Re: [OT] ATX-PSU and amperage on connectors...
On Wed, 27 Feb 2008, Michelle Konzack wrote:
> Am 2008-02-23 18:00:39, schrieb Henrique de Moraes Holschuh:
> > Well, I recommend you go to extreme levels of output harmonics filtering,
> > that alone will increase system stability a damn great deal. Some tests
> > using memory bit-rot testing a few years go (either by Ars Technica, or
> > Tom's, I don't recall) nicely illustrated why one would want to do so.
>
> In the specification "ATX12V" version 2.2 from 2005 is all documented
> what is required. Some of the requirements affect only the AC-PSUs and
> not DC-PSU's...
*If* every ATX12V PSU out there implements these requirements properly, all
I can say is that they are NOT good enough. OTOH, if the bog-standard
ATX12V PSU you get from China and Taiwan are not following all the
requirements properly, then maybe they are good enough.
> And yes, Filtering the 24Vin is heavy... since I have no real
> experience with it and need some stuff to read.
DC inputs are always MUCH better since they have a lot less high-frequency
noise than the AC power grid. So you have to filter out less noise in the
first place. That's why people like -48VDC and -24VDC power supplies so
much in anything but the consumer marketplace.
BTW: if this is the kind of stuff you are going to leave in the field inside
an airtight box, did you remember to add some EMP shielding and surge
protection? Otherwise, the first time you get an electrical storm dropping
a lightning bolt near your box, the entire thing will fry.
> Can you suggest Literature in english or german?
Sorry, no. It has been a long time since I did any power supply design.
> > And don't think for a single moment that you are designing for linear loads.
> > Computers are !@#$!@#$#@$ unhelpful loads, and the more efficient the
>
> I am simulation some PIGs on my programmable Load... :-)
Make these PIGs switch on-and-off at high frequencies, with some minor
filtering to simulate the motherboard VR, and you will have the load a
modern CPU does :p
My laptop actually *sings* in high frequencies because some of the crap in
the mainboard is not phisically stabilized enough and ressonates due to the
high-frequency load changes :P It is not induced noise in the speakers, the
components actually make noise because they are vibrating at high
frequencies(!). You can imagine the load profile that causes such a thing.
Must have a RMS power factor close to zero...
I will get a glue pistol one of these days and cover the damn chips and
capacitors with it to see if that dampens the noise enough to not bother me.
Drat :-)
> > Yes. http://en.wikipedia.org/wiki/ATX, look at the external links section.
>
> I have gotten from two peoples the ATX12V sepecs, one v1.1 from 2000
> and one 2.2 from 2005 and it seems, there is all described what I
> should know...
The links point you to online up-to-date ATX PSU specs.
> since I have the need for complete 36 DC-PSU's and I do not realy like
> to develop them only for me since it is quiet expnsive for only 36
> DC-PSUs.
Well, the open hardware crowd could help you, I think.
http://en.wikipedia.org/wiki/Open_source_hardware#External_links has links
to many open hardware sites.
> There are some real weird things in the ATX12V specification, since some
> cables of my 600W AC-PSU have only 0.75mm² or 1.0mm² cables (arround 40cm
> long) which definitivly do not support more then 10A whithou being
> transformed into a heating-cable but the connectors should support
> 3.3V/18A or 5V/24A on the AUX Power connector.
If you don't use the standard connectors, you can have different power
requirements for them (just use high-quality connectors made for power
transfer when dealing with power transfer, and made for signal transfer when
dealing with signal transfer).
Otherwise, use the rates specified in the spec. They ask for a 18A type
<whatever> connector even when you won't make more than 5A available there,
because a conector of that type with less than 18A must be made of shinny
crap instead of real metal :)
> I can buy DC-DC-Converters from China but I have never seen sone, which
> have an efficiency OVER 80% (ATX12V is talking about 66% but for AC PSUs
> which are a VRY BIG no for my usage in Photopholtaik-Systems)
You can get high quality stuff from China, but it is not easy. Just get
them from Europe or the USA because they either make it well, or do the
quality checks on the chinese stuff.
> The MAX8655 is one example... But you need to solder them InfraRed or
> something similar, which should no problem for me... guess why!
You're superman in disguise and have that cool heat vision thingie!
You or your SO likes to use super-duper-mind-boggling-powerful hair driers!
You are an alien in disguise, and can use pyrokinesis!
The possibilities are endless ;-)
> The only thing which is currently not solved is the 24 Vin (18-30V) to
> 12Vout (and between 10 and 50A) which must be very stable and resistible
> agains voltage waveing from the input... If this problem is solved, it
I'd kinda expect to find ready-made DC-DC PSUs that do it for the telecomm
market, but they won't be cheap at all.
> Also I try to get an experience with UPS-Electronic and I am searchin
> for this highly intergated solutions, which mean, I wan to put an USB
A proper UPS needs a bit more. If it is a DC UPS, it is easier (no need for
IGBT rectifiers and inverters), but you still need to at least have a good
battery charge gauging system, and battery charging circuitry. Even for
lead-acid this is not trivial (if you want anything that will make the
battery last for five years instead of junking them within one year, that
is).
If you make it an UPS based on Li-ION batteries, look for the SBS
specification in google. It will help you a *lot* to get SBS ICs to make
the entire charger, battery pack, and battery-capable VR.
--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh
Reply to: