0
\$\begingroup\$

I am working on a project that involves a DC-DC power step-down. The details are as follows :

  • Input : 48V , 15A
  • Output : 18V ( around 700W ; due to losses)

However, I could not get any buck converter directly for such high output current. My question is - Could I use a 48V (4S-5S) ESC (of proper rated current) along with a 3-phase full wave rectifier to get the desired DC output? Could the exact voltage that I need could be set with a PWM signal to the ESC?

I am posting this question because I tried this concept with a 4S ESC and a 160V MDS40 3-phase bridge rectifier, and it did not work. I currently do not have an oscilloscope, hence couldn't determine what exactly was happening. Also, I am operating the ESC at 50Hz.

Could anyone please point out what the problem is?

Update: 16th Oct,2019

I connected the ESC to my pc-oscilloscope and I am getting mostly zero voltage reading but there are funny little pulses (see video here) that are occurring in the waveform.

\$\endgroup\$
6
  • \$\begingroup\$ @Curious_Techie then remove the numbers, because a 100% efficient converter in principle does not exist, so these numbers just make your problem clearly answerable with a "no" without even thinking about the ESC aspect; in that sense, rather say "I need about 700 W output power", to give us an idea of what you're dealing with instead of specifying somthing impossible! \$\endgroup\$
    – Marcus Müller
    Commented Oct 6, 2019 at 10:29
  • \$\begingroup\$ I think the number to remove are the 15A. All other are more fixed by environment and specs. probably. \$\endgroup\$
    – Ariser
    Commented Oct 6, 2019 at 14:11
  • \$\begingroup\$ Please read and take time understanding this existing question and its responses to see why your idea is not really practical electronics.stackexchange.com/questions/31699/… \$\endgroup\$
    – Chris Stratton
    Commented Oct 7, 2019 at 5:07
  • \$\begingroup\$ @Curious_Techie couldn't have put it any better than Bruce Abbott did in his answer, combinew with Chris' comment: you're missing the smoothing energy storage to begin with. \$\endgroup\$
    – Marcus Müller
    Commented Oct 7, 2019 at 7:45
  • \$\begingroup\$ @MarcusMüller it's not the "smoothing" that is missing, it is the actual power conversion mechanism. Just smoothing out PWM with a filter does not make a remotely efficient buck converter. \$\endgroup\$
    – Chris Stratton
    Commented Oct 7, 2019 at 15:44

3 Answers 3

Reset to default
1
\$\begingroup\$

Reasons why it won't work:-

  1. This ESC is designed to work with a sensorless brushless DC motor. If it doesn't see a back-emf waveform from the motor it can't synchronize to it, so it will shut off to protect itself and the (non-existent) motor.

  2. Theoretically you could install custom firmware that runs 'open loop' with no motor present. However without any inductance to smooth the current flow the output will just be a series of pulses at full battery voltage. This might be OK if the load is a heater element or incandescent lamp, but not for anything that needs a proper DC voltage. If the device being powered has an input filter capacitor large enough to smooth out the pulses it will simply charge up to full battery voltage and you won't get any voltage reduction.

  3. So you add an inductor and capacitor, and now you have smoothed output. But you don't have any feedback to regulate the voltage, which will vary depending on load. At startup it will probably ring at the filter frequency, damped by the load resistance. If the load is fixed and doesn't mind going over-voltage for a short time you might get away with it, otherwise...

\$\endgroup\$
5
  • \$\begingroup\$ This still misses the point - it's not the smoothing that is missing, it is the actual power conversion mechanism, which is what real buck converters use their inductor for. \$\endgroup\$
    – Chris Stratton
    Commented Oct 7, 2019 at 15:45
  • 1
    \$\begingroup\$ With PWM you have 1/3rd of what you need. Just add an inductor/capacitor filter, and feedback to control the PWM ratio, and you have a full buck converter. Change the firmware to produce the same PWM output on all 3 phases (aka brushless to brushed ESC conversion) and you don't even need the rectifier. Not a good one since the PWM frequency would be low and loop response slow, but still... \$\endgroup\$
    – Bruce Abbott
    Commented Oct 7, 2019 at 18:57
  • \$\begingroup\$ Again, what is needed is NOT a filter. That's not the role of the inductor in a proper buck converter. The inductor is what accomplishes the actual power conversion between voltages. \$\endgroup\$
    – Chris Stratton
    Commented Oct 7, 2019 at 20:01
  • \$\begingroup\$ Please check the video that I uploaded @BruceAbbott . I think I am observing similar to what you've said. \$\endgroup\$
    – curious_techie
    Commented Oct 16, 2019 at 6:37
  • \$\begingroup\$ Yep, looks like it's trying start the 'motor' at a very low PWM ratio and low rpm. after a short while it gives up because it doesn't see any back-emf, then it tries again - exactly as expected. \$\endgroup\$
    – Bruce Abbott
    Commented Oct 16, 2019 at 6:44
1
\$\begingroup\$

The peak voltage of an ESC is not going to change. The effective output voltage is changed by changing the duty cycle. If you rectify the output you will have a pulsed output with a constant peak voltage. If you put a capacitor on the output, you will have a constant voltage that is comparable to the input voltage. The bottom line is that an ESC is not going to be very good at powering much of anything except a motor. It might also work as a light dimmer for incandescent lights.

\$\endgroup\$
1
\$\begingroup\$

The hardware of ESC is probably OK. The control though would not work, it's designed to sample things back from the motor. So if you have a low-level access, it could work.

upd: you need an inductor!!!

\$\endgroup\$
5
  • 2
    \$\begingroup\$ Even the hardware is wrong, there's no inductor \$\endgroup\$
    – Chris Stratton
    Commented Oct 6, 2019 at 14:16
  • \$\begingroup\$ The motor could be. But yes, I assumed it's external anyway. \$\endgroup\$
    – user76844
    Commented Oct 7, 2019 at 15:08
  • \$\begingroup\$ There wouldn't be any motor in the askers intended usage. The point is that a remotely efficient buck converter requires an inductor as an intimate part of the design to do the actual conversion, and this abuse of an ESC wouldn't include one. So no, the hardware is not suitable. At most there are maybe some FET's you could recycle. \$\endgroup\$
    – Chris Stratton
    Commented Oct 7, 2019 at 15:42
  • \$\begingroup\$ Like I said, you are right \$\endgroup\$
    – user76844
    Commented Oct 8, 2019 at 1:59
  • \$\begingroup\$ And i have no idea why this one was selected as THE answer. \$\endgroup\$
    – user76844
    Commented Oct 9, 2019 at 21:09

Your Answer

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

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