Suppressing transients from a pump

[Sparweb]

It seems this pump is causing voltage surges and sags when it turns on and off. It's a 100 Watt pump in a small water circulation unit. There is a bank of LED lights on the same feeder bus (28VDC) which flicker when the pump cycles on and off. After the flickering, the light intensity is normal. We have eliminated possible bad grounds or voltage drops on the power supply.

I didn't design the pump or install it, so I have to take it "as-is", while troubleshooting from the other side of the world. This schematic (attached) from inside the pump shows a FET built-in with the motor. The power is supplied at pin 2, the ground is on pin 9, and the installer connected pin 6 to ground so that the voltage divider R1/R2 would put half the supply voltage on the MOSFET gate. This is a P-channel MOSFET, so I thought the way to turn it ON is to ground the gate with NO resistance. Maybe since there is a positive Vsg this is okay?

I'm also worried about the lack of built-in coil suppression. Not even a diode across the motor terminals. There is no way to open the motor to access the hidden connections. I believe this is the cause of the voltage spikes. How they get out into the rest of the circuits though...

Wouldn't the FET block the spikes when it turns off? Is a spike from the motor fast enough to escape before the FET stops conducting? Is this circuit holding the FET open longer than normal?
Currently the gate is switched floating when "off", but through R2 it gets +28V like a pull-up resistor. Is this good for the P-channel FET?

There is also a possible switch "bounce" in the pressure switch that turns the pump motor on and off. Should I investigate this? (very difficult since the customer has to test for me.)


STF

 

[Skogsgurra]

I don't think that you need to worry much. You would have the same flicker also if the pump were controlled with a simple on/off switch.

If the flicker is unbearable, put a filter in the LED supply instead of worrying about the pump.

Gunnar Englund

http://www.gke.org

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

 

[Sparweb]

Thanks Gunnar,
I have MOVs selected for the lights, but still worried about the motor in case they don't do the trick.


STF

 

[xnuke]

Are you worried about the flickering LEDs or the back emf damaging the LEDs? If it's the flicker, use capacitors to filter the LED supply (as Gunnar recommended), and possibly add a voltage regulator. If I remember correctly, the MOVs only prevent damage from back emf spikes, they don't filter or regulate voltage.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[IRstuff]

If the flicker is during power-on, the power-on transients are probably from the stall current as the motor starts to move. Diodes don't help this condition.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[Sparweb]

The lights seem to flicker at the moment the pump turns off. Information is 3rd-hand, but for now it's the turn-Off theory. EMF spikes.
Worried about both LED damage and the customer's aesthetics. Also worried that the pump might not be operating correctly. The pump has been installed in other systems where this effect was not seen. Those systems have a "black box" that prevents me from seeing what emf suppression measures, if any, were taken.
Capacitors in addition to MOV's? This is getting more and more like an AC power supply filter, but I see your point.

Thanks again everyone


STF

 

[Skogsgurra]

There are zillions of little motors run off 12 and 24 V DC with minmum (no) problems. Inductive kick-back is close to nil at these low voltages. Also, you are not switching the excitation coil, which could produce som overvoltage. Just relax. No probs - I can assure you.

If you were running a highly inductive load at hundreds or thousands of volts and using vacuum bottle switches - then there would be some reason to be concerned. But you are far from that situation.

Gunnar Englund

http://www.gke.org

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