Does anyone know if it would be possible to push the L-Adapter to 24v @ 3A? I have a Burson Soloist 3x headphone amp and they have just released a 'Supercharger' for it which is quite expensive and is just a wallwart adapter. They have rated it at 3A but I cant see how a headphone amp would actually need almost 75w of power so Im a bit dubious it actually needs that much. But would it be possible to build an L-Adapter for this use?
Correct, it could be done with a higher value trimmer and a 22VAC secondary transformer. Regarding the Soloist's 3A DC input rating it certainly seems rich for what the open case pictures at Burson's website and elsewhere suggest. Almost only chips, switching supplies, no board level sinks. Eight TO-220s look thermally coupled to the chassis. And the two chip regs at the V6 and Muses rails.
1A looks more realistic to me, 3A maybe was the smallest decent quality wall-wart to get in the OEM market. Also why not rate it at 3X power for long term reliability. Its a cheap mass produced device nowadays.
If you want to know the Soloist's practical current consumption, you need to measure it. Either right there at the DC input or approximately by translating its wall power draw using a KILL A WATT style device to see it. Of course there is efficiency loss and power factor to consider when from there. But roughly.
In any case the L-Adapter will not be a bottleneck for 3A peak if there is any actual condition ever demanding that with the Burson Soloist.
1A looks more realistic to me, 3A maybe was the smallest decent quality wall-wart to get in the OEM market. Also why not rate it at 3X power for long term reliability. Its a cheap mass produced device nowadays.
If you want to know the Soloist's practical current consumption, you need to measure it. Either right there at the DC input or approximately by translating its wall power draw using a KILL A WATT style device to see it. Of course there is efficiency loss and power factor to consider when from there. But roughly.
In any case the L-Adapter will not be a bottleneck for 3A peak if there is any actual condition ever demanding that with the Burson Soloist.
My initial L-Adapter for my First Watt F6 carried the full 3.2A load for both channels @ +/- 23v, with a voltage drop about 4.5v across the regulator - it needed a pretty big heatsink and a quiet fan! This was further modded to use a separate L-Adapter reg (pair) for each channel (+/- 24v @ 1.7A) sinking about 15watts per channel
Excellent result - the quality of the capacitors makes quite a difference.
Excellent result - the quality of the capacitors makes quite a difference.
That was a good task for L-Adapter. First Watt F6 spec says 180W power consumption. Not trivial when wanting stabilized rails.
Wise move you spread the dissipation with dual mono L-A. Congrats for the nice result
*What specific reservoir capacitors you favored in this situation?
Ah, what caps, eh? Well, the long suffering F6 amp amp has undergone a number of 'upgrades' (with various versions of 'improvement', as you'd expect) - I first started with my favourites, 'ye old Epcos/Siemens B41550's with the basic MSR diodes and was quite happy with the overall sound - then changed up to the fancy lt4320 rectifiers (slightly clearer) with the same B41550s and the L-Adapters using the Elna Silmic with the Nichicon bipolars and operated with the reduced rail voltages okay and the definite overall improvement in the whole amp was quite a surprise.
I had some pcbs made with both rails and have a number of these regulators with different components and still trialing different combinations - for example, I've generally settled on different versions of the Sigma 22, the Super Teddy Reg or one of your high performance regs but some situations are better suited to this little gem and optimizing components gets the last little drop if performance, IMO.
Curiously, replacing the transformer with the higher secondary voltages returned the original rails (+/- 24V) but no really noticeable change to the sound (apart from my imagination, I think!)
Then I changed the 'workload' of the amp to the Wild Burro Betsy Speaker in the Caintuck Audio sized OB (Open Baffle) from 160Hz up via the diyA 6/24 Xover (yes, with the hypnotize reg!) and went through the caps routine again and surprisingly ended up with quite a different set of caps on the amps supply with the same Torroidy trannie and the lt4320 bridges but used the Nichicon Superthrough for the ripple caps and the Rifa 169s for the 'onboard' second filters on the L-A boards (the usual C-R-C setup) with the green Nichicon bipolars and the big Slit Foils on the output (only had the large 4700uF ones)
[Note here: I had a few different ancient Black Gate caps here and got a few of the AudioNote new version ones from HiFi Collective but even after some weeks (about 500 hrs use), I was a bit disappointed in the resultant sound - they just didn't suit my requirements here
Then I replaced this whole C-R-C setup with the single T-Network cap for the ripple cap and kept the L-Adapters the same with the Nichicon bipolars and the SlitFoil output caps and this was like stepping out of a basic sedan car into a sports car! As you would expect, the Fostex and Lowther drivers didn't appreciate the change!
It's still a 'work in progress' as I've just started using the RME ADI-2 dac instead of the Ayre Acoustic QB-9 dac and the extraordinary adjustments in the tone controls has started a 'new ball game' (sorry for the poor analogy!) for matching amps/power supplies with speakers and the room - I think it's quite possible to get similar results without the exotic components/cost but these 'more flexible' pro-audio type dacs aren't at all cheap so 'tuning' the power supplies is still well worth looking at - IMO, naturally.
Many thanks, Salas, with persevering with these simple yet elegant power supply designs over the last 20 odd years.
I had some pcbs made with both rails and have a number of these regulators with different components and still trialing different combinations - for example, I've generally settled on different versions of the Sigma 22, the Super Teddy Reg or one of your high performance regs but some situations are better suited to this little gem and optimizing components gets the last little drop if performance, IMO.
Curiously, replacing the transformer with the higher secondary voltages returned the original rails (+/- 24V) but no really noticeable change to the sound (apart from my imagination, I think!)
Then I changed the 'workload' of the amp to the Wild Burro Betsy Speaker in the Caintuck Audio sized OB (Open Baffle) from 160Hz up via the diyA 6/24 Xover (yes, with the hypnotize reg!) and went through the caps routine again and surprisingly ended up with quite a different set of caps on the amps supply with the same Torroidy trannie and the lt4320 bridges but used the Nichicon Superthrough for the ripple caps and the Rifa 169s for the 'onboard' second filters on the L-A boards (the usual C-R-C setup) with the green Nichicon bipolars and the big Slit Foils on the output (only had the large 4700uF ones)
[Note here: I had a few different ancient Black Gate caps here and got a few of the AudioNote new version ones from HiFi Collective but even after some weeks (about 500 hrs use), I was a bit disappointed in the resultant sound - they just didn't suit my requirements here
Then I replaced this whole C-R-C setup with the single T-Network cap for the ripple cap and kept the L-Adapters the same with the Nichicon bipolars and the SlitFoil output caps and this was like stepping out of a basic sedan car into a sports car! As you would expect, the Fostex and Lowther drivers didn't appreciate the change!
It's still a 'work in progress' as I've just started using the RME ADI-2 dac instead of the Ayre Acoustic QB-9 dac and the extraordinary adjustments in the tone controls has started a 'new ball game' (sorry for the poor analogy!) for matching amps/power supplies with speakers and the room - I think it's quite possible to get similar results without the exotic components/cost but these 'more flexible' pro-audio type dacs aren't at all cheap so 'tuning' the power supplies is still well worth looking at - IMO, naturally.
Many thanks, Salas, with persevering with these simple yet elegant power supply designs over the last 20 odd years.
More LEDs you used are fine in a specialized variation.
But messy to bodge on a multipurpose Vout L-Adapter standard board. Especially since a higher value trimmer also takes it to 20V plus territory if needed.
It's not too bad vs the LEDs only low dynamic resistance approach either, because there's the bypass capacitor across the engaged LEDs & trimmer Vref system. Lowering the AC impedance enough while filtering noise.
Not to derive more reference voltage from the trimmer than the LEDs should be a good rule of thumb for settings.
But messy to bodge on a multipurpose Vout L-Adapter standard board. Especially since a higher value trimmer also takes it to 20V plus territory if needed.
It's not too bad vs the LEDs only low dynamic resistance approach either, because there's the bypass capacitor across the engaged LEDs & trimmer Vref system. Lowering the AC impedance enough while filtering noise.
Not to derive more reference voltage from the trimmer than the LEDs should be a good rule of thumb for settings.
You can find RC time formula and calculator online. Digi-Key has a nice collection of online conversion calculators for instance.
But put that resistor horizontally between the LEDs and the capacitor to work right. Transistor's base stays at the RC node.
*https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-time-constant
But put that resistor horizontally between the LEDs and the capacitor to work right. Transistor's base stays at the RC node.
*https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-time-constant