Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)
-   -   H-Drive Strafing issues (http://www.chiefdelphi.com/forums/showthread.php?t=154921)

messer5740 07-02-2017 21:30

H-Drive Strafing issues
 
Hey CD
So we are experiencing traction trouble with our H-Drive chassis. When we attempt to strafe, sometimes the wheel just turns and doesn't catch on the ground. We were doing it on polished cement, so that may be an issue, but what to do you all think, what should we do?

Hitchhiker 42 07-02-2017 21:35

Re: H-Drive Strafing issues
 
When we did this in 2015, we found that on carpet, the robot sinks significantly into the carpet for the wheel to catch. You should probably try this out on carpet and see if this is the case. If it is, great! If it still doesn't catch, you may need to lower your center wheel.

Southpaw 07-02-2017 21:38

Re: H-Drive Strafing issues
 
Carpet will most certainly help, what wheel is your strafing drive?

Cothron Theiss 07-02-2017 21:40

Re: H-Drive Strafing issues
 
Is the strafing wheel at a set position? Is there any way you can increase the amount of normal force on that middle wheel? Carpet will almost certainly help. But it probably won't fix the problem.

messer5740 07-02-2017 22:58

Re: H-Drive Strafing issues
 
The center wheel has no applied force on it, but we do have the majority of our heavy weight items (battery, etc.) toward the center wheel. It is just an Omni wheel with rollers. We should be taking it up to the practice field soon

Joseph Smith 08-02-2017 06:31

Re: H-Drive Strafing issues
 
1 Attachment(s)
In 2015 our strafe wheel was in a spring powered suspension setup to keep it in contact with the ground. I don't know what your setup looks like, but it's very difficult to keep weight on a fixed center wheel without lifting up other wheels. One of the best methods I've seen is what 148 did in 2014 with their "rocker" module. Perhaps someone from their team can elaborate on it further.

GeeTwo 08-02-2017 08:15

Re: H-Drive Strafing issues
 
We tried strafe with H-drive in 2015, using a hard chassis mount. No matter how high or low we put it, it never worked right. Off season, we used a pneumatic cylinder to apply a steady force to it, and it worked first try. Springs can also work, I understand. The force on that wheel should be roughly between 20% and 35% of the weight of your robot, depending on how much strafe force you want and how much forward drive force you're willing to give up while you strafe.

messer5740 08-02-2017 09:59

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by GeeTwo (Post 1641936)
We tried strafe with H-drive in 2015, using a hard chassis mount. No matter how high or low we put it, it never worked right. Off season, we used a pneumatic cylinder to apply a steady force to it, and it worked first try. Springs can also work, I understand. The force on that wheel should be roughly between 20% and 35% of the weight of your robot, depending on how much strafe force you want and how much forward drive force you're willing to give up while you strafe.

Think it would work if we just ran it on the carpet? We are using a Vex versaframe H drive chassis

Chris is me 08-02-2017 10:04

Re: H-Drive Strafing issues
 
Is the center wheel rigidly mounted to the frame, or is it articulated? An articulated strafe wheel will do a much better job contacting the ground. One COTS way to do this with VersaFrame would be to use a drop drive module with only one wheel in it (I've never tried this, it just seems like it would work).

If your frame uses VersaBlocks, be sure you don't accidentally have the drop center in the wrong direction.

A picture would really help here.

pmattin5459 08-02-2017 10:44

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1641970)
Think it would work if we just ran it on the carpet? We are using a Vex versaframe H drive chassis

He said that with a hard mount, no. You need some sort of actuation. I would suggest the RoboWrangler way from the comment above if you don't want to use pneumatics.

GeeTwo 08-02-2017 11:13

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1641970)
Think it would work if we just ran it on the carpet? We are using a Vex versaframe H drive chassis

Quote:

Originally Posted by pmattin5459 (Post 1641993)
He said that with a hard mount, no. You need some sort of actuation. I would suggest the RoboWrangler way from the comment above if you don't want to use pneumatics.

Correct. It didn't matter whether we ran on bare floor or carpeting; we never got a useful strafe with fixed mounting, even when we tried a slot and use cable tension to hold the gearbox/wheel down.

messer5740 08-02-2017 11:20

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by GeeTwo (Post 1642007)
Correct. It didn't matter whether we ran on bare floor or carpeting; we never got a useful strafe with fixed mounting, even when we tried a slot and use cable tension to hold the gearbox/wheel down.

So you suggest Vex Versadrop?

Chris Hibner 08-02-2017 11:50

Re: H-Drive Strafing issues
 
When we did our H-drive in 2011 (which worked very well), our center wheel was on a simple suspension system. We put the center wheel on a swing arm and used a gas strut from McMaster Carr to provide the spring and damping. The trick was to size the gas strut to provide 1/5 of the robot's weight to the center wheel.

GeeTwo 08-02-2017 11:58

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1642008)
So you suggest Vex Versadrop?

That would be one way, and the only COTS design that comes to mind. Our off season version was home brew, and not at all optimized, more a proof of concept. In some way, you need to have the wheel able to move up and down, and apply a relatively steady force to hold it down as it moves. Pneumatics gives you the option of lifting the strafe wheel up (or at least not pushing it down) to maximize traction on your primary drive wheels.

EricLeifermann 08-02-2017 11:59

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by Chris Hibner (Post 1642017)
When we did our H-drive in 2011 (which worked very well), our center wheel was on a simple suspension system. We put the center wheel on a swing arm and used a gas strut from McMaster Carr to provide the spring and damping. The trick was to size the gas strut to provide 1/5 of the robot's weight to the center wheel.

If you don't have space for a gas shock constant force springs work well too.

bowmanb 08-02-2017 12:16

Re: H-Drive Strafing issues
 
We also used a pneumatic cylinder as a "spring" on our H-drive setup in 2015. We put the cylinder on its own air circuit and tuned the pressure to the point where we got the max down-force without lifting the "normal" wheels too much. ("too much" being a technical term meaning the point where things don't work correctly :) )

messer5740 08-02-2017 13:16

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by GeeTwo (Post 1642007)
Correct. It didn't matter whether we ran on bare floor or carpeting; we never got a useful strafe with fixed mounting, even when we tried a slot and use cable tension to hold the gearbox/wheel down.

What chassis did you use?

GeeTwo 08-02-2017 13:21

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1642044)
What chassis did you use?

We've modified and built atop the AM14U2 for three years (2015-2017). Does it matter?

Citabria42 08-02-2017 13:24

Re: H-Drive Strafing issues
 
What we did in 2015 is have 2 pneumatic pistons on either side of axle, to push down when needed with a single trigger pull.

messer5740 08-02-2017 13:27

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by GeeTwo (Post 1642051)
We've modified and built atop the AM14U2 for three years (2015-2017). Does it matter?

I think it might.... we are using Vex, so it may be more reliable (since it is a kit)

Mulcahy 08-02-2017 18:16

Re: H-Drive Strafing issues
 
We used a copy of 148's design in 2015. https://www.youtube.com/watch?v=uL-NnuZzOkU It is the easiest way to accomplish H drive... - no springs, no pneumatics, no extra controls, and it accommodates changes in floor height seamlessly. I wish I had invented it.:)

GeeTwo 08-02-2017 22:40

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by Mulcahy (Post 1642182)
We used a copy of 148's design in 2015. https://www.youtube.com/watch?v=uL-NnuZzOkU It is the easiest way to accomplish H drive... - no springs, no pneumatics, no extra controls, and it accommodates changes in floor height seamlessly. I wish I had invented it.:)

Your link is to a match not a design. What does that design mean, and how does it make H drive easier to accomplish?

Quote:

Originally Posted by messer5740 (Post 1642054)
I think it might.... we are using Vex, so it may be more reliable (since it is a kit)

I don't get your gist. OBTW, the AM14U2/3 is also a kit. Possibly not germane, but our off-season pneumatic strafe had a number of variables really screwed up because we were using what we had vice what was best for the application, but it worked first try. Barring further findings, I cannot and will not recommend anything other than some definite control over the applied force of the strafe wheel to the carpet, whether spring or pneumatics or synthetic muscle or cosmic pressure.

pmattin5459 08-02-2017 22:47

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1642054)
I think it might.... we are using Vex, so it may be more reliable (since it is a kit)

It doesn't matter. First off, both are kitbots, and both are somewhat reliable chassis. I like Vex better personally, more options for entirely COTs chassis. Anyways, you simply aren't going to get the right amount of compression between the strafing wheel and the ground without actuating it. The best way to do so without mucking about in custom manufacturing is to grab a versadrop module and slap it into the place currently occupied by your strafing wheel. Position it right, and you should be able to drop the strafing wheel by actuating a single piston.

At this point, though, if you want to delay it further, then continue to deliberate. But you've only got a couple weeks left, and if you want this thing driving well by the end of build, you're going to need to either order the parts you need very quickly and get it done, or just switch to a basic 2+2 or 6-wheel tank drive.

messer5740 08-02-2017 22:48

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by GeeTwo (Post 1642302)
Your link is to a match not a design. What does that design mean?



I don't get your gist. OBTW, the AM14U2/3 is also a kit.

Wait are we talking about the same thing here???? I'm talking about this-
https://content.vexrobotics.com/vexp...-top-trans.png
And andymark has a H drive kit??

pmattin5459 09-02-2017 04:41

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1642307)
Wait are we talking about the same thing here???? I'm talking about this-
https://content.vexrobotics.com/vexp...-top-trans.png
And andymark has a H drive kit??

Oh I thought you guys were using VersaChassis. Vex still sells that old stuff?

Anyways, you'll probably want to use that Robowranglers method of actuating the strafing wheels. Or switch to a tank drive. Given that you have your wheels in the corners, a 2+2 (2 traction in front, 2 omni in back) would probably work alright.

cbale2000 09-02-2017 08:37

Re: H-Drive Strafing issues
 
It is possible to make a ridged mounted H-Drive work, but not work well.

In 2015 we gave H-Drive a try without realizing the need for the center wheel to be articulated, by the time we were done we had offset the center wheel 1/16"-1/8" below all the other wheels on the robot, this resulted in some very odd rocking in the chassis and the strafe only worked about 75% of the time (because the carpet floors are never perfectly flat). It wasn't accurate enough to use during autonomous, but driving it manually it worked ok.

If I had to build an H-Drive again though I would definitely put it on a suspension.

GeeTwo 09-02-2017 09:27

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1642307)
Wait are we talking about the same thing here???? I'm talking about this-
https://content.vexrobotics.com/vexp...-top-trans.png
And andymark has a H drive kit??

Apparently not; you asked about the chassis, not the strafe mounting.

Getting the strafe wheel at the same height as the other four is NOT what you need. Even when we had about 1/16" of a drop on ours relative to the corners, it did not provide enough traction to move the robot well. By the time we got the drop deep enough to produce consistent strafing, we had problems driving in forward and reverse. There just is no right height for a rigid mount.

Peyton Yeung 09-02-2017 09:37

Re: H-Drive Strafing issues
 
2 Attachment(s)
Quote:

Originally Posted by GeeTwo (Post 1642414)
Apparently not; you asked about the chassis, not the strafe mounting.

Getting the strafe wheel at the same height as the other four is NOT what you need. Even when we had about 1/16" of a drop on ours relative to the corners, it did not provide enough traction to move the robot well. By the time we got the drop deep enough to produce consistent strafing, we had problems driving in forward and reverse. There just is no right height for a rigid mount.

I had moderate success with a fixed mount h drive in the fall of 2014 albeit the entire chassis was made of folded 1/8" polycarbonate. The flex in the PC acted like a suspension for all the wheels.

All in all I would definitely take a look at 148's robots both their 2010 and 2014 strafe wheel assemblies.

cbale2000 09-02-2017 10:07

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by Peyton Yeung (Post 1642422)
...albeit the entire chassis was made of folded 1/8" polycarbonate...

I hate to go off topic, but in all seriousness how did this drive not self-destruct immediately? We've used solid 3/8" polycarbonate plates for our drive system for several years and it barely holds up (it tends to crack a lot). We keep using it because of the weight savings but we have to be very careful with it.

GeeTwo 09-02-2017 10:13

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by Peyton Yeung (Post 1642422)
I had moderate success with a fixed mount h drive in the fall of 2014 albeit the entire chassis was made of folded 1/8" polycarbonate. The flex in the PC acted like a suspension for all the wheels.

polycarb = tough ≠rigid

Quote:

Originally Posted by cbale2000 (Post 1642434)
I hate to go off topic, but in all seriousness how did this drive not self-destruct immediately? We've used solid 3/8" polycarbonate plates for our drive system for several years and it barely holds up (it tends to crack a lot). We keep using it because of the weight savings but we have to be very careful with it.

I suspect you were using acrylic (plexiglass), not polycarbonate (lexan). See my post in the Quotes thread yesterday for a description of the difference.

messer5740 09-02-2017 10:22

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by Peyton Yeung (Post 1642422)
I had moderate success with a fixed mount h drive in the fall of 2014 albeit the entire chassis was made of folded 1/8" polycarbonate. The flex in the PC acted like a suspension for all the wheels.

All in all I would definitely take a look at 148's robots both their 2010 and 2014 strafe wheel assemblies.

So you didnt need to add suspension into the frame? We found ut if you drive forward slowly it strafes well.

Peyton Yeung 09-02-2017 11:53

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by cbale2000 (Post 1642434)
I hate to go off topic, but in all seriousness how did this drive not self-destruct immediately? We've used solid 3/8" polycarbonate plates for our drive system for several years and it barely holds up (it tends to crack a lot). We keep using it because of the weight savings but we have to be very careful with it.

It never ran against other robots but it drove well. Each wheel was a single reduction off of a CIM which kept the gear meshes simple. It was incredibly flexible (both a blessing and a curse) which made for constant contact with the carpet. Never had cracking with the 1/8" PC even with the cold bending for the flanges.

Peyton Yeung 09-02-2017 11:55

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by messer5740 (Post 1642440)
So you didnt need to add suspension into the frame? We found ut if you drive forward slowly it strafes well.

No but the entire frame acted as a suspension since the polycarbonate is flexble. Probably not ideal for a full contact games such as Steamworks.

messer5740 09-02-2017 13:44

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by Peyton Yeung (Post 1642488)
No but the entire frame acted as a suspension since the polycarbonate is flexble. Probably not ideal for a full contact games such as Steamworks.

So think we will be good once we put it on the carpet?

cbale2000 09-02-2017 14:14

Re: H-Drive Strafing issues
 
Quote:

Originally Posted by GeeTwo (Post 1642437)
I suspect you were using acrylic (plexiglass), not polycarbonate (lexan). See my post in the Quotes thread yesterday for a description of the difference.

I don't think so, unless the supplier screwed up and gave us the wrong stuff. We did waterjet it if that makes any difference.

Quote:

Originally Posted by Peyton Yeung (Post 1642486)
It never ran against other robots but it drove well. Each wheel was a single reduction off of a CIM which kept the gear meshes simple. It was incredibly flexible (both a blessing and a curse) which made for constant contact with the carpet. Never had cracking with the 1/8" PC even with the cold bending for the flanges.

I'm a bit surprised you were able to keep the gears lined up with the amount of flex I would expect it to have plus the weight of the CIM motors hanging off. Makes sense that it didn't drive in a competition though, I'm not sure even bumpers would save that frame from a game like this year or last. :rolleyes:


Quote:

Originally Posted by messer5740 (Post 1642553)
So think we will be good once we put it on the carpet?

The carpet will definitely help but I STRONGLY recommend not waiting till you get to your first competition to find out. Try to find someone nearby with a practice field or pick up a small section of carpet comparable to what's used on the field from a local carpet store. There are few things worse then going to a competition and finding out your drive system doesn't run like you expect it to.

Cothron Theiss 09-02-2017 15:40

Re: H-Drive Strafing issues
 
A picture of the robot in question would help us out a lot. I have built the exact same H-drive from the Vex Drive in a Day kit. It could barely even drive on linoleum or another hard surface. Driving on carpet was much more successful, but there's a large chance you'll still run into consistency issues. Now, you have a couple options open to you.
  1. Leave it as is. When you get to competition, you might be able to strafe; you might not.
  2. Take out the strafing module entirely. Then you'll just have a four wheel omni.
  3. Take out the strafing module and switch out your back two wheels for traction wheels.
  4. Lower the mounting point for the strafing module. You can do this in a variety of ways, from complex as welding the module to the current drive rails to as simple as getting clever with washer placement. This will mean there is more of the robot's weight on the strafing wheel, allowing you to actually strafe. However, this takes normal force off of your four main drive wheels. This means you'll have less traction when driving forwards, and your chassis may not sit flat on the ground.
  5. You can design and use some sort of actuating module like the VersaDrop or something completely custom to lower the strafing wheel when you need to strafe and raise it when you drive normally.
  6. You can design something like 148's actuating modules that use the the friction of the gearbox to provide downforce on the strafing wheel.

Those options are listed in rough order of difficulty. You and your team will have to decide which option is best for you.

alephzer0 09-02-2017 21:14

Re: H-Drive Strafing issues
 
My team ran an H-drive in 2015, and had this issue for a while. Our solution was to use a spring to pull the module down while simultaneously acting as a suspension allowing for the robot to not get completely shaken off-balance upon hitting bumps.
I agree that a picture or model of the drivetrain would be extremely helpful for us to see where a solution could be implemented.
Simply put, if I can't see it, I can't solve it.

Jono37 10-02-2017 00:29

Re: H-Drive Strafing issues
 
Our team is also going with a H drive this year. We have found that it isn't difficult to put a sideways module on a pair of articulated arms. The plates we're using are cnc'd but you can cut them out of an old AM14U if you need to. They just have to have the Toughbox's hole pattern. Here's a picture: https://scontent-syd2-1.xx.fbcdn.net...f5&oe=589F53C9
Hope this helps.


All times are GMT -5. The time now is 11:59.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi