# H-wheel Question

Hi all,

This is our first time going with an H-drive system and we are having a problem with our center wheel for our drive losing traction to the ground preventing us from strafing. It is currently being tensioned towards the ground with ~20 lbs of force using constant force springs.

Sometimes the wheel grabs and we move, but other times it completely slips and we go nowhere. Are we using enough force? Any other thoughts?

Thanks for the input!

It really depends on the weight of your robot, including whatever you’re carrying. If your robot weighs 120lb before battery and you’ve got a tall stack of totes and a battery, you may very well weigh 150lb. In this case, the 20# corresponds to only about 13% of the robot weight. If you’ve got wheel alignment problems or other friction in your system, this may not be enough to get moving. The more force you put on those springs, the more acceleration you can get in strafing, at a commensurate loss in your tank drive.

If you are more likely to slip when carrying a load than when empty, this is quite probably your problem.

Seeing as you are using a spring to force the wheel down I am going to guess it is safe to say that you have a pivot and some sort of hard stop. It is possible that the wheel needs to come down further for more “contact” with the floor. If possible make the stop adjustable so you can lower it down a bit more.

Also is the problem more of an issue when loaded with totes or does it matter loaded or unloaded ?

We had the same issue and a 1/16" drop in the strafe wheels did the trick.

Should have added that if your H wheel can go slightly lower that your regular drive wheels you shouldn’t have a problem with the H wheel unloading the drive wheels as long as the rest of the bot weighs more than the 20# of force you are applying to the H wheel.

Without seeing the setup you have this is all speculation, and should not be taken as fact. Only testing will give you the true answer.

Yes!

Also, does the surface make a difference? If your wheel is a bit high, you may run just fine on carpet, but not a smooth (e.g. tile) floor.

Does the direction of the strafe make a difference in the slipping? The geometry of the pivot may be such that the strafe wheel tries to push itself under the pivot in one direction, increasing its loading and giving it more traction.

The wheel is fully contacting the ground. When we lift the robot it can drop another half inch before bottoming out so I do not think the geometry is the issue.

As far as how load affects it, we seem to have the same issue loaded or unloaded with a stack.

All of the testing was done on a pretty large section of FRC field carpet.

This year was our first attempt at a H drive as well, although we went with more of a ladder drive.

We considered a drive like this, but discarded it because it appears that the two strafe wheels separated on a single axle would provide too much resistance to tank mode turning. Has this been an issue, or do you have some sort of differential working between the two strafe wheels?

Or (I just noticed the pneumatics) perhaps you can lift the strafe wheels completely off the carpet for tank-mode driving?

This is the setup on our practice bot. The competition bot has a better bracket system but I don’t have any examples to give.

do you lift those “ladder drive” wheels when not needed?
IF not, how is your turning?

Yes we can lift the drive as needed. We went this route for two reasons… The one you mentioned about tank drive, and the other so there would be no issues driving over the scoring platforms.

I’d warn against running a system like that; we used a fixed setup with a drop during the 2014 competition season, and was probably in my opinion our robot’s biggest drawback.

In the offseason, we actuated the center omni wheel using pneumatics to ensure we always had contact with the ground. If you don’t actuate it and just fix it in place, a few problems arise: you either mount it too high up, and because the carpet is not entirely level (in a perfect world, it would be, but sadly it’s not), there are a few patches where you will get stuck. If you mount it too low, it will always touch the ground, but not only will it cause ridiculous amounts of friction (slowing your acceleration and speed in the forwards/backwards direction), it will take a toll on the batteries as well. You have to take the bump of the scoring platform into account as well.

Just because the bot is able to strafe in your shop doesn’t mean it will necessarily strafe out on the competition field, as we found out.

As already mentioned, more force may be needed. Many teams utilze pneumatics to raise/lower the “H” drive wheels, thus giving it more down force. Also, the full raising of the wheels eliminates drag and wear. Omni’s do have drag, especially when at an angle not inlign with the rollers. Also the omni rollers wear quite a bit on the carpet. We had a few omnis replaced last year due to “Flat spotting” a few rollers . It’s just the nature of the beast.

roger

The rollers do not have bearings. The axial force on the rollers pushes them against the stops and creates a lot of friction. Same is true of the affordable mecanum wheels used in FRC

. This is a major factor in the “non-idealized” behavior of omni and mec in FRC
.

AndyMark calls them bearings, but I agree that they are more properly bushings. The more traditional omnis with a lot of short cylindrical rollers have brass bushings at the ends of the rollers, and the new DuraOmnis with the long tapered rollers have nylon bushings. In both cases, these rotate on steel shafts.