supression of back EMF from ups

[ultrasoundguy2]

Ive need to isolate wall voltage (powered from small generator) from a EMF generated by a UPS.
It is a large UPS, it is powering a power supply rated for 1560 of which is supplying about 300V and 0.5A to an instrument.
From what I understand, most commercial UPS have a back EMF which is normally insignificant, but however, can damage our source.

Varistors and flyblack diodes are ways of suppressing emfs but this is generally done in a circuit.
What would be the best way of doing this in a commercial product that runs off 120V AC.
Any ideas or thoughts on this?

 

[ultrasoundguy2]

commercial type varistor that is plug and play?

 

[ultrasoundguy2]

what about line filter?
those are more for frequency based filtering though?

 

[Skogsgurra]

EMF? Do you mean EMI?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ultrasoundguy2]

no, back EMF

 

[Skogsgurra]

Like Electro-magnetic Force?

I have lived with power systems and inverters for decades. Mitigating all sorts of problems associated therewith. But this is the first time I hear about this.

Reflected noise, harmonics, Electro-magnetic Interference and sometimes PWM residues that contaminates the grid - yes. But EMF - never.

I am genuinley interested to hear what kind of problem you are experiencing with this EMF.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ultrasoundguy2]

from what has been reported to me, when the powered on / plugged in, it can create a back emf that can create a voltage transient.
I didn't believe it much either but figured somewhere here might.
I have seen inrush currents on the order of 30amps though that trip breakers.
but this is a unique niche that Im in right now

 

[Skogsgurra]

Inrush current does not create any back-EMF.

Switching on a UPS can, depending on how the mains rectifier is configured, create a current surge when the DC link capacitors are charged. That usually creates a short dip in mains voltage and sometimes trips overcurrent devices. But it would be a very special case if it also creates anything that could be characterized as a transient back-EMF. Could you show us a simple sketch showing the arrangement? With L, N and GND plus UPS and the device that gets destroyed/disturbed.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ultrasoundguy2]

I believe its when its turn off that I think you get the back emf rather than on?
when you turn it on it creates a dip in the mains though (opposite direction)

The information is not something I have no direct experience with.
From what Ive been told, all commercially available (and wired) UPS fail under the test given for this application.

Im sorry for not presenting schematics and more information, I need to be vague intentionally.

With in inrush currents, there are inrush current limiters

http://en.wikipedia.org/wiki/Inrush_current_limiter

But there is no equivalent for the output?

 

[Skogsgurra]

I understand that you are having problems with "a commercial product that runs off 120V AC" and that the UPS is also connected to that 120 V AC grid. That, I hope, is not a secret that you cannot disclose.

Further, I understand that the output of the UPS (300 V 0.5 A) is not connected to the "commercial product that runs off 120V AC".

I also think that I have understood that you have observed inrush when switching the UPS on and that the inrush current is around 30 A and trips the supplying breaker.

If all that is correct, and if the UPS is a standard static UPS with a standard DC link, which, with reference to the modest size, seems plausible, I cannot think of any other failure mechanism than an inductive kick-back caused by a highly inductive supply when the breaker trips the 30 A.

Most grids are not highly inductive. But, if there is an upstream transformer involved, it may be the explanation. In that case any standard MOV will work very well. Remember to select it large enough so it can handle the surge energies.

But better, still, is to reduce the inrush.

If you are sure that the problem is at turn off, the MOV alone should work quite well.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[stevenal]

From IEEE 100:

EMF: See: electromotive force.
electromotive force: See: voltage

 

[Skogsgurra]

Close to the definition of Recursive:

Recursive, see Recursive.

I really hope that the OP is beyond that stage. On the other hand, maybe not.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ultrasoundguy2]

thanks skoggurra

 

[PersonalProfile]

I cannot assist with your problem, though for the doubters regarding UPSs "backfeeding" electrical potential:

http://mines.industry.qld.gov.au/as..._264_hazard_of_backfeed_on_some_ups_units.pdf

Regards,
Lyle

 

[Skogsgurra]

Lyle

I seem to be the only person that have questioned the use of "back-EMF", so I think that I shall comment on your link to the Queensland mining safety issues.

What I mean is: (see 6 Aug 14 18:16) "Reflected noise, harmonics, Electro-magnetic Interference and sometimes PWM residues that contaminates the grid"

And that is exactly what Gavin Taylor, Chief Inspector of Mines and James Catt, Inspector of Mines are talking about. There may also be some back-feed through filters that are not properly connected to PE/GND and which back-feed into the Neutral in case N is opened, which is - I think - common in some mine installations.

There is a distinct difference between EMF and EMI. At least in my part of the World. So, it is good to know if the OP really means EMF and not EMI. EMF would demand quite other measures than does EMI.





Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Compositepro]

The original question does not make amy sense. But the reason for it appears to be:
"From what Ive been told, all commercially available (and wired) UPS fail under the test given for this application."
What is this test?

 

[pckaakee]

We use to instal MOV`s on all relay coils and power supplys but now use transorbs as they seem to take more abuse and are more reliable. I always considered a MOV a sacrificial device, to be replaced occasionally (If i remembered to do it and had spares in the bin}