HELP! New team needs H-Drive help

Technical Technical Discussion
1

Hey CD
Our team decided to go with an H-Drive this year, but we are not quite sure where to start? We were thinking of buying the andymark Omni chassis upgrade and adding Omni wheels to the center wheels and adding a gearbox and frame to the center of the chassis. Is this a good start??? Any tips or cad helps!!

5740

2

I donā€™t want to burst anyoneā€™s bubble, but if youā€™re a (relatively) new team that doesnā€™t have a lot of experienced members when it comes to complex drive train programming, you might want to reconsider doing an h-drive. Unless youā€™ve come up with a design that is absolutely dependent on using an h-drive, you might be better off considering something else.
Take a look at this:http://www.simbotics.org/resources/mobility/drivetrain-selection
And if youā€™re still set on h-drive, come back and I can help you find some great resources!

1 Like
3

Seconded. Some things to consider:

  • Youā€™re having trouble just figuring out how to put it together and what to get.
  • You still have to program it.
  • Placement of the strafe wheel under or close to your CG is pretty critical.
  • Your strafe wheel really needs a floating mount, not a rigid one to deal with inevitable floor unevenness. (Yes, even on this ā€œflatā€ field.)
  • Your drivers have never driven one before, so you donā€™t even have a sense of how useful itā€™s going to be.
  • The time you spend figuring the above bullets out could be spent on designing a better gear mechanism or shooter, or just plain practicing driving the 6WD chassis youā€™d build instead.

My team has done experimental drivetrains during the build season before. It doesnā€™t work out well because you donā€™t have time to figure out all the issues in 6 weeks, the time you do spend troubleshooting it is stolen from the other more important functions of your robot.

4

Sorry I think Iā€™ve been calling it the wrong name! I missaid H-drive for a slide drive. Any tips on a slide drive?

5

When you say slide drive, what do you mean? If youā€™re referring to a chassis with 4 Omni wheels, one on each corner and in the same orientation with one or two Omni wheels near the center of the robot perpendicular to the rest of the wheels; then I think weā€™re all on the same page about what you meant.

6

If thereā€™s an omni wheel in the center to move it sideways, that would be an H-drive.

7

ok i am super confused right now. Im talking about the chassis that is like this with all omnis:

H

8

The picture is of an H-drive. Is this what you mean?

9

Yes, but we were planning on using the andymark omni upgrade chassis with all omni wheels

10

We did H Omni (full frame) wheels second year, and are returning to them in our fourth year for this gameā€¦ Our biggest issue was leveling all the omniā€™s so we turned to a center suspension which we will incorporate again this year we just used the standard kitbot chassis which we drill many holes through to lighten it. The center suspension is the trick, handles great and provides more needed stability.

Programming a bit tough but not insurmountable, its a good platform a driver delight, thatā€™s why we are going back to it this year.

11

Can you explain more about center suspension? Do you have any CAD for your previous design? Also, will you be using the andymark KOP omni chassis?

12

Ok, awesome. I wouldnā€™t if I were you.

H-drive, when executed successfully, can make an awesome drivetrain that combines speed and maneuverability better than mecanum, in my opinion. But unless you already have experience in building, testing and driving an H-drive, I wouldnā€™t do it. It is a very finicky drive system and the placement and weight placed on the center strafing wheel is of the utmost importance. To fully utilize the benefits of an H-drive, you would have to invest so much of your teamā€™s resources that the rest of your robot would be neglected.

On a completely unrelated topic, can anyone remind me how to post images with a reasonable size? Or are the handy little letters that dictate the size on imgur links only a feature of imgur?

13

We used a hinge on one side and a bolt with heavy spring through hole on the other, centering the wheels and adjustable pressure, we will probably go lighter than steel plate this time for center beam. So the suspension was one sided, but worked great and kept that center to the ground with really good contactā€¦hard to move laterally if bumped. So pretty darn stable and our fastest/most nimble drivetrain lots of power any direction.

just standard kit the two omnis fit right in if you remove the other wheels. Sorry no CAD that year.

Heres our 2015 reveal: https://www.youtube.com/watch?v=nmPpe_ZzDPY

14

H drive really isnā€™t as complicated to build or code as others are making it seem (we did it back in 2015), and I believe itā€™s probably well suited for the game.

With that said, Iā€™d recommend you just use mecanums instead. The floor is not flat, and thus traction on the center wheel is a serious issue. ā€œDead zonesā€, areas where you simply wonā€™t have the traction to slide, are not uncommon.

You need get around this by using a butterfly gearbox (I think thatā€™s what theyā€™re called? 987 used a custom one for their h drive in 2015), available here (http://www.vexrobotics.com/vexpro/all/new-for-2016/217-4824.html).

Also, keep in mind that relying on the single motor to the drive the lateral wheel will put a ton of stress on it, using two might be advisable.

Proper positioning of your lateral bar under your CV is nice, but not totally necessary since itā€™s possible to PID the robotā€™s heading if you have a gyro (this is also an issue with mecanum wheels I believe).

Anyways, if it were me Iā€™d probably go mecanum but if your heart is set on the slide drive then you have the tools and info to make it work

15

We used a Vex versaDrop on a grasshopper drive last summer.

We used chain to reduce the transverse wheel speed and dialed in the right down force by adjusting the air pressure.

If your transverse wheel is not centered, you can apply power to main wheels to compensate for your heading.

Dave

16

Messer,
Like many on this post, Iā€™m going to recommend you consider not attempting the H-drive unless absolutely necessary. A new drive should be something you work on in the offseason to allow your team builders and programmers the opportunity and ā€˜timeā€™ to get the concepts down.

Happy building.

17

For a First year team, I agree do a kitbot and KISS (keep it simple ___) our first year we were a defense goalie botā€¦Overbuild super durableā€¦ that got us to St Louisā€¦ second year is when we went H. Then Tank/track last year.

I think a good path is for first year teams should stick with KOP chassis just to see how it works. Its a great chassis. Very reliable and pretty effective. Then build on that in second year as you get a feel for what it takes to compete.

This is a good year for a defensive bot as the field is wide openā€¦ do one thing well as a rookie is my suggestion.