Searching \ for '[EE] Power supply' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: piclist.com/techref/power/actodc.htm?key=power
Search entire site for: 'Power supply'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Power supply'
2005\08\17@104508 by Mauricio Jancic

flavicon
face
Hi,

           I have a system that has to use two voltages, 5V and 12V. The 5V
need to supply 100mA max, and the 12V have peaks of 2A and 0.75A continuos
current. As you might have already guess the 12V are supplying a DC motor.



           I currently have a full bride and a small cap + a 7805 for the
5V section, and, coming from another winding on the transformer, another
full bridge + a 4700 cap. The transformer winding is 12VAC for the motor,
so, once rectified, the motor gets around 16~17 Volts, which was fine
because the motor speed increase and that was a good thing.



           Last detail: The motor works usually 4 times a day for 10
seconds max each day.



           Now, I need to simplify the design. I need to eliminate the
power supply board and I don't have much space on the board that the pic is
in. Everything is SMD I already managed to include the PIC supply on that
board but I having trouble redefining the DC motor supply for lower cost and
lower component count.



           At this moment, the speed of the motor is not a mayor concern as
it is the torque. I made a simple 1 diode rectifier, with no cap, thinking
that the motor doesn't really need a pure continuos current, but I lost a
lot of torque. I can increase the transformer output voltage, but I'm not
quite sure it's the right move. What do you think it would be the most cheap
but reliable approach to make the DC motor work?



           Kind regards,



Mauricio Jancic

Janso Desarrollos - Microchip Consultants Program Member

spam_OUTinfoTakeThisOuTspamjanso.com.ar

http://www.janso.com.ar

(54) 11 - 4542 - 3519



2005\08\17@122903 by Bob Blick

face picon face

>             At this moment, the speed of the motor is not a mayor concern
> as
> it is the torque. I made a simple 1 diode rectifier, with no cap, thinking
> that the motor doesn't really need a pure continuos current, but I lost a
> lot of torque.

Hi Mauricio,

The transformer loses efficiency when you only use a half-wave rectifier,
because the windings take on a DC current. The amount of current available
is probably only 15%. And since current=torque, that is a problem. You
will need to use 4 diodes or 2 diodes with a center-tapped transformer.

Cheerful regards,

Bob



'[EE] power supply'
2012\10\08@233620 by PICdude
flavicon
face
Added a subject ^^^

Couldn't you modify the power-supply by adding whatever protection  circuit to a small thin, long-if-necessary PCB in the middle of the  power cord, and heat-shrink it over?  I've done something similar  (different purpose) and used 1.5" heat shrink over it.  I found 4:1  shrink at Digikey, which brought the ends pretty nicely closed.

Cheers,
-Neil.


Quoting David VanHorn <.....microbrixKILLspamspam@spam@gmail.com>:

> All good suggestions, IF I could modify the existing circuits.
>
> What I need, is a power supply brick, to replace the existing power
> supply brick, that won't keep hammering a failed circuit.
>

2012\10\08@234802 by David VanHorn

picon face
Whatever I do has to be something that the end user can apply, and
that isn't too visually different.
I'm pretty much constrained to replacing the power supplies with
something that just shuts off till the input is removed

2012\10\09@002752 by PICdude

flavicon
face
Oh -- I thought you were shopping for power supplies, so I was  suggesting getting something COTS, you make the mods, and provide  those to your customers as ready-to-use units.

Alternatively, how about building your protection/shutoff circuit on a  small PCB with male and female connectors.  If you are still acquiring  PS'es, you can choose something with a different connector than what's  on the device to be powered and match connectors on your protection  PCB, so that they are forced to use it.

Now, I'm not sure how much is "too visually different".

Cheers,
-Neil



Quoting David VanHorn <microbrixspamKILLspamgmail.com>:

> Whatever I do has to be something that the end user can apply, and
> that isn't too visually different.
> I'm pretty much constrained to replacing the power supplies with
> something that just shuts off till the input is removed.
>

2012\10\09@004831 by David VanHorn

picon face
On Mon, Oct 8, 2012 at 10:27 PM, PICdude <.....picdude3KILLspamspam.....narwani.org> wrote:
> Oh -- I thought you were shopping for power supplies, so I was
> suggesting getting something COTS, you make the mods, and provide
> those to your customers as ready-to-use units.

These supplies are typically sealed, and modifying them would
invalidate all the safety certifications that I need.

> Alternatively, how about building your protection/shutoff circuit on a
> small PCB with male and female connectors.  If you are still acquiring
> PS'es, you can choose something with a different connector than what's
> on the device to be powered and match connectors on your protection
> PCB, so that they are forced to use it.

Nope.. These are delivered systems, and I'd have to add another box to
the chain, many reasons.

It's a little frustrating coming here for help, in that many people
try to second-guess the requirements and suggest solutions that might
well be viable in some applications, but not mine.  I've explored the
solution space, and I know what will and won't fly.

So, I'm looking for a simple universal input SMPS with 24V output at
2.75A to 3.5A, that will sense an overload, and shut down till the
plug is pulled.  I know that I'll need to work out connectors etc, but
the point of this exercise is to find someone who makes supplies that
behave like this, instead of the typical "Burp, Hiccup, or
Auto-Restart" modes

2012\10\09@080259 by Sean Breheny

face picon face
Hi Dave,

I don't know if what you are looking for exists as an off-the-shelf
product, but I have found that both MeanWell and XP Power (both makers
of power supply bricks) are willing to discuss custom versions. I
don't know how many you are planning on buying and what kind of NRE
costs you are willing to pay but I think that they might be willing if
you were going to buy at least a few hundred units and pay perhaps $2k
up-front cost.

Sean


On Tue, Oct 9, 2012 at 12:48 AM, David VanHorn <EraseMEmicrobrixspam_OUTspamTakeThisOuTgmail.com> wrote:
> So, I'm looking for a simple universal input SMPS with 24V output at
> 2.75A to 3.5A, that will sense an overload, and shut down till the
> plug is pulled.  I know that I'll need to work out connectors etc, but
> the point of this exercise is to find someone who makes supplies that
> behave like this, instead of the typical "Burp, Hiccup, or
> Auto-Restart" modes.
>

2012\10\09@083614 by alan.b.pearce

face picon face
> It's a little frustrating coming here for help, in that many people try to second-
> guess the requirements and suggest solutions that might well be viable in some
> applications, but not mine.  I've explored the solution space, and I know what will
> and won't fly.
>
> So, I'm looking for a simple universal input SMPS with 24V output at 2.75A to 3.5A,
> that will sense an overload, and shut down till the plug is pulled.  I know that
> I'll need to work out connectors etc, but the point of this exercise is to find
> someone who makes supplies that behave like this, instead of the typical "Burp,
> Hiccup, or Auto-Restart" modes.

I can't understand why the 'burp, hiccup' ones are a problem. I could understand if the PS went into constant current limit, but my impression is that the 'burp, hiccup' ends up supplying pulses of current, and not continuous current.

Unfortunately the foldback function you are after is a relic of liner power supply days. I have never come across a switch mode supply that does it, simply because they need to 'burp' to get any volts on the output side. -- Scanned by iCritical.

2012\10\09@101403 by David VanHorn

picon face
> I can't understand why the 'burp, hiccup' ones are a problem. I could understand if the PS went into constant current limit, but my impression is that the 'burp, hiccup' ends up supplying pulses of current, and not continuous current.


Read the article again.  The "burping" supply can deliver pulses of up
to 24W, with an average power of 1W, sufficient to sustain ignition.
I've been able to cause ignition of several PCBs (UL 94V-0 FR-4) with
one of these supplies, starting with an SO-8 fet that had a floating
gate, which might occur if you had a bad solder joint on a pull-off
resistor

2012\10\09@102349 by David VanHorn

picon face
> I don't know if what you are looking for exists as an off-the-shelf
> product, but I have found that both MeanWell and XP Power (both makers
> of power supply bricks) are willing to discuss custom versions. I
> don't know how many you are planning on buying and what kind of NRE
> costs you are willing to pay but I think that they might be willing if
> you were going to buy at least a few hundred units and pay perhaps $2k
> up-front cost.

The NRE isn't a problem, it's the UL/CE/VDE... certs.
If they don't already make something like this, then it will likely be
very expensive.

The problem is real, once I understood it, I found it pretty easy to
cause PCB ignition using a 2.7A rated supply.

Fuses don't help, there's not enough current over time to pop a fuse
that will also conduct rated current indefinitely

2012\10\09@135057 by Sean Breheny

face picon face
I just wanted to let you know that I have worked with XP Power before
on a custom design and at least the quote wasn't bad. We ended up
going with a different solution because our requirements changed, but
I think it might work in your case. I think it is likely that they
would not have to re-do all of the certification work, only the parts
which were affected by the design change. It might even be as simple
as changing the component values or tieing pins differently on the
switcher IC so that it has a different current-limit behavior. Even if
they have to completely re-do UL cert, I seem to recall that cost
about $15k for a typical case so it isn't ridiculous.

Sean


On Tue, Oct 9, 2012 at 10:23 AM, David VanHorn <microbrixspamspam_OUTgmail.com> wrote:
{Quote hidden}

>

More... (looser matching)
- Last day of these posts
- In 2012 , 2013 only
- Today
- New search...