heinzboehmer's 2002 Topaz 6MT Coupe

[heinzboehmer]

Are you using a maker space? I had access to one in Wichita, KS. It was outstanding! Wish I had something here in Houston. All the ones I've found require you to be a college student.

Did this at the machine shop at work.

I'm surprised about the college student thing. Over here, the maker spaces don't care who you are. All they care about are your monthly membership fees

[bigjae46]

Are you using a maker space? I had access to one in Wichita, KS. It was outstanding! Wish I had something here in Houston. All the ones I've found require you to be a college student.

[heinzboehmer]

Finally got myself in the machine shop.

3D printed fixture worked incredibly well. I could have probably made it half as stiff and it would have still been totally fine. Playing musical bolts to get to the bosses was a bit annoying, but no big deal

Machine operator worked okay. I do have experience on a mill, but this is my first time ever on one without supervision, so go easy on me.





Ended up removing 4.5 mm from bosses to new flat surface. This is 0.75 mm more than I had originally thought I would need, but that's totally fine with me. My brackets will be 4.75 mm thick, so the final assembly will sit pretty much exactly where the stock ones do.

I did forget one thing when designing the fixture. The strut bar surface is angled on two separate axes compared to the strut tower surface, so the fixture needs clearance on the sides as well to be square on the vise when using those two reference cuts I added. Got around that with some creative shimming, but this was far from the stiffest setup, as you can probably tell.

My "lightest of skims" turned into a 0.75 mm cut, which I'm not thrilled with. I fell into the trap of trying to get the surface flat with a subpar setup and removed more than I wanted to. I kinda wish I had just flattened the high spots with a file, but oh well. It shouldn't really make a difference in the end, but lesson learned.

With the final dimensions of the stock brackets known, I jumped into CAD and adjusted my brackets. No big design changes, just some planes moving around a bit:



Also added some extensions to the flanges that were too small for sendcutsend to bend:




And done! They're off to be manufactured now. Next up is finding someone who can weld these up for me.

[heinzboehmer]

Also fixture is done. Only took 15 hrs to print (!).



As expected, the surface finish on the cuts on the bottom isn't great, since they needed support material:



Again, I'm not going for a flatness record with this, so I'll just clean them up with some high grit sandpaper and it'll be fine.

Assembled (and terribly propped up to show how the cuts are useful):





After holding this piece in my hand, I'm very confident that it'll do great on the mill. It's extraordinarily chunky, I would be very surprised if it fails.

The biggest failure point I can think of are the plastic threads. I added a bunch of wall loops to those areas so they should be fine. Additionally, those are really only there to make sure the strut tower bracket is pressed up against the tops of the posts (i.e. flat), they're not taking much load. Most of that load will be taken by the fasteners threaded in from the bottom into where the stock strut bar studs usually live. I'm definitely not ripping those threads out.

And yes, those are class 12.9 bolts going into 3D printed threads. Gotta make sure the fasteners don't fail!

[heinzboehmer]

Do it!

+1! Put your new printer to work Obioban

I need to get into this as well. Soon.

[bigjae46]

I'm sure this is what you were referring to, but for others benefit...

I want to start doing this

Do it! All of the difficulty with compression molds is making the mold and then the math to calculate resin and carbon weights.

EDIT: Google will tell you the density of carbon is 1.7 to 2.0. Stick to 1.4. Learned the hard way. The worst part about working with carbon trimming and finishing. Minimal post work needed - maybe a razor blade and drilling. You do want to use a carbide drill bit (not carbide tipped) which are about $10 to $20 each bit. Use water or a liquid to contain the dust.

I have about 5 lbs of loose carbon strands. Just pay shipping if anyone needs some.

So no excuses!

[bigjae46]

Lol, if I didn't want to get this done ASAP, I would 100% take you up on that. A mold for this would fit in the printer, which would make it even easier to make out of CF.

Yes, a 3D printed mold eliminate the hardest part of making the mold. The only downside is it tends to not be as durable as a composite mold. Only becomes an issue if the plan is to make a LOT of parts.

[Obioban]

Lol, if I didn't want to get this done ASAP, I would 100% take you up on that. A mold for this would fit in the printer, which would make it even easier to make out of CF.

I'm sure this is what you were referring to, but for others benefit...

I want to start doing this

[heinzboehmer]

Drove my car around for longer today (including a decent chunk of highway driving) and WOW, it drives incredible. Pulling away from a stop and upshifting are buttery smooth now. And it's got torque at super low throttle inputs! Crazy.

Bryson and I were talking and we can't really come up with a good explanation for why my car wants so much more fuel down low. Maybe I have a vacuum leak or maybe he's got some restriction somewhere or maybe it's the flap? I'll smoke test my car if I can find a tester, but yeah, fairly stumped. Slideways let's flash some stuff on your car and see how it reacts

[heinzboehmer]

You know what I'm gonna say...I think I can make that in carbon fiber lol

Lol, if I didn't want to get this done ASAP, I would 100% take you up on that. A mold for this would fit in the printer, which would make it even easier to make out of CF.

[bigjae46]

Will this work? Maybe

Will this crack as soon as I tighten down the vise? Maybe

Will this shatter as soon as the end mill touches the aluminum? Maybe

It's unconventional for sure, but I figured I would give it a shot. I did find a bunch of youtube videos of people doing very similar things, so that gives me hope. If it doesn't end up working, I'll have to make/improvise a similar fixture with stiffer materials, so this part will still be very useful to have as a reference.

I was hoping to get to the machine shop today, but the print isn't ready, so I'll have to go in some day after work next week.

You know what I'm gonna say...I think I can make that in carbon fiber lol

[heinzboehmer]

I engineered cheese:



Following the success of the 3D printed front main seal and stud install tools, I decided to see if a 3D printed fixture would work for holding the stock strut bar brackets at the correct angle in the mill.

Here's some pictures of the fixture without the holes to better show the geometry:




Those cuts in the rear are parallel to the strut bar mounting surfaces. Surfaces are fairly flat on the brackets, but there are some casting imperfections:



I'm gonna give them the lightest of skims just to get rid of the bumps. The cuts in the fixture will act as a reference when setting them up in the mill. That setup is going to end up being pretty tall, but there's no good way around that thanks to the geometry of these things. If it ends up being too tall, I'll just touch them up on the belt sander instead. Perfect flatness is not crucial here.

So why make it all swiss cheese-like?

The answer, counterintuitively, is that the holes add stiffness. This part is way too big to print with 100% infill, so the holes are there to force the slicer to add more walls inside the part. Should drive stiffness up significantly:



Will this work? Maybe

Will this crack as soon as I tighten down the vise? Maybe

Will this shatter as soon as the end mill touches the aluminum? Maybe

It's unconventional for sure, but I figured I would give it a shot. I did find a bunch of youtube videos of people doing very similar things, so that gives me hope. If it doesn't end up working, I'll have to make/improvise a similar fixture with stiffer materials, so this part will still be very useful to have as a reference.

I was hoping to get to the machine shop today, but the print isn't ready, so I'll have to go in some day after work next week.

[heinzboehmer]

Now we’re talking!

I'm very excited to get all those parts installed.

I think I'll wait until I get my front triangulation brace installed though. Just in the interest of science.

[Bry5on]

Now we’re talking!

[heinzboehmer]

Flashed karter16's custom DME FW today to get extra data over CAN and it works perfectly (0x7D0 is the new message):



Log of a drive around town immediately after flashing the modified FW: http://datazap.me/u/heinzboehmer/csl...?log=0&data=15

Thought it was funny that you can see the lack of traction very clearly towards the very end:



That's a hard shift from first to second at redline. Note that the slope of the RPM curve is way slower when in second, even though throttle pedal is at WOT. Pink line tells the full story: DSC is kicking in.

With everything on CAN now, I was able to do multiple rounds of VE tuning without having wires running all throughout the cabin. Just tell gauge.s to start logging, drive around, import logs into tuning spreadsheets, flash car and repeat until satisfied. The only changes I ended up making were down low, but it seems to have resolved essentially all the remaining weirdness. The way it drives now is extremely close to how it drives with the stock airbox + tune.

Here's the changes that the VE tuning process resulted in, for reference:



Extremely happy with how the car is doing these days. Now I need to make the chassis feel as nice as the engine, especially since parts are starting to pile up