Announcement

Collapse
No announcement yet.

Das Beast: My E30 track / street build

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • dvallis
    replied
    Awesome. Thanks for the great info. The machine we're getting is from Northern Tool, also a clone of the X2, Harbor Freight etc. R8 spindle so tools galore. We plan to eventually upgrade it to CNC, so we'll be in touch.

    Nice flywheel work.

    Leave a comment:


  • whodwho
    replied
    I needed some machine work and the list was growing and with everything shutdown I recently picked up one of these for some of my projects that this can handle. It is a great little machine, it is a re-labeled Sieg X2,.same as the one at Harbor Freight and many other vendors. It will work fine in stock form with a good cleaning from all the cosmoline grease and will need to be gone over and tightened up and gibs adjusted on axis. It uses R8 tooling and is the same as a Bridgport so you can find good deals on tooling for it.

    A couple links that will be your goto sites
    https://littlemachineshop.com/
    https://allindustrial.com/
    https://www.cnczone.com/

    I am in the middle of converting one to CNC so when it comes time I can help answer questions you may have.

    A couple teaser pics of some recent additions and first project (drilled the bolt holes got the boring head to open the center hole on the flywheel )


    Leave a comment:


  • dvallis
    replied
    Well, we have been meaning to buy a mill anyway to THIS bad boy is on the way. Will be way easier to make all manner of aluminum parts. Might upgrade to CNC later. For now we want to cut metal, not mess with building a CNC.

    Leave a comment:


  • dvallis
    replied
    Prototyping the load cell mount on some drop before we cut plate.We don't have proper tools to punch 1/4" plate so that's the only PITA work.







    Data looks good. Very linear. We'll calibrate the plates separately and and keep a curve in the MCU for each one.

    Time to fab.

    Leave a comment:


  • dvallis
    replied
    Slight COVID delay. Back to weight scale fab.

    Here's the raw material. 40 square feet of 1/4" aluminum plate from the local metal shop. Slightly dinged so I got a deal.



    Circular saw made short work of it. Here we are cutting the 4' x 1' plates into squares.



    Cutting all done. Here are most of the blanks for turn plates and scales.



    Ready for drilling, tapping and counter sinking.



    Here's the final product for the top of one corner. 1/4" aluminum plate with 3/8" reinforcement and load stop all the way around. We'll put 4 more screws in the middle of each side next time. Now that we have the process down we'll make some steel fab jigs to speed up the work.






    Leave a comment:


  • Old525i
    replied
    Hats off to you sir, you are indeed a jack of all trades!

    Leave a comment:


  • dvallis
    replied
    Tuned on 93 octane pump gas

    Leave a comment:


  • pinki
    replied
    Congratulation !Run on E85 ?

    Leave a comment:


  • dvallis
    replied
    Got my run files from the last Dyno session. Really happy with the final results. Thumbs up for Roman at Boost Logic. Take your E30 there if you want a really good tune.

    Will probably race with 10 psi. Love that torque curve.



    20 psi is just for showing off :-)

    Leave a comment:


  • dvallis
    replied
    We did a lot of design today, and a bit of fab.

    First off, we wanted to understand how much the load on each cell would change if we were a bit off center on the scale.

    In the picture below

    R1 = F x (3A +B)B^2 /L^3
    R2 = F x (A + 3B)A^2/L^3

    F = 600 lb
    L = 12"

    Obviously if A=B, R1=R2=F/2 (do the math)

    Moving the load point left of center by 0.25", A=5.75, B=6.25

    R1 = 319 lb, R2= 281 lb (74 lb delta)

    With the load one inch off center left, A=5, B=7

    R1 = 371 lb, R2 = 226 lb (19 lb delta)

    For small displacements in the middle of the scale Delta_Load = 73.3X + 0.7

    Note this is a beam equation. We're on a plate. The 371 lb load is actually shared by 2 load cells, so 185 lb each. No problem.

    Summary: We're not going to overload the cells unless the tire is off center on the scale by 6". And even then, it just hits the bump stop. And this is worst case, since the tire is a distributed, not point load. So, all is good.



    Next we need turn plates to align the front end properly. There are two flavors of turn plate online: "cheap ass" (that's a technical term) and really freaking expensive. As in $600 - $1,800. I really just wanted buy a pair. Ungh.

    Rob aka "Pack Rat" (in a good way) pulled THESE our of some corner of the shop. Whaaa? They are bar stool bases. Large plates rotating on ball bearings. Nice.



    One had thrown bearings, but we fixed that, repacked them and resealed the unit.



    Alright. Gotta reload on aluminum plate, bar stock etc. and then we'll get back to it.

    Leave a comment:


  • dvallis
    replied
    Mechanical design is shaping up. Beam bending calculator says the load cell will only deflect 0.005" with 850 lbs applied to it, so we should be Ok with 1/16" travel until it hits the bump stops. (0.0625") I had to make some minor changes due to the reality of bolt length availability. still, looks good.

    Last edited by dvallis; 05-30-2020, 07:26 PM.

    Leave a comment:


  • dvallis
    replied
    I like how the User Interface is coming along. Here's the display with corner weights, total, percentages and cross weights. Pretty cool to have everything on one screen.



    We're back to working on suspension setup. Substitute for my 225 lb self in the driver seat.



    I didn't get a photo today (duh) but we have the chassis level and this known square frame attached. Strings are mounted level with the ground, about even with the wheel hubs (not as shown). We're using this as the known baseline measurement for ride height.



    Adjusted each corner of the suspension until we got even ride height on both sides, obviously with the front lower. You can see the string height. Ended with with a ride height such that the inner control arm ends are 1/4" above the outers, so no need for roll center correction. This should give us decent corner weight as a starting point.



    Leave a comment:


  • dvallis
    replied
    You're welcome. I could only find 100 kg (220 lb) planar load cells so three wouldn't be enough per plate. Could have made them but it's just one more thing to do. 4 x 220 lb gives me enough capacity. Even then, these won't be "drive on" as two load cells won't take full corner weight. Will have to jack each wheel and slide them under. Not a big deal.

    Leave a comment:


  • ForcedFirebird
    replied
    Thanks for posting this public!! Hopefully it's uploaded to r3v database and kept.

    Why not use 3x 220kg per pad for 660 per corner? Based on simple geometry, three points create a plane, four adds a variance. Pretty sure this is why the tire balance machines have 3x pads. Personally, I was cursing yesterday since the tire balance machine has 3x contacts, and the one leg just had to land on the concrete expansions joint, no matter how many times the stupid thing was moved back-n-forth 6". Grrr

    Leave a comment:


  • dvallis
    replied
    Finished the scale design. Pretty simple really.



    And now we build it. I had all the parts laying around. LCD first, just to get "proof Of Life"



    I'l save you the gory details of the MCU code. First power up ..... It lives! Swoosh. Nothing but net.



    Going to work on more suspension tomorrow.

    Leave a comment:

Working...
X