[EE] 0V...50V/40A low noise power supply

[Vasile Surducan]

Any relevant experience with high power/low noise (less than 50mV
ripple) power supplies here ?

thx,
Vasile
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Richard Prosser]

Vasile,

We design & manufacture Telecom supplies (48V up to ~3kW per unit,
with parallel operation possible well over the kA range. Noise
requirements are <100mV wideband, <2mV pshophometric. (sp?))

But I'm not specifically involved in that area. If you have a specific
question I could probably get an opinion however.

Richard P

[Steve Smith]

Yes what do you need to know

Steve

[Vasile Surducan]

There is one customer here interested to buy a few power supplies in
that specific range. One of the customer's vendor asked me if I'm not
interested in a small production for such devices (and I'm quite busy
now). Could you connect me please
with your marketing department for getting a quotation ?

thx,
Vasile

[Vasile Surducan]

I'm interested how is solved the weight problem for such supplies.
I guess one of the schematic choice is based on one or more switching
supplies connected in parallel followed by a low drop linear
regulator. Both the switching and the regulator should be
synchronised.
Probably using ferrite core and high frequency for switching is the
way for decreasing the weight.
I can't imagine 50/60Hz toroidal transformers in those supplies, or I wrong ?

thx,
Vasile

[Steve Smith]

Vasile
We make a module that's 3U high x 60 W and 250 D its 25 A 48v smps they are
designed for telecom application and are stackable up to 5 units wide
www.pbdesign.co.uk

regards Steve

[David VanHorn]

On 3/14/07, Vasile Surducan <picl...@gmail.com> wrote:
>
> I'm interested how is solved the weight problem for such supplies.
> I guess one of the schematic choice is based on one or more switching
> supplies connected in parallel followed by a low drop linear
> regulator. Both the switching and the regulator should be
> synchronised.

Spread spectrum modulation of the switchers?

I came up with that technique back in '91, tried to interest the boss in
patenting it, but he said "It's not what we do"... :(

That by itself dosen't really lower the noise, but it spreads out the energy
so that it's less of a problem at any specific frequency.

I had a fun incident about a year ago, with a switching supply at a VHF
repeater. It happened to have some significant energy at 600 kHz, which is
the split between transmit and receive at this repeater. When the VHF
transmit energy mixed with the switcher noise in diodes somewhere, we got
two noise products radiated, one was on the recieve frequency. Awful..

But linears aren't immune to oscillating, I've seen the 78XX series
oscillate at about a watt, near 2 MHz, due to failed bypass caps.

[Steve Smith]

Had a 78 go off like that yesterday somebody cunningly disguised the bypass
cap fault by fitting the wrong type...
S

-----Original Message-----
From: piclist...@mit.edu [mailto:piclist...@mit.edu] On Behalf Of
David VanHorn
Sent: 14 March 2007 06:20
To: Microcontroller discussion list - Public.

[Richard Prosser]

Vasile,
If you send an email to dc....@eaton.com someone should be able to
put you onto a distributor in your country.

Alternatively, try http://www.powerware.com and jump through the hoops
for "Power Solutions for..... Telecom. to see the products and
"How to Buy .... Find a Reseller" for a local agent.
I hadn't realised how difficult it was to locate this sort of info
until I tried!

Or if you let me know what part of the world you're in (or your
customer is in) then I'll ask our marketing guy for a local agent. We
have a second design centre in the UK (Bognor Regis) and a lot of the
European sales are handled through there. (But I can't locate their
website)

Unfortunately the DC side of the business is smaller than the AC side
so most of the resellers are more familiar with UPS's etc.

I take it the voltage requirement is either around 24V or around 48V
otherwise we will not have anything available.

[Steve Smith]

Richard:
We do 12,24,30,48,60,110,144 all in 60 wide 3U bricks 1300W /each 5 (max) in
a rack so if you get stuck we can do some strange voltages
Steve
www.pbdesign.co.uk

-----Original Message-----
From: piclist...@mit.edu [mailto:piclist...@mit.edu] On Behalf Of
Richard Prosser
Sent: 14 March 2007 19:18
To: Microcontroller discussion list - Public.
Subject: Re: [EE] 0V...50V/40A low noise power supply

[Richard Prosser]

On 15/03/07, Steve Smith <xy...@blueyonder.co.uk> wrote:
> Richard:
> We do 12,24,30,48,60,110,144 all in 60 wide 3U bricks 1300W /each 5 (max) in
> a rack so if you get stuck we can do some strange voltages
> Steve
> www.pbdesign.co.uk
>
>

Er, Thanks.
It was Vasile - or at least one of his customers that was enquiring. I
was offering assistance the same as you were - albeit with a more
limited voltage range it appears.

RP

[Vasile Surducan]

Thanks to all.
I've forwarded your emails to my customer.

Vasile