Dell XPS 9000 vs SW5548 - Page 8

 

On Sat, 27 Feb 2010 00:58:28 GMT, Johnny B Good


Oh, I see.  It's not a generic thing with SmartUPS2000's; it's an  issue
with your particular unit.  Well, if you got it at a hamfest, you probably
got it for a pretty good price, I hope, so it's worthwhile to put up with
its idiosyncracies.

And I suspect you'll get things sorted out.
--ron

 

On Wed, 24 Feb 2010 20:42:32 -0600, "Daniel who wants to know"



Active PFC has everything to do with a device not working on an MSW
inverter.  In fact, that's one of only a couple of reasons why a
device will not work on MSW.  The other reason is that the device
relies too much on the peak value of the AC sinusoid.  Some microwave
ovens are offenders here.

Just off the top of my head I can name at least 10 active PFC
architectures and there are invariably those that I don't know about.
I can show you one that for sure will NOT work on anything but a low
distortion sine wave.  The reason that I know is that I designed it.

http://www.neon-john.net/Induction/PFC.htm

No, I didn't intentionally design it not to work with non-sinusoids.
The controller chip that I selected determined that.  I've since
switched techniques since my product needs to be able to run on
inverters and cheap generators.

-----------

To the original poster.  Does your power supply use a 3 wire cord?  If
it does then try connecting the power supply to the inverter without
the ground wire.  Use a 2 to 3 wire adapter, a 2 wire extension cord
or something similar.

I've run into Dell power bricks that won't run on an inverter and the
problem was with the inverter neutral and not the active PFC.
Specifically, the problem is in the inverter and its lack of an earth
ground.

Most inexpensive inverters have the neutral sitting about 60 volts
negative with respect to earth ground.  That makes for a much simpler
and cheaper output stage.  The problem is that the Dell power brick
doesn't like seeing any potential between ground and neutral.  In the
instances where this problem has arisen (either with my computer or
those of people whom I've helped over the net), removing the
connection to the ground pin has solved the problem.

I don't advocate cutting the ground pin on general principle. The
easiest solution is to use a 2 to 3 wire adapter plug. You can also go
into the inverter and remove the ground wire, as it has no meaning in
this context.

John

 



Thanks for that idea.  But a true sine wave UPS is being delivered today,
so I'm not going to experiment with grounding issues.  But it'll be
interesting to measure the neutral-ground voltage on the UPS I've been
using.  Although it is non-sine wave, it is not what I would term "cheap",
so I would expect it to be properly grounded, unless there is something
wrong with the house wiring.  And the UPS does have a sensor that is
supposed to detect this kind of fault.
--ron

 


Add a little RC.  Nothing less than 10 ohms and a 5 mf capacitor in
series should do the trick, applied across that power line and you're
good to go.

You might try as much as a 25 ohm and 10 mf, or even as little as 10
ohms and 1 mf.  Rounding off those SW5548 spikes should be sufficient.

Otherwise exclude the UPS all together because your PC will never know
the difference.  Don't even bother asking APC, because they don't seem
to know squat.

 ~ BG