[iyonix-support] Re: Iyonix hanging (was Re: EtherK not starting up reliably)
- From: Steve Fryatt <lists@xxxxxxxxxxxxxxxxxx>
- To: iyonix-support@xxxxxxxxxxxxx
- Date: Mon, 28 Aug 2006 11:25:58 +0100
On 24 Aug, Dave Higton wrote in message
<19018d5b4e.davehigton@xxxxxxxxxxxxx>:
> In message <2679675b4e.steve@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
> Steve Fryatt <lists@xxxxxxxxxxxxxxxxxx> wrote:
>
> > Even allowing for the calibration (or otherwise) of people's DVMs, the
> > kind of voltage tolerances being thrown about on this list are utterly
> > meaningless in the context of ATX power supplies.
> >
> > Unless the standard spec has changed significantly since I stopped
> > designing the things three years ago, the 5V rail is usually given as
> > 4.75V to 5.25V. The nominal often sits a little high of centre (say
> > about 5.1V), due to the realities of transformer design, but even this
> > isn't guaranteed.
> >
> > Exactly where any given PSU will operate under the 'standard' Iyonix
> > load is impossible to say, but it could fall anywhere within 0.25V
> > either way of 5V and still be within spec. The exact voltage at any
> > time will depend on the loadings of most of the PSUs other outputs
> > when the measurement is taken (and hence on hard disc activity,
> > whether the CD is in use for reading or burning, etc), not to mention
> > component tolerances in the PSU.
>
> Don't forget that the requirement is that the voltage AT THE PINS OF THE
> CHIPS is within tolerance.
Correct, but the ATX spec is for the voltage on the motherboard at the
multi-way ATX header (for a new connector). For test, this is usually
measured using sense wires soldered on to the connector pins, so that
voltage drops in the interface card and wiring to the loads isn't taken
into account.
The voltage reaching the chips is a problem for the motherboard designers
to worry about. All the ATX spec guarantees (unless the PSU manufacturer
has given a better than standard spec) is that the voltage hitting the
motherboard will be between 4.75V and 5.25V, AS LONG AS THE LOADS ON ALL
THE PSU's OUTPUTS REMAIN WITHIN A VALID AREA OF THE LOAD CHART.
It's the last bit that is likely to be an issue on the Iyonix. From what
has been said on its power consumption, the PSU is always likely to be
operating somewhere on the edge of the valid load area. This will be a
bigger issue if the supplied PSU is replaced by a higher wattage one.
> The average voltage being launched from the PSU is higher than the
> voltage arriving at the chips because of various resistive losses.
> Then you need to allow for noise, which is "randomly" varying in
> magnitude and sign. To keep the voltage at the chip pins within
> tolerance, the average (i.e. DC) voltage from the PSU needs to be (a)
> within a closer tolerance (because of noise), and (b) slightly high on
> average (because of unavoidable resistive losses).
All of which is irrelevant if we're quoting "5.06V on a spare connector",
as seems to be the standard test around here. That's why I think the test
is totally meaningless. It may always be true on an unexpanded Iyonix,
but I would expect it to change when a second hard disc or similar is
fitted (I suppose I could check that, now I've got a second disc fitted;
if I get chance, I'll try to remember when I've got the current deadlines
out of the way and I can 'afford' to break the machine).
--
Steve Fryatt - Leeds, England
http://www.stevefryatt.org.uk/
---
To alter your preferences or leave the group,
visit //www.freelists.org/list/iyonix-support
Other info via //www.freelists.org/webpage/iyonix-support
Other related posts:
