Is there a way to keep a 4-wheel omni base from being pushed and sliding short of adding another wheel to the middle of the other two.
Slide screws between the wheel and the roller
This is what is generally called having “locked omnis,” or a “bling drive.” It works great. But I would only lock one set of the wheels, unless your wheels make up a perfect square.
@Got a Screw Loose is there a diagram or picture out there of locked omnis? I’m having trouble picturing exactly what that is, or even what a slide screw is.
Slide screw isn’t referring to a specific type of screw. It is referring to the action of putting a screw between the wheel hub and the small rubber rollers. I used the old motor screws and they worked well.
You literally just stick a screw in the rubber rollers. I found that the old motor screws were too small to hold, but that could be due to my slightly more worn out wheels.
Here’s a picture, if it helps. I just took it from another picture of my robot and cropped it, so it’s not of the best quality.
If you are preventing your omni wheels from behaving as omni wheels, why not just save yourself the trouble and not use omni wheels?
Isn’t the rubber on the omni wheels softer/more grippy than the big rubber wheels?
Because the rubber on the Omni wheels is significantly more grippy than the 4" traction wheels. Alternatively, 6 wheel drives with 4 Omni wheels on the ends needs a grippy middle to center rotation upon. The 4" wheels are not exactly the same diameter at the Omni wheels, so they would not make sufficient contact with the ground, hence locked omnis.
Sooo…I should only do one pair of locked omnis on my bogie drive, or would it be beneficial to lock all the wheels?
Is the bogie drive 4 wheel or 6? I forget. Unless you have it perfectly tuned, locking all of your wheels will reduce turning speed. I would only lock one set.
This has been tested empirically in the past without quite this much confidence, although, IIRC, the omniwheels did have the best grip on the foam tiles by a slim margin. How did you test relative grippiness? If you haven’t, it would be a nice little project to work out a test plan, record your measurements and share them here.
I ran a couple of tests. I built two identical robots and had a pushing contest. The omnis won easily. (On foam) A little less scientifically, I simply grabbed two wheels and pressed them against foam with approximately the same amount of prsssure and tried to slide them. Granted I did not run quantitive tests, my general qualitive tests were pretty decisive. Would you mind linking a couple of these other tests you mentioned? I would like to ensure I am not making some radical error.
They were in the Lost Forum. It’s been a few years. I doubt you made ANY radical error, but doing the numbers makes this all seem more like science/engineering.
Agreed. Maybe I’ll look further into it after I fail to qualify for Worlds.
The problem in going from such a test to general grip of the rubber is that the omniwheels have edges that can catch against the foam when the foam compresses some. So if the omniwheel rubber even has a slightly lower coefficient of static friction against the foam than the grip wheel rubber does against the foam, then the omniwheel will grip better. But that’s also specifically on the foam. The tops of the platforms will not compress nearly the same as the foam, and the material is different, so it is quite hard to judge based on the earlier pull tests which will grip the platforms better. New tests are needed.
I found several earlier threads discussing friction coefficients of various wheel types. They have some clever tricks both how to measure static and dynamic friction and how to build the best drivetrain for your needs.
TLDR; (as several people stated above) there are different ways that wheels can grip onto the surface and it depends on the surface material a lot. Real life testing is always the most accurate way to find out what is going to work the best. Don’t be afraid to experiment…