Bearing Block Question

[ttldomination]

Hello Everyone,

So, this year, my team ran a standard 6WD with #25 chain and things were never really smooth. We seemed to pop a chain about every other match, and we believe we've isolated the problem to our bearing block setup.

We use 8020, so we're somewhat limited to using standard pillow blocks mounted directly onto the metal. We attempted to drill holes directly into the 8020, but without proper tools, drilling offset holes is nearly impossible.

The pillow blocks we used were somewhat low quality sleeved blocks. And the sleeves could rotate, which allowed for the axles to sit at strange, unwanted angles.

So here's where the questions start flying.

One side of the argument is saying that the low quality, sleeved blocks were our problem. Perhaps something like this block would help with stability and keeping the axles lined up.

The other side of the argument is that the blocks were never really that much of an issue. Because of the force that a 6wd experiences while turning, the bearing blocks themselves would physically slide on the 8020. An issue that will occur regardless of the block we used, and we need to seek solutions to meet that end, even if it means going back to 4wd.

So, we're really stumped, and while both sides have their merit, we've never really been in this kind of territory before. This past year was the first year that we've run a 6wd system, so we're not entirely sure how to move forward.

Thanks for the input,
Sunny G.

[MrForbes]

You might look into building the frame out of something besides 8020. There are a lot of ways to build a frame.

[jreuter]

1073 has used a similar approach for several years, using an 6WD with 8020 chassis. Yes, we occasionally need to retension the chains when the blocks slip, but this is typically only once per competition. We use 35 chain, and have not had any breakages.

The only difference I see is that we use non-rotating axles, with bearings pressed into the wheel hubs, rather than rotating axles. Our shaft support blocks are home made, but like yours have two bolts to hold them down to the 8020. You do need to pay attention to making sure the axles are not at an angle, but a combination square works OK for us.

As a possible improvement, consider mounting a flat plate in front of the mounting blocks, using more 8020 inserts. Now you will have 4-5 bolts holding the block in place, not just 2. It's not as good as a through-hole mount, but is worth a try.

[JesseK]

A potential design is to use dead axles with 1/8" angle aluminum as axle mounts beneath the 80/20. You'd eliminate the low quality bearing blocks while also moving to more efficient bearings on the wheels.

We used this setup in 2011, and were able to easily drop the center wheel using a shim. We also had to account for welding imperfections, and do so by treating each wheel/mount assembly as an adjustable block. Each wheel had precise spacers between the wheel bearings and the angle aluminum mounts, which meant that once we mounted the two angle pieces, the wheel would never shift side to side (thus throwing off the chain). The angle block didn't perfectly line up with the frame members, but the sprockets did perfectly line up with each other. You can account for your own imperfections with 80/20 by drilling the hole in the angle aluminum relative to the 80/20 channel, rather than attempting to drill through the 80/20.

Once the sprocket alignment is set, you really just need a way to make sure the bolts don't slip in the 80/20 channel. This can be achieved with an imperfect hole through the 80/20 at the right spot -- a hole that doesn't need to go through your wheel mounts.

[jwfoss]

80/20 is a great material to work with but it does come with some difficulties as well. An option would be to use some flat plate with your axle holes or bearings and mount to the sides on the 80/20 rather than the top of the rails.

If you are not totally set on 80/20 for your chassis, and dont want to spend any extra money, you may want to consider throwing together a 6wd on the kitbot. Integrating an 80/20 superstructure onto a proven chassis has worked for a large number of teams over the years.

There are a number of things that would cause #25 chain to fail. From proper tensioning to misalignment in the sprockets. Be careful with these things, but remember that although it is heavier, #35 chain is also more forgiving.

[JamesCH95]

Originally Posted by jwfoss

There are a number of things that would cause #25 chain to fail. From proper tensioning to misalignment in the sprockets. Be careful with these things, but remember that although it is heavier, #35 chain is also more forgiving.

Came here to say this, #35 series chain is wonderfully tolerant, I wouldn't use anything else.

In 2011, and most other years for that matter, we used #35 size chain. We simply set our sprocket locations, made the chain as close to the right length as we could (sometimes using half-links), and were off to the races. No tensioning, no thrown chains, it proved to be very reliable.

[Jon Stratis]

My teams first year, we used 8020 for most of our robot... except for the main frame. We stuck with the kitbot frame, which is designed for a 6 wheel drop-center drive train. It's actually pretty easy to attach 8020 to the frame, too - you can add vertical members by simple tapping the hole that goes through the middle of the 8020. We're the robot on bottom in this picture:

http://www.chiefdelphi.com/media/photos/27516

It just goes to show how durable 8020 is... a puny little frame held up 2 robots... and one of them was on that thin little 8020 arm!

[rcmolloy]

Quoted from Phalanx from Team 1089

Chain pitch is either .250 (#25) or .375 (#35)

Formula for Center to Center Distance of Sprockets:
(B1/8) * (2*B4-B2-B3 + SQRT((2*B4-B2-B3)^2-(8/3.14159^2)*((B2-B3)^2)) )
B1 = chain pitch
B2 = Sprocket 1 teeth
B3 = Sprocket 2 teeth
B4 = # Whole Links

Formula for Number of WHOLE Links:
(2*E4/E1+E2/2+E3/2+((E2-E3)/(2*3.14159))^2/(E4/E1))
E1 = chain pitch
E2 = sprocket 1 teeth
E3 = sprocket 2 teeth
E4 = Center distance

These two formulas should help you actually determine the correct length for the perfect chain to not have tensioning. It should, somewhat, help your chain popping problems. Many teams run #25 chain for their drives and its possible to do. The right calculations just need to be brought in if not done so already.

[Peter Matteson]

Originally Posted by JamesCH95

Came here to say this, #35 series chain is wonderfully tolerant, I wouldn't use anything else.

In 2011, and most other years for that matter, we used #35 size chain. We simply set our sprocket locations, made the chain as close to the right length as we could (sometimes using half-links), and were off to the races. No tensioning, no thrown chains, it proved to be very reliable.

I wanted to 3rd this. We have (almost) always used #35 just because it's more forgiving. It's less to worry about and the weight doesn't kill us.

[sdcantrell56]

Sunny send me a message on facebook or email me and let's figure out how to get you guys taken care of. We definitely have some resources that could fix your issues.

[Garrett.d.w]

This year for a preseason build we had the group of newcomers build a 6 WD kitbot. They followed the model outlined in the Kitbot on Steroids video. This worked surprisingly well. They had it running in less than a day.

We plan to do something similar for the actual season simply because it will get you a working drivetrain in one day.

If you want to stick with 80/20, then you can place a rod in between the transmission and bearing blocks (this fits nicely in the slots and can be zip tied to keep it in). This anchors them relative two each other, which is all that you need. We did this a while ago when we built a small 4wd robot to practice driving against. After several impacts, the blocks had moved about a half inch, but we still could drive because everything was still in place relative to each other.

[Dick Linn]

Originally Posted by ttldomination

... The pillow blocks we used were somewhat low quality sleeved blocks. And the sleeves could rotate, which allowed for the axles to sit at strange, unwanted angles...

Are you using single, self-aligning bearings, or just something sloppy enough that you get radial play? How far are things cantilevered?

[Chris is me]

If you didn't drill through and didn't double your T-nuts, I'm going to guess it was sliding in place on the 80/20.

I would use 35 chain with pretty much any 80/20 drivetrain. If you're worried about weight, you really shouldn't be using 80/20 extensively like that, in my mind at least.

[JimC]

Along with the all the great ideas here for mounting the bearing blocks, we have had great results using Roll-Ring chain tensioners.

We have had the blocks move, but the tensioners keep the chain from jumping off.

http://www.renoldjeffrey.com/nmsrunt...=1046&sID=2750

You need to have enough space along the chain to use them, but it seems to work in most situations.

Jim

[ttldomination]

Originally Posted by Dick Linn

Are you using single, self-aligning bearings, or just something sloppy enough that you get radial play? How far are things cantilevered?

Yes, you called it. The wheels aren't cantilevered at all. We have two blocks on each side of the axle with 3 inches between them.

Originally Posted by JimC

Along with the all the great ideas here for mounting the bearing blocks, we have had great results using Roll-Ring chain tensioners.

We have had the blocks move, but the tensioners keep the chain from jumping off.

http://www.renoldjeffrey.com/nmsrunt...=1046&sID=2750

You need to have enough space along the chain to use them, but it seems to work in most situations.

Jim

That's a great looking product.

It seems to me that the easiest way to eliminate the problem would be to follow some of the suggestions and statically mount the bearing blocks with a couple of through holes and then use nice tensioners to tension the chain.

- Sunny G.