# Selecting Flyback Diode for Inductive Load

I intend to run a 48V DC water pump which uses this motor (http://www.leeson.com/leeson/searchproduct.do?invoke=viewProductDetails&motorNo=098382.00&productType=0) through a solid-state relay like this one (http://www.crydom.com/en/products/catalog/power-plus-dc-series-100-dc-panel-mount.pdf - DC100D20C) and need help selecting a flyback diode.

The pump motor is rated at 1/2 hp, and at its full load/RPMs (1800) can apparently draw 11.0 amps. But according to the pump retailer, with the head height I am dealing with it will probably be at around 170 watts or around 3.5A @ 48V. To build in a little margin for safety let’s say we are dealing with a 10A load, even though I am pretty sure it will never reach that level (even at the maximum head height supported by the pump, the retailer shows only 314 watts / 6.5A).

From what I’ve read I understand that the flyback diode needs to be able to handle the exact same current that was flowing through the motor the moment it was switched-off (via the relay) since the inductor will want to continue flowing that same current (even after switched-off) through the flyback diode until that stored energy has been fully dissipated.

So I know I need a 10A+ diode. But what about some of the other attributes:

• Breakdown Voltage: As I understand it, this is the voltage at which the diode will allow current to flow in the reverse direction. I don't think this should ever happen (right?), so the breakdown voltage should be at the very least higher than the expected battery bank voltage range. How much higher? Any sort of back EMF voltage spiking that might occur would be positive with respect to the flyback diode, right (and therefore the breakdown voltage is not applicable in that scenario)? There shouldn't (in theory) be any voltage spikes on the back/blocking-side of the diode. Although I guess if there were a voltage spike up-stream from the pump/motor I guess it would be better for that to flow through the diode (backwards) rather than the pump.
• Working Voltage: Does this need to be at or above the voltage range of the battery bank (i.e. 44V-52V)? Or does it need to be higher so as to accommodate voltage spiking? Or is it that with the flyback diode there is no voltage spiking (i.e. if the voltage is 48V with the pump switched on, then immediately after it is switched off it slowly decays from 48V down to zero via the diode loop)?
• Maximum Reverse Standoff Voltage - "the voltage below which no significant conduction occurs" ... from another S.E. post: "breakdown voltage is usually 10 % above the reverse standoff voltage" ... so it sounds like this is related to Breakdown Voltage above and as long as it is sufficiently high it shouldn't matter.
• Clamping Voltage: "the voltage at which the device will conduct its fully rated current" ... again, should this be low-ish? so that the full <10A can be flowed immediately with no restrictions? Or does this need to be 48V to ensure that current will only be cycled back through the motor at that voltage (and not at some other voltage that might damage the motor?)?

Thanks in advance for your help! There are a bajillion different diodes out there to choose from, and I'm just looking for a little guidance on how to select the right one in order to prevent voltage spikes from damaging the pump/motor, solid-state relay, and/or other components in the system.

Thanks!

Update: How about the Vishay Semiconductor VS-T40HF10 (https://www.mouser.com/ds/2/427/vst40hfseries-50776.pdf)? Rated for an average forward current of 40A, a reverse voltage of 100V, and a surge current of 600A. Relatively high forward voltage of 1.3V, and probably way overkill for my needs all around, but this would be installed in a remote/rugged location (outdoors, but protected) and I like that it is screw mountable and has screw terminals. I know I could get something that would work for like $0.30, but I also don't mind spending$20 for a more robust design that will stand up to abuse. Its classification as a "Power Rectifiers Diode" has me questioning its viability, but as long as it behaves as a diode and only allows current to flow in one direction then it should be fine. I'm not using PWM or switching this circuit frequently; probably on and off only once or twice a day.

All you need is a diode with a forward pulse current rating equal or greater than the motor draws at full load, and a reverse voltage rating comfortably higher than your DC supply.

In your case I'd suggest a 100 V 10 A surge current diode would be more than adequate.

A lowly 1N4003 diode would be quite adequate at 140 V reverse voltage and 30 A non-repetitive pulse rating.

However something like this Schottky device (even more margin) is only 10-15 cents on Digikey or Mouser.

• Motor surge is over 50A – Sunnyskyguy EE75 Feb 20 '18 at 23:21
• Thanks for your response Jack (and Tony). I will consider those options and make sure to build some extra safety margin in in terms of amperage. Check out the Vishay model I posted as an update to my original question and let me know (as a comment in original question) if you think that one would work. Thanks! – Trevor Feb 21 '18 at 2:56
• @TonyStewart.EEsince'75. The diode suppression is never exposed to the maximum surge of the motor if the system is designed to turn on the motor without contact bounce. And the second diode I specified would easily handle even this event if it occurred. – Jack Creasey Feb 21 '18 at 5:36

The type of diode you most probably need is a schottky diode. Those are the preferred diode type for flyback diodes.

What you really should look for :

• Reverse voltage > supply voltage (x2 or more to be safe)
• High forward surge current (you can easily find one with x20 the load current)
• Forward current near to the current of your load or more

I work as a technician for the railways and for our DC traction motors we use flyback diodes that could barely handle the forward motor current but their forward surge current is very high and those diodes rarely break.

• Yea, I've been looking at both Schottky and "regular". I will keep looking, but for now I think I've found one I'm interested in, primarily because the specs are within range and I like the form factor / mounting options. Check out the update I added to my original question and let me know whether you think that one is a viable option. Thanks! – Trevor Feb 21 '18 at 3:03
• That series seems more then adequate. I can see why you like the formfactor, nice choice if you don't mind the price. – Daniel P Feb 21 '18 at 7:42

From the motor datasheet @9.5V the loss is 107 watts which translates to around 0.9 ohms. So you can expect a start surge of around 53 A which is approx 5x the 11A max rating which is normal for a motor this size.

Unfortunately, contact bounce and power interruption can happen so the diode must be able to handle much more than 10 A and sustain some current until the pump stops. So the inertia of the motor will generate more power into the diode than a lowly 1N400x can handle at uncertain times.

For a $2 you are far wiser to choose a diode that can handle 50A for some unknown duration often rated in n*60Hz cycles vs Amps which is probably a 200V PIV rating. I would be surprised if the Crydom SSR does not have a zener embedded in its design to withstand this reverse voltage current as a diode in the forward direction. The diode becomes the brake to the Pump motor when switched off. How many Joules? You might have to calculate. I suggest a 40A Schottky Diode 200V 30 mJ , 121 A square wave, 0.5 mA Reverse leakage. TO-247 ~$5