1
\$\begingroup\$

I have a motor that draws a max continuous current of 60A, but has a peak in-rush draw over 120A for a few milliseconds. I want to power the motor using a DC-DC converter.

I purchased a converter that has a peak output of 60A and tested it with the motor. It worked, but there was a long wind-up before the motor achieved its required torque. Turns out the converter has a built-in mechanism to prevent in-rush current so we were not getting the instantaneous torque we wanted.

I was thinking of purchasing a supercapacitor bank and connecting it in parallel to the converter's output. My understanding is that the converter will slowly charge up the caps initially, and as the motors start-up, it will draw the in-rush current from the capacitor bank. Once the motor reaches its steady-state current draw, the DC-DC converter will be able to supply that and top-up the capacitors.

I am asking to see if my thought process is correct or is there something I am missing or have to be careful when designing/testing. With regards to the details, I will make sure that the capacitors and DC-DC converter are spec'ed properly to be able to handle the voltage and current draw.

Any advice will be appreciated!

\$\endgroup\$
5
  • 1
    \$\begingroup\$ Sounds like you need to charge the capacitors and then connect the motor (or command its controller to start). This is reasonable. \$\endgroup\$
    – user16324
    Apr 8, 2021 at 11:28
  • 1
    \$\begingroup\$ Sounds like an XY problem. Can't you just limit the current? \$\endgroup\$
    – winny
    Apr 8, 2021 at 12:14
  • \$\begingroup\$ I have used 5 Maxwell 250F x 2.5V to start a small car, several times. I'm not an official EE yet, but I think what you want to do will work well and provide the startup speed and torque you want from your motor, since my supercaps easily met the cold-cranking amps of the starter in my car, I'm guessing 150A with the D-cell sized supercaps used. You may want to keep them charged because charging time can be significant. I have left mine charged for months, and they keep their charge reasonably. \$\endgroup\$ Apr 8, 2021 at 16:56
  • \$\begingroup\$ More specifics would really help... What voltage is your motor taking? The 350F supercaps I used are $11 each... Is cost an issue? Are there any other challenges, like heat? The 350-400 Farad variety just might be good for your case (here's one, but they are only 2.5v to 3.0v max each, so knowing the voltage need of your motor is very important to properly-answering your question. Probably need balancing, just as for a Lithium-ion battery pack. Help us help you with the complete picture! \$\endgroup\$ Apr 9, 2021 at 15:10
  • 1
    \$\begingroup\$ @MicroservicesOnDDD I plan on a Mean Well SD-1000L-12 for my converter and an Eaton XVM-16R2656-R for my capacitor bank. My motors are 12V PMDC. \$\endgroup\$
    – adonis
    Apr 12, 2021 at 14:03

2 Answers 2

1
\$\begingroup\$

Using a supercapacitor is reasonable. Just make sure your power supply doesn’t complain about:

  • charging current during the initial start-up
  • back-powering when PSU is off

Why the concern? The capacitors will discharge through the power supply when primary power is removed. Diode isolation can prevent this.

As far as charging current, a large supercap bank will demand much more current than the PSU can deliver, tripping the PSU overcurrent protect mode until the charging can finish.

In short, the cap bank needs to be treated like a battery, with current-limited charge control and back-power blocking to protect the PSU.

\$\endgroup\$
8
  • \$\begingroup\$ I don't see how additional output capacitors could back-power a PSU. The output capacitors are charged up by the PSU and discharge through their bleeder resistors. Also, the PSU wont complain about the initial charging current. \$\endgroup\$ Apr 8, 2021 at 8:34
  • \$\begingroup\$ @PrathikPrashanth "the PSU wont complain about the initial charging current" I like your confidence. Especially since it is completely misplaced. Do you realize that capacitor inrush current is much more than that of a motor? \$\endgroup\$
    – Maple
    Apr 8, 2021 at 16:10
  • \$\begingroup\$ @hacktastical I think adding any capacitor in parallel to the load with a goal to limit inrush current is the opposite of reasonable. See BrianDrummond's comment. \$\endgroup\$
    – Maple
    Apr 8, 2021 at 16:31
  • \$\begingroup\$ @Maple It depends how the startup is sequenced. Yes, from a dead cold start, layering the capacitor charging with motor spin-up makes a bad situation worse. What should happen is, PSU on -> controlled-rate cap charge -> motor on. Then the caps provide the peak current in addition to the supply. \$\endgroup\$ Apr 8, 2021 at 16:44
  • \$\begingroup\$ Yes, but that is not what OP has in mind. Read carefully - the idea was to add supercap and then power everything together. Somehow the converter is supposed to "slowly charge up the caps", while "motor gains momentum". And only when that momentum happens the motor "will draw the in-rush current". That demonstrates total misunderstanding of what really happens, that is why I am worried that encouraging one part without explaining the error of the basic premise will not do much good. \$\endgroup\$
    – Maple
    Apr 8, 2021 at 18:19
1
\$\begingroup\$

I was thinking of purchasing a supercapacitor bank and connecting it in parallel to the converter's output. My understanding is that the converter will slowly charge up the caps initially, and once the motors gain momentum, it will draw the in-rush current from the capacitor bank.

Once the motor starts turning, a back-EMF is generated in the motor due to which an inrush current does not flow into the motor. The reason why, initially, an inrush current flows into your motor during startup is because it takes some time for the motor to spool up and develop a back-EMF before it draws a steady state current.

So adding super capacitors at the output and starting the motor as the DC DC converter turns on is of no use, as the DC DC converter will not charge up the super caps instantly and will go into current limit mode as it did with just the motor.

But what you can do is connect a super cap to the output of the DC DC converter and then connect the motor to the DC DC converter + supercap bank through a mechanical relay or a solid state relay. Then, turn on the DC DC converter and let its output voltage rise up. Then, close the relay. The supercaps, which are charged up to some voltage already, will take care of the motor's inrush current requirements.

\$\endgroup\$
6
  • \$\begingroup\$ Just FYI, most of the DC-DC converters have maximum output capacitance specified right in the datasheet, and it is nowhere even close to the supercaps. \$\endgroup\$
    – Maple
    Apr 8, 2021 at 16:26
  • \$\begingroup\$ Yes, constant current charging of the supercaps will be necessary. It might be more challenging when you electrically attach the fully-charged supercaps to the stationary motor. Probably needs many big FETs in parallel, like a spot-welder. I would love to see how many amps your solution ultimately ends up handling. \$\endgroup\$ Apr 8, 2021 at 17:08
  • \$\begingroup\$ @MicroservicesOnDDD the solution will handle as many amps as the supercaps and FETs / relay allows. It entirely depends on the choice of components. \$\endgroup\$ Apr 9, 2021 at 8:40
  • \$\begingroup\$ This 2.7 volt, 3,000 Farad supercap has an ESR of 0.26mOhm, and is rated for 2,000A! That just may weld your relay shut. That's what I was referring to. \$\endgroup\$ Apr 9, 2021 at 15:31
  • \$\begingroup\$ Ah, I see the problem now @MicroservicesOnDDD \$\endgroup\$ Apr 10, 2021 at 15:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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