rod end conflicts

[URracing]

Well after searching a few threads and talking with a few people at the competition, I think I can conclude that the only real conflict with using rodends in suspension is how some teams install them so transverse loads are being forced through the threaded shank, resulting in them bending at the base and eventually seizing the ball.

I've been battling a few ways to very SIMPLY adjust caster and camber so I'll mention it here just to get feedback (postive or negative, I could use anything) just to make sure I am doing the right thing before we get it there and find out otherwise in tech inspection.

Basically using an M8 rod-end FULLY threaded directly into the end of the a-arm, the bearing will be offset so it slides along a Grade 8.8 or 10.9 bolt that is fastened between two clevis tabs that are bolted directly to the frame rail. The bearing (along with ENTIRE wishbone assembly) simply slides fore/aft on these bolts while spacers/washers are used to secure the rod end in place once a setting is found. (I have a picture but can't figure out how to attach it to these threads).

Doing it this way, the rod-end is FULLY threaded into the a-arm (using aluminum a-arms with steel threaded inserts). It puts only longitudinal loads on the bearing shank, which is what we want from what I can see. I'm just curious to see if a 8mm rod end is going to be too small (both for the shank as well as the bolt running through the bearing and between the tabs. Not sure what most teams use but I think it's around the 10mm range? The rod-ends are just a standard quality as their compression/tension force required (before the ball seizes up) is 13500 newtons, so we should be pretty good since we shouldn't have that high of force being directed through the a-arm. (I realize aviation quality products should be used but for us their availability is quite bad for any decent price).

Now I understand the concept of using shims to adjust caster, but to me this method screws around with your articulation axis and I've actually seen some suspension buckle to an extent that this axis was not paralled to each other after shims were placed. BUT, we are using shims behind the clevis tab plates on the top and bottom to only to adjust camber. To get away from using multiple shims for each setting, we would just fab up one or two shims for each setting that we want.

The reason I like this setup is because we do not have any tire data to work with so we basically just purchase a tire that looks good and test with it until we find settings that we like. We don't have the sufficient support or the man power to put in the time to properly calculate slip angles and such, so once we get the car built, we find tires, and go!

Thanks a load for any input anyone has on this. Once I find out I have a good idea what I'm doing is going to be sufficient on these cars, it will make me feel better.

Cheers - Brett

[URracing]

Well after searching a few threads and talking with a few people at the competition, I think I can conclude that the only real conflict with using rodends in suspension is how some teams install them so transverse loads are being forced through the threaded shank, resulting in them bending at the base and eventually seizing the ball.

I've been battling a few ways to very SIMPLY adjust caster and camber so I'll mention it here just to get feedback (postive or negative, I could use anything) just to make sure I am doing the right thing before we get it there and find out otherwise in tech inspection.

Basically using an M8 rod-end FULLY threaded directly into the end of the a-arm, the bearing will be offset so it slides along a Grade 8.8 or 10.9 bolt that is fastened between two clevis tabs that are bolted directly to the frame rail. The bearing (along with ENTIRE wishbone assembly) simply slides fore/aft on these bolts while spacers/washers are used to secure the rod end in place once a setting is found. (I have a picture but can't figure out how to attach it to these threads).

Doing it this way, the rod-end is FULLY threaded into the a-arm (using aluminum a-arms with steel threaded inserts). It puts only longitudinal loads on the bearing shank, which is what we want from what I can see. I'm just curious to see if a 8mm rod end is going to be too small (both for the shank as well as the bolt running through the bearing and between the tabs. Not sure what most teams use but I think it's around the 10mm range? The rod-ends are just a standard quality as their compression/tension force required (before the ball seizes up) is 13500 newtons, so we should be pretty good since we shouldn't have that high of force being directed through the a-arm. (I realize aviation quality products should be used but for us their availability is quite bad for any decent price).

Now I understand the concept of using shims to adjust caster, but to me this method screws around with your articulation axis and I've actually seen some suspension buckle to an extent that this axis was not paralled to each other after shims were placed. BUT, we are using shims behind the clevis tab plates on the top and bottom to only to adjust camber. To get away from using multiple shims for each setting, we would just fab up one or two shims for each setting that we want.

The reason I like this setup is because we do not have any tire data to work with so we basically just purchase a tire that looks good and test with it until we find settings that we like. We don't have the sufficient support or the man power to put in the time to properly calculate slip angles and such, so once we get the car built, we find tires, and go!

Thanks a load for any input anyone has on this. Once I find out I have a good idea what I'm doing is going to be sufficient on these cars, it will make me feel better.

Cheers - Brett

[Christopher Catto]

yes. I think very many teams have rod-ends with load along their symmetry axis (tensile-compressive) and the judges are ok with that. there are teams who adjust the camber by elongating the rod-ends maybe on the inboard side (and at a guess they need to move one or both the inner clevises (i call them brackets) fore and aft to accound for the change.

shims or slots i regard as good and i believe the judges like this method. i used not to but then i used to be so anal about some things and then i learned my lesson from seeing many race cars. http://fsae.com/groupee_common/emoticons/icon_razz.gif

one method i sometimes recommend is for some things to have interchangeable parts (say the top upright mount) with preset camber. these can be in different paint or anodysing colour schemes. it is more expensive but it looks very pro and avoids much tinkering. but you need to design these with a bit of care during CAD since the differences between one part and another may be only say 1mm in offset and you'd want all the rest of the car to be accurate and symmetrical. at my uni the problem was always that we wanted to just interchange wishbones to adjust camber but were hampered by lack of money. so if your uni is like mine, tell them to wake up a bit and take brand awareness and advertising a bit more seriously! http://fsae.com/groupee_common/emoticons/icon_mad.gif

dont worry about rod-ends with axial load. Renaut F1 and Jordan F1 used to have them in their cars till 5 years ago. i know this is not a good explanation for an FSAE judge but if you explain that the load is correct (and let's face it, a rod end IS SUPPOSED to be loaded along its axis) then you will have no problem.

i would use aerospace grade. the play should be less. depends on how sticky your tyres are but check the ball fit after say 2-3 endurance distances. if they loosen up, put new ones but by then you will have spent more money. check these metric rod-ends out (you should be able to receive anything from Germany via post for reasonable prices but always buy more than you need)

Fluro: http://www.fluro.de/FLUROHoehnHome.htm
Rodobal: http://www.getecno.com/english/home.htm
also National Precision make them and they are closer to you

i personally like metric rod-ends. I mean, NAS bolts are amazing, especially the different grip lengths and the nice fine thread. But there is nothing easier than buying a box of UNBRAKO M8 socket heads and just designing your suspension to use these. Why use imperial if you can buy pretty much similar quality bolts? An FSAE car is not an F22 so I think NAS bolts are very OTT, despite what some judges say. http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif I see Enzos and Carrera GTs with metric bolts so I am not too worried. I wonder if the Eurofighter has all imperial bolts actually... http://fsae.com/groupee_common/emoticons/icon_confused.gif

[murpia]

Rod ends in axial loading is fine. Like a trackrod in your steering for instance, or a pushrod.

The inboard joints of a typical wishbone are loaded axially along the wishbone itself. If you want the asjust camber here, then you are correct in assuming you need to allow the clevis to slide. Just don't do that thing of welding little tubes to the ends of your wishbones so the rod-ends are no longer axial to the wishbones but perpendicular to the centreline of the car. Then they ARE in bending...

Regards, Ian

[Osth]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">The rod-ends are just a standard quality as their compression/tension force required (before the ball seizes up) is 13500 newtons, so we should be pretty good since we shouldn't have that high of force being directed through the a-arm. (I realize aviation quality products should be used but for us their availability is quite bad for any decent price). </div></BLOCKQUOTE>

We looked in to this for our car, and determined to chose standard grade rod-ends for the upper wishbones and tie-rods as they take much smaller loads than the lower wishbones and the calculated loads where lower than the recommened 1/4 of the static load rating for our rod-ends (Rodobal M6 series). Our '06 car had Motorsport grade (equivalent to aerospace grade I suppose?) rod-ends all around so we wanted to try to save some real money and some for the cost report.

The standard grade rod-ends worked well for a while but some of them started to wear out and had to be replaced before competition at Silverstone this summer and then we chose the Motorsport grade for the replacements. Our previous cars (03 and 05) have had the same problems, the standard rod-ends (M8s from SKF on these cars) wear out as they cannot take the vibrations in the suspension (this is also stated in the product information from Rodobal).

I should remind you that we ran over 5 test endurance races and a total of about 500 km of testing before competition so the standard grade rod-ends works for a while, but they will wear out for sure and then you have only saved money for the cost report as 2 standard grade rod-ends &gt; 1 motorsport grade. And maybe you could also downsize to M6 by chosing higher grades... should save some grams of weight and also some cost maybe.

[Fyhr]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Osth:
The standard grade rod-ends worked well for a while but some of them started to wear out and had to be replaced before competition at Silverstone this summer and then we chose the Motorsport grade for the replacements. Our previous cars (03 and 05) have had the same problems, the standard rod-ends (M8s from SKF on these cars) wear out as they cannot take the vibrations in the suspension (this is also stated in the product information from Rodobal).
</div></BLOCKQUOTE>

Were those SA 8 C or SAKB 8F? They do differ slightly, i'm curious to know what bearing surface material is used on SA 8 C.

We ran SAKB 6 F on last years car and i'm looking into alternatives.

[Osth]

Don't know exactly the type of the older cars SKF rod-ends, but I belive I can recall seeing old boxes of SA8C still lying around in our Formula room so I think that would be what was on the cars.

[Tim G.]

Very interesting thread here. The post which explains that using 'little tubes perpendicular to the chassis tube' makes it easy to see the bending introduced into the rod end.

Knowing this, and if one keeps the rod end aligned with the axis of the wishbone arm would the design judges still have a problem with rod ends? How can they knowing that the loads can only be axial.

Lastly, if one does this is there any reason to ensure that the rod end is fully threaded into the wishbone arm? Can one just use the rod ends to adjust camber and caster?

Thanks for enlightening me.

Tim

[DJHache]

If your inboard rodends are inline with the wishbone tubes then there is no bending in them. On the outboard end any rodend you put there will be in bending, even if it is fully threaded into the wishbone.

A simple free body diagram of the entire a-arm under braking loads should clarify this.

[Ben C-M]

This is incorrect. You should only use inboard rod ends as long as you don't use them for any type of adjustment. As you change their length(to adjust camber, for instance), you are forcing the a-arm angle to change, thereby bending the whole tube and your rod end as well. The judges will not like this.

I would suggest using spherical bearings instead of rod ends. It's not very much more work to use sphericals and you will get a much more precise a-arm with probably alot less chance of breaking. In terms of what is more suitable for the application, it's definetly sphericals.

[Brett Neale]

Also remember that if your pushrod/pullrod isn't pointed directly at the outer balljoint then that will induce bending loads into the wishbone, and hence the rodend.

[flavorPacket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Ben C-M:
This is incorrect. You should only use inboard rod ends as long as you don't use them for any type of adjustment. As you change their length(to adjust camber, for instance), you are forcing the a-arm angle to change, thereby bending the whole tube and your rod end as well. The judges will not like this.

I would suggest using spherical bearings instead of rod ends. It's not very much more work to use sphericals and you will get a much more precise a-arm with probably alot less chance of breaking. In terms of what is more suitable for the application, it's definetly sphericals. </div></BLOCKQUOTE>

sphericals are more expensive, harder to jig, harder to install and replace, etc. I'd suggest that using sphericals IS much more work (and more money). And rod ends are not that bad. The Bentley Speed 8 GTP car used them to win LeMans...

[URracing]

Well I believe we figured out our strategy using both rod-ends and spherical bearings. We are using a spherical bearing that will be pressed into an aluminum block at the upright end. This way (since our push-rod WILL NOT be in axial line with the bearing) we are not using a rod end at this point.

The a-arms are glued into the block while rod-ends will be FULLY threaded into inserts that are glued on the chassis end of the a-arm tubes. The M8 rod-ends will be attached to clevis plates on the frame using Grade 11 8mm bolts. The rod-ends then slide fore/aft on these bolts for adjusting caster while using shims behind the clevis plates to adjust camber. We are going to stick with standard Aurora rod-ends since they give us a 28 deg offset angle to match the frame that we have to work off of. (We were going to build a new and better frame but it costs too much and we can't factor in the time to produce it either, so we are using the original frame that was started for this car a couple years ago and just make changes as necessary).

This solves just about all the issues I can think of behind this subject. Although the rod-ends are threaded into the inserts, they are not used for any adjustment at all. There are no transverse loads being applied into the a-arm and the spherical bearing at the upright end should be capable of handling both the a-arm loads and the push-rod load. Adhesive tests (both shear and tension) have been done to make sure the glue we are using will be sufficient to keep it all together. My only last concern is making sure the 8mm bolt on the clevis tabs will be sufficient to withstand the loads applied on it. Haven't really seen what other teams use on these clevis systems so I'm not too sure about the hardware.

It's great to hear your opinions and suggestions. Thanks a lot and hope to see some of you in competition. (We should be sending a few guys from Regina to Michigan again this year to get a good idea of the competition).

[Osth]

Without knowing your forces I think your grade 11 M8 bolt holding the suspension to the frame will be more than enough. What I did for our car was looking in to my Machine elements handbook realizing that a bolt in shear basically is a rivet, and for rivets the critical parameters is the shearing stress and the pressure in the riveted hole which in this case is the rodend and finally the stress in the rivet head which we don't care about in this case.

We used M6 bolts of grade 12.9, had a fat 243 kg car and did 1.32 g in skidpad and had no problems so I think you will be fine, but look into your handbook for rivets and do the calculation to be on the safe side.

I'd like to see a picture of the mounting your are describing when it is finished, because I can't visualize it in my head http://fsae.com/groupee_common/emoticons/icon_wink.gif.

[URracing]

I would absolutely love to post a picture since in my opinion they really do tell a thousand words, but I'm not sure yet as to how to attach a picture to these threads. I see a URL can be done but I'm curious if there is any way to attach a file from my computer drive (like most forums have).

Cheers,

Brett

[Christopher Catto]

Guys.

1) Quality for quality, spherical bearings are CATEGORICALLY NOT more expensive than rod-ends. If they are, then your supplier is either cheating you or his rod-ends are fantastically cheap. The simple reason is that rod-ends have a spherical pressed in them, and they have all the casing and thread.
Sphericals are also not that hard to install or replace once you have done a few but I respect the opinion of others. In theory it's true that you can just weld a fitting and screw in your rod-end and not worry too much, whereas for sphericals you need to be good with your interference fit tolerancing.

2) Inboard rod-ends ARE OK for camber adjustment, just make sure you move the clevis (bracket) longitudinally to account for the change. Please don't make me repeat this. You are making an FSAE car, not an Ultima. So, nothing is impossible or particularly bad. Just THINK A BIT. Like don't have the rod-end out by miles. Trust some of the opinions from David (Hache) since he's been in FSAE for quite a while.
As you can see, F1 teams do it one way, Bentley another way and if you look at Nascar, it will be something else. If a judge is particularly stubborn, that's because he's not open minded. Not your fault. I would not go about impressing the judges all the time. I would make the car that is easiest to drive, tune and to prove everybody wrong just like Chapman proved Ferrari wrong when he installed engines BEHIND the driver.

3) Host your images here and copy/paste the link they give you: www.imageshack.us (http://www.imageshack.us)

http://fsae.com/groupee_common/emoticons/icon_wink.gif

[flavorPacket]

Chris,

Not only is the spherical we use more expensive than the rod ends we use, on top of that you have to add in the cost of the staking tools and finish machining of the housings, as well as the extra time it takes to do all that. For us, it's a very easy choice to use rod ends on the inboard side.

[URracing]

Just to add a quick note, yes, I agree that spherical bearings due tend to run a little more than rodends because of the machine time and material for the housing.

The main reason we went with rod ends is because we were able to get a 28 degree offset with an M8 rod-end for the inboard chassis mounts. This allows the a-arm mounts to be closer together and still get away with a wide frame and narrower track.

Thanks for the info. I think we are pushing forward now.

[buggaero]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by URracing:
Well I believe we figured out our strategy using both rod-ends and spherical bearings. We are using a spherical bearing that will be pressed into an aluminum block at the upright end. This way (since our push-rod WILL NOT be in axial line with the bearing) we are not using a rod end at this point.

The a-arms are glued into the block while rod-ends will be FULLY threaded into inserts that are glued on the chassis end of the a-arm tubes. The M8 rod-ends will be attached to clevis plates on the frame using Grade 11 8mm bolts. The rod-ends then slide fore/aft on these bolts for adjusting caster while using shims behind the clevis plates to adjust camber. We are going to stick with standard Aurora rod-ends since they give us a 28 deg offset angle to match the frame that we have to work off of. (We were going to build a new and better frame but it costs too much and we can't factor in the time to produce it either, so we are using the original frame that was started for this car a couple years ago and just make changes as necessary).

This solves just about all the issues I can think of behind this subject. Although the rod-ends are threaded into the inserts, they are not used for any adjustment at all. There are no transverse loads being applied into the a-arm and the spherical bearing at the upright end should be capable of handling both the a-arm loads and the push-rod load. Adhesive tests (both shear and tension) have been done to make sure the glue we are using will be sufficient to keep it all together. My only last concern is making sure the 8mm bolt on the clevis tabs will be sufficient to withstand the loads applied on it. Haven't really seen what other teams use on these clevis systems so I'm not too sure about the hardware.

It's great to hear your opinions and suggestions. Thanks a lot and hope to see some of you in competition. (We should be sending a few guys from Regina to Michigan again this year to get a good idea of the competition). </div></BLOCKQUOTE>

Hi,

Would you mind sharing the exact deatails of the glue. (Plus i'm running a parallel search for them)

[Old Greg]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by buggaero:
Would you mind sharing the exact deatails of the glue. </div></BLOCKQUOTE>

Given that he hasn't posted anything in two and a half years, don't count on a response.