48VDC FIELD POWER SUPPLIES BLOWS OFF

Status
Not open for further replies.
R

RNLENI

Member
Location
Michigan
Occupation
Engineer
I have 480VAC 3ph 60hz Power buses going out from my main cabinet to field. Each Buses are on different contactors and feeds certain number of 48VDC Power supplies. The 48VDC supply feed my conveyor rollers. Its noticed that when I operate the Estop , some times these 48VDC Power supplies blows off. I checked with manufacturer of the power supply and their preliminary observation is that this is due to overvoltage. But upon monitoring the 3 phase voltage and current in the panel , there is no voltage spikes noticed. Does anybody know what will be the possible reason the 48VDC power supplies are failing?
 
Was it overvoltage on the input or on the DC bus? Are these switching power supplies? Could it be back emf from the motors killing the powersupplies? Can you put a scope on the DC bus and capture the waveform when you initiate an E stop?
 
petersonra said:
What do you mean by "blows off"?

Does the estop remove power from the input of the power supplies?
Click to expand...
Yes. Each 480 vac power bus have few number of 48vdc power supplies . When Estop is pressed the entire 480 Bus is out of power as the the EStop opens the Power Bus contactor

The Power supply getting damaged. Components inside fails
 
Last edited:
Todd0x1 said:
Was it overvoltage on the input or on the DC bus? Are these switching power supplies? Could it be back emf from the motors killing the powersupplies? Can you put a scope on the DC bus and capture the waveform when you initiate an E stop?
Click to expand...

I put power analyzer at the main cabinet and didn't notice any voltage spikes at all. But manufacturer of power supply told its overvoltage.
They are switching power supply. the load side of the power supply is plenty of conveyor rollers. each roller only take 40 mA . Will there be a back emf that destroy power supply.? But I don't see it in any other similar systems

Yep I can put a scope and measure the DC Side voltage ,but I doubt is it a secondary side issues or 480v ac side issue
 
I am not a fan of shutting off power to the DC power supplies on a regular basis. Better to leave the power supplies on and let the contactors open the 48 V side.
 
RNLENI said:
I put power analyzer at the main cabinet and didn't notice any voltage spikes at all. But manufacturer of power supply told its overvoltage.
Click to expand...
I assume that the power analyzer was placed on the load side of the contactors and it's monitoring the bus feeding the power supplies?

It's possible that the sampling rate of the power analyzer in not fast enough to capture a very short voltage spike.
Perhaps there are inductors on the input of the power supplies, for example for filtering, and interrupting their current with the contactor creates a voltage spike which causes damage to internal semiconductors. A surge suppressor on the load side of the contactors might help if that's the problem.
Were any failed power supplies returned for failure analysis, or is the manufacturer just speculating that the damage was caused by overvoltage?
 
RNLENI said:
I put power analyzer at the main cabinet and didn't notice any voltage spikes at all. But manufacturer of power supply told its overvoltage.
They are switching power supply. the load side of the power supply is plenty of conveyor rollers. each roller only take 40 mA . Will there be a back emf that destroy power supply.? But I don't see it in any other similar systems

Yep I can put a scope and measure the DC Side voltage ,but I doubt is it a secondary side issues or 480v ac side issue
Click to expand...
Why do you think it isn't a problem with the DC side? Why not just measure measure it to remove the doubt - if indeed it is in doubt?
 
synchro said:
I assume that the power analyzer was placed on the load side of the contactors and it's monitoring the bus feeding the power supplies?

It's possible that the sampling rate of the power analyzer in not fast enough to capture a very short voltage spike.
Perhaps there are inductors on the input of the power supplies, for example for filtering, and interrupting their current with the contactor creates a voltage spike which causes damage to internal semiconductors. A surge suppressor on the load side of the contactors might help if that's the problem.
Were any failed power supplies returned for failure analysis, or is the manufacturer just speculating that the damage was caused by overvoltage?
Click to expand...

I was thinking the same surge protector. But we have many other sites wired same way but only this site has the problem. Not sure why. Do you have a recommendation for a surge protector we can add to a power bus
 
RNLENI said:
I was thinking the same surge protector. But we have many other sites wired same way but only this site has the problem. Not sure why. Do you have a recommendation for a surge protector we can add to a power bus
Click to expand...
If you are cutting the line side connection, lifting the 480 volts, then the conveyor motors will act like generators and try to back-feed your power supplies as they coast down. See petersonra's suggestion at post #6.
 
Since we don't know what in the PS is failing, dismissing a DC-side problem doesn't make sense. And just because the same PSs don't have problems elsewhere is not a good reason to avoid checking them.

My completely uninformed guess-
These particular supplies aren't happy with back EMF from the motors; maybe they're a very slightly different model from other installations or are missing a critical component (are they rated for driving motors? constant voltage or PWM drive?)

Easy band-aid fix- put a diode in series with each motor and see if the supplies stop failing. That would confirm the back-EMF issue.
 
zbang said:
Since we don't know what in the PS is failing, dismissing a DC-side problem doesn't make sense. And just because the same PSs don't have problems elsewhere is not a good reason to avoid checking them.

My completely uninformed guess-
These particular supplies aren't happy with back EMF from the motors; maybe they're a very slightly different model from other installations or are missing a critical component (are they rated for driving motors? constant voltage or PWM drive?)

Easy band-aid fix- put a diode in series with each motor and see if the supplies stop failing. That would confirm the back-EMF issue.
Click to expand...

Its all embedded 48VDC Motors inside the rollers. Are you able to suggest another way to check? each roller consumes ony 40 milli amps @ 48vdc Incase of back emf how can we block that and save the power supply?
 
Are there any other loads on the the 480V bus that feeds the power supplies? For example, solenoids or transformers would have inductance that could cause a voltage spike if their input current was interrupted suddenly.
 
RNLENI said:
I was thinking the same surge protector. But we have many other sites wired same way but only this site has the problem. Not sure why. Do you have a recommendation for a surge protector we can add to a power bus
Click to expand...
Put the scope on it. Why are you reluctant to try that?
 
RNLENI said:
In case of back emf how can we block that and save the power supply?
Click to expand...
The aforementioned diode, which for 40ma does not need to be large (a 1N4004 in series should do and they're almost as common as dirt). Don't know if that will "save" the PS, but it'll block a back-spike.

BTW, you can measure that back EMF by just putting a volt meter across a disconnected motor and spinning it (the motor, not the voltmeter :LOL:); that'll show you what happens when the load moves the motor.

Oh, and you haven't told us much about the PS itself and how many motors are on a single PS.
 
210608-1655 EDT

RNLENI:

You list yourself as an engineer, but you do not speak like one.

zbang told you to insert a diode in each motor. I would change that to one diode at the power supply output. Also put a suitable resistor load directly across the power supply output, and one after the diode. Use a diode with a controlled avalanche breakdown. Put one of these diodes in series with the output of the power supply, and another reverse biased across the output of the power supply on the load side of the first diode.

.
 
Status
Not open for further replies.
Top