Just want to open up a thread for general questions about the Bosch seat motor in all Kits this year 6 004 RA3 194-06.
Firstly I’ll just go ahead and disclose that I work for Bosch and that my opinions represent my own and do not reflect official statements from Bosch.
This motor is good as a general purpose motor if you want a ready to go solution. I know the star output isn’t so FIRST friendly but there are some solutions to make the job easier by either broaching a 3/8” hex, adapting to a general socket, 3D printing an adapter or cutting down hex shaft into a square that fits the profile. Some tips are available in a screen steps presentation (linked below). Once you solve that problem most teams even those with few resources can make it work.
It compares with the window and snow blower motor however has more torque at lower speed. There’s also a built in hall sensor and a thermal protector if you over work the motor. Since it’s a worm driven gear system it resists backdrive very well.
Here’s a few general links to find more information:
From your suggestions, my team had actually tried to mill down a 1/2" hex shaft to a square to fit. We had a nice interference fit, but the issue really is the amount of material you end up removing. The result is a very small shaft that, in our case, ended up sheering under load (the motor never stalled, we got about 10 turns before it broke). So, the lesson there is don’t use aluminum shaft if you go that route!
My team has had good results, both last year and this year, using 3/8" hex shaft after drilling out the plastic spline and broaching. However, none of our applications have required the full torque capability listed on this motor’s data sheet. We have used it to adjust tilt angle for various mechanisms that do not have a lot of overhanging load.
After broaching 3/8" hex, the plastic coupling is fairly thin. Has anyone used this method and loaded it heavily; e.g., torque > 100 in-lb?
We’ve also had a good experience broaching 3/8". Not sure if it reached full loading but it was pretty close.
I also tried adding a 1/2" socket bit and filled the gaps with epoxy. Haven’t torture loaded but it was holding up for some mid load prototyping.
Just remembered we did some torture testing. Information is in last years thread about half way through. https://www.chiefdelphi.com/forums/s...sch+seat+motor. I should point out that this was using a rotary broach which doesn’t quite give you the flat sides that a linear broach provides.
We swapped out our prior solution tonight for the ½” socket bit option in the screen steps live link. it worked perfectly for our application. We don’t really need the max torque the motor can put out, but rather the form factor and speed are what makes it a good fit for this particular application. Being able to get this working tonight after having our milled square solution break on Tuesday was really great - it’s nice not having to wait for additional parts (if we were to go with the 3/8" hex option, as we don’t have any of that in stock, only 1/2") or switch to a different solution. Having these options presented like this really made it easy to understand and made my student’s evening!
That’s awesome! I’m so glad it worked out. Keep us posted if it holds up. I think we all know that feeling all too well when something fails and we must scramble for a solution. I think next year we should have an adapter available.
We are using it for a lifting mechanism. I found some 6mm square steel stock, and some 5/16" steel spacers at our local industrial hardware supply. The square is a super tight fit, is easy to weld onto any attachment, and spins nicely in the spacers welded to a bracket. Overall, this was probably the easiest custom motor to mount so far. Having the right parts is the key.
Thanks for sharing! Plastic does creep over time with heavy loads so I would suggest possibly filling the gap with JB weld or some other high strength epoxy if you find slop start to develop. Doubling up on the motors will help distribute the load a bit too.
We decided to use it to lower our gear holder to place it on the peg. We machined down a hex shaft to a square to fit tightly into the star output. It fits very tightly and we are very happy with how easily it lowers and lifts the gear holder with the gear in it. The encoder enable us to lower the holder to a specific degree to drop the gear on the peg or lean forward to receive the gear then come back to vertical. The only issue we are having is we are noticing that our encoder is losing counts and after repeated cycles of rotating up and down, it will not rotate down as far or rotate up as far.
In the automotive application for this motor there’s actually a 12V input to the hall so the peak to valley of the wave is more than you get with a 5V signal. I wonder if that could be the reason you might loose some counts. Can you keep track of your counts and compare with actual? Should be ~175 counts per output revolution. Maybe put a limit switch to zero and reset your reference so count losses don’t add up over time. Since I don’t have much experience on the programming side I would be interested if this is a big problem for other teams and if they found a good solution.
We have noticed the loss of counts also. In our case it appears to do well when speed is at 50% or less. Going above that led to increasing loss of counts. We are planning on a home switch to re-zero counts every time our tilt returns to the up position. Haven’t tested it yet, but I expect it to work out ok.
I wonder if the frequency of checking might be the problem. 20 RPM free speed is 1/3 rev/s with 175 pulses per rev = ~60 pulses per second. If your frequency of checking is any less than that you’ll probably miss counts. Not sure how your program is checking but if you have other calculations happening at the same time it might be part of the problem. Keep in mind I don’t have much knowledge of FRC code. Best solution in that case is just the mentioned zero each cycle or count separately on an Arduino or the like.
I opened up one of these yesterday and found something that’s pretty different than the cut-away on the WPI site. That one had a plastic gear broached in a star pattern, the one we have seems to be a plastic gear driving (by some kind of rubber damper) a metal plate with an output that looks more like a plus sign. Needless to say this is far harder to machine an output shaft for than the hex or square recommendations in the resources.
We considered taking the whole back plate off and bolting the mechanism directly to the rotating plate but I’m concerned about compromising stability and the potential legality issues. Another thought was to tap the weird pointed output on the exterior of the motor and use red locktite to connect a drive screw to it…though this is suboptimal…
Anyone have any ideas of (preferably machining-free) ways to adapt this motor’s output?
Yes…that’s the site I was referencing on my post. On the bottom of the page you can see that the plastic gear is broached in a star (the “two squares offset at 45 degrees”) output. Our motor isn’t like that: the output is a plus sign, not a star.