Modeling Gears/Chains for Drivetrain

[drivetrainUW-Platt]

I'm attempting to model gear and roller chain in Solidworks and am finding the files I have recieved that allow for modifying chain and tooth counts to be frustrating. I was wondering what others have used for modeling gears given x chain and x teeth.

I want to design for the least amount of movement in the differential uprights for chain tensioning and I fell that by solid modeling with correct imputs can give me a very good idea of where things will sit.

To be specific I'm looking for a 12T and a 45T gears and what I believe to be 520 O-Ring Roller Chain(I will do some more precise measurements and compare to standards).

I am trying to get away from turnbuckels and incorperate an eliptical shaped insert into the upright that allows for holes drilled at different ofsets. To tighten the chain one would swap in an insert that moved the upright slightly farther towards the rear of the car. Any opinions on this idea would also be welcomed.
Thanks

[drivetrainUW-Platt]

I'm attempting to model gear and roller chain in Solidworks and am finding the files I have recieved that allow for modifying chain and tooth counts to be frustrating. I was wondering what others have used for modeling gears given x chain and x teeth.

I want to design for the least amount of movement in the differential uprights for chain tensioning and I fell that by solid modeling with correct imputs can give me a very good idea of where things will sit.

To be specific I'm looking for a 12T and a 45T gears and what I believe to be 520 O-Ring Roller Chain(I will do some more precise measurements and compare to standards).

I am trying to get away from turnbuckels and incorperate an eliptical shaped insert into the upright that allows for holes drilled at different ofsets. To tighten the chain one would swap in an insert that moved the upright slightly farther towards the rear of the car. Any opinions on this idea would also be welcomed.
Thanks

[Denny Trimble]

I'm such a nice guy... (http://students.washington.edu/dennyt/fsae/520sprocket.SLDPRT)

This model (SW 2004) has a "full tooth" profile, but most sprockets have shortened teeth.

http://www.gizmology.net/sprockets.htm

Also, make sure you consider the resolution of your chain tension adjustment, as well as the range. Last year I made some eccentric bearing mounts for the diff, and the range was good, but the resolution wasn't fine enough.

[rjwoods77]

I have never had any luck with eccentrics. I have had real good luck with screw adjust or spring loaded chain tensioners. Thats not to say eccentrics dont work good because they are on almost every atv(but atvs have a foot worth of chain guides in addition) there is but my personal experience is to use tensioners. I like them because they help cushion drivetrain snap but that was what was needed for baja and might be completely different from formula. If you do eccentrics, make sure that you have the ability to make sure the eccentrics eccentrate in unison. Cocking is a possibility and that is the usefullness of dentents but I beleive that is the limitation Denny was talking about. Only so many detents you can have.

[drivetrainUW-Platt]

thank you for the gears.
here it what I'm talking about, the ofset bushing are seen in red at the top of the differential uprights, the hole would be drilled at various places for different tensions, and they would be in sets so that it wouldnt be off from side to side

http://i22.photobucket.com/albums/b309/duwem/dtrainmockup.jpg

[drivetrainUW-Platt]

what do you guys think about that bushing idea?

[Denny Trimble]

I don't see a problem with it, we use similar inserts for camber adjustment on our uprights.

Just make sure the distance from the lower pivot bolt to the hole in the insert is the same, either by having curved inserts or holes at different vertical offsets from the centerline of a straight insert.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by drivetrainUW-Platt:
what do you guys think about that bushing idea? </div></BLOCKQUOTE>
Workable, but here are two alternatives.

1. You use four longitudinal bolts to attach the diff carrier to the square tube frame between engine and diff. So the top and bottom rails of the frame each have two fore/aft holes in them, preferably with a "plug" welded into the SHS, and the diff carrier has longitudinal holes where the red inserts are, and also at the bottom. Then you use four equal length spacers (drilled aluminium round bar) to push the diff carrier back to tighten the chain.

2. I also like the idea of an idler sprocket on the bottom chain run. The idler sprocket can pivot on one bolt and adjust via a second slotted hole/bolt, or two slotted holes and bolts, or other... - the idler has less load on it than the diff carrier, so adjustment is easier. This quietens the normally loose bottom chain, allows the diff carrier to be mounted much more rigidly, and doesn't change driveshaft angularity with chain adjustments.

Up to you? http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[Storbeck]

I've always wondered why more teams don't use an idler sprocket rather than make thier whole diff, and often brake system moveable. It seems like it would be simpler.

[Mark TMV]

because of the mechanical losses associated with idler sproket.

[rjwoods77]

We are using a cvt which makes a jackshaft necessary. Also we need to make the c-to-c adjustable for belt wear and differetn c-to-c belts. After dealing with the turnbuckle deal on the old car I though that rotating/hinged diff brackets were really not a great idea. We cant put a tensioner on a cvt belt but we are going to modify the c-to-c distance by placing shims(ala gm double a-arm circle track) under the bearing blocks (the clutch centers are at the same height from the ground so it is easy to know how much to shim) or we might put the engine on sliders (ala go-karts). Our final drive train is using a tension spring plastic bearing roller (ala sprint cars).

Why would you want something that has the most torsion/bending in the drvetrain to be moveable. Lock it down and adjust it somewhere else. The chains slack side is great because there is very little force on it unless you get reversals like Denny mentioned in a previous post. Even then they shouldnt be that bad if you limit the amount of distance the slack side has to gain speed on the snap with a tensioner. I disagree that tensioners of a good design have any real losses to them. There is no need for a sprocket. A roller or even dragging the chain along a guide will cause only the slightest of losses. In baja, a slack side tensioner quickens the already painfully slow response of the engine. Less loose stuff to load up before you are going straight again. The real nice chainguide strips actually ride on the rollers of the chain and not the side plates. On and on and on.

Other reason for tensioner:

Can adjust tension in a fraction of the time over eccentrics, spacers or bushings.

If you ever change chain or gearing you have to go through that process over and over instead of rotating the spring loaded tensioner(ala all fwd car timing belts) or turning an adjusting screw a couple times.

Chains do strech, wear, break. If they do you can replace very quickly with a tensioner.

Just cheaper and simpler to do.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mark TMV:
because of the mechanical losses associated with idler sproket. </div></BLOCKQUOTE>

How much? 0.1 hp???

Z

[Jersey Tom]

We only put out 1hp so .1hp loss is pretty substantial. In an effort to lose weight we're using a leaf blower engine. Turbocharged, obviously.

[Bowtie Man]

I looked at doing the idler chain tensioner lastyear and noticed that under engine braking the spring would have to be extremely strong to keep tension on the chain (providing you're tensioning from the bottom up) and if you're tensioning from the top down again the spring would have to be huge.

One thing I'd make sure you do with the insert idea is make sure you drill the holes along the arc of rotation.
As well I would consider pivoting from the top and adjusting the bottom so that in the event of the systm loosening the chain would still be in tension under acceleration.

Luc

[TG]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Denny Trimble:
I'm such a nice guy... (http://students.washington.edu/dennyt/fsae/520sprocket.SLDPRT)

This model (SW 2004) has a "full tooth" profile, but most sprockets have shortened teeth.

http://www.gizmology.net/sprockets.htm

Also, make sure you consider the resolution of your chain tension adjustment, as well as the range. Last year I made some eccentric bearing mounts for the diff, and the range was good, but the resolution wasn't fine enough. </div></BLOCKQUOTE>

ha, now I see why all of your awesome renderings have that bronze light in them. Anyways, do the teeth have any specific profile to them? All I can see is that you put in a spline between two points and put a tangency relation to the round. I know you can do a curve by table for involute curves (I have a simple table like that but it isn't on a server). Thanks.

[drivetrainUW-Platt]

First, lets keep baja outa this, this is formula sae not forumla minivan.....

I will have the holes machined in that insert in some sort of an arc to account for the pivot point.

I like the idea of having the inserts on the lower tension side of things, and hopefully if things work out, we will have duel outboard brakes so there wont be as much direct braking force into the differential and the chain.

MSOE did a nylon idler and had no success with it from the pictures I saw on there website, I believe they eventually went to a gear.

As far as an idler goes, you would need a strong spring and that requires more mounts and if you want it to be fixed instead of on a spring you come across the same problem of making it adjustable and staying where you want it to be.

The amount of time that our cars are actually driven at competition, they chain shouldn't stretch enough that it would need to be retensioned.

I think this is a much simpler approach to the "western australia" essentric rotating bearing assembly.

[rjwoods77]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by drivetrainUW-Platt:
First, lets keep baja outa this, this is formula sae not forumla minivan.....

I will have the holes machined in that insert in some sort of an arc to account for the pivot point.

I like the idea of having the inserts on the lower tension side of things, and hopefully if things work out, we will have duel outboard brakes so there wont be as much direct braking force into the differential and the chain.

MSOE did a nylon idler and had no success with it from the pictures I saw on there website, I believe they eventually went to a gear.

As far as an idler goes, you would need a strong spring and that requires more mounts and if you want it to be fixed instead of on a spring you come across the same problem of making it adjustable and staying where you want it to be.

The amount of time that our cars are actually driven at competition, they chain shouldn't stretch enough that it would need to be retensioned.

I think this is a much simpler approach to the "western australia" essentric rotating bearing assembly. </div></BLOCKQUOTE>

Mounts shmounts!! Your mount could be that lower engine support bar. You could use a spring loaded linear tensioner if you wanted. Would be real easy. Sorry to bring it up baja again but the chain has pretty decent intial stretch. Besides, the we shouldnt have to do this or that argument is the one that always bites you in the ass.

[Wright D]

I bought an old drafting book from a used book store a few years ago, and it is awesome. In the book the authors out lined how to draw all sorts of different gears and chains. The text I picked up cost five bucks; there are lots of these types of books around. Since the advent of cad no one wants them. However; the information they contain is still vary relevant. I used my text to draw our sprockets and chain, as well as the gears in the diff, and our sump pump drive.

Pictures of my efforts can be seen here:
http://formula.engr.arizona.edu/2005.html

When modeling a chain; if you take the time to draw the sprocket yourself, then you can use a tool called feature driven pattern to speed up the assembly time.

[Bubba]

We CNC our rear sprockets. I used the Machinery's Handbook Vol. 26 (probably any volume will have the same information) to properly draw the tooth profile, then created a simple circular array from that. The alignment and seating of the chain rollers between each tooth is still perfect on our 2005 sprocket.

As for modelling the chain properly, I would recommend you:

1) get a hold of manufacturer's prints of a single chain link, create sepearate parts, create sub-assemblies of links, create a full assembly with all of your links properly referrenced to the sprocket.

OR

2) do what I did and grab a pair of calipers and the chain to measure plate thicknesses and offsets (this will vary from brand to brand and style - whether it is a standard roller chain, O-Ring, or another variation of the O-Ring with a nice corporate buzz word attached to it).

On another note, I hate seeing idlers of chains. It's painful to look at a source of friction loss that could be avoided wihtout adding any weight through a pivoting/sliding differential.

UWO Racing (http://www.eng.uwo.ca/sae/)

[Jersey Tom]

http://www.camnetics.com/GearTrax.htm

Geartrax add in for Solidworks. I HIGHLY reccomend it. Makes modelling all sorts of gears and sprockets and what not super easy. Input a couple parameters and it busts out your model.

No drawing of tooth profiles necessary

[drivetrainUW-Platt]

that chain I have on there is one that I got off of 3D content central (Solidwork's part library). It uses a single link and then patterns it around a profile. I have the links of our chain drawn up, so I plan on just copying there method only with a different link.

[Jersey Tom]

Why model the chain?

[drivetrainUW-Platt]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jersey Tom:
Why model the chain? </div></BLOCKQUOTE>

First, so I can accuratly figure out where those bushing are going to be, and second for the bling factor in the design report