Automotive Rear Upright/Wheel Carrier

It is fancy, but not as hard to make as it appears. Took about 5-10 minutes. Sure looks the part though doesn't it? I'm trying to reduce any turbulence in the pour but problem is my pouring technique and initial slosh in the cup is anything but...! This needs work.

With my supply air pressure back up to snuff, it worked like a champ on this go. So glad I built it!

The brushed batch was new and a bit thicker on this part. I started a new thread here.

http://forums.thehomefoundry.org/index.php?threads/dip-coating-polystyrene-foam.573/#post-11356

I've been saying for a long time I need to work on coatings. -Time has come.

Completely forgot. Have a look at the new thread I started.

I wasn't too concerned because of it being a sample part but I didn't realize that seam was as bad as it was. Just goes to show, the pattern needs to be near perfect or the ability of the coating to reproduce surface flaws will betray you.

No! I'm considering making changes to get it right!!!! Already have a theory and plan of action.

No. Alcohol is one of the few solvents that will not attack polystyrene. I can at least say so when it comes to Ethyl and Isopropyl. I have not (intentionally) tried Methanol.....but it probably does not attack it either. I've painted Shellac on Polystyrene which is reduced with Alcohol and it will seal the foam to the extent it can be coated and painted (sparingly) with things that would otherwise attack it.

Best,
Kelly

 

Going to need a little further explanation on those Joe.

1. When you say each color is a different casting process, like die, permanent mold, shell, sand, etc?
2. What are the limit lines and average value in each color? Are those the values of the respective mechanical property measured at different locations on a given part? For a given alloy?
3. What do the values at the limits on the color bars correspond to?
4. What alloy(s)?

Best,
Kelly

 

Thanks Joe,

Makes sense now and certainly relevant given the intended alloy and purpose of the subject casting for this thread. The mean yield and ultimate stress are fairly consistent with the values I've seen suggested for sand casting, but I was more accustomed to those being expressed as increased safety factors (most conservative for sand casting) against published data.......but you arrive in the same place as far as design margin. The scatter appears to be circa +10% of the mean except elongation which is a bit disconcerting as I'd rather bend than break. What that data means when applied to my process may be another matter.

I'm going to give it my best and see where I get to. I'll have some time over the holidays and hope to make good progress. I have four more foam patterns about half completed. The next pour will be the application of (my) current best practice in prep of pattern, mold, metal conditioning ....and if the result is worthy, heat treatment.

Best,
Kelly

 

Just to let you guys know, I haven’t been loafing. I decided to make four more patterns; #3-#6. I’ve said it before…….once patterns become cheap, so does experimentation. I’ve added some more simple machining fixtures to help with reproducing the patterns but there is still a lot of features and work in them, so although not exactly cheap, they’re certainly more plentiful.

I was pleased to see my local Home Depot is now stocking Foamular 150 4’x8’ sheets in 2” thickness so I bought two sheets. Previously they had only offered Foamular 250 which is 25kg/m3 versus 20kg/m3 of the Foamular 150, so these patterns will be made entirely of the 150 material whereas the previous patterns about 50/50 of the 150/250. First order of business was resizing that stock and making the bodies. I won’t go into the details but making these shells repetitively and identical is the hardest part of the pattern build. Because of the complex geometry, the butt joints are not 90 degrees and require some hand tuning before gluing.



There are several refinements in pattern construction and assembly since Pattern #2. The biggest change is how lower section of the casting that houses the pivot shaft of the upright interfaces with the rest of the pattern. Below shows the previous and new version.



The old design cast surprisingly well on Pattern #2, but the sharp included angle collected pattern coating that was hard to dry, hard to remove after casting, and also ran the risk of flaking small coating pieces as the pattern was displaced by molten metal. As-cast it was also somewhat of a stress riser. I had to hand tune the previous joint with a 2” spindle sander sleeve and even so, I just didn’t feel I could hold a reasonable positional accuracy. The new approach with the butt joint eliminates most of these issues. I’m not crazy about the continuous (instead of staggered) butt joint in the foam pattern but I can control the joint fit, accuracy, and repeatability so much better, I think it’s an improvement.

Here are the individual pieces. I plunged with a bottom cutting router bit for the two diameters. Using router bits makes for a better finish on the bores than Fortsener bits. I then used this simple fixture to cut the profiles, and assembled them…..



....…and used another simple fixture to machine the internal radius and core windows which are two slots this time instead of three holes drill Forstener bit. Unfortunately, the width of the pivot support cylinder must finish at 6.125” and my foam stock is only 2” thick, so it required four joints. I would have preferred three.



The internal core windows were moved to the top edge of the roof of the largest core cavity where previous mold instability (leaker) occurred. I also think this was the source of the surface flaws in both the outer wall immediately above the suspect area and the inclusions in the bearing bosses. An air void is a very undesirable thing in an unbound sand LF mold. During the pour there was once again a slight burp back and this may have been when the advancing metal front reached the air void. I think the surface flaws and inclusions were from displaced pattern material, broken bits of coating, formation of oxides, and the air bubble buoyantly traveling late in the pour.



Some further refinements were made to the assembly fixture to quickly position and secure the patterns. It locates and helps hold the positional tolerance of the various casting features. This fixture also does double duty to cut the interface of the main body with the respective the bearing carrier cylindrical bosses and enabled good quality joint fit.



The brake and parking brake caliper have different mounting hole centers. The last upright had a different wing for each caliper making for a right-and left-hand casting. On this iteration I made the wings the same and the third boss can be left, or machined off if desired. This means a single casting can accommodate either, and the casting machined as a LH and RH version. So I took the easy route, laid out the wing on some thick foam, plunge cut the boss locations, and then used my hot wire to slice identical wings. My 1” plug cutter knocked out the bosses and made short work of these features.



With that I used the assembly fixture to glue the caliper wings and bearing bosses in place....getting pretty close to complete patterns now.




So, I need to cut the core windows, detail them up with wax fillet, add gate/sprue, dip coat, and……..you know what comes next…..

Best,
Kelly

 

But then he would have to design the part in cad. Take the cad model and put it in a cam program and generate the g code then onto the mill.
Right now I think his current method works out great without the need to learn cad/can and the expenses that come with it....
ROCK ON KELLY!!

 

I do want to build a CNC router. Without argument, improvements in accuracy and repeatability are there for the taking.

That's really the primary issue for me right now. When dreaming up a part, if your natural inclination is to dream via 3D solid modeling in your favorite CAD software, then that may be a time saver.....but that's not how I dream. Could change though.

I would extend a healthy challenge to the assertion of less total invested time via CAD CNC. I'm sure from your comments you would be surprised by how quickly I can knock out a simple MDF pattern and fixture. Once I do, the cycle times for cutting are literally a few seconds. If you are talking the typical low end CNC router spinning an 1/8" bullnose bit with a 1 to 1 1/2" cutting length, for the size and complexity of parts I'm doing (and planning), I think you would quickly find you couldn't reach many of the features with the cutter, and with any surface driven step over program on that size bit, your run times would be >10hrs hours if not a day or two. If you have somewhere to let the machine run for that time I guess you wouldn't care, as long as the software was stable enough that it didn't crash. The real time advantage to the pin router is cutter selection and the amount of stock that can be removed and pattern surface area cut.....in seconds. If you're CAM software is basically 2D with height control, no big deal but if you were to have a cutter library, cutter changes, and complex surfaces, it can be a different ball game.

I still want one!

I think Cojo and several others have one started here.

Thanks Pat. I get by. It's therapy for me.

Best,
Kelly

 

Yesterday was a pretty light day but I did finish up some pattern work. I cut the core windows in the previous two patterns by hand and then tuned them to the ID of the embossments with an Exacto knife. Given I had 24 holes to do on these four patterns I said the heck with that, made myself this little mini-router base for my die grinder:



Then blasted a couple holes in some MDF and taped them in position onto the patterns.



Cro-Magnon, but fast and effective. Pattern fab is complete. I detailed the one in the foreground.



These patterns are assembled from 26 individual pieces. However, 12 of those pieces are in the brake caliper wings. You can see they were simple piece parts, and were very easily assembled, but it’s a good thing I’ve become proficient at laying wax fillet because there is more fillet work in the wings than the rest of the pattern.




I was planning ahead because after my New Year’s Eve celebration last night, laying wax fillet is a more fitting task that running machinery.

Best,
Kelly

 

@garyhlucas I've been using alibre cad for about 10 years now, the picture in my avatar is something I designed in the first two weeks of learning cad. I'm pretty new to cam and cnc in general but do ok. I use sprutcam for generating my g code, but had considered using cambam in the past. Now a days it seems fusion 360 is the way to go. They say it's free for students and hobbyists but you never know when that might change.. I only wish there was more time in a day so I have more time to get better with it..
Too many hobbies and interests....

 

Select something challenging and useful to make, start a thread, cut some patterns, cast the parts, post some pictures and videos...

Best,
Kelly

 

Thanks Cojo.....I'm workin it!

I think that's right. For me the CNC makes sense for complex surfaces and blends. For instance, if I could machine this part as two pieces with the parting line something similar as shown, that could be a big benefit.....



....but like I said, the run time and finishing work from whatever step-over chosen would still be a consideration. That parting line would have to be perfect and of course, there is deep machining required from each side. That part is 7 3/4" tall as shown. That would require a lot of vertical axis travel, distance under spindle, and some thoughtful fixturing.....or break it down into more pieces and assemble.

Best,
Kelly

 

I have "mill wizzard" for the quick stuff, but still find it easier sometimes just to throw the part on the bridge port and do things manually.