the BOM says, the SuperTeddyReg is far superior to any LM chip.
That is, for those wishing to attain the highest possible sound quality of any given amp.
I considered the SuperTeddy but I seem to recall a pair was needed for dual rails, hence a quad for a pair of JC-80's. @46 euros each (6 for the pcbs) these fell off the end of my 'bell curve' but his web site notes "we believe that such high end products don’t have to be outrageously priced." I would love to listen to JC-80s powered by a quad of these but four of them would nearly double my current total project budget. Then I would need to SERIOUSly need to consider a blowtorch and shortly thereafter to consider . . .
What I have shot for is Everyman's JC-80. I am really curious how good it might get but it is in competition with the BAL-BAL 'wire'. At this point it is 2x the cost of that and climbing. With superTeddies it would jump again and I would grow skeptical.
I don't wish to influence power chain decisions and welcome variations. I agree that the superTeddy likely beats any LM chip but . . . it better!
What I have shot for is Everyman's JC-80. I am really curious how good it might get but it is in competition with the BAL-BAL 'wire'. At this point it is 2x the cost of that and climbing. With superTeddies it would jump again and I would grow skeptical.
I don't wish to influence power chain decisions and welcome variations. I agree that the superTeddy likely beats any LM chip but . . . it better!
Again, the TeddyReg may be skipped, but I wouldn't skip Salas Shunt.
You may start with LM regulators and later add a better option.
Anyhow, SOTA cannot be attained in a very low budget.
The late Allen Wright wrote that the power supply is responsible for 50% of the sound quality of an amp.
You may start with LM regulators and later add a better option.
Anyhow, SOTA cannot be attained in a very low budget.
The late Allen Wright wrote that the power supply is responsible for 50% of the sound quality of an amp.
Wolfsin would hope no option was skipped, merely that otherz lash up the more expensive configurations and bring them to the 2012 Denver CanJam. The Salas should be on the list, Do has the Borbely already, and I will take a Kubota to bring up the low end. This is looking more like a bazaar than a GB, Do
Those Teddy regs are in the same class as Paul Hynes regs I guess... Very expensive class...
Home Page
I bought AVCC shunts for my BII dac and they're amazing but really pricey! Especially I'm running a Dual Mono BII, that means I had to buy 4 of them...
Do
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.