Bellcranks

[Tim Yeung]

Hey guys,

I've recently been given the project of designing the bellcranks for our car for the 2006 competition. Last year, we got our bellcranks CNCed out of aluminum but they still required machining once we got them because the CNC couldn't cut on one side. So after talking to my suspension leader he mentioned that there were possibly other ways of doing it. Like CNCing two pieces of aluminum and using adhesive to put them together to make one bellcrank and avoid the extra machining. He also mentioned that we could make it out of steel sheet metal just welded togeterh and have the bellcrank hollow in the centre.

I was just hoping you guys could point me to some good information or give me some of your past experience with the subject.

Thanks for your time,

-Tim Yeung

[Tim Yeung]

Hey guys,

I've recently been given the project of designing the bellcranks for our car for the 2006 competition. Last year, we got our bellcranks CNCed out of aluminum but they still required machining once we got them because the CNC couldn't cut on one side. So after talking to my suspension leader he mentioned that there were possibly other ways of doing it. Like CNCing two pieces of aluminum and using adhesive to put them together to make one bellcrank and avoid the extra machining. He also mentioned that we could make it out of steel sheet metal just welded togeterh and have the bellcrank hollow in the centre.

I was just hoping you guys could point me to some good information or give me some of your past experience with the subject.

Thanks for your time,

-Tim Yeung

[Denny Trimble]

Our bellcranks this year were paper-thin and separate plates, out of aluminum. They weren't connected together, except by the four bolts for all the connections. No problems.

http://students.washington.edu/auto/photo_galleries/200...uild%20Pic%2016.html (http://students.washington.edu/auto/photo_galleries/2005carbuild/pages/2005%20Car%20Build%20Pic%2016.html)

[Mike Cook]

Our team pretty much did the same thing as Washington. This will work fine when mounting the thing in double shear. However our fronts were in single shear and needed to be one piece.

[Travis Garrison]

Bonding is definitely an option, but you have to really pay attention to how you do it. You want to avoid peel at all costs, tapered lap joints are a good option, thats how this upright was done:

http://dot.etec.wwu.edu/fsae/v35/bonding/pages/Upright%...%20at%20once_JPG.htm (http://dot.etec.wwu.edu/fsae/v35/bonding/pages/Upright%20-%20Inside%20both%20halves%20at%20once_JPG.htm)


I wouldn't bank on a bonded part being simpler, and there may not really be much of a need for it on a bell crank. I'd listen to Denny and Mike, two piece bellcranks seem to work very well, and they are dirt simple to make.

If you do want to do a bonded part, I'd be happy to share what experience I have doing bonded parts...

-Travis Garrison
University of Washington FSAE

[Wilso]

We did two piece CNC aluminum front and rear originally. They are both in single shear also and work just dandy. However the fronts ended up interfering with the shocks, so I made three piece steel parts. Super simple machining, easy to assemble and weld. Made them both in a day after waiting 7 weeks for the CNC'd ones. The pics of the front are pre-lightening via the grinder. Took off about 2 oz per bellcrank and they were still way overkill.

Front:
https://webspace.utexas.edu/ca61483/www/Fcar/DSC00445.JPG
Rear:
https://webspace.utexas.edu/ca61483/www/Fcar/DSC00444.JPG

[Agent4573]

We do a 3 peice welded bellcrank. It's basically 2 pieces of sheet steel and a bearing housing we make on the lathe. Then all three are just welded together. It's a really simple process we normally give to our freshman to do. We used to do the whole machined aluminum, but the steel is just as light, stronger and much much easier to make.


*EDIT* We also mount the bellcranks in single shear and have never had a problem with them.

you can see one of them here, just under the vise grips. http://www.rfr.rutgers.edu/gallery/Building%20RFR052/DSCF3580%20(Large).htm

[yofa]

you should get a slot cutter for your cnc.

[TA]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Agent4573:
We do a 3 peice welded bellcrank.

you can see one of them here, just under the vise grips. http://www.rfr.rutgers.edu/gallery/Building%20RFR052/DSCF3580%20(Large).htm </div></BLOCKQUOTE>

Went through the whole gallery -- kudus to your exhaust fabricator, that's really nice work. Also, extra style and originality points for using a sawzall as a tirechanger http://fsae.com/groupee_common/emoticons/icon_wink.gif

Cheers, Ted

[Tim Yeung]

Hey Denny,

After going over some ideas with my suspension leader, we were a little confused as to how you can constrain your bearing from moving between the two plates without having the piece constraining it rubbing on the two plates.

Thanks a bunch,

Tim

[Denny Trimble]

It's just a matter of shoulders that hold the OD of the bearing axially, and washers / spacers that constrain the ID of the bearing axially.

I don't have time to dig up a CAD model, but take apart a cartridge bearing bicycle hub to get the idea http://fsae.com/groupee_common/emoticons/icon_smile.gif

I don't know where I'd be if I hadn't grown up in a bike shop...

[SD]

I was more wondering if the two plates were constrained to the middle bearing housing by anything more than a press fit?

Thanks for all the info Denny.

[D-Train]

We have used Chrome Moly bellcranks for the past few years now. Simple construction - just 2 plates with the appropriate holes, connected by some tubing that holds the needle bearings at the pivot point. We get the plates laser cut, so manufacturing is really easy.

Out of curiousity, what do the aluminium bellcranks weigh each approximately? From memory ours are around the 100g mark just for the plates and tube (not bolts, spacers, etc.

[SD]

I don't know the weight of our finished bellcranks and it would take me a while to get my hands on a decent scale. Here is a picture of our CNC'd + some manual milling bellcranks.
http://www.formulaubc.com/gallery/albums/2005-Production/DSC00340_Medium.jpg


http://www.formulaubc.com/gallery/albums/2005-Production/DSC00328_Medium.jpg

[SD]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Dwayne:
We have used Chrome Moly bellcranks for the past few years now. Simple construction - just 2 plates with the appropriate holes, connected by some tubing that holds the needle bearings at the pivot point. We get the plates laser cut, so manufacturing is really easy.

Out of curiousity, what do the aluminium bellcranks weigh each approximately? From memory ours are around the 100g mark just for the plates and tube (not bolts, spacers, etc. </div></BLOCKQUOTE>

Do you have two sets of needle bearings? (one set for radial loads, the other for axial)

[Jersey Tom]

What do you mean the CNC "couldnt cut on one side."

[SD]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jersey Tom:
What do you mean the CNC "couldnt cut on one side." </div></BLOCKQUOTE>

Check the above pictures, our CNC sponsor was unable to machine the "slots" in the bellcranks. This meant we had to do it manually afterwards. (And do 2 more setups to get the angles all correct)

[SD]

The weight of the our rear 2004 bellcrank is 148 grams. That is with the bearing but no bolts.

Our 2005 bellcranks should be slightly lighter (the ones pictures above).

[D-Train]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by 5teN:
Do you have two sets of needle bearings? (one set for radial loads, the other for axial) </div></BLOCKQUOTE>

We use the needle rollers for radial loads, and brass bushes for the axial loads.

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by 5teN:


Do you have two sets of needle bearings? (one set for radial loads, the other for axial) </div></BLOCKQUOTE>

We always used needles for both. For radial loads we used drawn cup needle rollers and used a shoulder bolt as both the bellcrank axle and the inner race.

Ben

[Denny Trimble]

Single shear with threads in bending? Oy!

Did you at least drill it out to save some weight?

[ben]

Ours were double shear.

Ben

[Lars Kanter]

Something I've always been curious about - how do you hold the needle bearing in the bellcrank mount? How do you make sure that the bellcrank faces do not rub againts the inner faces of the mount? Since there's no inner race on the needle bearing, you don't have anything to sandwich between two shoulder bushings. This is probably painfully obvious, but I just don't see it.

[Jarrod]

needle roller thrust bearings.

[Cement Legs]

Yeah we were going to go the same route until last minute when we decided to use a pair of cartridge bearings instead of the 3 bearings for each rocker. Do you run into any bending issues with the needle bearing carrying your moment?

[Jarrod]

we run the bellcranks double shear, so no bending, but on all open wheel circuit cars I have looked closely at, they run a large diameter hub (ie at least 50mm, maybe up to 75mm) with needle rollers and thrust bearings either side, and the bellcranks are probably 25mm or so wide. So long as the moment is reacted over a wide enough base you shouldn't have a problem. If you are running small diameter centre hubs it may be an issue.

[Omer]

hi
how did you design your bellcrank(rocker right) geometry?
I'm using adams/car to design my suspension,and this program uses the forces method to find roll center(and not geometry way),any way by changing the geometry of the rocker i changed my roll center location and movement,so it has importance,further more adams shows me that by changing caster i change my roll center as well,things i would have never realaize with a 2d program.
using miliken way to calculate load transfer he asks for istallation ratio, because it's a 3 dimension mechanism i wrote a matlab prog that calculates the IR with dependency on rocker dimensions and a-arms length and location,do you think i'm walking in the right direction?
we are a first year so pls be gentle.

[Jarrod]

changing rocker ratios will likely be changing your roll centres as the car will be rolling and pitching differently, possibly due to different load reactions due to push/pullrod loadings too. I haven't got far enough into adams to determine how sensitive it is to that kind of thing. I use susprog3d in combination with CAD (NX3 in our case, but it is the same whichever package you are using) Susprog is about AU$300, and you can get it up an running in no time. There are a few little tricks in some of the parameters, but it isn't too bad, the guy who developed it will answer emails within a day usually. Much easier than writing your own matlab code, and a lot more verstaile I would imagine.

[Omer]

thanks for the reply
but i am still confused about the things i need to do first.
1.you pick a tire and a rim.
2.you design your upright.
and then what should i design first?
a-arms mounting points to the frame?,shock mount on the frame and bellcrank mechanism,or steering rack location?
or you should look at the big picture right from the begining which make it even harder.
by the way monash sits in melb?been there for several weeks and loved it.

[Jarrod]

Tire, rim, upright, brakes etc, then wishbone geometry, and a-arm points, find a suitable place for your steering rack, as getting the rack to fit in a good spot that suits driver ergonomics and steering column requirements can be tricky, then work your shock and bellcrank around that, then go back to the start and iterate about five times through the whole process. Others may prefer a different sequence, I think I changed the sequence on each iteration. Keep in mind through the process whether you would prefer pull or pushrod, front or rear mount rack, can you get chassis tubes and nodes where you want them. Building a mock up chassis out of bits of wood can really help to visualise your chassis geometry, and it can be made and modified really quickly, and can identify ergonomics problems much more quickly than CAD.

melbourne as a whole is OK, our little corner in clayton is not exactly a prime location though.

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Omer:
how did you design your bellcrank(rocker right) geometry?
do you think i'm walking in the right direction?
we are a first year so pls be gentle. </div></BLOCKQUOTE>
Omer,
Why do you want rockers?
Do you know they are not necessary?

Z

[Omer]

I understand (I hope i do...) that you can create increasing wheel rate with only mounting the shock in an angle,mounting it this way also brings them outboard which is good because you get air cooling it down and bad because of aerodynamic.
I guess i want it because it really looks cool.
and it reduces your shock dimensions.
I saw your name on a lot of threads and would love to hear from your experience the advanteges ans misadvanteges.
thanks

[jack]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">...I guess i want it because it really looks cool... </div></BLOCKQUOTE>

oh man, Z is going to have fun with that...

PS: don't say that to the design judges http://fsae.com/groupee_common/emoticons/icon_wink.gif

[Z]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Omer:
I guess i want it because it really looks cool.
</div></BLOCKQUOTE>
I rest my case. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Omer,

Good to see that you are being honest with yourself.

I've ranted on about rockers quite a bit, so press "Find" button at top-left of screen, then "Advanced Search", in "Search Words" box type 'rockers', in "Author Name" box type 'Z', then press "Search".

Z

[BryanP]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> I guess i want it because it really looks cool. </div></BLOCKQUOTE>
That was our rationale for this years car. http://fsae.com/groupee_common/emoticons/icon_smile.gif Next year ( Hey , wait a second, that's next week)we will go for direct mounting. Now if we can just get a sponsor to buy us a set of the Ohlins....
regards
Bryan

[D-Train]

One of the real downsides to bellcranks (rockers) is the fact that as the bellrank moves, the motion ratio also changes as the push/pullrod and shock loads do not act tangentially to the rotation of the bellcrank, right?

Has anyone thought of alternative methods to translate the rod movement to the shock and maintain a constant motion ratio?

[yofa]

constant? constants don't exist in racing.

it's not necessarily a bad property and unknown. if you use this knowledge correctly, you can make it a rising rate setup without having to make trick rising rate springs.

[Frank]

excuse me Z,

your profile dosn't say who you are,...

do you mind me asking what uni you go to?

are you from industry?

frank

[Mike T.]

Dwayne, the relationship between the angles formed by the pushrod or shock with it's leg of the bellcrank from the pivot is what matters, as that's what determines how the pushrod or pullrod force is transmitted to the shock. It's not difficult to get extremely linear rates with a bellcrank setup, its a matter of properly analyzing the motion of the system throughout the range of travel, ensuring that each incremental step matches the desired ratio. For the past 2 years, our cars have had 1:1 linear ratios to within 1% as designed, and within 2% physically tested (.01/.02 variation in the 1 ratio).

Frank, similar hostility has been expressed towards Z in the following threads:

http://fsae.com/eve/forums/a/albumcomments/f/125607348/m/40710430521/p/1

http://fsae.com/eve/forums/a/tpc/f/8356059423/m/86910941621/p/1

And a link to some articles that reveal his identity is on the second page of the thread in the first link.

Mike Trumbore
University of Washington FSAE '03-'06

[Frank]

im not showing hostility

my guess is ex fsae working in (racing or auto) industry, we have bets here on who it is...

[markocosic]

Frank - do you/can you get BBC TV in Brisbane?

If so, there's an entire series called "Grumpy Old Men" that might be worth a watch. (there's even a "Grumpy Old Men" handbook to accompany the series) It'll explain the general inconsequential low-level grumpiness that Z and every other post-WWII/pre-dotcom male exhibits. http://fsae.com/groupee_common/emoticons/icon_wink.gif

Seriously though; Certain things Z says will be provocative purely for his entertainment, but in the main he's only provocative to make you think.

[Chris Boyden]

I think his name is Zorro, the masked bandido. He's out to save FSAE from bellcranks.
Eric Zapletal?

[Z]

Close, but not quite, Khris. (At least people spell Z correctly! http://fsae.com/groupee_common/emoticons/icon_smile.gif)

Frank,
Shortly before Z went troppo!!! (http://fsae.com/eve/forums/a/tpc/f/8356059423/m/52010404621) (The nice men in white coats say I am getting much better now. I hardly ever offend anyone these days http://fsae.com/groupee_common/emoticons/icon_smile.gif)

Z

[Frank]

eerrrrr

I'm 30, does that make me "pre-dotcom"

in fact here's a funny story....
in my first week of uni, I asked a lecturer
(sic) "sir, what exactly is windows?"
he replied "are you from another country"....

well, now i write NC postprocessors, program CNC, write webs, obnoxiously big spreadsheets, and spam this forum to bits...

BUT I'm still grumpy, just ask the UQ team....

luv and kissus

Frank