switching power supply start up

[zappedagain]

I have an application where I'm considering using 10 switching power supplies, each rated to 50W. I have various voltages that I need to supply, and it turns out I can stack 10 supplies in a smaller volume than multiple output power supplies require (about 2/3 the volume by first estimates).

While the total power dissipation is 500W, the in rush current of each switcher is spec'd at 37A (at 230VAC), so I'd have a 370A surge for one cycle of the mains. At 50Hz, that gives an A2t of over 2700, and that is more than a slow-blow 15A fuse can handle. At 115VAC I only need a 6A fuse, so these two demands are diverging rapidly.

I suspect that mains with standard 15A wiring wouldn't give me a 370A surge... would the in rush current then last multiple cycles? I suspect that would cause the brown out circuitry in other equipment on the same main to trigger.

Is this fairly typical and thus a reason not to use multiple switchers, or should I keep looking at other fuse manufacturers or maybe circuit breakers?

Thanks,

Z

 

[Skogsgurra]

Humbly suggest a simple 'soft start' circuit consisting oc a suitable resistor that limits your inrush to a value that the 15 A fuse can handle.

The resistor drops voltage for a few periods and when all capacitors have charged, the voltage drop is small enough to pull in a relay or small contactor thats shorts out the resistor to avoid extra losses.

Works well for large transformers. Haven't tried it for SMPS.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[zappedagain]

That is an interesting technique; maybe I need to find a switcher that already has that feature built in.

I found this on the Electronic Design site:

"Because of the negative resistance characteristic associated with the inputs of many switchers, insufficient soft-start time will result in high input currents, causing the input voltage source to be pulled down. With a lower input voltage, the switching regulator requires even more input current to correctly start up. This, in turn, will pull the input voltage lower still, resulting in a latch condition. When in a latch condition, the input current is relatively high, the input voltage is pulled low, and the switching regulator’s output never reaches its regulated value. Because of the charging of the output capacitor, this high startup current can exist even without a load on the switcher’s output.

Besides minimizing the high startup current associated with switching regulators, a soft-start function also can be used for power supply sequencing. In some situations where multiple regulated voltages are present, it may be necessary for one or more voltages to come up after the main power has come up and stabilized."

In the last paragraph I'm starting to see the benefit of the multiple output power supplies.

 

[Warpspeed]

There are two problems here.

First the initial charging of the large reservoir capacitor that runs at peak mains voltage. That can certainly be softened with a starting resistor, a negative temperature coefficient thermistor, or SCR phase control.

The switching supply itself also requires a soft start circuit to very gradually ramp up the output voltage at turn on. Together a system can be built that has a greatly diminished inrush current. If you really wanted to, the maximum full load input current need never be exceeded at initial power up.

 

[zappedagain]

I wonder if off-the-shelf (UL approved) solutions exist for this. I'll be going for UL approval on the system, so I'd rather avoid any custom mains circuitry as that bogs down the approval process.

I just noticed that the article I quoted is from 1998 (

http://electronicdesign.com/article/power/simple-soft-start-circuitry-provides-long-startup-.aspx);

I'd expect this problem to have been solved long ago. I suspect rapid start up time must be more of a market driver than in-rush current.

Z

 

[itsmoked]

The reason this isn't coming clean is because 10 switchers is not a good idea. You should re-think this. Do you really need 10 different voltages?

500W is easily within power factor control realm of availability. Once a supply has a PFC front-end it's a small step to inrush limiting.

If this is LED lighting there are better ways to go than 10 drivers.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MacGyverS2000]

There are sequencer chips, too... if the supplies offer enable pins, consider controlling them with an external chip.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[zappedagain]

I have five different voltages, at a variety of watts.

37A seems to be the default spec for in-rush current. I'm even seeing it on PFC power supplies. I wonder if that keeps the in-rush specified within one of the regulations so they power supply designers can treat it as a 'don't care' spec because they know they will have a lot of margin.

From the data sheets I've seen so far, it looks like the enable pins only enable the output, so I suspect they wouldn't help for in-rush current; although I agree that seems like a major reason to use the disable.

John D

 

[DHambley]

If 10 supplies are indeed the cheapest way to go, I like mcgyver2000's suggestion of some sort of sequencer. It's a new circuit design but maybe it could be simple: 10 triacs, one to the input of each power supply, all controlled by a simple CMOS discrete circuit?

 

[itsmoked]

Why not use one nice supply or the most loaded voltage and a handful of DC-DCs?

http://www.murata-ps.com/cgi-bin/cd4power.storefront/EN/Catalog/1016

http://www.xppower.com/page.php?pagename=DC-DC&lang=EN

http://www.acopian.com/con-details.html

And a favorite:

http://www.astrodyne.com/Product_HighDensityPowerSupplies.asp

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Warpspeed]

Have you tendered out a custom wound transformer with five suitable secondary voltages?

Five bridge rectifiers, and five filter caps come pretty cheap too.
The "Kiss" principal at it's finest.

 

[itsmoked]

"[red]of[/red] the most loaded voltage

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottyUK]

Check Vicor's modular range: a common front end converter with active PFC and inrush control and individual regulators fed of the main bus. Great for distributed supplies with very high currents at low voltage, or for multi-rail applications.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[itsmoked]

I can't believe you got that post in between mine and my correction Warp!! Me->

<-Warp

Scotty I thought of Vicor first but all their stuff seems to be 150W and up.(?)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MikeHalloran]

I think the Vicor modulars use a 48VDC bus. Maybe some of the smaller loads (not revealed) could just use resistors and Zeners.

I think I know where the 37A spec comes from. I built a show display that contained a 1HP fan, started directly across the 115/60 line. It pulled ~48A for a couple of seconds, and reliably tripped the facility's 15A breakers. Feeds with 20A breakers were available, but obscenely priced.
It was a dual voltage motor, so I wired it to start on just one winding for 6 seconds, by which time it was rotating fast enough to produce some back EMF. Then the second winding came in and the remainder of the start was normal. ... and the peak current was limited to ~38A, and the facility's 15A breakers didn't trip.

Yeah, I should have bought the fan with a 230/3 motor and driven it from an inverter. I know that _now_.


Mike Halloran
Pembroke Pines, FL, USA

 

[ScottyUK]

Vicor do have some smaller stuff there - a 600W front end with multiple smaller converters would be feasible from stock parts. Maybe worth a chat with their apps guys. The range is bigger than I remember it!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[DPolak]

Vicor - you may have problems with EMI....
Plus - if you connect few Vicors - they just explode (sometimes, not always) during on/off.
(I think because too much PHDing inside).

10 switchers - wow! Must be very hi-tek project.
I'll go for sequencing...
thx