Automotive Rear Upright/Wheel Carrier

Discussion in 'Lost foam casting' started by Al2O3, Nov 25, 2018.

  1. JoeC

    JoeC Copper

    Beautiful work as always.
    What is the function of this part, I have a concern if it is a suspension part.
    Such parts are considered safety critical and subject to seroius process control, 100% x-ray, maybe FPI, hardness, tensile, microstructure, destructive, functional tests, and on and on to keep the lawyers away - at least in automotive supplier world
    The oxide folds from the melted foam would be a pretty severe stress raiser if the part is subject to fatigue loads.

    Now for a homebuilt car, myself I would drive it and just keep an eye on it

    I don't want to be negative but take a look at that steel part, a ton of engineering in it, and it is steel for a reason (or maybe it is ductile iron?)

    Here is a steering knuckle being x-rayed


    Low pressure casting


    Appropriate process control - there is no such thing as overkill


    One steering knuckle maker in Michigan extracts castings with a robot and if the computer reports any out of process parameters, the robot passes the casting through a saw to scrap them rather than deposit them on a conveyor to sort later

    OK, I dont like how negative this sounds, just go ahead and give er!

    If I was the Wright Brothers I would be worried about what could go wrong and still on the ground....
     
  2. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    It is a suspension part. The fella I'm making it for has obsessively lightened his car 600lbs+ from 3200lbs and it still looks like the street car from which it started. I told myself I wasn't going to make any critically stressed parts and did offer similar cautions to him that you expressed. He was considering using a weldment which have their own set of issues so I told him I would describe exactly what the part was, how it was made, and yes, use at his own risk.......but that's the way it is when you modify and build cars. I have no intentions of manufacturing them for the very reasons you cite. It's a one shot deal.

    He is an engineer and an attentive mechanic so I'm sure he will keep a close eye on them. I suggested he polish them. Not for the bling but so as to easily identify cracks.

    Not at all different than my initial thoughts and sentiments. I will say there are a sh!t pile of high power kit and replica cars with sand cast aluminum uprights running around. I own one of those too. I'm not aware of a single cast upright failure. Mine is 540Hp V8, 2500lbs, sticky track tires, and I can tell you I've beat the piss out of it!!!

    With respect to the specific parts of this post, my car is a street car. I'll stick with the OE uprights myself. It's a high power car with big soft tires. I was told the OE parts are malleable steel. Funny thing, my first remark like yours was "sure they're not ductile or malleable iron"? Nope, steel. Not sure of the source for that info. There's strength and then there's toughness, and steel is darn tough, if that's what is really is. Oddly, tough materials are probably more important in street cars because of pot holes and impulse loads. For race cars strength and lightweight matters. For one-off track use I'd probably have chosen a fabricated 41xx series Chromoly. Requires more care when welding and very strong but when they fail, they often fail catastrophically.

    Thanks Joe.

    Best,
    Kelly
     
  3. JoeC

    JoeC Copper

    Have you seen this trend?

    Concurrent FEA, fluid flow and solidification modelling ends up with a creepy looking bone structure

    Lends itself to 3D mold printing

    So there you go, creepy looking as heck but you can see insane strength with the least weight

    I have a nice article but it is too big to attach, what link method works now (still bitter about Phototbucket)

    Creepy Upright.jpg
     
  4. JoeC

    JoeC Copper

    If this gets too annoying feel free to just tell me to go pound sand

    For fatigue loads, the sharp feature at the what I guess to be a bearing race concerns me, can you transition it into the main body more smoothly so it isnt a possible stress raiser

    The thick isolated boss may be prone to shrinkage, hard to feed through the thin walls, hard to feed,

    Could open up voids when the holes are drilled, can it be pocketed, maybe thinned and supported by gussets

    Had a bad experience with fatigue failures on a VW engine mount, took years off my life and now I see every part in terms of fatigue

    Worked with Alcan fatigue experts, it is a very complex issue, the smallest things matter

    features.jpg
     
  5. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    No worries Joe. Always open to comments from pros. That transition won't be there in the go live part. I just stuck this pattern together with limited fixturing as a trial and development part. If I get this one to cast I'm just going to cut it up to examine it anyway. The patterns for the real part are yet to come and will be more refined.

    Yah you'd have to have read all my ramble in the earlier posts but it's only thick like that because the fella I'm building them for couldn't give me the mounting details for his brake calipers (yet). The deliverable part will have webbing of similar wall thickness to the rest of the casting with cylindrical mounting bosses on the proper bolt centers. If you have a look at pictures in post #15 you can see I already whittled some mass out that area of the trial part for the very reasons you cite.

    Best,
    Kelly
     
  6. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Is that a theoretical part or a real part? There's theory, and then there's practice.....sometimes the two coincide ;). How big is your article? email it to me. Do you still have my email? If not message me here

    Best,.
    Kelly
     
  7. OK, am I the only one who really does like the residual chemical smell from the foam? I try to avoid sniffing it because I'm sure it's not the best for long term health.
     
  8. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    You and Tobho Mott. See post #14.:)

    Best,
    Kelly
     
  9. I'm not so sure, the Esteemed Mr. Mott is given to literary license and was quoting Francis Ford Coppola and further called the sand in question "stinky", which although it may have been an effort to win over his skeptical audience, I'm still on the fence as to whether he truly does love the smell, and has been trying to simulate it in a cologne for his wife, as I have.:rolleyes:
     
    Tobho Mott likes this.
  10. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Finally had a break in the weather and managed a casting session Saturday morning. I had a visitor as well. Our own AlPuddle stopped by so he can attest. I managed two lost foam pours; the subject of this thread automotive upright and another intake manifold lid.

    I didn’t take any pictures of the lid other than the final casting as it was a more or less a play for play repeat of the previous thread.

    I did have an issue with my compressed air source that was somewhat problematic. I updated my lost foam rig thread with those details. Though I could see some evidence of inadequate packing in several areas of the upright, I’m not sure the results would have been much different.

    So here’s the upright sprued and placed and packed in the flask.

    11 Sprued.JPG 12 In Fask.JPG 13 Full Flask.JPG

    The A20 crucible was charged with about 15lbs of aluminum. Since this was a trial part, it was just wheelium and recycled automotive castings.I wasn’t particularly accurate hitting the cup during the pour and also had a bit of a burp fairly deep into the pour, which may have a been a first for me. Sorry, no video. Here’s the result immediately after demolding. Superficially, it appeared to be a successful pour.

    14 Demold.JPG

    And after a little brushing and clean up

    15 Exterior Clean Up.JPG


    Upon closer examination, it was a mess. It appears that the hydrostatic pressure of the pour floated the loose sand cores upward and inward. As a result, the portions of the casting that are essentially the roof of the unsupported cored areas (note 1) are paper thin. So is the cylindrical feature through the center of the casting (note 2). Both of these features were ¼” wall on the foam pattern. The float also added thickness to the lower wall (note 3). The bore through the lower cylindrical feature was probably insufficiently packed but also destabilized by the shift of the other cores.

    16 Sectioned.JPG 17 Sectioned.JPG

    Some of this could be partially addressed with a change in orientation of the pattern in the mold, but with the core movement and apparent mechanism, any confidence my former experiment suggested was possible with an unsupported loose sand core has been diminished. I think a change in part design is also required as this is obviously nowhere near an acceptable result given the nature of the part. It takes a lot more effort to make these patterns than the simple 4 cylinder engine block I referenced in post #1. I’ll have to see what the next iteration yields and reassess.

    Not all was lost. There was some learning and cause for encouragement. The part only weighs 6.75lbs.

    18 Weight.JPG

    Assuming the core shifts still netted the same total volume of metal in the casting, that is very encouraging given the target weight was 8lbs. That means I have some room to maneuver with a little design optimization. If the cored areas were stable, the part would very likely have filled and perhaps cast successfully and given the complexity of the part, that aspect is encouraging as well.

    I’ll also need to make sure my compressed air supply is at full force for any future endeavors as this is a very important aspect for packing lost foam molds.


    Best,
    Kelly
     
    DaveZ likes this.
  11. PatJ

    PatJ Silver

    Not a bad attempt at all, especially considering such a complex part.
     
  12. Amazing how the core just seemed to shift rather than fall apart.

    Did you try to get mud inside the cavity? I assume not because that was not the idea and know your earlier tests worked great. Be a great place for a dip solution.

    Can you make a large opening into the center cylinder? Maybe press a steel shell into it if it needs to be continuous?

    Just thinking out loud.

    Still learning going on and I'm enjoying watching.
     
  13. FishbonzWV

    FishbonzWV Silver Banner Member

    Interesting.
    I can see 1 & 3 happening but #2 is quite the surprise. I'm having a hard time fathoming the reason.
    Number 4, you win the prize for the largest leaker ever!
    Vertical orientation next?
     
  14. Looking again, it is interesting that the loose sand core moved as a unit, upward and inward. Your feed nozzles were too small to resist the buoyancy of the core, which was compacted enough to stick together. So we know a very large feed nozzle will correct that, the question being how small can the feed nozzle be and still keep the core from moving. The feed nozzles may have just been too close to the grain size to allow a locked column of sand, so holes double in size might work. Are the holes 10x or 25x the grain size? Maybe one larger feed nozzle in the center would prevent movement where two smaller ones didn't because there is more sand locked together away from the edge of the loose sand column.

    Another thought might be to coat the inside of the chamber with sheetrock mud prior to installing the bottom panel so you could get a mud membrane on the upper sections and through the holes to give a rigid shell to avoid float.

    Very interesting issues.
     
  15. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I did not try to get mud coating inside the cores. I don't think that would affect anything except the surface finish and they would be hard to dry out. I think the fact that the core shifted intact means the two small inlet and outlet holes did allow the cores to adequately packed, however, are analogous to a bound core with an insufficient core print surface area to prevent core float. While the aluminum is molten, there is a couple inches (Inches AL pressure that is) more static pressure working on the core and it only has those two small 3/8" diameter loose sand plugs to hold it in place. The "burp" I mentioned far into the pour may have actually been when the roof of the foam pattern in those cored areas became replaced with molten metal and the cores abruptly shifted position.

    I too was comparing it to my earlier experiment and I think this arrangement is actually more analogous to the "pill" in that experiment which did the exact same thing....floated the core. The engine block had the intersecting sand cores of the four cylinder bores of the cylinder to help hold everything in place......which I think is instructive and leading my thoughts for the next iteration, and hopefully a moment of enlightenment.

    I'm with ya Bonz. At first that was my reaction too, but if you think of it as just buoyancy of the core floating in the molten aluminum, the whole core is just trying to float to the highest possible location within the molten aluminum thus the shift inward as well as upward. I think the core shift exacerbated the leaker at "4" but frankly, this was surely aggravated by my low compressed air supply pressure and low energy vibe.......so besides getting my supply air issue worked out, I have a theory and new approach.

    I don't think an orientation change alone will prevent the core from floating if the core is still completely unsupported. It will just float (up) toward a different location of the casting. I think I need a small window in each side of each cored area. Like shown below.

    19 New Core.jpg

    If you subscribe to the notion that vibrating the loose sand essentially binds it (and I do subscribe, albeit weakly bound), then each window is analogous to a core print that connects the respective cores to themselves and the external mold, thus providing stability and hopefully immobility within the molten aluminum. I'm thinking the windows will be a small slot I cut with a 1/2" diameter router bit, or perhaps a round hole, and maybe embossed a bit to reduce the extent to which they becoming stress risers.

    I also completely forgot to exploit the fact that my new LF casting rig allows the orientation (tipping on the trunions) of the mold to be changed during vibration and I intend to apply that combined with the pattern orientation when vibrating/packing the next mold.

    There is at least one inconvenience to this approach. With all the cores connected together, the core windows/prints in the two trapezoidal cored areas will expose working parts of the inside of the upright assembly to the ambient, rain, dust and the elements. The cylindrical areas hold lubricated bearings so I'll have to make rubber plugs to seal those holes.......a small price to pay if it makes the part cast well. I'll try to further develop this train of thought this week.

    Best,
    Kelly
     
  16. Rasper

    Rasper Silver

    I can't help but worry about aluminum uprights. 7000 series aluminum, maybe I could listen to. But 356?

    Of course it depends on the use. The Nascar boys used to use the 1967 and '68 Cadillac front uprights. They were about the strongest ever made. (Like the '37 and '38 Cad/LaSalle gearboxes.) A broken upright at 175 mph is a major disaster. Whereas for a show car, it won't ever happen. In a street machine, it depends. The way I drove back in my day, it would have been a disaster as well. Back in the 1950's, when there were no cars on the roads at night and almost no cops, we used to road-race until dawn. We could outrun the few cops there were. The country roads around Richmond, Virginia were like the wild West must have been. Driving used to be fun. It's hard to imagine now.

    Richard
     
  17. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    The upright that is the subject of this thread has the interface dimensions and parts compatibility for a DeTomaso Pantera. The picture below is an aftermarket billet version of same.

    Montage.jpg

    The Lamborghini Miura had a cast aluminum upright as shown below. Higher strength alloys were (more limitedly) available, and anything may have been possible with Lambo, and though I can't say with certainty, I doubt they bettered A356 in material properties in the 1960s.

    Miura Upright Montage.jpg

    The picture below is a sand cast upright hub assembly in the style of the original Ford GT40 upright.

    GT40 Original.jpg

    The one shown immediately below and others very similar in design to the original are used on all the GT40 replicas. Obviously influenced my design mods from the OE to the cast aluminum Pantera upright of this thread. My Pantera weighs about 400lbs more than my GT40 replica but the fella's Pantera who wants these cast uprights Pantera is about on par weight wise with the GT40.

    MDA Upright GT 40 original copy.jpg

    I have about a dozen other examples of very low production volume road and track going aluminum sand cast rear uprights (also many Chromoly weldments). The comments offered about casting flaw populations, process integrity, and criticality of the part are certainly valid concerns for a cast aluminum upright. Not at all trying to be dismissive of those concerns but the deed has been done many times before......as I mentioned in the first post, not necessarily my first choice but a choice. -But I digress.

    If I get things sorted out with the casting, I may consider other alloys, but still be aluminum. I need to see how far down the road I can get this one before any decisions about use.

    Best,
    Kelly
     
  18. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    OIF, You sneaked this post in on me while I was posting #35. I think the sand is about 50 mesh so the core feeder openings are probably on the order ~>25x-30x grain size. Per my commentary in post #35 we're on the same wavelength regarding increasing the number and size of core windows in an effort to immobilize the cores. I don't think a mud interior shell would do much for strength though......at least on at the thickness I apply them. The cores could also be packed with bound sand...but they'd still need to be immobilized and bound sand for a lost foam caster.....what fun would that be ?:rolleyes: The next attempt will be a more concerted effort.

    Best,
    Kelly
     
  19. I know 356 is a common alloy used in aftermarket cylinder heads. I remember reading several years ago that control arms on Corvettes are A201 or A206 alloy. I can't remember which it was, but I do remember that this alloy has a small amount of silver in it. You might want to research this alloy for future uses.

    Tom
     
  20. Interesting pictures with healthy openings to the cavity.:rolleyes:
     

Share This Page