1
\$\begingroup\$

I have a 36V 800W DC brushed motor (MY1020). It is controlled by a 10-50V 40A motor controller (Cytron SmartDrive40) and an Arduino UNO. I use a 36V 34A 1200W power supply (AC/DC). the user's manual of the driver says the following:

"If a power supply that cannot sink current is being used (example: bench top and AC to DC switching power supply), the input voltage will rise when the driver is regenerating (motor is slowing down). Thus, it is important to connect a battery with same voltage in parallel with the power supply to absorb the current generated by the motor. Else, the input voltage might rise to a level where Enhanced SmartDrive40 will be destroyed permanently or the power supply trigger protection mode."

I wonder why a battery is adviced instead of a resistive load or something similar. I don't want to buy an expensive 36V battery just to use it as a " buffer tank for the brake energy". Also, powering a battery permanently is not good to me. Is there any shields that could do the job? Would it work with 3 old used 12V batteries?

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Sure you could use a resistive load ... rated for 800 Watts at 36V. It's a bit inefficient though. \$\endgroup\$
    – user16324
    Commented Nov 29, 2017 at 20:54
  • \$\begingroup\$ What chemistry are the 12V batteries? \$\endgroup\$
    – ThreePhaseEel
    Commented Nov 29, 2017 at 23:48
  • \$\begingroup\$ @BrianDrummond - even more fundamentally, you cannot use a resistive load unless the driver is designed to dump to one. Think for a minute about what would happen if you connected a resistive braking load across the power supply. \$\endgroup\$
    – Chris Stratton
    Commented Nov 30, 2017 at 2:03
  • \$\begingroup\$ @ChrisStratton Exactly, that's why it would be a bit inefficient! \$\endgroup\$
    – user16324
    Commented Nov 30, 2017 at 9:20

2 Answers 2

Reset to default
2
\$\begingroup\$

The reason that they suggest a battery is efficiency (And back-EMF protection, cheers Tony).

The battery can store the energy generated from regeneration, and then it can be used when you next demand power. You could configure a capacitor bank for the same thing, however the quantity of stored charge would require quite a large capacitor bank and would be expensive. It also allows the motor to draw surge current.

A resistive load will waste a great deal of energy, as commented by Brian, 800W is quite a lot, and if you were working in regen mode for a long time it could burn whatever load you have designed (unless you accounted well in terms of cooling)

And YES, you could use three 12V SLA (lead acid) batteries in series as you suggest (if they are lead acid), be sure they can handle the required regen current though. (This is what I would do if you have them availiable - they cope well with a floating voltage on there all the time :-) )

\$\endgroup\$
5
  • \$\begingroup\$ It's not efficiency, but rather protecting the drivers from damage from BEMF \$\endgroup\$
    – D.A.S.
    Commented Nov 29, 2017 at 21:34
  • \$\begingroup\$ Its a fair comment, I had failed to consider the back EMF, too used to dealing with motors suited to a protection diode (and nowadays brushless DC) - but I didn't want to add that to my answer just after you had mentioned it in your answer slaps head \$\endgroup\$
    – Rendeverance
    Commented Nov 29, 2017 at 21:38
  • 1
    \$\begingroup\$ dont be too hard on yourself ;). It can reduce some surge losses in the SMPS by essentially being much bigger than a supercap. But dont forget 3 old car batteries will get un balanced voltages \$\endgroup\$
    – D.A.S.
    Commented Nov 29, 2017 at 21:51
  • \$\begingroup\$ Yeah, I get that, just completely passed my brain when replying to the OP lol :) \$\endgroup\$
    – Rendeverance
    Commented Nov 29, 2017 at 21:52
  • \$\begingroup\$ The most obvious issue with a resistive load is that the motor driver would have to be designed to dump to that, rather than the power input. You can't just connect the resistor across the input supply, without constantly burning nearly your peak braking power's worth of energy in it. Do note though that you can't safely dump braking energy to a battery that is already fully charged. \$\endgroup\$
    – Chris Stratton
    Commented Nov 30, 2017 at 2:00
1
\$\begingroup\$

The controller spec recommends as follows:

 6 – 18 cells NiMH or NiCd battery.   
 3 – 6 cells LiPo or Li-Ion battery.    
10V – 45V sealed lead acid battery.
10V – 45V power supply (Must be in parallel with a battery with same voltage).

Since you are using 36Vdc 3 SLA's or 10 LiPos will work as they suggest and 3 old car batteries will work if they are balanced externally.

Consider what your motor DCR is and measure it. This is the surge current it can generate when full braking from full speed. It may be up 10x the rated current for I=P/V = 800W/36V = 22 A for a 220 A max surge current.

  • verify by measuring motor DC resistance then compute 36V/DCR= Imax

Meanwhile your PSU is only rated for 34A continuous.

The reason for this is to provide storage capacitance in the order of 10k Farads with an ESR much lower than SMPS. Consider that a new car battery with 700 CCA means a 5V drop/700A=7mohm , which is how they measure CCA at 0'C.

  • You will need to provide active balance to each battery to prevent over/undervoltage in series unless perfectly matched while being charged by your SMPS.

When Back EMF over drives the controller there is a risk of blowing the bridge FETs from excessive voltage.

Your "mileage" will vary depending on load accel/deacceleration profile.

\$\endgroup\$

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.