86% Efficient AC-DC ATX supply

auto_robotics · 10-06-2004, 10:40 PM

I've been trying to avoid the sine wave inverter + ATX supply route because of 1) inefficiency and 2) cost. I think I just found a good way around the efficiency problem ...

I just came across the Antec Phantom 350 ($155):
http://www.antec.com/us/productDetails.php?ProdID=24350
with 86% efficiency at full load.

Now combine that with the Samlex 600W sine wave inverter ($279) I was looking at:
http://www.samlexamerica.com/product...roductsID=7016
with 85% efficiency, and I get 73.1% overall!

Considering the Keypower 250W seems to be the only available 12V input higher wattage DC-DC supply and it's only 65% efficient, DC-AC-DC wins here.

Only issue is the $434 vs. $2XX something the Keypower costs.

I thought people would like to hear about this. Anybody with any better finds?

BTW, I need the higher wattage for some engineering experiments I'm planning on running, so please, no cries of OPUS!

mushin · 10-06-2004, 11:34 PM

OPUS! Opus solves EVERYTHING!
j/k

There's an interesting thread going on about skipping the DC-DC or DC-AC-DC altogether and using cheaply acquired deep cycle batteries instead. You'll still need to isolate your charging system, but it may be worth a look.

hcker2000 · 10-06-2004, 11:43 PM

Yea if you got the cash to waste on that junk u were talking about just buy an opus.

Ricky327 · 10-06-2004, 11:56 PM

And OPUS says :

High Efficiency power supply (extended battery life).

I wonder whats the exact figure...I hate it when they cant say the damn figure.

Can we say that if a PSU is 100% efficient and can run the PC for 120mins. Lets say OPUS is 85% efficient and a DC-AC-DC is only 75%. The OPUS can run the PC for 102mins and the DC-AC-DC can run for 90mins, now thats only 12mins more running power...wow big deal

Thats why I can never understand the DC-AC-DC efficiency bashing...or am I missing something here?

So anyone know the exact figure for the OPUS?

auto_robotics · 10-07-2004, 12:07 AM

Oh, I have no doubt that the venerable OPUS is still the way to go for <150W apps, but that's not me.

From OPUS's site:
http://www.opussolutions.com/modules...=article&sid=6
The supply is 90% efficient. So that kicks ***, but again, not for higher wattage apps.

My power supply is isolated out the wazoo. I just installed an auxiliary Optima D34/78 for the computer and a Hellroaring isolator and surge protector between the two batteries. Still need to power the computer though, so that's my current project.

Thanks for the replies. I would love to hear about any DC-DC supplies with >250W watts and high efficiency (70%+). I've found some, but those are always 24/48V input and to convert to that, you lose still more efficiency. At that rate my poor ol' Camry will be using most of its engine power to run the computer.

mushin · 10-07-2004, 12:08 AM

I recall my opus advertising 95% when I got it (last year). Not an exact figure, but it's ballpark. I always thought the extra efficiency was a nice bonus to lesson the strain on the alternater, never so that my batter will last longer... car batteries aren't designed for deep discharging, and I don't run the computer with the ignition off.

edit: duh, of course even if you have a nice isolated 12V power, you still need a unregulated DC PSU

Bad me! Though at that point, you just need to get 5V and 3.3V... Well, short of making your own, I havn't seen any DC-DC units greater than 250W. Good luck on you system, sounds like you've got quite a setup going there.

Ricky327 · 10-07-2004, 12:14 AM

95% sounds alot better to compare with

But really I dont think DC-AC-DC solution is that really bad the way people described it.

Sure DC-DC is an ideal

Ricky327 · 10-07-2004, 12:16 AM

Look for MrP's PSU design, not sure about the exact power output or efficiency but it does look promising if you need more power.

Tony S · 11-22-2004, 10:29 AM

Opus at 90% efficient? It's possible because of the low wattages, but with all the problems people are starting to have I doubt that figure. I'm beginning to think the Opus has been out long enough that we're now seeing it's useable life expectancy... or the techy term MTBF. I think that MTBF is getting shorter.

MrPerfectionest · 11-22-2004, 12:27 PM

Really at this point in the game there just isnt a very economical high powered solution because in general it doesn't make sense to need this kind of power in a car for every day purposes.

I am curious what you are experimenting with that you need such high power.

Secondly, is a build your own option out of the question?

I have designs (mentioned by Ricky327) that I am kicking around but generally if 250-300W is required then in small quantities it is still going to cost in closer to $100 just for the parts.

ColdPhreze · 11-22-2004, 12:38 PM

Tony, we can't really blame OPUS for most of the problems users are experiencing. I have been keeping up with those threads, and it seems a recurring theme that many having trouble have been overloading their OPUS for some time.

When I first signed up with mp3car and learned of people saying OPUS would handle over 150 watts, I warned them they would have troubles in the long-run when they overload an OPUS, as my friend did. It is why I decided not to use an OPUS with my setup: I didn't want any problems as I was going to draw much more than 150 watts.

As a rule-of-thumb, you will severly shorten the life of a PSU if you draw more than 80-90% of it's stated capacity. Following that rule-of-thumb, you can safely draw 120-135 watts from an OPUS (these a approximate, but are a good practice when designing a PC). Also, remember that you could be pulling more current on one rail than it is designed for, and much less on another, still be within 120-135 watts, but the PSU won't last becuase that one rail being stressed will die.

For now, the *best* solution for a high power pc is either DC-AC-DC, or the Keypower 250W. Please don't flame me. Just because you have been using an OPUS for several years at over stated capacity, doesn't mean it is the norm. We are seeing the norm unfolding on this forum, and anyone who designs power supplies will agree overloading one will cause it to die prematurly.

ColdPhreze

eCar™ · 11-22-2004, 02:55 PM

I'd expect the efficiencies of the KP and Opus to be closer than what you think.

Opus rates theirs at "optimal load". KP rates theirs at "typical load".

Neither one says what that load is. And you didn't mention what your typical load would be.

I wish someone who had both (if there is someone with both), would do some testing (at various loads).

eCar™ · 11-22-2004, 03:04 PM

Additionally the Salmex doesn't mention what load that it's 85% efficient at. You woudn't be running it at its full load (wher it is likely to be most efficient), if your other options include Opus.

Tony S · 11-22-2004, 07:38 PM

"Just because you have been using an OPUS for several years at over stated capacity, doesn't mean it is the norm." ColdPhreze, I guess you read my old sig line. My system's changed since then but anyway I've not overloaded my Opus. I agree that most users pro'lly are. The average person (are we average here though?

) doesn't know about electronics and electronic components so this doesn't surprise me. Luckily sites like this exist where we can help each other out. Still, short of overloading, I found it odd that now all these problems arise. Coincidence? Maybe. We'll find out as time goes on.

MrPerfectionest · 11-22-2004, 09:22 PM

Quote:

As a rule-of-thumb, you will severly shorten the life of a PSU if you draw more than 80-90% of it's stated capacity. Following that rule-of-thumb, you can safely draw 120-135 watts from an OPUS (these a approximate, but are a good practice when designing a PC). Also, remember that you could be pulling more current on one rail than it is designed for, and much less on another, still be within 120-135 watts, but the PSU won't last becuase that one rail being stressed will die.

Where did you arrive at this 80-90% value from? Is that just from your experience?

In general you can't just go on a blanket statement about total power drawn. You have to look at each rail independent to the total power draw to see if you are violating anywhere.

I haven't looked at the full Opus spec in awhile but if a supplier says that this product will do this kind of current on these lines given a certain input voltage and a certain temperature then I fully expect their product to do that or else they are lying.

As a consumer should you err on the side of caution? Perhaps. But sometimes that isn't always so easy to do.

User Name		Remember Me?
Password