Not there yet on the ball valve control, and missing oxygene

Yesterday night at the Electrolab was dedicated to two things: testing the larger gears for the ball valve to make up for the lack of torque of the stepper motor we are currently using, and welding with oxyfuel gas a part of a whipped cream siphon used to unload nitrous oxide cartridge and a standard plumbing link.

You can see in this video the result of the test with the larger of the two gears that Ryan had 3D printed in PLA and sent to me from Finland to France. Unfortunately, it is still not providing enough torque to rotate the ball valve.

I'm now thinking that we should seriously consider finding a stepper motor with more torque, rather than going for still bigger gears, which will in the end mean that a lot of time (and pressure in the combustion chamber) will be lost while the ball valve will be rotated painfully slow.

For the oxyfuel gas welding, I had purchased a (supposedly) brand new 60L oxygene canister to replace the depleted one in the blowtorch at the lab. However, as this video shows plainly, this canister turned out to be empty... quite a disapointment.

The bright side is that once again I learnt a lot about a technique I knew nothing about before that night. Not only was the welding metal to use discussed extensively (tin was abandonned in favour of silver, because of the high pressure to withstand), the way to weld was also discussed (pre-heating the 2 pieces to weld at 400°C beforehand?), but we also dipped the two parts to be welded together in chloridric acid to remove all copper oxides, which do not fix well to welding metal. And I also learnt about the basic security precautions, and appropriate configuration of the workplace, in order to weld efficiently and securely. My thanks go to our tech guru Zenos and to deputy tech guru Eric for guiding me through all this!


ryan.pulkrabek's picture

Too bad. I am surprised that with this much torque, we still could not open the valve.

At what position do you have the Pololu trim pot set? If you haven't adjusted it at all, we could turn it up to provide more current to the motor, which would provide more torque. Just be careful not to turn it up too high, then the driver will burn up. I have learned this the hard way.


I agree that we shouldn't make the gears any larger. Pretty soon, they won't even fit into any rocket. We could test to see if an absurdly large gear works, just so we get the feeling that something would work.

If nothing seems to work, perhaps we can upgrade to NEMA 23 motors, which is capable of almost 2x the torque.

I will play around with this setup at home too. I just need a battery with more voltage to test.

Damien's picture

I'll try toying with the trim pot next Thursday.

In the meantime, I will also consider which bigger motor to purchase: apart from the NEMA 23, do you know of cheap stepper motors which provide more than 1Nm holding torque? Just to compare the torque/price ratio.

ryan.pulkrabek's picture

I don't immediately know. I will begin searching, though.

something to do in the mean time is to empericaly test the output torque of the stepper as well as the required torque to actuate the valve with the gear applied. i'll also have some points on the plumbing system to talk about during the meeting this sunday to do with a different set-up than the one we currently plan to employ

Damien's picture

I had an Anakin-Skywalker-before-the-pod-race moment last Tuesday at the lab, see for yourself!

I spent half an hour lifting various weights with the motor to pinpoint its moving torque, and finally I got to playing with the trim pot.

The video shows what happened after 2 x 1/4 turns. It's wooooooorking! ;o)

Now I have to modify the Arduino sketch to make the valve open/close completely. I'm afraid that if we increase speed we will lose torque, and even if it worked well in this test, we are dangerously close to the torque limit of the NEMA 18 motor. But for the moment, I will stick with it.

ryan.pulkrabek's picture

Great news!

We can easily go with a NEMA 23 motor, which is close to twice the amount of torque as NEMA 17. Additionaly, I can design a larger gear to attach to the valve, and a somewhat smaller gear to attach to the motor.

Do you need help modifying the sketch? We can upload the correct one for you to GitHub. Then you can play around with the speeds to find the limit.

ryan.pulkrabek's picture

Another thing about the larger gear of this assembly: it does not need to be a complete circle. If it only needs to rotate 90deg, then we can design it to be an arc of about 100 deg.