"Auto" Input Power Supply

[ckramer]

We use a couple different 24VDC power supplies to drive components in testing instrumentation we manufacture. Recently we've been told that using a supply with "Automatic" voltage inputs (120V/230V) puts our equipment at risk when used on a 230V circuit and one leg drops, causing an "under voltage" situation, blowing up critical components because the supply cannot switch down to 120V fast enough. We have never experienced this in the 7yrs I've been on the project and never had any of our international customers report this problem.

We work mainly with 24V/20A DC switching supplies, and some smaller 24V/1.3A, mostly Omron and Sola.

A few of our instruments are used in countries with less than perfect infrastructure (Indonesia) that experience frequent brown/black-outs, not one has lost a PS.

I am aware that power supply manufactures do make 120V or 240V input power supplies (seperate spec's), but I and cannot find any information pertaining to the above scenario.

Any help is appreciated.

Kramer

 

[itsmoked]

That's nonsense! Don't worry about it.

The supplies don't control their output based on the input. So no matter what happens on the input they don't 'over volt' the output.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

Someone is seriously misinformed.

There are lots of auto-ranging power supplies on the market now. All that really means is that they have an input rectifier that takes any incoming AC source down to DC and then a chopper that provides a constant DC output voltage regardless of the rectified input voltage (as long as it is higher of course). The only reason there is a limit on the voltage range has to do with the current ratings of the components on the rectifier; if you design for 5A at 240V and the line drops to 120V, the input current doubles for the same output current.

But in the scenario you described, losing one leg of a 240V input, your output would be nothing because you would have an incomplete circuit on the input. Now if you GROUNDED one side of a 240V circuit (in the US by the way, because it would not work this way overseas), the input voltage would indeed drop to 120V. But doing so would also likely trip the 240V circuit breaker. If you had fuses and the ground fault caused only one fuse to blow in the 240V grounded leg, you may still end up with a completed 120V input to the PS.

So if we get past all the "what ifs" and somehow manage to instantaneously drop the input voltage to 1/2, there is a remote possibility that the PS might not react fast enough and allow a brief moment of low voltage output. But damage to the downstream devices? Not likely. Long before the current was able to spike high enough to cause damage to downstream devices, it would trigger the crowbar circuit of the PS and shut it down.

 

[ckramer]

Looks like I was seriously misinformed!

The original concern came from our parent company, of course they failed to mention the supply was a custom linear design - developed by their EE. This design had the switching set to 190V, so in a "brown out" situation the PS would switch to the 120V circuit and blow a fuse. By the time this info trickled down to me it was turned into a "serious concern" about the way we are supplying power to our current equipment!

I appreciate all of the help, maybe next time I'll wait to get the full story!

Kramer

 

[itsmoked]

Thanks for the report.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[rtronics]

Showing my age here, but, I have seen several instances of the older style 120/240 auto-switching power supplies blowing up during indeterminate mains voltage variations.
These are exclusively the type which senses the incoming voltage, then voltage doubles if necessary to get the DC bus of 330 VDC. So, voltage double the 120AC=>330VDC, straight through 240VAC=>330VDC, oops, voltage double 240VAC, give way too much volts, smokes.
Newer designs, just bridge rectify, the DC whatever it is, is converted to required output volts. But the long term reliability hasn't impressed me so far, I've seen lifetimes of 1 to 3 years 24/7 on these.

Ray