2
\$\begingroup\$

I hope this thread is allowed. Based upon my previous topic, which turned out to be the exact opposite of what I needed to accomplish, I have a few questions.

I have a 240W 24V Supply: https://www.us.tdk-lambda.com/ftp/Specs/dpp120-240.pdf

I have a coil with a 750W turn on rating and a 5W hold in rating: http://www.kvc.com.my/EnterpriseChannel/SharedResources/Datasheet/0/?ProductId=1000066755&Filename=SCHNEIDER-LX4-FH024.pdf

The device will also be used for some sensitive equipment, so I want to be safe and not allow too much voltage fluctuation.

50ms (closing time) * 30A= 1.5C (Thanks @DaveTweed) To be safe, 1.5C/4V= 375,000uF

What voltage rating? I've read to derate 2-3x.. for a .4F 50V cap and above, I'm looking at $1-200 per capacitor. Am I playing it way too safe with these numbers, or is this really what I need?

Ok, so if my above power supply uses a forward folding method of current limiting, do I need to use a resistor/thermistor to ensure the capacitor itself doesn't overdraw my supply after it's emptied? I will probably have at minimum a minute of time for the cap to recharge.

As far as wiring, since there are other items in my circuit, I want the capacitor in parallel with my coil rather than directly off my supply, correct? If I were to need a thermistor/resistor would that be in series with the + side of the coil, prior to the capacitor?

\$\endgroup\$
  • \$\begingroup\$ You won't need to derate that much : 25V would be dangerously close but 35V would have enough margin. Alternatively, and for a good deal less money, consider trickle charging a couple of SLA (lead acid) batteries off 24V and powering the contactors from them. \$\endgroup\$ – Brian Drummond Aug 27 '15 at 10:34
  • \$\begingroup\$ So at first I was going to put the capacitor straight across the contactor, which I realize doesn't make any sense as the capacitor will not charge when the control relay is off, which is when it should be charging. i.imgur.com/SRv9YPb.png So I'd put the return side straight to ground so the contactor is released but the capacitor can charge. i.imgur.com/LQRPc7L.png Will this also eliminate the potential for reverse current induced by the contactor releasing, or can that still potentially be an issue? \$\endgroup\$ – C.J. C Aug 27 '15 at 21:11
1
\$\begingroup\$

This sounds expensive but it will work. If your 24Vdc supply is REGULATED which most are these days then your caps could be at a lower voltage rating . As long as the WORKING voltage rating exceeds the 24v psu then you are OK .I suggest 35volt caps which will save you money.The reason I said caps and not cap is that I suggest you use several in parallel.Paralell caps will cost even less and lower values are easier to find.Also the total ESR will be lower meaning that normal electros could be used saving more money.When you experiment with your caps yopu will find that you will get away with significantly less than the previously stated 375 millifarad.I think 375 millifarad is a sensible starting value.Once you have established your minimum total capacitance then upfactor it for aging temp ,tolerances.Doubling it would give peace of mind.If you want to know if its pulled in properly or not then on the fly inductance measurements work well because the inductance change is large and hence easy to detect reliably.

\$\endgroup\$
  • 1
    \$\begingroup\$ Thank you. Price is of no concern. Durability and stable voltage/current is most important. \$\endgroup\$ – C.J. C Aug 27 '15 at 1:02
0
\$\begingroup\$

Why not taking 1kW power supply? 50msec is serious time. And the capacitance you mention is serious capacitance, you can't just charge it by turning on the power supply, because you have to limit the current.

On the other hand, the difference of two orders of magnitude is strange. Are you sure it's 750W and 5W?

\$\endgroup\$
  • \$\begingroup\$ Ok, after looking at datasheet, i would argue you can reduce the capacitance by three times at least. \$\endgroup\$ – Gregory Kornblum Sep 26 '15 at 8:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.