Problems turning
 

Well we shipped the robot with 2 main problems, we are 5 lbs over weight, and the robot doesn't turn.

im working on a sollution to the weight problem, but i have a question about the turning problem.

if you watch this video:
http://video.google.ca/videoplay?doc...15467259458065

you can see that we have issues with turning, we have to take a run and then quickly reverse one side to get it to turn.

i was thinking as a simple and easy sollution would be to replace all 4 wheels with the omni wheels available from andymark. as well as premounting the wheels to aluminum sprockets from ifi robotics so that its a simple swap at the competition.

my questions are as follows:

will this help us turn?
what are the disadvantages?
how much weight can we expect to loose by swaping to the aluminum sprockets?

Thanks alot.

- Bochek

Re: Problems turning
 

Yes, it will help you turn, but you will get pushed sideways. A lot. A more cost-effective/easy solution would be to just switch out two wheels for omnis. The only thig is that you would pivot around the two wheels that still have sideways traction, but if you replace your back wheels, it will allow you to pivot around the front of your robot... which could be a huge advantage while positioning around a target such as a tube.You can reasonably expect to lose between a pound and two pounds by switching out the sprockets.


-Cody C

EDIT: Your drivers have learned to cope with the decreased ability to turn very well. Good job.

Re: Problems turning
 

Omni wheels will make turning much easier. The major disadvantage to that system is that while you can turn yourself with less effort, so can everyone else.

I would suggest only changing out either the two front or back wheels with omnis. This way you can still turn easily, but you can resist a little pushing from another robot. Note that your turning center will shift towards the non-omni wheels in this case.

The sprocket weight thing can be calculated based upon the size of your sprocket and the densities of the metals. If your current sprockets have large hubs, the weight savings could be significant

Edit: so a lot of this repeats Cody's post, but the sprocket calculations still stand.

Re: Problems turning
 

Bochek,

You didn't tell us what your drivetrain consists of, and your video doesn't give us much insight into it either. However, based on a previous picture you posted of your kitbot, I'm going to assume that you're using the standard kitbot setup with 4 kit wheels, a single motor Banebot 56mm tranny, and 2wd.

Your problem comes with the fact that your robot requires more power to turn than the single motor setup provides. The kit wheels provide a lot of traction in both the lateral (front/back) and transverse (side) directions. You need to either:
a) increase the supplied turning force.
b) reduce the required turning force.

You can solve (a) by adding more power with more motors, or by reducing the gear ratio to reduce your speed and increase your torque. With your existing weight problem, adding more motors doesn't help your cause.

You can solve (b) by reducing the transverse friction in your wheels. Most teams accomplish this by adding one set of "omniwheels" to their drivetrain. For 2wd setups, you would add the omniwheels to the non-driven wheels. Andy-Mark makes a nice omni-wheel, but they are not the only ones on the market. There are cheaping and lighter alternatives, such as ones found at McMaster.com or Omniwheel.com You can also reduce your required turning force by reducing your wheelbase, or powering all 4 wheels.

There are several whitepapers, as well as numerous threads on this site that explain this problem in more detail. Some simple changes should be able to be made at your competition to have you up and running in no time. Best of luck,

BEN

Re: Problems turning
 

The omni-wheel placement depends on where they want the center of rotation to be. If they wish to pivot around the front, place the omnis in the back. Vice versa for pivoting around the back. If they wish to pivot around the center (as well as reduce opposing robots ability to pivot them compared to other omni placements), the best solution would to be place the omni wheels on opposite corners of the drive.

Re: Problems turning
 

From what I've seen on other forums it would seem that all omni-wheels causes you to be pushed, I would suggest doing what li'l Lavery said and if you have a 4 wheel base, put on 2 omni-wheels in opposite corners of the robot, this allows for easier turning and you're still hard to push. On the aluminum bearings I am no help (sorry I'm an EE guy.)

(The wheel design looks like this the parenthesis are omnis and the straits are regular wheels.)


()____________________||
()____________________||
()____________________|| <----these are regulars
()____________________||
()____________________||




||____________________()
||____________________()
||____________________() <---these are omnis
||____________________()
||____________________()

Re: Problems turning
 

Since surface area has no bearing on friction, I would assume that this would be a bad Idea. By putting the two "sticky" wheels beside each other, you are reducing the amount that the wheel has to be pulled sideways when turning, and increasing the amount that it rolls along the turn. This decreases over all turning friction.

If you were to place them on opposite corners, it would do nothing to the amount that the sticky wheels have to be pulled sideways to follow the circle of your turning radius, and since One wheel has the same amount of friction as two, I wouldn't expect this to solve your problem at all.

Re: Problems turning
 

One of the best solutions to turning difficulties is very often overlooked.
Lower you gear ratio, significantly
It gives your robot the force to turn against friction, and saves energy while doing it.
Also, you will probably find it makes the robot much more controllable anyway.
This is just a suggestion, and seeing as it doesn't help you with weight either, might not be the best solution.

Re: Problems turning
 

sorry, the drivetrain is the banebots 56mm gearbox with a CIM and 4wd.

basically the standard kitbot chassis but with added sprokets and chain to make it 4wd.

so replacing the rear wheels with omni's could prove a good solution then?
and i would still keep it 4wd correct?

- Bochek

Re: Problems turning
 

Yes, that would keep it four wheel drive. My question is, how are you operating your bot. If you are running a single motor and gearbox, think about adding another and running it tank style, this makes turning much easier without purchasing the omni's. The only drawback would be your wheelbase, the larger the wheelbase on your 4WD system, then the harder it is to turn if with a tank drive system.

Re: Problems turning
 

Thanks for clarifying the drive train.

With that information, omniwheels are probably your best choice in this situation, due to weight and time contrainsts. If the weight of your robot is equally balanced between the front and the back (a 50/50 weight distribution) then it doesn't matter which side you put the omniwheels on. Basically, it would come down to the driver's preference as to which end of the robot gets rotated. I would suggest looking at some video of FRC176 this season. They are using omniwheels on the back of their robot, and you can see how it significantly changes the style of driving.

If the robot is significantly heavier on one side or the other, it is advantageous to put the omniwheels on the lighter side of the robot. With your traction wheels on the heavy side, you will transmit more power to the ground, and the robot will turn even easier.

Having the 4wd setup has already provided a huge step up in performance, as you probably realized during testing with the lighter kitbot setup. Mounting a chain and sprockets to some of the cheaper, generic omniwheels is a little more complicated than just a drop in solution. I feel that the Andy-Mark 6" omniwheel is your best solution, as those wheels are a direct replacement for the existing 6" wheels.

One other solution, which may not be widely accepted here, is to find an old 6" skyway wheelchair wheel. These are the old kitbot wheels which teams used in previous years. They are not as grippy as the new kit wheels, and could provide you the right balance between side slip and lateral grip. With the proper wheel hub, these are also a direct drop-in solution.

Best of luck.

BEN

Re: Problems turning
 

This will work well (we've used it in the past).

A inexpensive alternative is putting zipties on the rear set of wheels. Putting about 10-20 zipties on each wheel (equally spaced) will reduce the traction of those wheels. The more zipties, the less traction. As with Omniwheels, as your traction goes down it becomes easier to push you around.

Re: Problems turning
 

From experience with vex bots, it definitely improves turning. Also look at 494/70 last year.
Omni placement all depends on how you want your robot to drive and how you want your robot to react to defense.

Re: Problems turning
 

where you turning at full speed?
Try telling your programers to make the wheels spin in opposite directions when turning.

also you might want to consider sliding the wheels closer together to get smoother turning it really helped us out since we had a 4wd robot last year that jumped around like a bull

this year just with replacing tires and moving the wheels closer it turns like an RC car

Re: Problems turning
 

uhh we had a bad experience with zip ties in the Philly regional

In New Jersey they said it was perfectly legal; while in philly we ended up shaving all of the treads from our tires and waxing them every match becuase they said nothing mechanical could be added to the tires.

Re: Problems turning
 

This years KOP wheels are flat. We drive on a 3d surface. The edges of the wheels tend to dig into the carpet. The skyway wheels Are slightly v shaped and not as grippy. The tend to turn better. Unfortunately, having the power of 2 cims to a side and tank drive makes things much better. We are using skyway wheels this year. They are cheap and adapters are easy to make. We also bought IFI AL sprockets for weight.

Re: Problems turning
 

Sorry to hear that you weren't allowed to use them.

Last year I saw a good number of teams here in New Hampshire using Zipties.

As long as they can't damage the carpet, I don't understand why they can't be used. Can anyone post a rule as to why zipties cannot be used on the tires?

Re: Problems turning
 

If you are using the stock banebots gear boxes, the first thing that I would do is replace the existing planetary gear set in the stock unit to the 4:1 gear in the change kit (available from banebots.) This will give you a 16:1 ratio vs. the stock 12:1 ratio. This will not negatively impact your weight issue, and depending on how accessible your gear boxes are, this swap can be made in 10-15 minutes easily. In addition, if you have reduction between your gearbox shaft and the sprocket at the tires, you can change the sprocket sizes to further increase your reduction. You will be giving up some speed everytime you make a change. If you do decide to buy omni wheels, you may want to buy extras. Several teams have had issues breaking the omnis in a 4WD configuration. The kitty-corner wheel configuration may help the breakage issue. The simplest test to perform on a drivetrain is to, on carpet, push against a wall. You should be able to spin the tires without tripping the breakers. Some simple things to check are:

1. Are you using the right fuses (40Amp.)
2. If so, are the fuses tripping when you are turning? If so, you will have to make a mechanical change.

Re: Problems turning
 

You should definitely read Chris Hibner's white paper called "Drive Train Basics" which describes the math involved with creating a drive train that turns. http://www.chiefdelphi.com/media/papers/1443

Re: Problems turning
 

As with most things involving robots, there are tradeoffs. The omnis give you turning definitely, but they also give way in either the front or back. Try this:

If you take 2 omni's and put them on opposite ends meaning 1 on the front right and one in the rear left side of your robot, it reduces the lateral friction by 2 that you experience when turning. It also keeps a traction wheel in both the x and y axis. Im sure teams have done this before, but this is a fairly effective way of using omnis and keeping some of that traction.

Re: Problems turning
 

well as stated many times before, omni wheels. But i am going with a new idea for you guys to ponder on. I see that you have the kit chassis, so if you have a little extra chain and weight if possible for some more sprockets, i say add a wheel in the center of the ones and make it a 6wd chassis. The kit chassis is designed to have the middle wheel lowered for ease of turning. It will really help with your turning solution.

Re: Problems turning
 

We are running a similar drive setup this year as well, 4WD with 4 12:1 BaneBots gearboxes, with 15/24 reduction giving us 19.2:1 at the wheels. We are running 6"x1" IFI tractions in the rear, and AM omni-wheels in the front. Our wheelbase, however, is much shorter than your's as far as I can tell in the picture.

Re: Problems turning
 

Dom,
This has come up before but I do not understand the inspector's decision without talking with them. As long as the mods meet the rules they should have been allowed. I have recommended this fix for teams who were having a problem and couldn't find a fix quickly. Tywraps obviously are allowed since they are supplied in the KOP. More accurately the rule is...


<R34> ROBOT wheels, tracks, and other parts intended to provide traction on the playing field may be purchased or fabricated (“traction devices” include all parts of the ROBOT that are designed to transmit any propulsive and/or braking forces between the ROBOT and the playing field). In no case will traction devices that damage the carpet or other playing surfaces be permitted. Traction devices can not have surface features such as metal, sandpaper, hard plastic studs, cleats, or other attachments. Anchors (i.e. devices that are deployed/used to keep one’s ROBOT in one place and prevent if from being moved by another ROBOT) can not use metal in contact with the carpet or other playing surfaces to “stay put.” Gaining traction by using adhesives or Velcro-like fastener material is not allowed.

As long as the lock part of the tie is positioned to the inside of the wheel, there will be no "hard plastic stud" in contact with the surface. A little hot glue to hold it in place doesn't hurt either. There is other possibilities as outlined above but this does provide a quick fix.

Re: Problems turning
 

Note that the center wheel does not need tot be powered for this to work.

It will definitely not help with your weight issue. 5 pounds is far too much to be had by putting holes in aluminum.

What is the material used for the ramp surface? It looks like aluminum with holes - if so consider a much lighter material, like 1/8" aviation plywood (sold at hobby shops). Far lighter, brutally strong. Or somehting like that.

1/2 pound per ramp can be lost by holing the ramp frame

We lost 1/4 pound in wire this year by just cutting things to length.

good luck with that

Don

Re: Problems turning
 

Ahhhh the four wheel pneumatic drivetrain...
Such a horrible idea...
We used that last year and never will again.
Broke so many chains jerking around the field.
We installed castors but they only make the jerking smoother.
My advice, use two omni-wheels, its the only way to save yourself.

Re: Problems turning
 

Not true. One wheel with 60 pounds on it has the same amount of friction as 2 wheels with 30 pounds each. However, one can assume that the weight of the robot will be approximately equally distributed across the four wheels, or at least the weight on the omnis will be the same as the weight on the traction wheels. Therefore, the traction wheels have only half as much turning friction as they used to.

Another solution would be to move your wheels closer to the middle of your robot. This increases turning capacity without decreasing traction. However, it causes balance problems, as your wheel base is shorter. Add a few trick wheels or casters that hover just above the ground to fix this problem.

Re: Problems turning
 

Re: Problems turning
 

Yes.

While the second scenario you illustrated will get you closer to a "zero friction" turn, it also changes your turning radius from dead center in the robot to inside of the axis between your back two wheels.

Putting the omni's in opposite corners still gives you the problem of turning scrub, but it will significantly reduce the problem as half of your total weight is being distrbuted along wheels which experience little to no axial frictional force. For most situations, this should be enough to kill the bouncing robot problem. This also allows you to keep your 0 degree turn radius, and keep a solid wheel on the ground on each side of the bot at all times.

So really, its all up to driver experience and preference. I have been told that drivers who like to use one stick drive will be much better on a "power slide" bot (omnis on same side) and drivers who prefer two stick will be much better with a 0 degree turn radius.

Re: Problems turning
 

well today we ordered 3 andy mark 6" omni wheels,

the reson for 3 was incase one breaks.

we will be replacing the 2 rear wheels with the onmi wheels.

does anyone know the difference in weight between the ifi aluminum sprokets and the steel KOP wheels?

- Bochek

Re: Problems turning
 

The KOP wheels weighs .8 lbs. I believe you were trying to ask about the sprocket that was provided in the kit. If you are wondering if it would help you lose weight when you switch those steel sprockets to aluminum ones, yes you are on the right track.

Also, you are already going to lose .6 lbs by planning on swtiching those kop wheels to AM omni wheels. Those AM omni wheels weigh 5 pounds.

If you need help figuring out how to lose more weight and if you have detailed/close up pictures of the robot, feel free to contact me. I will be more than happy to help you guys and give some suggestions.