Modifying Cheap Stepper Motor To Be More Powerful (28BYJ-48) 

Thanks, glad you liked it :)

Thank you ;) I’m working away this week but I’ll have a new video as soon as I can :)

Glad it was helpful!

Just fixed a typo. I meant “cool” not “cook”

Thank you, glad you liked it :)

Glad it helped, please let me know how it turns out, sounds like a nice project.

Thank you! Yes the case is not an easy feat, I’d did slip a couple of times into my other fingers. These 5V ones seem to use something like 40-42 SWG and the bobbin is only half wound, about 22 ohm resistance IIRC.

Thank you

Thank you so much :)

Thanks for the info, where in the gear reduction stages was it?

I’m glad you found the video useful. I’ll try my best to help with your questions, ask away!

I have two brand new steppers and drivers . I got the basic code to upload from another RU-vidr (paul Mcwhorter) I designed and printed a custom litho and base to rotate it the stepper got a little warm to the touch without having a load on it. Then after letting it run for an hour it stopped working. I can hear and feel vibrations from the motor and that’s all. Soooo I installed the second stepper and driver. The first setup was ran like I see on RU-vid, arduino stepper and driver. The second time I followed Paul’s instructions and used an external power source that came in the arduino kit with the wall plug making sure to have everything set to te 5volt pins . It ran for an hour and just sits and hum like the first one. Please help. I have new steppers and drivers coming in because I now have a few people that want me to make one for them. And I want to use the arduino nano

Ok so I guess the first questions I have are: Do you have a part number for your stepper motor? Do you have a part number for your driver? What voltage are you trying to drive the stepper at?

28byj-48 5v dc stepper And the uln2003 driver

I’m just using the 5v supply that came in the kit until I come up with a better idea because I also need to run led lights for the litho

Thanks, you too!

Just here to fix the fridge...

@@JustAnotherMakerChannelI feel like they could be another undisciplined example of generation z, seeing their comments history and videos they seem to show no interest nor understanding in your content from what I’ve seen. I feel like they’re using your comments section as a free advertising platform.

​@@BritishEngineercry more

Yes, they did seem strange, I remember having issues trying to ramp the speed too quickly too!

Yes, I think I’ve seen one like that driving a microwave turntable.

Si, pero los de microondas son de un bobinado y giro aleatorio si no llevan una carraca. En cambio hay ahora otros con doble bobinado para elegir la dirección de giro. Estos últimos si pienso que son lo mismo basicamente aunque la diferencia es el alto voltaje de 220v y que solo sacan 4 cables.

Thanks for commenting, I think they were somewhat more reliable, I do think there were still some occasions of missed steps/jitter, but I was abusing the motors I was testing so that could have contributed to the issues!

@@JustAnotherMakerChannel I did the modification after commenting and also increased the voltage to 17V (my motors are 5V version). It works great. More torque and no more missed steps from what I can tell =)

Yes that is correct the red cable is unused and can be removed if you want!

​​@@JustAnotherMakerChannelsame for six wires also , is it possible to use a six wire stepper motor from a printer that can be used as a generator or Dynamo

Yes I can share the model, what format would be useful for you, I modelled it in fusion 360. It should be quite accurate though I think the gears are just an approximation, I.e. the teeth are not accurate involute profiles.

@@JustAnotherMakerChannel Fantastic! Thanks for being willing to share the model. Fusion 360 format works perfectly for me. And no worries about the gear teeth; I can work with approximations.

I've just uploaded the files and you can find them here: www.thingiverse.com/thing:6620639 Sometimes Thingiverse is not very good and the link might not work until it decides to, you might have to check back in an hour or two.

@@JustAnotherMakerChannel Great! Thank you for sharing the link. I've managed to download the files successfully.

Thanks for the tip!

I think it would depend on the current limit set on the stepper driver, theoretically it should produce more torque at 12V than 5V

Thanks :)

Thanks and that’s an interesting question, I think if you drive them slowly you should be ok. Probably the bigger risk is if you’re going ‘fast’ then the inertial energy of the vent has to go somewhere and that might be the gears. But to be honest because the gearing is so high I doubt you could go fast enough to cause any damage!

I can't stall them at 5V but if I lower the ULN2003 voltage with a buck converter I can adjust the stall torque as I desire. 3.3v seems about right.

Cool, thanks for the info!

These are tough little motors, at 31V 0.8A it took a few minutes to die at 168 degrees!

@@JustAnotherMakerChannel feels like a battery per leg is the way to go.

Good questions! Unfortunately the backlash is the product of cheap gears, it’s no more or less before the modification. Probably the best course of action is to use these steppers for something like a lead screw, e.g. on a 8mm lead screw, 4 degrees is 0.09mm, or a 4:1 reduction with helical gears would net 1 degree backlash if the helical stage is close to zero backlash. You can also pick the motors with the lowest backlash from the batch you buy for the more critical elements of a project. Hope this helps :)

@@JustAnotherMakerChannel my first guess would have been that the plastic gear would be the root cause. with several!! transmission stages, each causing potential error. I learned that precision error is "not accumulative" which correlates a little with your "best in batch" theory. However, if the material of the gear wears out over time, its a false favor. lead screws rules ya. certain leads might even reduce the effect as one needs minimum force to move.

The stepper driver is used to limit the current, so you can drive the motors at higher voltages to reduce the chance of skipped steps under load and rely on the current limiting to try and prevent the motor from getting too hot. To adjust the current in the video I just adjusted the current limiting potentiometer on the stepper driver.

Thank you :) The change in voltage is mainly due to the change in driving mode. The original driver runs as constant voltage, whereas the modification allows the stepper to be controlled as constant current. Although the modified motor driver is running at 12V the motor will not have a constant 12V across the coils, this is mainly why I included the power values as they are more representative of what's going on, as an example looking at the first modified test @4:08, the motor is using 30% less power overall but producing almost 4 times the torque, since this test is performed at 50mA per coil and the original unmodified motor had a current draw of 340mA with a constant 5V applied we could in theory run the modified motor at 5V and still achieve 50mA per coil and still see an increase in torque, it's just easier to find stepper drivers that work at higher voltages as they can generally accelerate the motors faster by using a higher voltage. I hope that was useful :)

​@@JustAnotherMakerChannel Then, 1- you take the 5V version. 2- you cut the red wire conection. 3- You drive it with higher voltage but limiting the current. And then you get the increased torque. You don't need to buy a 12V version of the motor. I this correct?. And the second question is: You need a power supply which allows you to limit the current, right? The driver ULN2003 won't be enough to control the current if you want to avoid overheating.

This is correct but you need to break the connection on the circuit not just cut the red. You need another driver to drive as a bipolar stepper.

For the modified version I used a DRV8825, hope that helps :)

I can’t run the modified motor at 5V, the DRV8825 driver I am using has a minimum motor voltage of 8.5V. The current it adjustable so it can be anything up to 2A per coil.

Well the key trade off is that you need to use a different stepper driver than the one supplied with them, such as a A4988, DRV8825, TMC2208 etc. and these cost more, typically £2 - £8, these also typically prefer a higher voltage then the supplied driver, for example the DRV8825 needs about 8.5V for the motor power which can be more challenging if your project is battery powered. I haven’t used these much, I was planning to use them to improve my hexapod but the backlash in the gears is just too much so I may make a little automatic coil winder or something with them and then try rewinding one with different wire sizes to see if they can be made more powerful.

​@@JustAnotherMakerChannelthanks! If you do that, I'm sure it will make a cool video. I had assumed the extra stepper driver would be a cost for sure, but I'm curious if a similarly spec'd driver would face pros/vs/cons in this new bipolar mode vs the original mode, assuming it could support both. I am not too well-versed in stepper motors. :P or is the bipolar mode the preferred mode for higher-quality, serious drivers?

Yeah mainly it’s about cost, unipolar motors (original) tend to have lower power and torque because only half of the coils are energised, but the driver can be really simple and cheap because it doesn’t need to change the polarity of the voltage supplied to any of the coils. Whereas bipolar motors require a little more complexity from the driver as they need to change polarity, but for us it’s easier to use a driver like this because modern types are really sophisticated, we only really need 2 wires, one for step and one for direction, and they typically have current limiting that you can set, work on a bigger voltage range (so you can drive faster) and they can do micro-stepping (they precisely control coil currents to hold the rotor between positions, giving more steps per revolution), and even determine if there are faults with motor or circuitry, really the cost of these drivers is definitely worth it! Because of the tremendous popularity of 3D printing the cost of bipolar motors and drivers are significantly cheaper than they were 10 years ago, making the cost benefit of unipolar motors and control not as significant as it used to be.

​@@JustAnotherMakerChannelThank you for this excellent and detailed response :) this clears up my confusion!! and thanks for the cool video, it is inspiring me to spend some more time with my steppers. You'll definitely have a new subscriber for me!

@@JustAnotherMakerChannel I had the same idea but the payoff vs the time required just wasn't there for me - too many irons in the fire as it is.

Sorry for the late reply, if you check out my ‘how do stepper motors work’ video there is an example at 11:30, just substitute the A1 & A2 for orange and pink and the B1 & B2 for blue and yellow. Hope this helps :)

The power figures are comparable, I.e. the higher voltage and alternative driving mode produce more torque whilst consuming less power and thus producing less heat.

Generally bipolar motors cost more, DRV8825’s can be bought for as little as 40p, they use fewer IO pins and provide greater control. Using an unmodified motor at twice the voltage would cause double the current, so 4 times the power through the same amount of winding, vs the modification where a lower power consumption can produce a higher output torque.

What does this mean?

@@JustAnotherMakerChannel Field Oriented Control AKA Vector Control, which is a clever way of controlling the stepper drive currents to provide numerous performance benefits. However no-one in their right mind would be considering this for the kind of application these very cheap gears steppers would be used in.

Thank you, I wasn’t sure if they meant field oriented control, fact or crap, free of charge! Yes it’s used more in BLDC motors.