Hubs with built-in tripod joint [Archive]

Goost

10-20-2015, 08:27 AM

Sorry, but NOT good IMO. Many weak points...

Z

Sorry Joshkb, sometimes Z will write you a personal textbook to help, and sometimes you get this.

I think they look fine, and you seem to be well capable of manufacturing it. Neat use of the 3d printing. Some things I notice, overall it seems fine:

1) you could add some webbing in the upright to help torsional stiffness. better to build a closed shape for that in general
2) the brake hat could be relieved between bobbins if you only have 4, similar to the brake disks
3) those bearings - can't tell what type they are, but in general you can't preload them much - will need to add positive retention to the spindle nut anyway to pass tech I think
4) what keeps the inserts from sliding out? friction? Wouldn't trust that. seems your cad shows holes drilled for a faceplate there?
5) the step/ridge that presses on the outboard bearing may be too small diameter - sometimes large bearings have a decent radius/chamfer that might try to wedge onto this.
6) brake hat could have larger radii at the base to reduce stress without much change in weight. FEA may not show it but in real life you might have an axial load there

less useful:
a) the CAD for the wheel might be improved, just to be careful about space
b) same note, based on experience you may want to ensure >0.25" clearance between the wheel shells and components to account for possible deflection
c) brake caliper mounting may be convenient there, but I expect it isn't optimal for loading. Check out how the bearing forces change during braking as you change the location of the brake caliper the bearing forces change
d) is the material between the tulip housing and the hub empty? Could you bore it out from the outboard side?
e) do you have boots that fit the nut? As Mitchell suggests, there are better options. The ones that fit those nuts are surprisingly heavy

Joshkb

10-20-2015, 09:00 AM

Thanks Goost, very helpful. Let me try to address your comments:

1 - The uprights havent been made and I agree, the added weight wouldnt be much so why not go stronger?

2 - The idea here is that the aluminum surface may bet work out eventually, which I assume will happen before the rest of the hub has minished its service life. In that case, we can rotate 45 degrees and have fresh surfaces. I just don't think our team is at a point where we should spend a whole lot of time on weight reduction. Remember, this is our first car..ever.

3 - They are SKF 61816 DGBB 80x100x10..can you explain what you mean?

4 - The inserts have a very tight press fit which we were able to achieve without temperature difference due to the inserts not being closed, so they can flex a bit when clamped on. They are very tight. I guess time will tell on that one...

5 - This is a good point, as these bearings do have a fairly large radius. Hopwever, I am following SKF reccomendartions for this diameter, so I will leave that alone.

6 - We are running 10" wheels, which makes the whole brake systrem a challenge. Between an 80mm spindle to accomodate the tripod housing and inserts, as well as the radial clearance required between our calipers and inner wheel, we have less than a millimeter to play with there. I wish we could fit larger radii.

a,b - We have roughly 5mm clearance from components to wheel, so a little tighter than you reccomend, but still comfortable, I believe.

c - Is this a front caliper vs. rear caliper comment, or regarding the deflection in outer race bearing housing induced from braking force at caliper connections? Let me know..

d - I bored 100mm from the outboard side and saved ~ 1.6 lbs, I'd have to check the model weight. I think that'
s what you're getting at

e - My appologies, we are actually making custom aluminum nuts from the material we bandsawed off of the blanks to make these hubs, so these are not the final nuts.

Thanks for your helpful comments, Austin.

Josh

scotty young Taylor Race

10-20-2015, 02:47 PM

Thanks Josh ,,

Nice work . I will second Mr. Goost on some type of retaining device on the inserts. A little heat in the housing and your tight press assumtion just cost you endurance..The insert you have in the model needs to be flipped around . and the pin used to keep it from
rotating in the housing .

MegaDeath

10-20-2015, 05:02 PM

Josh,

Don't let anyone (Z) on this forum scare you from pursuing an idea just because they think it's a bad idea..... We ran one piece aluminum rear hub/tripods from 2011 till I graduated in 2014 and never had a single problem with them wearing out or getting excessively sloppy. That was on both 4-cylinder big power, little torque and 1-cylinder big torque, little power engines.

To speak to their potential reliability; when I graduated in 2014, our 2012 car had about 2100 miles on it and it's still driving.

In summary, with proper design and manufacturing you can absolutely make it work....... Even if Z thinks that it is "NOT a good idea"

Cheers!

Goost

10-20-2015, 08:11 PM

2 - The idea here is that the aluminum surface may bet work out eventually, which I assume will happen before the rest of the hub has minished its service life. In that case, we can rotate 45 degrees and have fresh surfaces. I just don't think our team is at a point where we should spend a whole lot of time on weight reduction. Remember, this is our first car..ever.

Neat idea. Personally never seen that surface fail, except once when that was the least of the issues compared to everything else that broke...

3 - They are SKF 61816 DGBB 80x100x10..can you explain what you mean?

If I recall DGBB are OK for axial loads, maybe not as good as an angular contact version?
Anyway, I meant that the desired pre-load for the bearings is probably lower than desired to keep the big jam nut in place, so to keep the nut from walking away it needs some type of secondary retention, a pin or something.
That nut is also considered a "suspension fastener" so it needs a positive locking mechanism anyway to pass technical inspection (since you are machining your own it might be especially scrutinized).

4 - The inserts have a very tight press fit which we were able to achieve without temperature difference due to the inserts not being closed, so they can flex a bit when clamped on. They are very tight. I guess time will tell on that one...

As Scotty says, I wouldn't trust the press fit. Tripod bearings can generate some odd loads, even pulling on the inserts even though it seems like they should just roll smoothly.
A plate on the spindle face, or a pin as he says may be useful. Also, if the inserts are cylindrical you may want a pin or something else to keep them from rotating.

c - Is this a front caliper vs. rear caliper comment, or regarding the deflection in outer race bearing housing induced from braking force at caliper connections? Let me know..

Hm, I'm not thinking about deflection.
So, as you rotate the location of the caliper around the spindle, the couple between the bearing and the brake caliper is the same magnitude during braking - BUT the bearing's force component changes direction.
You can use this force to offset bearing loads during braking. Perhaps reduce the total bearing load after you sum forces from from the vehicle weight, lateral force, longitudinal force, AND this caliper/bearing couple.
(Example| during straight-line braking, assume the weight transfer is large compared to longitudinal force, no scrub radius or KPI, and look at front uprights:
If the caliper is on the back of the upright, then the vertical forces from the weight transfer and caliper/bearing couple sum "in opposite directions" to create a smaller total bearing load.
If the caliper is on the front of the upright, then they sum in the same direction - creating a larger total bearing load.
For your design you consider braking and cornering (which typically produces the largest bearing loads) and your specific upright geometry, so the ideal solution changes.
For similar reasons, the ideal bearing location is not centered in the wheel, rather it is slightly inboard. but your upright is past that stage now I think.)

d - I bored 100mm from the outboard side and saved ~ 1.6 lbs, I'd have to check the model weight. I think that'
s what you're getting at

Exactly! Big radius in the bottom of that pocket?

Hope it helps!

jd74914

10-20-2015, 08:46 PM

Neat idea. Personally never seen that surface fail, except once when that was the least of the issues compared to everything else that broke...

You might want to machine those before you assemble if that's something you are worried about. I'd be a shame to need to disassemble just to make a few pockets.

If I recall DGBB are OK for axial loads, maybe not as good as an angular contact version?
Anyway, I meant that the desired pre-load for the bearings is probably lower than desired to keep the big jam nut in place, so to keep the nut from walking away it needs some type of secondary retention, a pin or something.
That nut is also considered a "suspension fastener" so it needs a positive locking mechanism anyway to pass technical inspection (since you are machining your own it might be especially scrutinized).

Austin makes a good point here. You definitely need a positive retention mechanism (pin, etc.) for tech and to make sure it doesn't back out with the low axial loads.

Just so you know, we have used this bearing with good success. We calculated it's lifetime to be a little short for the application and assumed we would need to replace it, but in 3 years we haven't had a bearing failure. Note that after a full competition and training season we have seen some bearing getting a little rough. Our design focused more on getting the radial preload correct since our bearings are not axially constrained (probably not the best design, but it works well enough).

10-20-2015, 08:48 PM

Josh,

To clarify, you have a VERY BAD DESIGN.

Part of the reason so many other Teams persist with such bad designs is "because Racecar". I have heard too many FSAE students say "But it's supposed to fail like that, because it's a racecar...". Yes, literally!

But the main reason for your bad design is that you are simply copying everyone else. If this is not so, then please provide a comparison of above design against another ~half-dozen obvious alternatives, with OBJECTIVE NUMBERS for cost, strength, stiffness, lifetime, etc.

Anyway, here are some aspects of the design that are crucifiable offences (my dear old dad would have nailed me to the wall if I suggested them :)). Worst offender first, then in no particular order:

1. COST - You start with two ma$$ive billets of expen$ive alloy (upright + axle), then machine away almost all of them! The end result is a flimsy upright/wheel-hub assembly that, despite still being overweight, would quickly fatigue fail from the stress raisers you put in it (Goost's #6, at the highest stress-reversing area of the axle!). Except that the wheels will be flopping around so much that you won't be able to drive the car hard enough for the loads/cycles to do their work (see why below).

In short, a very expensive way of building heavy and weak parts, which are also perhaps the most crucial parts of the car! "But, hey, that's how everyone else does it!"

I most certainly would not buy a part like that. And any cost-conscious Production Engineer at any (good) company that you might work for in the future would not like it either. The profligacy you are practising here is common in the Racecar world, but NOT good elsewhere.

2. UPRIGHT - Too much material in the wrong places, and not enough in the right places. So, overweight + understrength!

How did you come to such a design? Did you have any strength/stiffness targets? (Yes, I know, cost was no issue...) More importantly, how many alternative designs did you consider? What are the numbers?

3. AXLE & BEARINGS - Just a "...light press on the spindle and a slight slip in our uprights", eh? And "...an inner race spacer to snug up against.", with the snugging-up done with an aluminium nut? And you are "...following SKF reccomendations for this [shoulder] diameter", but seem not to have noticed the word "minimum", which probably applies only to HARDENED steel axles...

I very much doubt you would get through a single 22 km Enduro with the design as shown so far, let alone any significant testing. And that is even if you use the world's most expensive aluminium alloy (the usual "racer" approach to solving such problems).

4. BRAKE DISC - Is it a good idea to drill a "speed hole" in the most highly stressed part of the disc? Or, put the other way around, where do you think the first cracks will start to appear in the disc?

5. OVERALL - You are a very-first-time Team. So why not buy one of the many small-car axle/bearing/hub units that are available off-the-shelf, in a size just right for your needs? These are much stronger and stiffer than yours, would have a lifetime of around 100,000 miles (cf. MegaDeath's ~2k miles), have a similar mass to yours (or perhaps less, after some machining?), and be much, much quicker and cheaper to get done.

If you think you can do a better job, then try next year. But compare objectively, with real numbers.

Z

Jay Lawrence

10-20-2015, 10:29 PM

Josh,

Don't get disgruntled by taking the above as how it comes across ("EVERYONE EXCEPT ME IS A FUCKING MORON!!!!").

I completely understand the desire to 'engineer' your car as much as possible, because being a student engineer is what that environment promotes. This creates many FSAE cars with many 'highly engineered' components that get cobbled together to create many crappy cars. As a first year team, just make the thing as cheaply as you can. This should drive you to simplicity, which will drive you to lightness and a quick build. You will learn the most (and earn the most points) by driving/testing the car. Unfortunately with designs like yours, it's very hard to have a backup plan and typically the lead times and costs can mean that you have designed yourself into a corner (in which you can't drive the car).

Joshkb

10-22-2015, 01:58 PM

Thank you for all of the feedback, everyone.

The radius on the back of my rotor hat could have been larger, I agree.

For the bearing shoulder on my hub, it is small to fit our rotor buttons. A radial circuit analysis showed that 5mm of clearance between caliper and wheel + the radial distance required for our brake pads to not interfere with the buttons left a small gap between that point and our 80mm spindle. The brake buttons were then located as far out (radially) as possible, and the remaining gap was the space available for this shoulder. My though process may have been flawed, but that is what I did to determine these dimensions. I probably should have valued this shoulder size more and modified our buttons.

Our main goals for the upright was to keep the stress under 300 Mpa during simulation, which we did in SolidWorks Simulation. Z, where do you think I have too much material, and not enough?

The material for each hub was $80, the machining, yes, quite expensive had I not done it myself.. and even then, it cost us time.

Josh

10-22-2015, 08:21 PM

Josh,

This would be easier if you gave more details. This is supposed to be "engineering", after all. So,

1. What is your engineering specification for "a light press on the spindle and a slight slip in our uprights"?

2. Do you you realise that there are proper standards for describing such "round-peg in a round-hole" fits? (Hint: look in the bearing handbook...)

3. How "snug" is the pre-load compression on the two inner races? (FWIW, many cars' workshop manuals have the torque figure for this "nut that does the snugging-up" as the highest of the whole car. Typically, several 100s of ft/lb or Nm.)

4. More importantly, what is the difference in widths between the Upright's shoulder (ie. that separates the two outer-races), and the "inner race spacer"? (This spacer is not shown in your CAD image, so I guess its importance is a recent discovery? Its width is important because it detemines how floppy your wheel will be.)

5. Have your teachers given you any formal "Design of Bearing Assemblies" lessons? (I guess probably not, and this thread is a result of their failing, not yours. You should tell them that.)

6. What are the uprights made from? (If not good quality steel, then what aluminium-alloy do you plan on repeatedly stressing to 300 Mpa?)

Anyway, you have now backed yourself into a corner that is difficult to escape from. So many little details that need fixing. I would almost be inclined to start from scratch. Perhaps just buy the core parts off-the-shelf...
~o0o~

...the upright ... where do you think I have too much material, and not enough?

You lost this race as soon as you chose to put the tripods INSIDE the uprights. Even with the "thinnest ring" DGBBs, namely the "68xx" series, you now have well over a foot (314 mm+) of circumference of bearing that has to be wrapped in upright. So no matter how thin you make the part of the upright that surrounds the bearings, there is inevitably a lot of it!

Worse yet, if you do not make that bearing support reasonably stiff, then the flimsy outer-race of the thin-ring bearing deforms, and only a very few of the balls must carry all the load. Rapid failure follows. (Followed by students cheering "Yey! At last! It's now a real racecar."!)

Look at how your upright transmits the many different loading patterns (= cornering, braking++) into the bearings. Look at how your crude "slabs" connecting the upper and lower BJs to the bearing-support-ring feed the loads into that bearing-support-ring, Specifically, look at the deformations!

A folded sheet-steel upright, similar in outside dimensions to yours, but fully "closed/boxed" and made out of maybe 0.6 - 1.0 mm thick sheet, would give your 68xx bearings a much better chance of survival.

Then get rid of the stress-raiser in the live-axle (ie. between outer-bearing and disc-brake flange). Look at the abundant prior-art that shows how to do this. (Hint: large fillet radius between axle-OD and disc-flange ... and then an appropriately shaped SPACER between outer-bearing and flange.)

And, your "toe-base" is probably much too small...

Z

Joshkb

10-22-2015, 11:10 PM

Z,

1 - The bearing inner races have a .0002-.0005" interference fit on our spindles. For our uprights, they have not been made, but we do not have an internal mic that large, so i'll have to measure some other way, or go by feel using an op-stop program that adds .0003" to the diameter per pass + spring pass. Any smaller and I don't get good tool engagement.

2,5 - I believe the standard fit for this bearing is a k5. Sadly, no, our professors and program in general does not cover bearing assemblies, different fits etc. We have a helpful machinist who we can ask these types of questions, but honestly, I'm not sure I could name a faculty member I would go to expecting a helpful lesson on picking fits.

3 - Rather than controlling the torque at the nut to provide axial preload, I have been guided towards a length-based tensioning method, which makes more sense to me. When tensioning the nut, i'll have force coming from the fit between my inner race and spindle, which leaves me guessing what component of that torque is actually from the bearings being preloaded. Let me know if you have a correction on this, I fell I am missing something.

4 - Bearing center-to-center distance is 35mm (~1.38") and the upright has a total width of 45mm (~1.77"). I've been guided by a SKF applications engineer to make my spacer on the order of .001" smaller axially than the outer race shoulder. Yes, you are right, I had initially planned on preloading the system using the nut.

6 - We had 7075 - T651 donated for these uprights and are planning on using that material, but then you knew that..because racecar. Our simulation stress resulted between 220 and 240 Mpa.

I will thicken the walls of the bearing housing and do a study on deflection of the area.

Thanks for the input, Z.

Josh

10-25-2015, 07:51 PM

Josh,

The most disappointing thing about your approach to this design is that, regardless of the hub-assembly's eventual performance, your Team is now very likely locked into this particular design forevermore. Each year from now until the end of time will see essentially the same design repeated, albeit each year having some small and inconsequential changes. This despite the fact that absolutely no objective engineering analysis of the pros/cons of your current design against the many alternative designs was made, or will be made. No comparative analysis of total mass, strength, stiffness, lifetime, cost, etc...

Of course, this is not unusual, because it is pretty much exactly the same approach taken by most other Teams. Namely, copy what everyone else is doing, and when the wheels fall off, cheer "Yeay! Perfectly optimised!!!"

Your current design can be made workable, and done right you should be able to complete a comp with it. To that end I was going to give you a few more suggestions on fits, preloads, slight changes to parts, and so on. But I really do not think it is a good idea to keep perpetuating this irrational approach to design.

So good luck, and let us know how it works out. :)

Z

P^squared

10-26-2015, 12:18 AM

Josh,
How much do your uprights weigh right now ? You should not mind adding material to places where you cannot project what the actual loads/deformations might be.

Z,
The reason teams follow this 'tripods inside the hub' design is because then they don't need a lot of steel in their hubs to accept splines.

The obvious, simplest way of doing this is to use solid axles, but that will need a lot of systems to be modified/simplified, and not an option for Josh right now.

Using flange couplings(welded on the tripod side) to connect the tripods to the hubs is far simpler, but that won't greatly reduce the hub OD/ bearing ID.

Any designs that might be missing out ??

Will M

10-28-2015, 02:38 PM

Josh,

1. COST - You start with two ma$$ive billets of expen$ive alloy (upright + axle), then machine away almost all of them! The end result is a flimsy upright/wheel-hub assembly that, despite still being overweight, would quickly fatigue fail from the stress raisers you put in it (Goost's #6, at the highest stress-reversing area of the axle!). Except that the wheels will be flopping around so much that you won't be able to drive the car hard enough for the loads/cycles to do their work (see why below).

In short, a very expensive way of building heavy and weak parts, which are also perhaps the most crucial parts of the car! "But, hey, that's how everyone else does it!"

I most certainly would not buy a part like that. And any cost-conscious Production Engineer at any (good) company that you might work for in the future would not like it either. The profligacy you are practising here is common in the Racecar world, but NOT good elsewhere.

Z

That's a negative Ghostrider.

I cannot speak to the floppiness of such a design but is IS well designed for manufacturing.
During my FSAE time (2011) we arrived at a near duplicated design (see pics) precisely for its manufacturability.
And now I do this kind of stuff professionally.

The key is to assess FSAE designs relative to the real world industry that FSAE most closely resembles.
And it is not the automotive market but rather low-volume high-mix (LVHM) manufacturing.

So yes machining it from billet 7075 is expensive material wise and possibly machine time wise.
However you will only be making one of each of these so your most of the processes traditional used in the automotive world don't work well.
In a LVHM environment material cost is only one factor; tooling and design are often the biggest cost.

For those uprights I would estimate less than $75 of material and about 45 min of CNC machine time.

So let's see what our options are.

For a larger volume a near net shape process would be a good idea but not here.
So no die casting or stamping because of the tooling costs.
These would still need secondary machining so your equipment costs will be higher.

Investment casting could work if you 3D print the pattern, no way if you mold a wax pattern.
But this is rarely a cost effective process for one off parts unless it absolutely must be cast.
Still have secondary machining.

Sand casting with a wood pattern.
Cheap but typically would not be acceptable in quality or performance for this kind of part.
Secondary machining and embedded silica particles! yay!

Electron Beam Additive Manufacturing.
Could make it out of stainless steel or titanium, so that is cool.
But either way the cost will be an order of magnitude higher.

Fabricated steel "box" upright
Sheet steel and a small thick wall tube for the spindle bore.
Welded together, either self fixturing or a small jig.
You probably can reach a weight parity, and material costs will likely be lower.
But now your design is more complicated (a welded assembly vs billet).
How much did that extra design time cost you? A good designer can cost over $600 per day burdened rate.
Designing for a single item made from a homogeneous material is more straight forward than a welded structure.
Especially if you want to set up a simulation or need to create drawings and BOMs.
Remember you only have $75 of material to start with.
Also you now have 3 process steps (sheet metal cutting, welding, final machining) compared to just one.
If your labor is free like in FSAE this might be a great idea, but a skilled welder is also expensive.

tl/dr version.
Or you can just hog it out of billet.
Yes your material costs are higher but you will have less cost from tooling, equipment, labor, and design time.
In the LVHM world making parts from billet is a totally reasonable thing to do.

-William

(Also Josh you do need positive retention for the hub. Check you the linked photos)

Originals here of the 2011 parts:
http://media.wolfpackmotorsports.com/displayimage.php?album=1&pid=972#top_display_media
http://media.wolfpackmotorsports.com/displayimage.php?album=1&pid=970#top_display_media
http://media.wolfpackmotorsports.com/displayimage.php?album=1&pid=967#top_display_media
http://media.wolfpackmotorsports.com/displayimage.php?album=1&pid=969#top_display_media
http://media.wolfpackmotorsports.com/displayimage.php?album=1&pid=940#top_display_media
http://media.wolfpackmotorsports.com/displayimage.php?album=1&pid=955#top_display_media

10-28-2015, 09:12 PM

William,

I have no idea what a "negative Ghostrider" is, but I am certain that "Profligacy 1.01" is indeed now a core subject of the modern Engineering curriculum (which explains why it is so widely practised in the western world's "LVHM" industries = shiny junk that doesn't work well). And "Critical Thinking 1.01" has long been abandoned as too "old-school" for this modern age...
~o0o~

I cannot speak to the floppiness of such a design but ...
...
The key is to assess FSAE designs relative to the real world industry that FSAE most closely resembles ... low-volume high-mix (LVHM) manufacturing...
So yes machining it from billet 7075 is expensive material wise and possibly machine time wise. However ...

I would say that the key to assessing FSAE designs is that the BEST DESIGN is the one that helps the Team score MOST POINTS at the competition. (Unless your goal is to lose? Is it? :()

So the most important factor is reliability, which here includes fatigue-life and strength (because no-finish = no-points). Next in importance are things like stiffness (important for hub-assemblies), speed of manufacture, mass, and cost, with all of these having to be assessed very objectively with carefully estimated NUMBERS that reflect "point-scoreability". Least important, although too often given undue weight by FSAEers, is "bling".
~o0o~

Anyway, your analysis of the various options is, sadly, and very typically in Engineering these days, a distorted and subjective one with next to NO NUMBERS. For example,

Fabricated steel "box" upright...
You probably can reach a weight parity, and material costs will likely be lower.
But now your design is more complicated...
How much did that extra design time cost you? ...
Designing for a single item made from a homogeneous material is more straight forward than a welded structure...

It only takes a bit of experience making these things to know that the subjective bias you have given to the above analysis is hogwash!

Specifically, your suggestion that the design of a folded-sheet-steel upright is somehow "more complicated" than that of a billet-machined upright is baseless. Why? I am sure that many people who have done it both ways will argue the opposite.
~o0o~

And another example,

Also you now have 3 process steps (sheet metal cutting, welding, final machining) compared to just one.

Since when is it NECESSARILY the case that "A + B + C > D"!?

This is a really STUPID ASSUMPTION, although, again, very common in modern Engineering. (For those struggling with this, try A = 2, B = 7, C = 0.0015, and ... D = 9,274,357!)

The above is the sort of nonsensical argument used by soap salesman flogging their product to dimwits, or politicians trying to get you to vote for them. Yes, sadly it does work well, and on too many people. Ahh..., H. Sapiens, the "wise one"!

But you will never win an FSAE comp using that sort of reasoning.
~o0o~

Or you can just hog it out of billet.
...
In the LVHM world making parts from billet is a totally reasonable thing to do.

That "LVHM world" is mostly about profligacy and bling. Selling junk to dimwits.

FSAE is supposed to be about engineering a fast car that scores maximum competition points. Many different ways to do this (eg. as one alternative not covered above, buy hub-assembly off-the-shelf...). Following the latest profligate fashion is not one of them.

Z

(PS. I remember this issue coming up before. There I suggested you students look at the countless, cheap, "Made in China" bicycles available in your supermarkets. These have many parts that look as if they are billet-machined, or forged, or investment-cast++, but they are actually sheet-metal fabrications. These parts are lightweight, strong, stiff, (so structurally very efficient), but obviously also very quick and cheap to make. And they look good. Go look at them, and learn...)

BillCobb

10-28-2015, 10:36 PM

Negative, Ghost Rider, the pattern is full. (Request for a fly-by to an Air Boss).

theTTshark

10-29-2015, 07:12 AM

Negative, Ghost Rider, the pattern is full. (Request for a fly-by to an Air Boss).

Murica... https://www.youtube.com/watch?v=vdHBsWXaHN8

Danschwind

10-29-2015, 09:56 AM

In the past, my team (s) built cars with steel (welded) uprights, off-the-shelf hubs, aluminum uprights and aluminum non-tripod-joint-built-in hubs.

If I have to choose, I would choose aluminum uprights and aluminum hubs with built in tripod joints, just like OP did. But, I would improve design a little, although that is part of the process of becoming a better engineer: making mistakes.

Will M

10-29-2015, 10:27 AM

Z,

1. LVHM – We must not be thinking of the same industries. An example for engine cooling. What I am referring to is making 100 units for each of 1,000 different engine cooling packages for a dozen different industries instead of 100,000 units of 1 engine cooling package for one automotive plant. The design and production processes for each are handled very differently.

2. Z, you brought up comparing the proposed design to actual industry. Do not switch the conversation around.

3. I did not pretend to have any more numbers than I actually had. It is easy to make up numbers to get whatever outcome you want (see below, though I do feel those are realistic estimates). You analysis is exactly as subjective and biased as mine, in fact my first post had more cost related numbers than yours.

4. If you want to do this objectively I will happily compare cost estimation models.

5. Your post script makes my point for me. You are comparing a mass produce product with one which is intended to be a one off. You could make paper clips in a job shop and you could build prototype cars on an assembly line. But neither would be a cost effective decision. I would encourage students to look at custom high end bicycles; that is a better comparison.

Billet Aluminium Box Steel
Material 75 Material 50
Hr. $/hr Hr. $/hr
CNC Maching 0.75 50 0.1 50
Sheet Metal Cutting 0 50 0.1 50
Welding 0 40 0.75 40
Material Handling 0.1 25 0.5 25

Engineering Cost 3 50 3 50
Design Cost (Drawings) 1 40 2 40

Total 305 333

*Sorry if the table doesn't line up right.

-William

10-29-2015, 09:54 PM

William,

Your analysis is exactly as subjective and biased as mine,...

I gave NO analysis. Certainly nothing I would call an "objective numerical analysis".

Instead, I suggested to the OP, Josh, that HE should do a much more objective analysis of his current design, and also of the many alternatives. I then gave him some hints (as did Goost) as to where he should look for the biggest Enduro-killing flaws in his current design, and how to fix those problems.

I note that Josh is the one person here who seems to be taking those criticisms the best. He seems to be genuinely pleased to have any apparent weaknesses of his design pointed out, so he can make the necessary corrections before comp. (BTW, Josh, widen "toe-base" to at least 6". Easy via a channel-section bolted to upright and carrying the 2 x outer-BJs. Shims between channel and upright adjust camber-angle with no toe-change.)

By comparison, many other people here seem have the mindset,
"Yep, looks great. Go for it. It looks a lot like what we did, which turned out a bit crappy, but was good enough... Yeah, no point thinking too hard about it, just keep copying everyone else and hope you muddle through..." :)
~o0o~

4. If you want to do this objectively I will happily compare cost estimation models.

YES! This is what all Teams should be doing, so I give my version of your (Will's) Cost calcs here. (Changes emboldened.)

UPRIGHT - ............ Billet-Aluminium .......... Folded-Sheet-Steel
$Material - .............. $75 ................................. $10 (Steel is cheap, and strong...)
$Processing (Hr. x $/hr)
CNC Machining - .. 0.75 x 50 ...................... 0.10 x 50
Sheet Cutting - ...... 0.00 x 50 ...................... 0.10 x 50
Welding - ............... 0.00 x 40 ...................... 0.75 x 40
Material Handling - 0.10 x 25 ...................... 0.40 x 25 (Slight rounding.)
Engineering* - ....... 3.00 x 50 ...................... 3.00 x 50
Drawings* - ........... 1.00 x 40 ...................... 1.00 x 40 (See note*.)
TOTAL COST - ..... $305 .............................. $250

* These Eng+Drawing Costs are "one-time", so are somewhat misleading in that they amortise over multiple units (eg. if all four of your uprights are identical). Here these one-time Costs are the same for both approaches, so make no difference.

Except that saying that it "costs" $115 to make each Billet-Aluminium upright, and only $60 for each Folded-Steel one, is a more realistic comparison of their relative costs (ie. Steel is ~HALF $s of Billet).

Also, everyone can argue about the exact size of all the numbers in above table, but the only way to really know is to carry out the various processes MANY, many times over. For example, more practice with welding makes you faster, but CNC times stay pretty much the same.

More importantly, as per earlier post,

... the most important factor is reliability, which here includes fatigue-life and strength (because no-finish = no-points). Next in importance are things like stiffness (important for hub-assemblies), speed of manufacture, mass, and cost, with all of these having to be assessed very objectively with carefully estimated NUMBERS that reflect "point-scoreability".

So "Cost" is only one factor that should be objectively assessed, with several other more important factors to be also considered for the overall "point-scoreability" of the various designs.

By my reckoning good folded-steel designs can win ALL those assessments. Fortunately, this makes the final decision very easy. No subjective "judgement calls" needed.
~o0o~

Your post script makes my point for me.
...
I would encourage students to look at custom high end bicycles; that is a better comparison.

Indeed, many high-end bicycles are exactly the sort of expensive-junk-sold-to-dimwits that I was getting at. Who honestly believes that a $50,000 titanium-plated, carbon-nanotube, funny-Italianish-named, bicycle will make them a THOUSAND times faster than if they are riding a $50 Chinese special? That sort of high-end fashion junk is no different to the $1,000(+++???) jeans that air-heads buy, that come with the knee and bum-holes already in them. :D

Yes, there is a brain-dead market out there, and those fools should be separated from their money. But winning FSAE is NOT about making money in a fashion industry (well, maybe 10 points are allocated to such in Design Event). It is about engineering a reliable, and moderately fast, small car.

Z

theTTshark

10-30-2015, 06:57 AM

Z, have you ever been able to concede a single point to anyone who was born after you?

jd74914

10-30-2015, 09:44 AM

I don't have time to comment further right now, but one thing in that cost estimate really stands out to me-I want to know where you can get welding done for $40/hr. I was a power generation industry guy, but around here the job shops are $75+/hr. If TIG welded I figure 5 mins/inch weld which comes out to maybe 2 hrs of welding or $150 instead of $30. Maybe labor costs are much lower in other parts of the country/world.

Will M

10-30-2015, 11:45 AM

Z,

I gave NO analysis. Certainly nothing I would call an "objective numerical analysis".
…
By my reckoning good folded-steel designs can win ALL those assessments. Fortunately, this makes the final decision very easy. No subjective "judgement calls" needed.
Z
So my qualitative analysis was not quantitative, and my quantitative analysis was not rigorous, but your “reckoning” constitutes a compelling argument?
Well I remain uncompelled and stand by my analysis.
I priced the material, I really doubt the uprights are identical, and the Folded Steel design needs more drawings (flat, folded, welded, machined).
It is absurd to change the model assumptions to give your desired outcome and then claim it is more realistic.

Bringing up performance will not convince anyone about the cost aspect.
Nor will I be convinced by hyperbole* or veiled insults**.

I contend that you offered an opinion (a VERY BAD DESIGN) and demanded objective analysis.
When offered a qualitative analysis you demanded a quantitative analysis.
When offered a quantitative analysis you changed the model assumptions.
I have seen nothing to change my position.

*(Who honestly believes …a THOUSAND times faster)
**( crappy, fashion junk, air-heads)

Josh,

Yes, you need to do you own analysis.
But I think you are on the right track.

jd74914,

Good eye.
Those $/hr are estimate internal costs.
To have them made outside I would also bump the CNC cost up as well.

-William

MCoach

10-31-2015, 03:56 PM

YES! This is what all Teams should be doing, so I give my version of your (Will's) Cost calcs here. (Changes emboldened.)
UPRIGHT - ............ Billet-Aluminium .......... Folded-Sheet-Steel
$Material - .............. $75 ................................. $10 (Steel is cheap, and strong...)
$Processing (Hr. x $/hr)
CNC Machining - .. 0.75 x 50 ...................... 0.10 x 50
Sheet Cutting - ...... 0.00 x 50 ...................... 0.10 x 50
Welding - ............... 0.00 x 40 ...................... 0.75 x 40
Material Handling - 0.10 x 25 ...................... 0.40 x 25 (Slight rounding.)
Engineering* - ....... 3.00 x 50 ...................... 3.00 x 50
Drawings* - ........... 1.00 x 40 ...................... 1.00 x 40 (See note*.)
TOTAL COST - ..... $305 .............................. $250
Z

Having designed and manufactured both sheet steel and billet aluminum designs. I'll have to agree that billet is a reliable way to go. The dimensional control over bearing bores, ball joint positions, and other critical geometry is much better. The reduction in operations are nice from a competition stand point, less handling for things to go wrong. Also from a competition stand point, making an upright that is a "competitive weight" to what a student can easily have done from billet requires incredibly thin steel.

Z, what do you think is a reasonable upright weight?

I would also like to add some cost to your analysis of heat treating of the steel upright (most likely an out-job) and fixturing for both items if I'm to play along and assume that you're treating this as a real business endeavor.
Looking up the material required for the uprights we produced, $56 dollars for 6061 aluminum and $22 for 4130 steel was required (resale cost for both) for each, so in the end, you're really breaking even based on that.

But, aren't we supposed to be discussing the hub? Talk to me, Goose.

From the current industry that I participate in, steel billet uprights are very much the norm. I really don't think there is any argument here beyond exaggerated bench racing where the pet concepts always win.

11-01-2015, 07:46 PM

MCoach,

I really don't think there is any argument here beyond exaggerated bench racing where the pet concepts always win.

So, given that you have "designed and manufactured both sheet steel and billet aluminum designs", why not present here your engineering analyses of the two? You know, with actual NUMBERS? Of comparative masses, ultimate strengths, stiffnesses, estimated MTBFs, etc?

This lack of real engineering on this thread (and on the Forum generally) is getting boring. A comment like "From the current industry that I participate in, steel billet uprights are very much the norm." is NOT an engineering justification. Sheep think like that.

Anyway, it seems that the majority of you young people are beyond help. Yes, you can keep swallowing the claptrap that your teachers have fed you, namely that any piss-poor job you do is worthy of yet another gold star. And then, when the wheels fall of your car mid-comp, you can all cheer "Yeay, more gold stars for us!".

For those very few of you who are interested in using an engineering approach, I have done some more homework and the material cost of a sheet-steel 4130/40 upright is actually closer to $5. Material cost is even less for mild-steel. Both are much stronger and stiffer than an equal mass billet-aluminium upright of the "typical" design (like OPs).

Yes, I have done the calculations! But I won't give you the other numbers here. Whenever I have done that in the past, everyone else shuts-up and the thread goes cold...

Z

(PS. I look forward to the honest billet-lovers out there being true to their beliefs, and "hogging-out" a full FSAE chassis from a single billet of 7075. Yep, ~2 tons at ~$20/kg = ~$40,000, plus some (!) machining time, so you should save a bundle over those damned expensive fabricated steel frames! :D)

MCoach

11-02-2015, 09:07 AM

Just because I can quip my 2 cents (2.8 AUS cents) of my experiences does not mean that I need to divulge my entire report on these. The pictures of the designs can freely be found online with some digging, do your own analysis from there.
I think Josh's design is on the right track: holds the wheel on the car with at least one clamping device, allows the tire to spin, and allows some stopping device to transfer force to the tire. Seems to check those boxes.

"A comment like 'From the current industry that I participate in, steel billet uprights are very much the norm.' is NOT an engineering justification. Sheep think like that."
Is nothing more than a comment. It is not saying "Gokart Fun Rentals runs XYZ therefore we need to because it's obviously a good engineering decision!" Do not attack the straw man, he is only in Oz looking for a heart, I hear.
I think OP's design could use a little weight. Then again, I know I was a little shy on giving away design details when he contacted me. I think the bobbin count could be reduced. I know I spent a long time fighting that point before moving concepts when previous leaders had been convinced that anything less than 8 brake bobbins would fail. Less bobbins --> less weight on the upright, reduced part count, less rotor weight. In terms of assembly...Josh, can the brake rotor be removed without pulling the hub from the upright?

Being attacked for being reserved is getting boring. Being attacked for sharing is boring. Why does anyone bother to share when they are just going to get torn down and pulled apart at the seams for not building a steel barrel, powered by an air cooled motor, that is infinitely rigid, makes 3Gs of downforce at 0mph, and backed by volumes of online forum posts that decry anything but?

Z, childish wars do not determine who is right, only who is left. That is why the threads go cold.

11-02-2015, 08:19 PM

MCoach,

In short, your advice to Josh (and other newbies) is,

* DO only what the majority are doing.
* DO NOT attempt any engineering analysis that estimates the relative merits of different designs.
* ABSOLUTELY DO NOT try to build a car that is faster than those currently running around.

Because that would make the work of other FSAE Teams look really stupid, eh?

Z

tromoly

11-02-2015, 11:02 PM

http://www.mmaplayground.com/forums/i/pi/467288_1.jpg

My $0.02, it's a first-year team with their first-ever car, let them walk before sprinting.

Josh, if you're still checking this thread, hope you got something out of it.

11-03-2015, 06:53 PM

Tromoly,

In ten years time Boston could be a tenth-year-Team that has still not finished a single Enduro. It is quite common...

(Although, as noted earlier, I think Josh has a better chance than many of being successful, because at least he is asking questions.)
~~~o0o~~~

Further to the subject of "Subjective Opinions vs Objective Facts".
================================================
The appropriate saying here is that "opinions" are like "@r$#oles", in that everyone has one.

Each person has their own unique opinion on any issue, and these opinions are generally as fickle and variable as the wind coming from said source.

But objective facts are CONSTANTS. They stay fixed, regardless of anyone's opinion.

For example, put one hundred people in a room, ask them to measure the "mass" of a single given upright, and the result should be one hundred equal numbers, within some tolerance range. If students are doing the measuring, then there might be some "outliers" because some students have not yet learned how to use the weighing-machine properly. If FSAEers are in the room, then, and based on much of this thread so far, there will likely be some useless non-numerical results, such as "it's waaay too heavy!". But such erroneous and non-objective results can simply be ignored.

The key point is that good progress happens when objective thinking is used. The alternative is religion and ideology, and any "progress" there is generally downhill, fast.
~o0o~

So, anyone care to share their objectively measured numbers for their uprights (mass, stiffness, ultimate-strength, MTBF, etc.)? :)

Z

MCoach

11-03-2015, 06:57 PM

https://www.youtube.com/watch?v=y1oxR_ShmYE

theTTshark

11-03-2015, 08:16 PM

https://media.giphy.com/media/5hHOBKJ8lw9OM/giphy.gif

Will M

11-03-2015, 09:59 PM

Not even the hard deck can save you from Maverick!
https://www.youtube.com/watch?v=z_BEJmY911s

Josh,

Just to further derail this thread.
My two cents is to not lose sight of the project management aspect.
If your design meets your basic requirements* then move on.
You could try to optimize it but how does that affect your overall timeline?
If it delays you a week to save a few kg on a component will that significantly affect your tuning and testing time?

How are you currently managing the project timeline?
It looks like you're past the design phase and into the build phase for this year.
For you next car you might want to do something like use a gantt chart to pick a design freeze date.
Start by getting a design for each critical system / component.
Then work on improving the designs until you hit the freeze date.

There are already several good threads on project management.
http://www.fsae.com/forums/showthread.php?5841-FSAE-Project-Management

Just food for thought.

-William

*safely holds the wheel on

Charles Kaneb

11-03-2015, 10:02 PM

An interesting item to do here is to make a toleranced drawing of a fold-and-weld upright.

If you have a location tolerance on the flat pattern of +/- 3 mm on any bend, go find how much extra material beyond the ends of the hub bearing bores has to be left to be able to actually mill them once the welding is done. Now add a bend angle tolerance of +/- 1 degree and determine how far off the two faces will be from each other (this is a good time to decide how to get those two faces parallel). Next, estimate how far the main section and backer will distort when heated to 1500 deg C, fused together, then allowed to cool to room temperature. Finally, calculate how much the position of the UCA and LCA bolts affect camber or KPI (+/- 0.060" on an 8" distance between UBJ and LBJ gives what camber angle tolerance?) and how much clearance you need between the edge of the upright and the wheel.

Texas A&M had billet front and welded-from-sheet (aluminum) rear uprights in 2012. The front uprights came out of the mill, were deburred, had bearings pressed in and spherical spacers made up, and worked as designed as long as the car was used. The rear uprights took days of tack-and-tack-and-tack-and-tack welding to reduce distortion, and took what whiltebeitel would describe as "special toolroom methods" to get the holes in the right places with enough metal around them to hold the load.

Lifespan is determined by stress level and number of cycles; even aluminum parts with no endurance limit can outlast the car.

Tim.Wright

11-04-2015, 02:24 AM

Come on Erik - your 2 most common recommendations of increasing bearing spacing and increasing the toe base are as subjective as anyone else's.

At what point did you ever define what is adequate in terms of camber and toe stiffness for these vehicles?

11-04-2015, 07:05 PM

Still no useful numbers [insert sound of tumbleweeds blowing through the desert].....
~o0o~

But interesting is that rather than well-reasoned arguments supported by objective numbers ... instead we have so many links to the Grand-Wizard of the Scientology cult*!!!
(* Or replace "cult" with - religion, ideology, superstition, witchcraft, black-magic, voodoo, or Optimum-G (<- this last one based on another thread now running :)).)

Ahh..., surely a sign/omen/portent of things to come... :D
~o0o~

Charles,

As any trade-school student could tell you, the final dimensional tolerances of folded-sheet-steel uprights can be exactly the same as that of billet-machined-aluminium uprights. Or better.

University engineering students will likely learn more useful things if they spend less time lounging around their ivory-towers dreaming up excuses for why "it cannot work", and more time out in the real-world finding out what ACTUALLY DOES WORK. Study the "prior art".
~o0o~

Tim,

... your 2 most common recommendations of increasing bearing spacing and increasing the toe base are as subjective as anyone else's.

At what point did you ever define what is adequate in terms of camber and toe stiffness for these vehicles?

I gave those numbers very clearly on this Forum at least as far back as 2005, and many times since then. You should pay attention.

Which reminds me, I am still waiting for a rational explanation from you supporting your claim that high-R% cars have too slow LART to win a race. Are you still working on that?

Z

Joshkb

11-04-2015, 07:31 PM

Back from exams...

Here is an updated version of my upright. Main changes include an uneven spacing from bearing center to LCA and UCA connection points, thickening of bearing walls and walls that lead to the LBJ clevis surface. This allows for a wider toe-base (105 mm center-to-center). The local area around the UBJ clevis surface has also been strengthened by adding material. The two weight reduction holes in the top half of the part save 99 grams. Maximum stress during FEA shows 145 MPa concentrated at the inner, lower clevis surface, specifically on the radius, seen below. The next lowest concentrations are 80 MPa seen in other locations of the upright.

The forces used for this simulation were: 8000 N into the lower clevis surface, 5000 N away from the upper clevis surface, 3000 N at each caliper tab. The upright was fixed at the bearing surfaces.

As for the hubs:

I am thinking the best course of action is to tap three holes in the face of the hubs. When a small (M4-5) screw is fastened into these holes, the cap will prevent the zinc sleeves from 1 - rotating, and 2 - removing themselves axially. Other suggestions are welcome, but I think this is a viable option which solves this issue.

Josh

816

Joshkb

11-04-2015, 07:32 PM

Image is very blurry:

Top (red) value = 1.451 E8
Bottom value (dark blue) = 1.449 E4

tromoly

11-04-2015, 08:33 PM

I am thinking the best course of action is to tap three holes in the face of the hubs. When a small (M4-5) screw is fastened into these holes, the cap will prevent the zinc sleeves from 1 - rotating, and 2 - removing themselves axially. Other suggestions are welcome, but I think this is a viable option which solves this issue.

That's what Monash has done in previous years, I found a picture years ago but can't find it at the moment, in my opinion it's thinking in the right direction.

Jay Lawrence

11-04-2015, 08:35 PM

Josh,

Your calliper mounts look a little light on to me, and your toe base still appears to be very small, but apart from that it looks like a decent 'standard' billet upright.

Joshkb

11-04-2015, 09:01 PM

Thanks for the feedback Tromoly. Seems like the simplest solution.

Jay, I wish I could do FEA or just hand calcs for the caliper tabs in bending, due to the force from the caliper trying to center itself around the rotor. I have no for values for this, and feel I would be guessing if I came up with a value. The floating system should handle this, but we both know it will not absorb 100% of the axial motion of the rotor. One of the issues I have with thickening the caliper tabs (7.5mm current) is trying to fit a nut between the back side of that face and the rotor, + 2 threads showing..

Josh

Fyhr

11-05-2015, 06:11 AM

As for the hubs:

I am thinking the best course of action is to tap three holes in the face of the hubs. When a small (M4-5) screw is fastened into these holes, the cap will prevent the zinc sleeves from 1 - rotating, and 2 - removing themselves axially. Other suggestions are welcome, but I think this is a viable option which solves this issue.

We've had this solution for many years, it works well. Now I missed the "lid" in my quick picture-grab yesterday, but it solves both your points, stopping rotation and axial motion. It also conveniently held the CV-boot and the inner race of the bearing.

The current car has a slightly less elegant solution for the boot, but seems to work fine anyway.

Joshkb

11-05-2015, 07:14 AM

Fyhr,

How does your hub mount to your wheels? Is the left image missing a component (other than rotor)?

Are you running 61816 bearings?

Josh

Fyhr

11-05-2015, 07:45 AM

Fyhr,

How does your hub mount to your wheels? Is the left image missing a component (other than rotor)?

Are you running 61816 bearings?

Josh

The left one used inboard rotors, the wheels mount straight to the hub, see attached image. I don't recall the exact bearing number, but they are SKF deep groove ball bearings, you should of course do the free body diagrams and life-time calcs for your application and select the bearings from that. I snapped a picture of the hat I mentioned too.

craigorydean

11-05-2015, 07:34 PM

11-05-2015, 09:09 PM

Josh,

Thank you. At last, something resembling engineering. :)
~o0o~

Now the bad news. Although you are approaching this the right way (ie. trying to analyse stresses, etc...), most of your analysis is wrong. Not unusual, because most other students make the same mistakes, or bigger (ie. education system down the tubes...).

[Mini-Rant] I never had the luxury of using "FEA". Back in my day we did all stress and deflection calcs by hand. Maybe because we had to do it all long-hand we made sure we made the right assumptions to begin with. No point spending ages on useless calculations. Anyway, it boggles my mind how BADLY done is most student FEA these days. So easy to get useful results, but instead just rubbish. What a waste... [Rant Over :)]

In short, your FEA is quite useless. Do NOT believe any of it. But it also easy for you to get better results.

THE BIG MISTAKE -> You have made an initial assumption, possibly encouraged by your teachers or the FEA manual (?), that the upright should be FIXED at its bearing-mounting-surfaces. Why? And how will you estimate the DISTORTION of this reliability-critical bore surface, if right from the start you prevent it from moving!?

Any structural analysis is simply a case of putting "the structural body" inside a static Free-Body-Diagram. That is, it is a problem in the field of "Statics". There is no need for any "constraints" (which are useful in Kinematics), just "forces in equilibrium" acting on the body and squashing it from different directions. You cannot make any progress here until you clearly understand this point. You must know where the various external forces are coming from, and what are their magnitudes and directions. (This Analysis of Wishbones (http://www.fsae.com/forums/showthread.php?11179-analysis-of-wishbones) thread may help you, but it is quite long and not immediately necessary for your current problem.)

Nevertheless, it seems that FEA programs want you to give them constraints. So, what to do?

THE FIX -> Firstly, sack your teachers and ask for your money back! :)

Next, realize that an upright has 6 Degrees-of-Freedom of movement in 3-D space, so it ONLY needs 6 x linear constraints. Adding any more than the minimum number of constraints allows the constraints themselves to carry many of the loads, rather than the structure carrying the loads. Take this far enough and you won't even need an upright at all, because the artificial constraints (ie. inside the computer, not in the real world) do all the work! This is all too common in the design of FSAE spaceframes, and explains many of those abominations.

One way of doing the constraints is to realize that the upright sits between the forces acting from-road -> to-car-body, and the equilibrating forces acting from-body -> to-road. The second lot of these forces, namely those coming "from-body", reach the upright via the suspension links. Conveniently, you should be able to identify here six linear force Lines-of-Action acting "from-body" to the upright. These are [...drum-roll...] the six suspension tubes! Check: 2 x 2 tubes per wishbone + 1 x toe-link + 1 x spring-damper, or pull/push-rod (aaack!) = 6. Make these the six linear constraints.

Now try some realistic load patterns. Please try this one. To ease the description, I refer to your FEA screenshot at bottom page 4 and assume it is looking at the inside of right-rear upright, with front-of-car to left of image. Do a pure braking force analysis first (because easier to describe here), then only later try adding cornering forces.

The assembly of wheel+axle+upright+brake-disc+caliper has a single force acting on it through the wheelprint, with this wheelprint-force having a vertical upward component, say, Fz = 600 N, and a rearward longitudinal component Fx = 1000 N (just use round numbers for now). Further up the chain of equal-and-opposite force-pairs this wheelprint-force is transmitted to the upright ITSELF as three much larger forces.
1. An up-left force on upper-caliper-mount (pointing to ~10:30 o'clock in your image).
2. An up-right force on lower-caliper-mount (pointing to ~1:30 o'clock).
(Note: Forces 1 + 2 = Vertical upwards force on LoA through centre of caliper pads. It may be more realistic to include the caliper in your analysis, because it actually does constrain/reduce distortion of the upright.)
3. Distributed down-right forces on both bearing-mounting surfaces (distributed from ~3 to ~7 o'clock, but mostly around ~5 o'clock).
The 6 x suspension tubes provide the equilibrating forces. Hope this makes sense, else do the FBDs.

Lastly, and most usefully, get your magical FEA to draw a pretty picture of the DISTORTED upright when subject to these forces. You are mostly interested in the distortion of the bearing bores because that is what kills the bearings.

I would really like to see such a picture, especially of your original, thinner-walled, upright. :)
~o0o~

Also...

Webs inside your speed-holes will greatly stiffen the upright against above distortion. See Pontus' pics for such webs, and I think Goost also suggested this. Basically, leave a "floor" in the speed-hole.

Wider toe-base! The standard prior-art solution of bolting a channel section to the end of the upright that carries 2 x BJs is a reasonable way of reducing toe-slop that comes from BJ-slop. For a given amount of BJ-slop (= outer PLUS inner-BJs!), doubling toe-base = half toe-change.

I repeat Scotty's advice, page 1, "The insert you have in the model needs to be flipped around, and the pin used to keep it from rotating in the housing."

Z

Joshkb

11-06-2015, 01:46 PM

To properly apply forces to my bearing structure in FEA, which I agree is where it should be applied, should I really be distributing the face along the whole circumference and depth of the hole, or only along half the circumference, specifically the half in the direction of the force? The other half seems to just be..."there." Not sure if applying force tot he whole bearing bore would be a valid assumption. Thoughts?

Josh

Ahmad Rezq

11-06-2015, 02:54 PM

Josh,
I think the loads will be applied to half of circumference in case of (Fx and Fz).
Fy will be transmitted from the bearing to the bearing shoulder on the upright

http://help.solidworks.com/2015/english/SolidWorks/cworks/c_Bearing_Load_Distribution.htm

Someone correct me if I'am wrong.

tromoly

11-06-2015, 04:29 PM

Josh, since you have the hub and bearing models, why not include them in your FEA? You can set them as hiddens (I think that's the word? Been a while since I've used SW FEA) so stress analysis isn't done on them, but they can be used to transmit loadings.

I'll leave this here for you to look at:
http://dpcars.net/atom/a251.jpg

jd74914

11-06-2015, 04:55 PM

Your analysis is done in Solidworks, right? All of my terminology is for ANSYS and I personally recommend using it, or at least something a little more powerful than Solidworks FEA, just for the extra options. I haven't used Solidworks in a while (since the COSMOS days) so I'm not terribly useful on that front.

My advice is to run analysis as fully assembled as you can. By this, I mean that I would put the upper and lower ball joint brackets into the model. You can then apply constraints to the bracket/upright interfaces to force them together just as the bolts would or alternatively you can make sure they form compression-only supports which are aligned by the bolt holes (sorry for the poor explanation here). Including the ball joint brackets in your model allows you to better constrain it since you now do not have to assume each is fixed. You can fix the ball joint bolt holes as compression-only supports (just as bolts are since they are not glued to the interface) so they do not limit rotation. You want the tie rod attachment point to limit rotation just as in the real car by making it a compression-only support. You now need a constraint to stop upward movement in the z-direction just like the spring/damper unit does. To do this, put a compression-only support or fixed constraint on the ball joint mount face which opposes upward motion (ie: the one that holds the ball joint with the a-arm that has the damper or push/pull rod mount).

If you can get all of that set up, you can move onto loads.

For loads, calculate (or estimate) the loads you will have at the tire/ground interface. You are going to have an Fx (longitudinal acceleration), Fy (lateral acceleration), and Fz (vertical/normal force). You now want move this force onto the central axis of the bearing since this is where it is going to be reacted. Use a bearing load to apply the force components to the bearing surfaces. Bearing loads distribute radial force components onto only the compressive side using projected area and distribute axial components uniformly like an ideal bearing would theoretically do. Applying the force this way will allow you to see the bearing surface deformations as well as the angular bearing axis misalignment, both which are important when looking at bearing life.

You can load up the brake bolts in a similar way as they will react the same. Go through the same process and assign each bolt a specific bearing load based on the braking force at the pad and moment equations to transform it onto the attachment bolt axis.

Once you get a model running, you’re going to want to do a mesh refinement study to be sure your results are grid independent. To do this, refine you mesh more from the base mesh and run the same exact analysis again. If they results are vastly different, your solution was not grid independent and you should refine the mesh until the solution does not change. Note that grid independence does not mean the solution is correct, just that the math was solved correctly. Since linear FEA is pretty simple compared to other numerical analyses, when you find your solution is grid independent you can usually assume it is also physically correct (assuming your boundary conditions are correct).

To learn more about what I wrote above, look up the “ANSYS Static Structural Analysis (Chapter Four)” on Google and download the first link. Note, it is not current but is easy to find and has some good pictures and descriptions.

Sorry for the long post, I hope some of that makes sense. I was trying to write fast so I cheated and didn’t draw pictures, but if you’re unsure just post up another picture of your FEA with constraints and the forum can critique. Also be sure to put up your assumed loads.

---Edit #2---
Tromoly's way of doing it is pretty good too. He applies the same BC I am suggesting, but does it in a more physical way. There are a few extra constraints.

Adam Farabaugh

11-06-2015, 04:59 PM

There are a lot of wrong ways to do analysis with the bearing races and the interface to the hub. EVERY method that has been mentioned so far is ONLY useful for relative comparisons between your models, none will give you realistic numeric results for deflection or stress. Break out NASTRAN and MMPDS, and get comfortable with contact modeling and FORCE1 cards if you want to really understand what's going on. Then again this is what safety factors are for...

Ahmed: a bit warmer, but those distributions assume a rigid load applicator
Tromoly: your upright is overconstrained.
Jim: as far as I am aware, ANSYS bearing loads also assume a rigid applicator. Of course if you decide that's a reasonable assumption, then that's fine. Also grid dependence is usually a product of using constant strain (linear displacement) elements. Try using quadratic element formulations and solve your models faster and more accurately.

Z's method is closest, but doesn't solve the problem that you still need to constrain something! Any RIGID BODY in 3-space needs 6 constraints, however the upright is not a rigid body, upright strains are what we're interested in anyway! A FBD around the upright ALONE has the forces from links to chassis PLUS forces at the bearing races! Can't ignore those.

Relatively realistic modeling process:
- Solid meshed upright. Prefer hex-mesh, tets are constaint strain elements and thus require much finer mesh to get good results.
- rigid links (use RBARS) with nodes at actual joint locations. Can fix the chassis nodes, or better yet integrate with your chassis CBEAM model! but make it useful before complicated
- RBE2 spiders between BJ lug hole edges and RBAR outboard nodes. This is unrealistic, but you won't get the mesh fine enough to learn anything that you can't learn by doing bearing stress hand-calcs
- Modeled bearing race and hub, likely with plate elements to make it easier and faster
- Linear contact between bearing race and upright surfaces using gap elements
- Apply tire forces and moments via rbe3 to hub rim interface. Extra points for using your rim plate model and maybe even a tire plate model (!!!)

I'm not saying everyone _should_ do this, but if you're _not_ doing this or similar than your numbers are suspect. Maybe not trash, but if you strain gage your uprights in high-gradient areas they won't agree with the model.

Tying into the thread topic a bit more. Sheet welded plate uprights are SIMPLE to model because you can just use CQUAD4s. Will solve literally orders of magnitude faster than a solid model. And you can learn more because "geometry" is easy to change by just changing the property cards.
Why has my team never down sheet welded uprights? Because we are TERRIBLE at welding. In fact before January 2015 we were not allowed to weld in our school! On the other hand, we are damn good machinists if I may say so myself.

Takeaways:
- Solidworks FEA is USEFUL but it sucks for getting real numbers
- Make it useful before you make it complicated
- Garbage in, garbage out

Adam Farabaugh

11-06-2015, 05:06 PM

Of course, since you're a first year team, the past to best success probably involves not doing any FEA at all and just building the thing. I'm not joking. You'd rather have a floppy car that runs rather than a rigid car that doesn't.

jd74914

11-06-2015, 05:20 PM

11-06-2015, 07:32 PM

Adam,

Z's method is closest, but doesn't solve the problem that you still need to constrain something! Any RIGID BODY in 3-space needs 6 constraints, however the upright is not a rigid body, upright strains are what we're interested in anyway! A FBD around the upright ALONE has the forces from links to chassis PLUS forces at the bearing races! Can't ignore those.

Read my post again! I gave essentially the same method as you and Jim, which is to use the six suspension links as the "constraints", and then load the upright at the caliper-mounts and bearing-surfaces.

I also clearly indicated where the "forces at the bearing races!" should go. Namely (quote from my post ->), "3. Distributed down-right forces on both bearing-mounting surfaces (distributed from ~3 to ~7 o'clock, but mostly around ~5 o'clock).", for the particular braking-only case I was describing.

I stand by my view that,
* shortly after students were given calculators ... they could no longer do simple sums,
* and when they got word-processors ... they could no longer spell, or write, OR EVEN READ (!),
* and when they got CAD/FEA/CFD+++ ... they could no longer solve the simplest engineering problems, such as doing simple FBDs!

Grooooaaannn..., mummmble... :(
~~~o0o~~~

Josh,

I await a colourful deformation diagram of your first upright under the load conditions suggested in my last post.

Z

tromoly

11-06-2015, 08:22 PM

Tromoly: your upright is overconstrained.

It's not my picture/upright, I was using it to show the upright/bearing/hub assembly with a piece standing in as the tire.

But just for discussion, would you be talking about the toe link having too many constraints on it?

11-06-2015, 09:36 PM

Tromoly,

"...would you be talking about the toe link having too many constraints on it?

Hard to see clearly, but it seems all three corners of the upright in your image (ie. 1 x top-BJ and 2 x bottom-BJs) are constrained in ALL THREE X, Y, and Z directions.

A typical toe-link can only "constrain" its upright-BJ in the car-coordinate ~Y-direction (ie. in direction of toe-link tube). Any X or Z constraints at the toe-link-BJ will artificially "stiffen-up" the upright.

Result = worthless FEA.

Z

tromoly

11-07-2015, 01:29 AM

Result = worthless FEA.

I went back and looked at the suspension arm the upright attaches to, and actually for that given application it's not worthless.

http://www.dpcars.net/atom/a259.jpg

The arm is on an Ariel Atom, which for some reason has the toe link welded to the suspension arm, so it actually is constrained properly for that particular vehicle. The toe link design is still garbage (way to go Ariel!), though.

Will M

11-09-2015, 08:32 AM

The Ariel Atom has some odd design choices*.
The fellow at DPcars.net documented his efforts to deal with them here: http://www.dpcars.net/atom/
That said it does look great!

-William

*example: regressive shock geometry, the "hole" in the bottom of the chassis where he put braces

Joshkb

11-09-2015, 11:18 AM

Ahmad Rezq

11-09-2015, 01:20 PM

Josh,
I don't know at which point you are in now regarding (Billet upright vs sheet metal upright).
But
How about designing a bent style bracket (laser cut + bending) to be mounted on the upright.
One bracket for both control arm + tie rod joints.
We manufactured 10 brackets with almost 10 EGP ~ $1.25 per bracket (Material which is 4mm St.50+Laser cutting+bending process) and save some time in manufacturing.

Joshkb

11-09-2015, 04:11 PM

Hey Ahmad. That sounds like good way to make these. Should I use this design, I will likely waterjet steel and bend the shape, adding a flat plate at the right side. The curved surface is more difficult to support..

Josh

Adam Farabaugh

11-09-2015, 05:48 PM

While that's common in industry (have a look at knuckles on a lot of cars and you see them dodging the rim/tire) it looks heavy and or floppy. Try this:

http://imgur.com/TgUa0lw

P^squared

11-09-2015, 06:35 PM

Trying to pull off that kind of a shim is similar to teams doing swan neck wing supports. Have fun FEAing to make it safe/stiff !

Also, would you prefer some bump steer over compliance steer ?

MCoach

11-09-2015, 06:36 PM

Before I do FEA, let's get the preliminary design correct, and handle wall thicknesses and webbing later.

Does this look like what people are suggesting I do to increase my toe-base? With 10-inch wheels, it seems to be one of only two options, the second being a side-mounted toe link, rather than beside one of the two BJ. I like this solution the best because it is easier to limit bump-steer.

I only spent 5 minutes on this design so I can come back and iterate, or leave, the design.

Probably way too late at this point, but you can sacrifice a little camber compliance of the upright by shortening the height of the UBJ and toe pick up to be able to spread them out in the wheel further. That would stiffen up your toe compliance. You know, if you really need it. Certified 5 minute analysis.

11-09-2015, 07:28 PM

Google "images of racecar suspension uprights" and amazing how many pics of FSAE stuff you get. This one from SJSU (San Jose State?).

http://aswmachining.com/wp-content/uploads/2014/12/SJSU-FSAE-SR5-006-1024x768.jpg

As Adam and MCoach suggested, by moving the "channel-section" (at left of image) closer to wheel centre (towards right) by ~1/2" allows you to increase toe-base by ~1" (edit: while keeping everything INSIDE the wheel shell).

The channel-section, which is bolted to the upright, but bolts not seen in image, can be machined-aluminium, or extruded-aluminium C-section, or folded-steel as suggested by Ahmad, or even a steel RHS ~20 mm x 65 mm (x 2-3mm thk) which is cut lengthwise to make two channels 20 mm x 30 mm. Extra speed-holes optional...

Z

Joshkb

11-09-2015, 10:58 PM

Ya, I would really like to keep the knuckles inside my wheel so I don't have to extend them so far, which I really don't think is a good idea. The design posted here has a toe-base of 120mm (4.7").

Josh

832

MCoach

11-10-2015, 12:50 PM

Ah, didn't even see that Adam said the same thing. Nice.
I think our rear has a toe base somewhere on the order of 5 - 6", also running 10" wheels.

If you're worried about stiffness and reacting forces from the contact patch, take a look at the Calspan data for the tires you are using. If you check out the pneumatic trail, that should give you an idea of where to point your caster trail. If they coincide there is essentially no distance for a moment to be formed. Work smarter not harder. Of course the trail and contact patch center will move around, but you'll have some general location to aim for.

Fyhr

11-10-2015, 05:24 PM

Google "images of racecar suspension uprights" and amazing how many pics of FSAE stuff you get. This one from SJSU (San Jose State?).

*image*

As Adam and MCoach suggested, by moving the "channel-section" (at left of image) closer to wheel centre (towards right) by ~1/2" allows you to increase toe-base by ~1" (edit: while keeping everything INSIDE the wheel shell).

The channel-section, which is bolted to the upright, but bolts not seen in image, can be machined-aluminium, or extruded-aluminium C-section, or folded-steel as suggested by Ahmad, or even a steel RHS ~20 mm x 65 mm (x 2-3mm thk) which is cut lengthwise to make two channels 20 mm x 30 mm. Extra speed-holes optional...

Z

Classical mechanics teaches us that this movement increases the forces in the lower A-arm by a factor of however much closer to the wheel center it is. And a symmetrically spread toe-base like this gives a larger lever for the lateral forces (mechanical trail, whichever direction you choose), so perhaps a trade-off curve is in order?

11-10-2015, 07:54 PM

Josh,

Before I do FEA, let's get the preliminary design correct,...

Only have time for a quick comment here (lots more details here, but...).

With a design like the SJSU one, or your latest, there is really NO NEED for a separate clevis at the "single-BJ-end" of the upright (ie. the right side of SJSU pic, or your lower-BJ). This is because camber adjustment is done by "shims under the channel-section/upper-clevis" (at left of SJSU pic).

So the "single-BJ-end" of the upright can have its BJ mounted directly inside a pocket machined in the upright. This makes this BJ-mounting both stronger and lighter than it would be with a separate clevis.

This is especially useful for a lower-BJ (as in your design), because the lower-BJ carries significantly larger lateral loads than the upper-BJs. It is also useful when that lower-BJ takes the vertical spring-damper loads. In fact, that lower-BJ should be significantly larger than the 2 x upper-BJs.

Also, doing the upright as above means one less clevis you have to make per side, and two less bolts to fiddle with, etc...

Z

Ahmad Rezq

11-11-2015, 05:08 AM

842

I found this curve while searching for toe compliance.

Ahmad Rezq

11-25-2015, 04:52 PM

Reading Erik post #49 in page 5
I'am wondering how you feel Erik after watching the upright stress analysis provided by SOLIDWORKS ?

https://www.youtube.com/watch?v=2DL0W2JCK5M

JT A.

11-25-2015, 05:01 PM

Reading Erik post #49 in page 5
I'am wondering how you feel Erik after watching the upright stress analysis provided by SOLIDWORKS ?

https://www.youtube.com/watch?v=2DL0W2JCK5M

Oh boy....

Solidworks must use some pretty terrible wheel bearings!

11-25-2015, 07:42 PM

I'am wondering how you feel Erik after watching the upright stress analysis provided by SOLIDWORKS ?

https://www.youtube.com/watch?v=2DL0W2JCK5M

Ahmad,

GROOOOAAANNNNN!!!!!! :( :( :( (... and worse ...)

That 5 minutes of supposed "engineering analysis" exemplifies just how far down the crapper this whole education system has slid!

(On a side note, it is hard to know who is most responsible for this slide. Is it the computer software companies who produce such nonsensical tutorials? Or is it the University academic who first taught the tutor in above clip? Or is it all the other University academics who nowadays happily let their students follow such nonsense, with no corrections? Either way, it seems that all are blissfully supporting each other, with no one pointing out the blatant stupidity of it all.)

Anyway, the "blatant stupidity" of the above analysis is that most of the middle section of the upright could be pruned away to nothingness with NO adverse effect on strength or stiffness. This is because the FIXED CONSTRAINTS at the bearing-housing carry most of the loads that realistically would have to be transmitted to the upper and lower BJs by the middle part of the upright structure! Note how there is no distortion of anything between the three main fixed constraints, indicating negligible stresses there. Utter rubbish!

Note also how the poorly designed "double shear clevises", at upper and lower BJs, also show no distortion of the floppy cantilevered bit. Not surprising, given that the floppy bit is thoroughly constrained by its bolt-hole! Completely back-to-front analysis.

And note, as someone else did on the link, that the braking loads are in the wrong direction.

And, if I am going to get the least bit picky, that the distortion of the caliper mounts, as shown, and with the assumed "fixed constraints", would in reality be much less, because, in reality, the caliper acually does constrain its mounts to a significant degree (ie. the caliper would prevent the top mount from bending quite as far sideways).

All up, an UTTERLY USELESS analysis of a badly conceived upright.

Z

(PS. Thanks, Ahmad, for providing evidence that supports my many rants on this subject. But I am not sure it makes me feel better? :))