Protecting a bench PSU from reverse voltage from a DC motor

[adauphin]

I would like to use the GPP-4323 to power small DC motors, but I'm concerned about any reverse voltage from a motor spinning down.

What would be the desired approach to this?

Does it consist of a simple in line diode or other means?

[Benta]

No need to be concerned. A good bench supply is already protected against reverse voltages, and a DC motor spinning down is about as harmless an a large electrolytic cap.

[analityk]

There isn't reverse voltage, it is reversed current flow. It occurs when PSU decrease voltage during DC motor are connected. In this moment DC motor work like power source, polarities are not changed but voltage on motor is higher than in PSU. Your PSU should be ready to absorb this excessive current.

[floobydust]

Be concerned, some bench power supplies will fail if subjected to backfeed - that is where the load creates voltage far above the setpoint.
A DC motor will have generator-action due to its momentum including the mechanical load, so revving the motor at say 24VDC and then dialing down voltage to a few volts will take time for the voltage to drop as the motor spins down.
Another example is if you are charging batteries, dialing down the power supply voltage below the battery voltage, or switching the PSU off (or a mains failure) with the battery connected, can cause backfeed and damage the PSU or worse.

We don't know the internal circuitry for what protection diodes are hopefully present. GW Instek is good and usually includes these. I haven't seen any teardown pics or schematic. It's a case of "know your power supply" before you blow it up. I'm saying there is a risk, best to ask GW Instek. I did not see special mentions in the manual about battery charging etc.

With -ve voltage spikes due to inductive impulses, there is almost always a reverse clamp diode across the PSU output to protect for this. If you wrongly connect a battery backwards to the PSU output, this is the diode that smokes.

I also see firmware V1.16 2022-07-01 is out.

[Benta]

Be concerned, some bench power supplies will fail if subjected to backfeed - that is where the load creates voltage far above the setpoint.

Please back this up with documentation. The manufacturers would have their hands full with warranty repairs.
I've never heard that that's the case.

[wizard69]

Be concerned, some bench power supplies will fail if subjected to backfeed - that is where the load creates voltage far above the setpoint.

Please back this up with documentation. The manufacturers would have their hands full with warranty repairs.
I've never heard that that's the case.

There is a lot going on here that we don't know but it is fairly easy to damage a bench supply with an electric motor.   It all depends upon how much energy there is available in the motor.   If the mechanical load is significant you basically have a fly wheel generator or gravity generator.   It then becomes a question of how robust the protection is on the power supply.   

This is why motor drives will often have huge resistors to dissipate that energy or in more modern systems a way to return that energy to the rest of the system.   To look at this another way look at what happens to your brakes when going from 70 to 0 in a fast stop.   That energy turns into redhot heat.  Now I'm fairly certain this motor is much smaller so maybe the potential for damage is much lower.

[wizard69]

What would be the desired approach to this?

The potential of power supply damage depends upon many factors so you may get away with it.  However in my opinion you would be better off using a cheap rectifier based power supply.   Save a bench power supply for the more traditional uses.   It is a good habit to get into.

[macboy]

The output transistor of a power supply can be damaged by applying an external voltage higher than the voltage at the unregulated side, that is the voltage on the transistor's collector (or drain). This is an issue if, for example, a battery is connected and the power supply is turned off. A power supply may also make use of multiple transformer taps, so a lower set point may result in using a transformer tap of lower voltage than the battery (or other external voltage). This should not be an issue on most supplies though, since the choice of tap is virtually always done based on the actual output voltage not the set point itself.

Another potential concern is due to the possible presence of a down-programming circuit. This may be present to also a quick transition to a lower voltage setpoint, discharging any capacitance on the output. The risk is that the kinetic energy in the motor will need to be bleed off by this down programming circuit. Unless the motor/flywheel mass/speed is very large, this should not be an issue. In any case, most power supplies do not have down-programming.

I don't foresee any other issue with the motor back EMF.  This will be the same polarity and no higher than the previously applied voltage so there are not many avenues for damage.

[coppercone2]

I fried mad power supplies with solenoids (even after I refurbished power supplies with new silicon power transistor, I still had fails). Nice short circuit on the regulation transistor to bring full rail voltage out to the PCB.

Other forum members seem to confirm that solenoids like to fry supplies.

[floobydust]

Be concerned, some bench power supplies will fail if subjected to backfeed - that is where the load creates voltage far above the setpoint.

Please back this up with documentation. The manufacturers would have their hands full with warranty repairs.
I've never heard that that's the case.

In the analog world we are talking about the two protection diodes like the old LM317 circuit. One for backfeed, another for reverse-polarity protection. Essential in any bench power supply.

Are these still there in modern digitally-controlled bench power supplies? It's hard to tell and an expensive lesson to find this out.

E.g. Keysight EDU36311A family:
"The power supply’s OVP circuit will program the output to zero whenever the overvoltage condition occurs. If external voltage source such as battery is connected across the output, and the overvoltage condition inadvertently occurs, the current from the voltage source will sink through the internal circuitry; possibly damaging the power supply. To avoid this, a diode must be connected in series with the output as shown below."

Rigol DP800 "Do Not Provide Power for the Active Load."
"In order to avoid the anti-irrigation current which leads to the power control loop out of control and damages the powered device, this power supply can only provide power for the pure load without the current output function."
(That's such terrible English)

The GW Instek manual is pretty crappy and has no advice/warnings if the PSU can charge batteries, has backfeed/reverse-polarity protection etc.
A DC motor spinning down is a short-term generator and a similar situation to a battery in that the PSU can unexpectedly see voltage not from itself.
If you want to smoke out a junk design bench power supply, just connect a 12V battery and play around.

[macboy]

I fried mad power supplies with solenoids (even after I refurbished power supplies with new silicon power transistor, I still had fails). Nice short circuit on the regulation transistor to bring full rail voltage out to the PCB.

Other forum members seem to confirm that solenoids like to fry supplies.

I'm not surprised. A solenoid is essentially a big inductive coil. The voltage kick from disconnecting it is not only reverse polarity, but potentially thousands of volts.
Solenoid!=motor

[macboy]

No need to be concerned. A good bench supply is already protected against reverse voltages, and a DC motor spinning down is about as harmless an a large electrolytic cap.

Excuse me, but ... this is just ...

"A good bench supply is protected"? In other words, you acknowledge that there are supplies that are *not* protected. Still, you say "don't be concerned". Erm ...

Regarding the facts: One can never assume that a bench power supply will tolerate reverse voltage, except when it is explicitly stated in the datasheet. Even if it does, the question remains to what magnitude, and inductors and motors can generate huge voltage spikes that no beginner will ever expect.

There are power supplies whose datasheets even explicitly say you should avoid reverse voltage, otherwise the power supply could be damaged. And I know people (not personally but from other electronics forums) who destroyed their power supplies by applying reverse voltage (unintentionally).

And yes, yes, professional power supplies will probably have good reverse voltage protection. But this has nothing to with the original question.

A motor spinning down does not apply reverse voltage. The voltage will be no more than the voltage previously applied (to power it) and the same polarity.

As suggested above, it will be roughly equivalent to having a large capacitance (and a resistor) across the output terminals. Any supply that didn't survive turning down the setpoint with that type of load is just garbage. Some will and some will not survive having that load attached when the supply is abruptly powered off. Good ones will, no questions.

Reverse voltage is another matter. Most supplies have a reverse biased rectifier diode across the outputs to attempt to short out a reverse voltage.

[MrAl]

There are two types of 'reverse' 'voltage'.

1.  This is where the voltage of the load goes higher than the supply happens to be putting out at the time.
Example: you have your supply set to 10 volts DC output and the load jumps up to 20 volts DC.
2.  This is where the voltage of the load goes negative which could draw much more current.
Example: you have your supple set to 10 volts DC and the load voltage reverses and goes down to -10 volts DC.

For #1 a series diode usually works, but it also depends on how high the load voltage jumps.  If the diode is rated for 200 volts than it should protect up to +200 volts on the output, but a resistor of some value on the output helps too just after the diode, across the load.

For #2 a series diode does not help at all because when the load voltage goes negative the diode still conducts.  Does not help at all.
For this you need a reverse parallel diode of high current.  This would mean connecting the cathode to the positive output and the anode to ground, right across the output of the supply.  The diode never conducts unless the load voltage goes negative, then it shunts the current to ground and may blow out the diode if the current is very high.  Sometimes the diode will blow short so it will still protect, but it's better to have a fuse between the diode and the load so that if the load goes negative the diode conducts and draws a lot of current and the fast acting fuse blows.  It's not a guarantee but it usually works.
Case in point: you set your supply to around +14v output intending to charge a lead acid battery such as that found in your automobile.  You accidentally connect the battery backwards.  The diode conducts, the fuse blows.

So it's a little tricky but with a little work and a few small parts you can protect it.

A drawback with the series diode is that the output voltage is not as well regulated anymore because the diode drops a different voltage with a different current.  The ultimate solution then is to break the feedback connection internally and connect it to the cathode of the series diode, if you can.  That will allow the regulation to take place again, but you may also need a capacitor and some load resistance.

[--Oz--]

Add a diode across the output of the supply (not in series).
And if you want across the motor. (unless you need to spin it in rev).

[MrAl]

Add a diode across the output of the supply (not in series).
And if you want across the motor. (unless you need to spin it in rev).

Hi,

I think you should read the post just before yours.  A little more detail helps here :-)