It seems like dual mono builds tend to double the total capacitance (and VA for that matter) rather than dividing the total power requirements equally and therefore halving the spec. for each supply. Is this a matter of "more is always better"?
For example, a build I'm considering typically employs 300va with 120,000uF total capacitance. The plan is to split the supply into dual 160va toroids and I'm having difficulty justifying doubling down on caps for a total of 240,000uF. If I reduce the total to say, 160,00uF, is this a reasonable approach?
For example, a build I'm considering typically employs 300va with 120,000uF total capacitance. The plan is to split the supply into dual 160va toroids and I'm having difficulty justifying doubling down on caps for a total of 240,000uF. If I reduce the total to say, 160,00uF, is this a reasonable approach?
1. Build dual mono.
2. Listen.
3. Connect these two psus (together).
4. Listen.
5. ...-)
Couple ways to interpret what you are hinting at. If you mean doubling up on a single channel, then I think I get you--bigger=better. But all else equal, would one 300va 120kuF stereo be comparable to two 150va 60kuF monos, or is the implication that the bigger one will be better, stereo or not?
Short way: Build ONE psu for all.
If dual mono: it is sufficient to halve the values. Because the values are not determined by ear. And most audio-signals are nearly mono or mono: psus are generally not designed to feed highly unbalanced, single-channel signals.
Much more important is the quality of parts. And whether you build up the capacitance by means of one, less or many capacitors,
And a tip: If you do build a "+ and - psu" (exacter: two psus;-)-: use for each rail ONE secondary winding only.
My opinion. My experience.-)
This is one of those counter intuitive and really fascinating situations-- the PSU is greater than the sum of its parts. One parameter driving my choices is the trafo height; I'm trying to keep true to a 1U envelope, so around 1.75" tall. This makes the dual-mono strategy very attractive because the smaller toroids can get me close to that minimal height. I guess you really nailed it in suggesting I connect the dual supplies to access better performance without even knowing my constraints 🙂
Okay, now this would seem to complicate any and all scenarios! Trying to picture this setup. Also, more smaller caps or few giant ones in your experience?
Hmmmm...-)
Several caps - generally when parallel connection of components - are to set at ONE point - of circuit;-) Then it benefits of smaller sizes of caps: more exacter "current", "flow". For sound reasons no screw terminals!
Are you building a HiFi or a PA amplifier?
Gotcha, you want the close proximity advantage plus the current capacity of parallel. HiFi, mainly Pass stuff and maybe TPA3255. I'm going to make some major tweaks to the current setup and post here for better scrutiny. I think I understand your comment on configuring secondaries so I'll try that and hopefully you can chime in and tell me if I did it wrong!
Not all that beefy, 25w classA Pass. What I am trying to do that is somewhat novel is pack a Nutube B1, M2x and dual toroid PSU into a 1U billet aluminum block. Rowland style, if you will. It's taking some pretty extreme fitting but I'm convinced it's doable. However, the PSU is the trickiest space wise and both the amp and preamp circuits will have to totally be reconfigured to deal with that. The long and wide boards won't fit. Here is the concept in its current form...
Attachments
Ahhh, now I see. Got to admit that the NP B1 is a favorite of mine. Just resurrected one last week. I tried using a B1 in an integrated amp that also used the nutube in each power amp channels. In my case it was a class D power amp. Sounds pretty good but still needs some work.
I don't envy you trying to pack a bunch of caps into the power supply. One class A power amp that I had redesigned for me had 4 large caps, but on the circuit board there was a 1,000uf cap (per channel). I am not an expert, but it was said to be that way to have some of the capacitance needed really close to the output transistors.
I don't envy you trying to pack a bunch of caps into the power supply. One class A power amp that I had redesigned for me had 4 large caps, but on the circuit board there was a 1,000uf cap (per channel). I am not an expert, but it was said to be that way to have some of the capacitance needed really close to the output transistors.
Take this with a huge grain of salt but, if you're building a Pass Labs or First Watt preamp or amplifier I've found that increasing the power supply capacitance indicated in the original design by 50% to 80% is a good thing. More than that and (for me, and perhaps me alone) the bass starts to sound a little wooly. A lot more than that and I hear bloat in the lower range. With dual mono, split the capacitance between the two supplies accordingly.
In my Aleph P 1.7 build, increasing the power supply capacitance from 4,000 uF to 7,200 uF per channel made a really nice improvement. My F5 Turbo V3 amps were originally built to the 160,000 uF standard specified in NP's white paper; a year or so later, I added 90,000 uF per channel (45,000 uF per rail) and really loved the change. My F6 was built as a dual mono amp from the get-go with 96,000 uF capacitance per power supply channel. Compared with a friend's stock F6 (with 120,000 uF in a single supply), I believe mine sounds a little more substantial, with an improved soundstage and stage depth.
Your mileage, as they say, may well vary.
In my Aleph P 1.7 build, increasing the power supply capacitance from 4,000 uF to 7,200 uF per channel made a really nice improvement. My F5 Turbo V3 amps were originally built to the 160,000 uF standard specified in NP's white paper; a year or so later, I added 90,000 uF per channel (45,000 uF per rail) and really loved the change. My F6 was built as a dual mono amp from the get-go with 96,000 uF capacitance per power supply channel. Compared with a friend's stock F6 (with 120,000 uF in a single supply), I believe mine sounds a little more substantial, with an improved soundstage and stage depth.
Your mileage, as they say, may well vary.
Is that an engineering term? Smells like something else to me.
There are very sound reasons that some capacitors have screw terminals.
I'm not sure what the original M2 has for capacitance but the power rating is 160w compared to 180w for F6. So if you ended with 120kuF maybe I can dial back from 160kuF quite a lot! 100-120k is probably a good range and will certainly help with the space constraint. Good to know, thank you-
Oh dang! Glad you caught that, I was about to place the order lol. Okay, so my original version was about perfect at 160kuF total. Phew 🙂
Do you have a thread on this? D with the Nutube sounds like a really cool mix. One problem I might (likely) run into with the low power class A is I ultimately plan to drive a pair of Purifis and they really need more power. But if start going higher power on the class A it's gonna become unfeasible to cool; even with fans probably.
You could compare your amp with your amp: connect both psus together.
In general, I recommend comparing with itself. Already the noise behavior of each device is different. The hearing gets used to this. A comparison even after a longer time out - for example for soldering work -, is problem-free.
Alongside: not the supposed power of an amplifier corresponds to its sound, its "power" and speaker control. The choice of components is decisive. Unfortunately, only about 10% of electronic components are usable for audio applications. Lots of testing/listening work.
A tip: components generally as small as possible. Except for some capacitors.
Sorry, I have written here once for self-perception;-)