Clinostat - Continuous rotation servo motors burning out

[BlueHeron]

Doing a project closely based on this post: https://www.sciencebuddies.org/science- ... -clinostat

As we are in Canada, we didn't buy the kit associated with it, but built our own, using these motors Wishiot Feetech 9g FS90R Mini Micro Servo 360 Degree Continuous Rotation Servo Motor (sold here: https://www.amazon.ca/Wishiot-Continuou ... hdGY&psc=1)

The seeds being used are tomato seeds (from the TomatoSphere project).

The issue we are having is that the motors seem to be burning out after less than 24 hours of continuous running. Software controls were put in to monitor if the motors had stopped spinning (using image detection of a unique QR code-like image). Our setup has 4 clinostats running simultaneously, and we've had to buy 8 motors so far. When a motor quits, it is very, very hot. We are fairly confident the setup and control of the motors is appropriate, from an electrical circuit point of view (our team has a reasonable degree of knowledge and support in this area), but we have not used servo motors like this before.

Our goal is to be able to run the clinostat continuously for several days (2-4 days). But our motors aren't lasting that long! Ideally we would like to repeat the experiment several times too, but replacing the motors for every repetition isn't ideal.

Anyone have any thoughts? Are servo motors of this kind designed to run continuously like this? We have tried to adjust the speed of the motors to be as close to 10 rpm as possible (this is a bit finicky, and is not exact). If they were spinning a bit faster, would the motors do better?

Thank you!

[calixte]

Hello!

From what I can gather, the servo is frying. Looking through the amazon description you gave it seems the the Feetech FS90R servo motor is more suited for short, strong, and precise movements, so having it running for over 24h is not great. I think you should change to a stronger servo motor or even use a DC motor instead. Good luck with your project!

[BlueHeron]

Thanks for your thoughts!

We agree that we think we are frying the servo motor and have considered switching to a DC motor instead (but that does add some complication as we think we will need some gears to slow down the motor, not to mention a different power supply than the one we currently have).

To the Science Buddies team: This is an awesome project that can be really interesting for students.

However, we are wondering if there should be a warning on the description of the clinostat project that warns that many servo motors are not designed to run for long periods of time?

For students doing the experiment just once and exactly as described and growing very fast growing cress, maybe it won't be as much of an issue. But if anyone wants to repeat the experiment several times, or is using slower growing seeds, and is using a servo motor like the one linked to in the description may run into the same issues we are experiencing with the motors simply burning out after 12-48 hours of continuous use.

[bfinio]

Hi,

I'm the project author - thanks for reporting this. The reality with cheap servos is definitely that you get what you pay for. I did test this project with a servo that didn't burn out, but maybe I just got lucky or some brands are worse than others. In my experience the first failure mode for these cheap servos is the tiny plastic gears stripping when they're over-loaded, but that isn't really the case for this project which has continuous use with a pretty low load.

In your case, it may be worth buying slightly nicer (i.e. more expensive) servo motors designed for continuous use rather than going with DC motors and gearing. I'm not sure what vendors ship to Canada, but in the US, SparkFun and Adafruit are usually reliable vendors for more selectively sourced/higher quality electronic parts than the stuff you can buy in bulk on Amazon.

This is a pretty popular project, and so far we haven't had any other reports of servos burning out. Hopefully that means this is not a widespread issue and your case is just a fluke/some particularly low-quality servos. I noticed that the ones you linked to on Amazon don't have any reviews at all - definitely check the reviews when buying on Amazon, and be on the lookout for comments saying that some parts were broken right out of the box or died very quickly etc.

The servo connection to the Arduino is pretty standard, so you're right that it's unlikely that you're somehow breaking the servos with the wrong PWM signal or power supply, but I'm also happy to take a look if you want to post your code and pictures of your circuit, just in case.

Thanks,

Ben

[BlueHeron]

Glad to hear that no one else is reporting similar issues! it definitely makes sense that quality parts could make a difference. Some members of our team are fairly new to electronics projects of this type, and none of us had any previous experience with servo motors, so we were doing our best to follow instructions from the project when we didn't have any previous experience. Maybe some text in the project description to say to look for a quality servo motor (and how you would know it is a quality servo motor), especially if you'd like to do many runs of the experiment. I think we followed the initial link about servo motors listed in the experiment description, which took us to Amazon.com, and then did our best to look for something similar available in Canada, so we had (mistakenly) thought that what we could get on Amazon would be sufficient.

Because we've run into a time crunch, and our servos were (somewhat) cheap, we have gone the route of just buying more. If we had more time, we would definitely look at getting better motors, or changing the motor type. (We are also running 4 motors at once (partly to reduce the number of control seed sets we need to plant, and partly to get more results faster) so would need to buy 4 motors if we upgraded.)

Overall, it is a great project that is highly appealing and very interesting! I would definitely recommend it to others (if they buy a good servo motor)!

[bfinio]

Ok - if you are already committed to using those motors, then the solution to preventing them from overheating (or at least, extending their lifespan long enough to do the experiment) would be to run them intermittently, giving them a little time to cool down. The challenge there is finding the balance between giving the motors enough time to cool down vs not pausing for so long that the plants start to have a gravitropic response. For example, on one extreme, I'd say 1 second on/1 second off probably isn't much time for the motors to cool down, and on the other extreme, 1 hour on/1 hour off is probably way more than enough time for the motors to get back to room temperature, but you may start seeing directional root growth. My gut instinct for motors of this size is that maybe alternating 5-10 seconds on or off would be OK and help extend the life.

Another option is to try cooling the motors simply by aiming a fan at the whole project. That MIGHT help keep them cooler and running longer. Finally - this probably isn't a great idea, but could be worth a shot - is to carefully try drilling tiny air holes in the plastic case of the motors to get better air flow and cooling inside. Normally that is a bad idea for at least two reasons: 1) It would be very easy to slip, drill too far, and damage the inside of the motor, and 2) now you're exposing the previously-sealed inside of the motor to dust contamination. BUT in this case, where you just need the motors to survive a few days for a project, it might be worth trying.

I also should have asked earlier, but just to be 100% sure, how are you powering the servos? Running them with too much voltage will also burn them out quicker.

[BlueHeron]

Thanks for the suggestions of ways to keep the motors cooler, and some guidelines for what might work to keep them cool!!

My guess is that a cooling fan won't do much - the first round we had the setup in our basement where is was 18 C, so cold the seeds didn't germinate at all, but the motors burned out. (But maybe more air circulation would have helped.) We moved it to a room where we can get the temperature to 23 C with 50 % humidity and saw about the same lifespan for the motors, and seed growth!

But having the motors stopping to cool down periodically is something that we could relatively easily incorporate into our system. Drilling the cooling holes is also an interesting idea - that will be a last resort for us I think.

Good point about problems if the power supply is not correct. That is a part of the project that is way outside my area of knowledge. We are not using the breadboard set up as proposed, but have, I believe, a 5V power supply connected using a system that I believe has significantly higher specifications that what is required. The electrical engineer on the team has been supervising this part (and although anyone can make mistakes, I'm personally highly confident this part is correct).

Unrelated to our motor problems, others might want to know that putting food colouring in agar seems to make it not set properly. We're automatically taking pictures every hour and will then use image processing software to check for root growth, but our background was white and the roots are white, we needed better contrast, so we tried blue food colouring the agar -twice. Both times, even with more agar, it melted and dripped off the petri dish in a matter of minutes. Will now have to colour the bottom of the petri dish, or the back of the clinostat.

[bfinio]

Ok - as long as the power supply is 5V then you should be OK.

For the image processing, maybe it would be simpler to just switch to a black background so you have better contrast with the roots? I'm a mechanical engineer so I'm definitely not very familiar with working with agar and what can mess it up.

Please keep us posted and let us know how it goes!

[BlueHeron]

Black as a contrast colour for the roots makes a lot of sense!! I will see if we can get some black plastic to glue into the inside of the bottom of the petri dishes - we already have the bit for attaching to the motor stuck on with double sided tape, and taking it off usually breaks the petri dish.

Will definitely let you know how it goes!