Stepper Drivers Burning Out

[Justin Dion Martinez]

So I have set-up a Smoothieboard (5x) with an X-Carver and modified the XYZ stage with a tilt (E) for a positioning system. The issue I am having is when I am working on the system in a powered down environment, I reconnect and a random driver will stop working. I have read about people burning out the driver because the have a disconnect the wiring to the motor driver while powered on and over heating issues. I do not believe that these are my issues. I have triple checked the wiring for loose connections, installed locking connectors, with polarity on the board,  using ring terminal blocks for all connections in the system, using emergency stop to disconnect power to motors. and I have 3 fans on the ICs to ensure that they are not heating up.  

My questions are:

Can power to logic cause this back EMF issue or moving the gantry (e.g motors generating power) ?

Do these drivers have a particularly high sensitivity to ESD? 

Do I need to disconnect the motor cable, motor power, and logic power when working on the system to stop this from happening?

Is there any suggestions to isolate the drivers to combat this issue?

 

[Arthur Wolf]

Hello.

On Tue, Aug 9, 2016 at 8:37 PM, Justin Dion Martinez <mart...@oregonstate.edu> wrote:

So I have set-up a Smoothieboard (5x) with an X-Carver and modified the XYZ stage with a tilt (E) for a positioning system. The issue I am having is when I am working on the system in a powered down environment, I reconnect and a random driver will stop working. I have read about people burning out the driver because the have a disconnect the wiring to the motor driver while powered on and over heating issues. I do not believe that these are my issues. I have triple checked the wiring for loose connections, installed locking connectors, with polarity on the board,  using ring terminal blocks for all connections in the system, using emergency stop to disconnect power to motors. and I have 3 fans on the ICs to ensure that they are not heating up.  

My questions are:

Can power to logic cause this back EMF issue or moving the gantry (e.g motors generating power) ?

Yes, definitely, for *some* motors. Different motors generate different voltages when moved unpowered, do you have the specs for your motors ?
 

Do these drivers have a particularly high sensitivity to ESD? 

Unlikely.
 

Do I need to disconnect the motor cable, motor power, and logic power when working on the system to stop this from happening?

Ideally yes.

Is there any suggestions to isolate the drivers to combat this issue?

Adding some diodes would solve the issue too.

Cheers.
 

 

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Courage et bonne humeur.

[Justin Dion Martinez]

I am using NEMA 17 motors. 

[Arthur Wolf]

That's just a size, we'd need to know much more about them ( in particular current rating, inductance, etc )

--

You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Justin Dion Martinez]

􀀁model# SM42HT47-0406B

Rated Voltage: 12V

Current/Phase: 0.4

Resistance/Phase: 30 Ohms

Inductance/Phase: 25mH 

[Arthur Wolf]

Yes, "rated voltage: 12v" generally means it's the "bad" type of steppers, that common drivers are not really designed for, and that will generate a lot of power when moved by hand.

Here's a factoid from the #reprap irc channel :
<gthx> ideal stepper is (for reprap printers and similar small CNC using microstepping drivers on 12-24v supply) NEMA17 size, rated 1.5A to 1.8A or less, 1-4Ω winding resistance, 3 to 8 mH, 62oz.in (0.44Nm, 4.5kg.cm) or more of torque, 1.8 or 0.9 degrees per step (200/400 steps/rev respectively), for example the kysan 1124090/42BYGH4803 or the rattm 17HS8401 or Wantai

To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsubscrib...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Courage et bonne humeur.

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Triffid Hunter]

Those motors will definitely over-volt the drivers and destroy them when moved by hand, some friends of mine were mistakenly using similar ones and having the same problem, when they 'scoped the motor during manual movement, they saw peaks of 90v, which is far beyond the driver's absolute maximum of 35v.

Furthermore, the drivers on smoothie will have a lot of trouble driving those motors because they're designed for low fixed speed applications with a constant voltage driver, whereas smoothie's drivers are constant current and optimised for high variable speeds, ie serious CNC applications rather than star trackers.

Also, smoothie's drivers can sometimes struggle to produce such low currents, leading to noisy operation.

You want motors with a current rating of 1.5-2A and the lowest winding resistance and inductance you can find, eg the ones that Arthur mentioned.

I personally like the 17HS8401 from Changzhou RATTM - http://www.aliexpress.com/store/product/5pcs-NEMA17-78-Oz-in-CNC-stepper-motor-stepping-motor-1-8A/907217_523595194.html

ps: with constant current drivers, motor "voltage" is a red herring - it's simply the voltage required to push the rated current while the motor is stationary. As soon as the motor starts moving, it needs more voltage due to 1) back-EMF and 2) winding inductance, and if the driver doesn't have extra voltage available then the motor will work rather poorly - noisy and weak usually. "12v" motors are totally mismatched for smoothie and CNC applications.

To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsubscrib...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Courage et bonne humeur.

[Justin Dion Martinez]

Thank for the feedback,

I would like to not have to replace the motors if possible. Has anyone had success with transient voltage suppression? I saw that some people have been placing diodes and bigger capacitors to try and fix this issue. 

Does anyone know where to place the diode to ensure no back EMF? 

From the diagrams online I see a diode across the motor when the motor is modeled as an inductor. Does that mean if you have a 2 phase motor you place a diode across the primary(A and `A) and then another across the secondary (B and `B), e.g. 2 per motor? 

Does anyone know what specs when choosing a diode are vital? 

"Maximum Clamping Voltage" needs to be less than the stepper motor drivers max voltage, but I am not sure about the importance of: 

"Typical Clamping Voltage" - I assume as close to the max clamping voltage as possible, but not sure on the effects on the normal operation. 

"Working Voltage" - I get that this needs to be above the normal operations range and below the max voltage range, are there any other considerations?

"Breakdown Voltage"-  is this the same as typical clamping voltage? what is the difference ?

[Triffid Hunter]

You could use some 30v zeners in anti-series to solve the over-volting issue, however you can't change that your motor specs are totally wrong, thus they will always run rather poorly.

Don't use ordinary diodes to shunt excess voltage to the power rail, you'll still over-volt everything because you're not catching inductive kick but rather back-EMF or generator effect from the motors' mechanical movement.

If you insist on using those terrible motors, you should at least use a 24v supply so the drivers have some small chance of being able to maintain current into the motors at speed to prevent loss of torque and clipped microstepping waveforms.

[Justin Dion Martinez]

Thank you for the concern about the motors, unfortunately I am stuck with them for now.

I am using a 24v supply, and the application is not demanding. I am creating a positioning system to characterize inductive charging coils. The target device lays on the base plate and the charging coil is suspended from the gantry.  The weight on the gantry is nominal and speed is not a concern. The motors have sufficient repeatability and accuracy my only issue is the board breaking randomly. 

I plan on using Vishay TVS doides # P6KE27A 

http://www.vishay.com/docs/88369/p6ke.pdf

My only question now is placement. . .  

Where do I place the diode to ensure no back EMF? 

From the diagrams online showing how to suppress transient voltages in motors, I see a diode across the motor leads when the motor is modeled as an inductor. 

For stepper motors does that mean if you have a 2 phase motor you place a set of diodes across the primary(A and `A) and then another across the secondary (B and `B), e.g. 2 sets per motor? where a set is a diode in one direction paired with another in the opposite direction? 

Is my issue uncommon? If not, it seems like for the next iteration of the Smoothie board adding some additional transient voltage suppression (e.g. external to the driver IC) would be a desired safety feature by many. 

Thanks again for all your thoughts.

[Arthur Wolf]

Hey.

On Thu, Aug 11, 2016 at 6:32 PM, Justin Dion Martinez <mart...@oregonstate.edu> wrote:

Thank you for the concern about the motors, unfortunately I am stuck with them for now.

I am using a 24v supply, and the application is not demanding. I am creating a positioning system to characterize inductive charging coils.

That's sort of ironic :)
 

The target device lays on the base plate and the charging coil is suspended from the gantry.  The weight on the gantry is nominal and speed is not a concern. The motors have sufficient repeatability and accuracy my only issue is the board breaking randomly. 

I plan on using Vishay TVS doides # P6KE27A 

http://www.vishay.com/docs/88369/p6ke.pdf

My only question now is placement. . .  

Where do I place the diode to ensure no back EMF? 

From the diagrams online showing how to suppress transient voltages in motors, I see a diode across the motor leads when the motor is modeled as an inductor. 

For stepper motors does that mean if you have a 2 phase motor you place a set of diodes across the primary(A and `A) and then another across the secondary (B and `B), e.g. 2 sets per motor? where a set is a diode in one direction paired with another in the opposite direction? 

I believe you need two diodes per coil ( there are two coils in each motor ).

If you can find the schematic for the "rambo" board, it has those diodes installed.

Is my issue uncommon?

Yes.
 

If not, it seems like for the next iteration of the Smoothie board adding some additional transient voltage suppression (e.g. external to the driver IC) would be a desired safety feature by many. 

We do plan to have this protection on the v2 smoothieboards, not really for your problem mainly, but more as a protection for when users disconnect the motors while the power is on.

Cheers.
 

Thanks again for all your thoughts.

On Wednesday, August 10, 2016 at 10:07:50 PM UTC-7, Triffid Hunter wrote:

You could use some 30v zeners in anti-series to solve the over-volting issue, however you can't change that your motor specs are totally wrong, thus they will always run rather poorly.

Don't use ordinary diodes to shunt excess voltage to the power rail, you'll still over-volt everything because you're not catching inductive kick but rather back-EMF or generator effect from the motors' mechanical movement.

If you insist on using those terrible motors, you should at least use a 24v supply so the drivers have some small chance of being able to maintain current into the motors at speed to prevent loss of torque and clipped microstepping waveforms.

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jetguy]

The right way is you basically are rectifying the output back to ground and the V motor source. This is why when you turn a stepper motor in the off position, many power the board.
See this https://github.com/watterott/SilentStepStick
https://github.com/watterott/SilentStepStick/blob/master/hardware/StepStick-Protector_v10.pdf

Again, the 4 outputs on each driver go to 8 diodes (2 per terminal) to rectify the output back to source. This is the same structure the driver already has internally, you are just duplicating externally using bigger and better rated diodes.
I don't think TVS diodes are the right diodes for this problem. Yes, they are great for static protection, not so much inductive kickback.

[Jetguy]

Again, the first place is to read the driver schematic  http://www.allegromicro.com/~/media/Files/Datasheets/A4982-Datasheet.ashx?la=en

[Triffid Hunter]

The problem here is that his motors put out up to 90v when turned manually, so dumping back emf to the supply rail is useless - he needs crowbars to turn that directly into heat and avoid over-volting the drivers. TVS diodes or zeners should work