Re: pic: Prototype Drivetrain
 

This is actually incredibly helpful, thankyou four your bluntness. I was toying with the idea of putting them on the outside, but thought that the advantages gained in turning speed would make this worth it because of the initial difficulty of control. If you guys had a hard time even with the wheels in a normal spot, I shudder to think how bad this will actually be. In the past years we have always had all sticky wheels, and this year with our drop-omni's was the first where we had omni's as far as I know. Your personal experience with this layout helps a ton, and I'll revise it hopefully this week, though A.P. Chem is looming...

sgreco, With the exception of the 8 wheels, is it still 2 omni's on the outside for that drive?

Re: pic: Prototype Drivetrain
 

OK... This is really starting to annoy me...

Seriously, if you want a zero turn radius robot, you don't need to put 2 wheels in the mid and hope this will work...
you just need that all the wheels will get the same amount of torque and speed, and that all of them will be in the same distance from the center of the turning radius ( Even if it is zero turning radius there is still a center-point) . This will apply for 4WD.
For 6WD you need that the 4 outer wheels will be just like the 4WD logic i said above and that the 2 center wheels will be exactly in the center between the two other wheels at each side.

Another thing, when you design a drive-train take to consideration the specific game, for most teams this drive-train will be a disaster for LUNACY and BREAKAWAY since in those games the space in the center of the robot was important for the mission. you want that your drive-train will take the least space as it can on the robot, and weigh as low as possible so all the other resources can go to the other robot mechanisms. But don't forget that the drive system should be as much maneuverability as you need to the certain game, easy to control for the driver, and that will almost need none program correction as possible (unless you MUST).

Re: pic: Prototype Drivetrain
 

No, they're all 6-inch plaction. The point I was making is that you don't "lose" maneuverability by putting all of your wheels on the outside. The two center wheels are lowered 1/8 of an inch (I think it's an 1/8, Chris can correct me if I'm off). I was basically just pointing out how the traditional style of 6/8 wheel skids, all wheels in line with each other and middle wheel dropped, works really well and has no issues with maneuverability when built well.

Re: pic: Prototype Drivetrain
 

You may not understand how difficult it is to fix drivetrain problems in software. We programmers can fix many things. A squirly drive is not one that is easy to fix (depending on how bad it is). Take our Lunacy bot. 4 wheel swerve, wide, geared fairly fast. It was really hard to drive, because it always wanted to go one way or another. Even with the trailer, it was still hard to keep it going straight. We attempted to fix it in software (and Jim wrote the code, he's a Chrysler programmer) and it did not work. There was too much lag between input and motion with the slippery wheels, and it was too hard to correct. For you to fix your drive in software, you would need to do PD or at least P on both sides to get them to exactly the same speed, tune it very well, and use the gyro to correct for drift. I don't think it would be possible with the wheels where they are.

Move the centers to the outside. It will work better, as many others have said.

Re: pic: Prototype Drivetrain
 

apalrd, trust me i know programing can be annoying...
I was in the programing team last year (LUNACY) and also the driver, and we also had trouble with our drivesystem (we used 2 regular drive on the read and 2 swerve in front that can go about 50 degrees each side to total of 100 degrees) and we couldn't fix all the problems in software so eventually i had to fix it with my driving...

For sure tank drive is the easiest to program and control so if you want to make the life of the program team and driver easier you should aspire to normal tank drive... Though if you aspire for maneuverability you should go for a swerve drive or mecanum (or even nonadrive like 148 and 217 had this year), there are some more configurations that have high maneuverability but they are less common in FRC (like KIWI drive)...

If you cant build a certain drivetrain that you haven't ever used yet before the season start, then don't try to do it within the 6 weeks.... our team tried to do swerve drive this year and we tried to build and program it within the six weeks and it was a total disaster... we eventually went to mecanum...
So my suggestion is that if you can build and test it before January go ahead, but as i stated before, there are many drive systems out there that have lots of maneuverability (from what i understand that is what you aspire for) so try using them instead of making a new system that you not sure that will work.

Generally i suggest that you will build it so all of us will learn a lesson. Many people including me don't think that it will gain what you aspire for but go ahead and prove that we are wrong... If not then what is FIRST about?

Re: pic: Prototype Drivetrain
 

Yeah, I hope to come up with 5 or 6 different drivetrains between now and September. Once we're back in school, we'll most likely build the most promising one or two.

Re: pic: Prototype Drivetrain
 

yeah, I'm the programmer this year.

We didn't have to correct for drive while driving, or autonomous. Since we have a 6wd articulated center, we just removed pressure on the center wheel so it touched under its own weight and that essentially made it a 6 wheel flat, so it didn't really want to turn. We had P control on speed for both sides, so it didn't drift at all (it never went faster than 4 ft/sec in auto, but only 1.2 when doing the kick portion).

I worked on our four-wheel swerve last year. 4 drive and 4 swerve motors, all independent, plus the annoyances of regolith. I only wrote the steering code.

Re: pic: Prototype Drivetrain
 

I have a feeling this idea will be shot down, but go ahead and try your design anyway. All this talk is very theoretical, and stuff sometimes works differentlly in theory that it does in reality. I realize that this is physics, and theoretical physics is rarely wrong, but there is always that chance that it might work better than expected with the right programming ("right" is not the same as "perfect")

Re: pic: Prototype Drivetrain
 

While trying this idea wont hurt anything, you really should take the advice of everyone on here and not waste your time building a sub-par drivetrain. Sure it will move but not with any form of control. The suspension is added complexity and points of failure for no added benefit as well.