What determines whether a switching supply will properly dri

  • Thread starter William Sommerwerck
  • Start date
On Mon, 21 Dec 2009 03:47:12 -0800, "William Sommerwerck"
<grizzledgeezer@comcast.net> wrote:

What makes you think I take anything you say seriously? Or that I'm bothered
that you so crudely disagree with anyone here, or am pleased when you do?

I couldn't care less. You are less than nothing, Mr. Allison.
If you "Couldn't care less" why'd you post this?
 
Rheilly Phoull Inscribed thus:

"William Sommerwerck" <grizzledgeezer@comcast.net> wrote in message
news:hgjnk8$fvk$1@news.eternal-september.org...
Just a wild guess .. a load resistor in parallel with the device
will fix the problem. It seems that some SMPS need a minimum
load to function as expected.

That's what I suspected. However, the energizer is not a "light" load
-- it
runs slightly warm. Whereas the CD player draws next to nothing, and
the supply has no problem with it.

Nevertheless, I might just stick a 200 ohm resistor across the power
jack, just to see what happens.


Perhaps a few hundred mF as well ??
I've seen a lot of laptop SM PSU simply cease to supply power to the
load because the output caps have developed high ESR. Adding a
suitable cap across the output often will confirm the problem.

--
Best Regards:
Baron.
 
On 20/12/2009 8:04 AM, William Sommerwerck wrote:
The interesting thing (finally!) is that most of the switching supplies
won't power the STAX energizer. The energizer's red "low voltage" light
comes on, the energizer tries to turn on (briefly flashing the green "good
voltage" light), but then drops back, and the cycle repeats, producing a
"thump-thump-thump" in the earphones.

Only one of the switching supplies I have will power the earphones -- and it
often has to struggle a bit to get going.

It's not an issue of voltage or current capacity, per se. What, exactly, is
going on?
My guess would be that the switcher is generating lots of hash, & it's
upsetting the oscillator in the phones. Try looking at the DC with a 'scope.

--
W
. | ,. w , "Some people are alive only because
\|/ \|/ it is illegal to kill them." Perna condita delenda est
---^----^---------------------------------------------------------------
 
On 21/12/2009 12:38 AM, William Sommerwerck wrote:
Some switching supplies do no do well in the presence
of a RF field. Does the STAX emit this type of interference
through the power lead?

The STAX energizer is an audio product, and it's unlikely any switching
circuits in it operate above 50kHz or so.
Not so. See if you can guess how it generates the HV it needs for the
electrostatic field.

--
W
. | ,. w , "Some people are alive only because
\|/ \|/ it is illegal to kill them." Perna condita delenda est
---^----^---------------------------------------------------------------
 
William Sommerwerck wrote:
I would find it interesting that a supply of 1000mA would go into
current shutdown. If you had a peak hold DMM or some other
way to plot inrush current draw, it would be helpful to know what
the device is actually drawing peak. If it is an over current situation,
maybe you can get a NTC resistor to cut down the peak surge
current. It might be a challenge to find one with that low of rating.

On the flip side, if its shutting down from not enough load, you need
to put a small light bulb or such in parallel with the device for some
additional start up load.

I think is the issue is "too much" load. The power supply operates a
Discman -- which draws 1/10 the current -- with no problem whatever. Ergo...

I don't have a peak-hold meter. However, I did measure the battery current
with a Fluke 87 -- it's around 180mA. But at the moment you connect the
battery, the digital bar-graph display goes WHAM!, all the way to the right
in a fraction of a second. I suspect the unit is drawing several times 180mA
at turn-on.

The AC adapter input is spec'd at 300mA -- presumably the maximum capacity
required, rather than the average drawn.

If I can find a matching plug/socket pair, I might experiment with small
series resistors (1 to 5 ohms) to see what happens. At the moment, it's not
a pressing issue.


You don't need a peak-hold meter.
You need a DC current probe and a dual-channel storage oscilloscope.
Conjecture is no substitute for measurement.
 
You don't need a peak-hold meter.
You need a DC current probe and a dual-channel storage oscilloscope.
Conjecture is no substitute for measurement.
Agreed, absolutely. But I don't have a dual-channel storage scope. I'd love
to have one of those fency new ones with the LCD (???) displays, but I can't
justify the cost, because I don't use it enough. My 25-year-old Philips is
gathering dust, as it is.
 
William Sommerwerck wrote:
You don't need a peak-hold meter.
You need a DC current probe and a dual-channel storage oscilloscope.
Conjecture is no substitute for measurement.

Agreed, absolutely. But I don't have a dual-channel storage scope. I'd love
to have one of those fency new ones with the LCD (???) displays, but I can't
justify the cost, because I don't use it enough. My 25-year-old Philips is
gathering dust, as it is.


Ok, then use a resistor as a current measurement and your scope in a
dark room.
You're gonna find it difficult to diagnose a problem if you can't be
bothered
to even attempt a look at the symptoms.
 
You don't need a peak-hold meter.
You need a DC current probe and a dual-channel storage oscilloscope.
Conjecture is no substitute for measurement.

Agreed, absolutely. But I don't have a dual-channel storage scope.
I'd love to have one of those fency new ones with the LCD (???) displays,
but I can't justify the cost, because I don't use it enough. My
25-year-old
Philips is gathering dust, as it is.

Ok, then use a resistor as a current measurement and your scope
in adark room.
You're gonna find it difficult to diagnose a problem if you can't be
bothered to even attempt a look at the symptoms.
That's a pretty rude thing to say.

The original posting was a question about why most samples of a particular
switching supply would not turn on when powering a device whose drain was
well below the supply's stated capacity. There was no "problem", as such --
merely an issue of curiosity.

The conclusion was that there was sufficient current inrush to force the
switching supply to briefly cut off. I've since found that if I simply leave
the supply on, it will eventually "come 'round", though it might take 10 or
15 seconds.
 
William Sommerwerck wrote:
You don't need a peak-hold meter.
You need a DC current probe and a dual-channel storage oscilloscope.
Conjecture is no substitute for measurement.

Agreed, absolutely. But I don't have a dual-channel storage scope.
I'd love to have one of those fency new ones with the LCD (???) displays,
but I can't justify the cost, because I don't use it enough. My
25-year-old
Philips is gathering dust, as it is.

Ok, then use a resistor as a current measurement and your scope
in adark room.
You're gonna find it difficult to diagnose a problem if you can't be
bothered to even attempt a look at the symptoms.

That's a pretty rude thing to say.

The original posting was a question about why most samples of a particular
switching supply would not turn on when powering a device whose drain was
well below the supply's stated capacity. There was no "problem", as such --
merely an issue of curiosity.

The conclusion was that there was sufficient current inrush to force the
switching supply to briefly cut off. I've since found that if I simply leave
the supply on, it will eventually "come 'round", though it might take 10 or
15 seconds.



Conclusions with no measurement are called conjecture.
My pet peeve with newsgroups is the massive amount of baseless
input that's presented as FACT.

Problem is that people asking the question probably can't sort
the good info from bad advice. If they could, they wouldn't have
needed to ask the question.

In a perfect world, all electronic designs would be "robust"
and every bit of advice would be good advice.
Sadly, it ain't so.

When you substitute part of a system without knowing
the issues and experience symptoms you didn't expect,
it's time to worry whether it's a good idea.

Conventional wisdom suggests you should beef up the current limit
so it doesn't cycle.
Be prepared for the possibility that the "correct supply" has
soft-limit features
that control the startup current and that substituting with one that
doesn't have a soft limit may prove very stressful to the system.
You might want to implement a softer limit rather than a harder one.

You have a scope...plug it in and find out what's happening.
A two-minute experiment is way better than all the baseless conjecture
you're gonna get here.
 

Welcome to EDABoard.com

Sponsor

Back
Top