Choosing motors for perfect stall detection



  • Hi guys, I want to buy some motor from Aliexpress and I remember in one of the posts on this forum DC42 wrote that for a perfect stall detection you need high inductance motors witch is the opposite of achieving high speeds. Now, I'm wondering if somebody know what will be a ballance between these two factor, so I can achieve some good speed but still to implement the stall function. These are the motors I was looking at:

    https://www.aliexpress.com/store/product/EU-free-20PCS-Wantai-Nema-17-Stepper-Motor-42BYGHW609-56oz-in-40mm-1-7A-4wires-CE/2853014_32834260377.html?spm=2114.12010612.0.0.55dd4faaHCDVC7

    Thanks!


  • administrators

    Please note, I am not certain that high inductance motors make stall detection easier, although I think it is likely. It may be the ratio of inductance to resistance that is more important. If you do choose high inductance motors, use 24V power not 12V.

    Also I am not sure whether 1.8 or 0.9deg motors are better for stall detection. We don't have enough experience yet.

    I think stall detection during printing on a delta printer is going to be difficult or impossible, because usually one or two of the carriages are moving only slowly.



  • In this case I'm lucky. I use the H-bot mechanics and I power my duets on 24V. I have not manage to implement the stall detection with my current motors but, I keep getting false detection and when I stop the motor sometime I don t get any answer. But anyway, I think I'm gonna go with these motors for now and see how they behave.
    Now can I ask you something else in this topic? When will be available a firmware update for PanelDue because I don't want to see anymore on my screen "17minutes left of filament". And also i get 30 hours for a print that it's done in 2 hours. Thanks!



  • @dc42:

    Please note, I am not certain that high inductance motors make stall detection easier, although I think it is likely. It may be the ratio of inductance to resistance that is more important. If you do choose high inductance motors, use 24V power not 12V.

    Also I am not sure whether 1.8 or 0.9deg motors are better for stall detection. We don't have enough experience yet.

    I think stall detection during printing on a delta printer is going to be difficult or impossible, because usually one or two of the carriages are moving only slowly.

    What does it mean high inductance? 2.8mA it is high?



  • No, it is low. This motor does not have a high impedance. High impedance means you have a low current for the same power, like 0.5A instead of 1.7A…



  • I think it's about inductance not impedance. At least, this is what I understand.



  • Yes, I think that stall detection relies on inductance, but both are related: to increase the inductance, you need to increase the number of turns of the coils, so you increase the resistance.

    BTW, impedance is made by inductance (reactive part) and resistance (active part). See https://en.wikipedia.org/wiki/Electrical_impedance#Inductor_and_capacitor



  • In terms of our applications, it seems that 5 mH is about "average" inductance for us. So, IMO, "low" is below 5 mH, and "high" is above 5.



  • @bot:

    In terms of our applications, it seems that 5 mH is about "average" inductance for us. So, IMO, "low" is below 5 mH, and "high" is above 5.

    Awesome. This is the answer i was waiting for. well… now comes the hard part. there are few sellers who offer 5mH nema 17 motors, so the only option will be by order. but.... i guess i can t make a special order for 5 or 10 motors....



  • Forget Aliexpress! They may not sell high impedance motors there, but any good seller does. 5mH is not really high; you can find some with 50mH inductance:

    low impedance: https://www.omc-stepperonline.com/hybrid-stepper-motor/nema-17-bipolar-45ncm-64ozin-2a-42x40mm-4wires-w-1m-cable-and-connector-17hs16-2004s1.html

    high impedance: https://www.omc-stepperonline.com/hybrid-stepper-motor/nema-17-bipolar-18deg-40ncm-566ozin-04a-12v-42x42x39mm-4-wires-17hs15-0404s.html

    They even have 2 other intermediate values for that 40-45N.cm torque range.

    (Don't know if you can reach the site, I have troubles myself).


  • administrators

    Don't choose very high inductance motors, or you will find that you don't have enough drive voltage to run them as fast as you want to. Use the calculation at https://duet3d.com/wiki/Choosing_stepper_motors#How_to_work_out_the_power_supply_voltage_you_need to check the suitability of your motors before you order them.



  • Also, keep in mind moving by hand a motor with a high impedance will generate a high voltage, which can destroy the electronic!



  • oh got it. Nice, today I have learned something new
    I ll keep you posted with my decision guys. Thank a lot!



  • @dc42:

    Don't choose very high inductance motors, or you will find that you don't have enough drive voltage to run them as fast as you want to. Use the calculation at https://duet3d.com/wiki/Choosing_stepper_motors#How_to_work_out_the_power_supply_voltage_you_need to check the suitability of your motors before you order them.

    I did some calculus with these types of motors:
    https://www.aliexpress.com/item/Hot-No-Tax-to-EU-10-PCS-Nema-17-Stepper-Motor-42BYGHW609L20P1-X2-4000g-cm-1/32792872368.html?spm=a2g0s.13010208.99999999.262.rpduqP
    I power my duet at 25 V. Making the calculus, I took the conclusion from the point 6 from wiki that the sum of my two back emf is 21.22 so I'm way good under 25V. This means I can easily go for higher inductance motors. The calculus was made for 250mm/s travel speed on a corexy.

    If I remake the calculus for 5mH inductance motors I'll have a sum of 28.8V witch is a bit bigger than 25V so Maybe I won't achieve a travel speed of 250mm/sec.

    With 4mH inductance motors I'll have 25.36V witch is a bit over limit so….Maybe I'll achieve something like 230mm/sec travel speed but also I'll be able I think to implement more easier the stall detections function.

    These been said I think the right motors for me will be:
    2 phase
    1.8 degree/step
    current: 1.7A/phase
    resistance: 1.5 ohm/phase
    inductance: 4mH/phase
    holding torque: 40N.cm
    I also run my motors at 1A current. I saw that this is the best proportion between noise and power.


  • administrators

    Those sound very much like the motors I use on my delta. I managed to get homing working using stall detect, but I haven't managed to get stall detection during printing working yet. I think it may be something that is difficult to do on a delta.

    I have put a spreadsheet to calculate maximum motor speeds before torque is list at https://www.dropbox.com/s/5z66rgjc8gptn5o/StepperMotor.ods?dl=1.



  • @dc42:

    Those sound very much like the motors I use on my delta. I managed to get homing working using stall detect, but I haven't managed to get stall detection during printing working yet. I think it may be something that is difficult to do on a delta.

    I have put a spreadsheet to calculate maximum motor speeds before torque is list at https://www.dropbox.com/s/5z66rgjc8gptn5o/StepperMotor.ods?dl=0.

    I have downloaded the sheet on my computer and replaced the values in the yellow boxex but the green ones are not changing. I also can't modify the values directly on the link you have provided.


  • administrators

    Yes you will need to take a copy because Dropbox will quite rightly not let you modify my files. I've changed the ?dl=0 at the end of the link to ?dl=1 so that if you click on it, it will prompt you to download it. I suspect that what you have downloaded may be the HTML rendering of the sheet.



  • @dc42:

    Yes you will need to take a copy because Dropbox will quite rightly not let you modify my files. I've changed the ?dl=0 at the end of the link to ?dl=1 so that if you click on it, it will prompt you to download it. I suspect that what you have downloaded may be the HTML rendering of the sheet.

    Now it will download automatically but when I try to open with Excel 2010 it says that some parts of the sheet are impossible to be read….something like that. After I press ok it opes and when I try to change values it behave the same. Green cell value don't change. Maybe a C++ little program written in CodeBlocks will be more efficent?!


  • administrators

    You could download Libre Office, which is what I used to create it.

    I'm hoping to get a Javascript version implemented on either my Escher 3D site or on reprapfirmware.org.



  • Thank you for the tool, David. It is very useful and confirmed my previous calculations were correct.


Log in to reply