250watt PSU Idea

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hmm. Where did the whole DIY thing go?

First of all, if you're serious, besides Pressman's book you should also check out "Power Supply Cookbook" by Marty Brown and "Switching Power Supplies A to Z" by Sanjaya Maniktala. The latter gives ALL the math, and explains loop stability criteria in great detail (that chapter alone is worth the whole book). Pressman/K. Billings also have some errors sprinkled here and there (but i believe they're intentional), and they have the habit of sending you hunting thru the whole book for a formula.

With those three books as reference i have made myself a few Excel tables which help with design. If you're interested i'll PM you a copy... they will be posted to the general public too but only when i feel they are complete.
 
Hmm. Where did the whole DIY thing go?

This is more than I bargained for, and you now are questioning whether or not I will proceed any further.

It is disheartening to my original idea to find switching power supplies such as the meanwell at such wonderful prices and abilities. But I still think it would be nice to develop home audio PSUs with their own advantages, or at least look at this application.

I very much like the sound of the switching supplies book with the math. I have 4 power supplies to goof with, listen to and compare, and this thread is very early on. I might not have said that (when coming on like gangbusters), but that is the original plan and where I am in it.

I come from a world of audio, riddled with objective ignorance, subjective fantasy and political strife. What better way to cope with this than to see for oneself?

My inspiration is still alive, because there is still a kind of power supply that I might like to see.

Cheers,
Mark
 
MeanWell have a good rep generally but i've seen some of their supplies using questionable grade capacitors. IMO this is the first requirement in making a SMPS that will last any reasonable amount of time - not some silly audiophile grade capacitors, just high quality low ESR capacitors like Nichicon or United Chemi-Con, AND optimized board layout that doesn't place the caps right next to the heatsinks.

Before it sounds well, it needs to measure well. If it measures well but doesn't sound well - then you're probably measuring with the wrong equipment.

I could pay for a MeanWell myself - but why should i when the stuff sold locally is overpriced and i can't find the configuration i want, and from most places shipping to Romania is either expensive or takes a month. I buy old/busted ATX power supplies for $2 a pop, and there you go, plenty of parts to cook my own.
 
Last edited:
I haven't gotten the MeanWell yet (its in the mail). When I do, I want to compare it to the Antek PS-3N30, which has some ripple (2-3 V), but is an unregulated supply. I am getting (right now) some pretty hard and fast bass. The thing I haven't heard yet are the (my) dynamics with the MeanWell, a.k.a. regulated. This is something I have been wondering about.
 
p.s. I thought adding cheap, durable, low ESR caps sounded like a great idea to put in the supply.

And a banded (EI) transformer sounded like another one.

Actually, some of the thoughts now are in the direction of what happens in a computer PSU. In this case, the idea tending toward a supply with cards for each unit being powered. Perhaps each card could have one voltage, or a small range.

Another idea was one transformer with two DC supplies running off of <it>. Each DC supply being its own "card". One big one (amp) and one little one (other audio device), and maybe another little one (3).

But these are two totally different approaches.
 
Last edited:
I started using the MeanWell yesterday and today, and I've just finished some switching back and forth with the Antek. The biggest difference I can hear between them is that the MeanWell is helping across the frequency range. This makes me guess that the 1/10th-as-much ripple is helping in this regard.

The MeanWell has a fan in it, and it is loud. It isn't as loud as me, but no one else is home.

Using a 200-watt PSU for 15 watts on average... Can I just disconnect the fan and not deal with a thermal issue? Or put a little resistor in its place. The fan changes in speed (I presume off a control circuit).

I suppose what I'm ultimately going to do is a PSU that does not have a fan.

As far as cabling goes, though, I like that. I may be a little oversized, but I am using 12-gauge to connect either supply to the amp. When I went from 18-gauge to 12 (only tests I tried), the difference was noticeable.

I would like to make a fat cable with barrel connector ends. Do they make barrel connectors, like with a big shell/jacket?

Cheers,
Mark
 
Using a 200-watt PSU for 15 watts on average... Can I just disconnect the fan and not deal with a thermal issue? Or put a little resistor in its place. The fan changes in speed (I presume off a control circuit).

No, you need to have some air moving, but im sure you could disconnect the fan if you are sure the average (~ 2-4 min) is less than say 20% of max. There is the MW SE series I linked to above designed for convection cooling.


For a connector system it's probably easiest to use molex ( like on PC's) and multiple wires w 20 AWG. I think each pin/wire s/b good for 7 A, so x 3 pins each direction.
There are some cool new >20A connectors for batteries, but are difficult to source at the usual places.
 
regulation depends on the dynamic peak currents X interconnect and wire resistances. most connectors are rated for a certain temp rise due to I^2 R loss. so I can see them as dominate in your case except for longish wires. Note the multiple Voutput connections on the supply, those are not extras there for convenience.
 
Last edited:
I was going to say that the connector looks like traces, or the wiring going into the amp. Should I be worried about regulation across the cable? I had not considered 10 amps (Peak) being a problem for the contacts in these connectors. But that is also weird about an "audio PSU". You don't use the power that you would if you just needed a PSU.

For example, a 300 watt PSU for audio will be used for as much as 30 watts average, with really tough speakers and playing it very loud. Conversely, a 300-watt PSU as a PSU would have to provide 300 continuous watts. Do you think it could melt?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.