heinzboehmer's 2002 Topaz 6MT Coupe

[George Hill]

I've seen a handful. I guess its pretty common on track cars where engines live in high RPM with the harmonics and banging gears?

Who knows though, maybe its from people not retorquing the damper correctly or reusing the bolts. That would seem likely to happen more often on race cars than a street car I'd imagine.

All I know is it can't be good for the damper to be moving around (which it obviously is by the damage to the crank gear adapter).

[heinzboehmer]

Nice work! I missed your question, I used a carbide bit in a dremel and ground into it enough to get an easy out and then spun it out. Last one I did was a racecar so I cheated and took the front end off.

Nice! I probably should have taken the front end off as well. Was too lazy to drain the coolant...

Do you see these broken frequently? I wonder what factors influence the failure.

[George Hill]

But I got it out of the crank!

Nice work! I missed your question, I used a carbide bit in a dremel and ground into it enough to get an easy out and then spun it out. Last one I did was a racecar so I cheated and took the front end off.

[heinzboehmer]

Forgot to add clearance for the crank damper e torx bolt heads in the press cup bolt guide, duh. Fixed:



Also, this sucked:



But I got it out of the crank!

I ended up center punching the dowel and then very slowly drilling it out. This took absolutely forever, since I was being very careful to not damage the crank (I know it's hardened, but still).

I have absolutely no idea what I was thinking, but at one point I decided to hammer a torx bit in to try and spin the dowel. Of course, the tip of the hardened bit broke off inside the dowel as soon as I put any torque on it. Managed to fracture the stuck bit piece with a center punch and then fish/drill the chunks out with the drill bits. Extremely stupid of me, but I got the job done in the end.

I managed to do all of this without removing the radiator by using hex shank drill bits on my stubby impact driver. Drilling is not exactly what that tool is made for, but the dowel is fairly soft steel, so it didn't complain much. Made sure to give it 10s or so of full throttle spinning off the car every once in a while to make sure the electronics stayed cool. A bit more jerry-rigged than I like, but it did work. I should probably get myself a right angle drill...

Also, a motivational picture:



That's one of the damaged seals installed successfully (!) using the modified tool. At that point, I had not cleaned up the crank at all (as is obvious) and yet, the pre-damaged seal went in perfectly. Gives me hope that a new seal will go on smoothly as well. Will get to that soon.

[heinzboehmer]

Also, here's the design for the bolt guide that goes in the cup. This won't be taking any load, so can be printed with a low infill %

[heinzboehmer]

You can increase the end taper OD by .1-.2mm or something without having to create an overlap. That might be enough to get you over the transition.

Also, I don’t see a downside to a small file or sandpaper as long as it doesn’t reach the sealing surface as you pointed out. Good luck.

edit: also krytox on install can help.

Was installing with Krytox on the timing cover edge of the seal. Crank surface I was leaving dry as per the repair manual. Hopefully the increased OD of the tool helps out.

I don't have a good suggestion on your seal issue, but are you using a BMW seal?

Tried both Corteco and Genuine seals yesterday. They appear to be identical.

The bigger problem is your broken damper pin. Do you have a plan for that?

It looks like the crank adapter has some marring on it too. I would dremel that piece out of the crank. Then take some very very fine wet sandpaper on a solid piece of stock and knock down the high spots. Same with your damper if you can.

This is the replacement pin you want. They are sold in packs of 50 so I have a lifetime supply. If you want one (or 2) send me your address and I'll send some.

Yeaaah, it's been that way since I bought the car, so my plan was to ignore until I do the head gasket. The sprocket (11211401036) is surprisingly affordable, so I was planning on swapping it with the head/timing chain cover off.

I don't know why I didn't think of just buying a new pin, but really appreciate that link and your offer to send me some! Looks like McMaster delivers tomorrow, so that should work.

How do you get the broken pins out? Screw extractor? Very careful drilling/dremeling?

May want to use a +/- 45 deg biax? Carbon is an option as long as it is not being elongated.

Yeah that could work. Gonna start with the laser cut parts cause they're dead simple to make, but will consider carbon for a V2.

[George Hill]

This won't work. The lip would increase the OD past the limit of how much the seal can stretch:


Modified geometry in grey, cobra tool in blue. Outer edge of the highlighted red region is where the cobra tool ends.

I'm also starting to think that my crank is part of the problem. The outer edge has been slightly marred by previous mechanics, which is likely contributing to the seal catching. Terrible pic, but you can kinda see it here:



I'll take some super fine sandpaper and try to knock down any high spots, but I likely will only do a super light pass. I really don't want to scratch the sealing surface. Fortunately, I have all these failed seals I can install to use as protection for the crank.

Additionally, I'm planning on increasing the taper OD by 0.5 mm. Hopefully that helps gets the seal over the edge without it catching.

I don't have a good suggestion on your seal issue, but are you using a BMW seal?

The bigger problem is your broken damper pin. Do you have a plan for that?

It looks like the crank adapter has some marring on it too. I would dremel that piece out of the crank. Then take some very very fine wet sandpaper on a solid piece of stock and knock down the high spots. Same with your damper if you can.

This is the replacement pin you want. They are sold in packs of 50 so I have a lifetime supply. If you want one (or 2) send me your address and I'll send some.

[bigjae46]

Carbon might work! Part is gonna need to be strong both in tension and in compression. It'll be seeing loads from the E86 braces, as well as loads from the stock strut bar. Compression case for reference:

May want to use a +/- 45 deg biax? Carbon is an option as long as it is not being elongated.

[Bry5on]

You can increase the end taper OD by .1-.2mm or something without having to create an overlap. That might be enough to get you over the transition.

Also, I don’t see a downside to a small file or sandpaper as long as it doesn’t reach the sealing surface as you pointed out. Good luck.

edit: also krytox on install can help.

[heinzboehmer]

They are in the US. I had a homemade tool and your post inspired me to get the correct one. Got it in (2) days from my dealer.

Oh nice! When I checked, most online places said they would ship in July, so I assumed they had to come in from Germany.

I have yet to use my design, but it does bolt right up to the harmonic balancer.

What about carbon fiber? How would the part be loaded? Might be able to increase the strength by using a unidirectional layup.

Carbon might work! Part is gonna need to be strong both in tension and in compression. It'll be seeing loads from the E86 braces, as well as loads from the stock strut bar. Compression case for reference:

I think I'm gonna experiment with modifying the design of the press tool slightly. I plan on adding a lip to the part that bolts to the crank, so that the seal has nothing to catch on.

This won't work. The lip would increase the OD past the limit of how much the seal can stretch:


Modified geometry in grey, cobra tool in blue. Outer edge of the highlighted red region is where the cobra tool ends.

I'm also starting to think that my crank is part of the problem. The outer edge has been slightly marred by previous mechanics, which is likely contributing to the seal catching. Terrible pic, but you can kinda see it here:



I'll take some super fine sandpaper and try to knock down any high spots, but I likely will only do a super light pass. I really don't want to scratch the sealing surface. Fortunately, I have all these failed seals I can install to use as protection for the crank.

Additionally, I'm planning on increasing the taper OD by 0.5 mm. Hopefully that helps gets the seal over the edge without it catching.

[bigjae46]

Also, simulation seems to work out if I make these out of Titanium...

Potentially a future upgrade?

What about carbon fiber? How would the part be loaded? Might be able to increase the strength by using a unidirectional layup.

[George Hill]

Looks like there's no stock of them in the US, so at least the laser cut part will get to me sooner.

They are in the US. I had a homemade tool and your post inspired me to get the correct one. Got it in (2) days from my dealer.

[heinzboehmer]

Learned my lesson and ordered four from FCP this time.

Okay, well, attempted to install a new seal four times today and failed four times. Man, this job really is not for me.

The failure mode was always the same: seal would be pressed in slightly crooked, catch on the crank and the lip would get ruined. New seals on the way, supposedly here mid week.

I think I'm gonna experiment with modifying the design of the press tool slightly. I plan on adding a lip to the part that bolts to the crank , so that the seal has nothing to catch on. Additionally I'll model in a sleeve for the bolt on the cup tool. This should help keep the press surface parallel to the timing chain cover.

More to come...

[heinzboehmer]

Shipping company has held the crank holder tool hostage for a few days, which means work on the front main seal is stalled. Supposedly delivers today, so should be able to wrap that up soon. I did go ahead and start disassembling and cleaning, but more on that later.

While I was waiting, I finished up the design for the strut bar brackets:



Orange piece is 4.75 mm thick 304 SS and grey piece is 3.18 mm thick 304 SS. I would have loved to try these out of aluminum, but unfortunately sendcutsend doesn't bend anything besides 5052, which is not gonna cut it.

I plan to weld the two parts together along the red lines:




No welds on the angled interface so as to avoid weakening the metal.

Clearances are stupid tight in that area, so I printed some prototypes for a final fit check:





Hell yeah, things are finally coming together! I'm really happy with how stock this all looks. Hardware isn't final, of course. I'll order studs of the correct length for the real part.

One thing I don't think I've mentioned is that this design requires camber plates where the studs can move independently of the top hat (e.g. Turner street plates). There is not enough space to slot neither the E86 brace ends nor the brackets, which means that the brackets are essentially constrained to one spot. The studs in the stock top hats would require the bracket to move along with them when setting camber, which is not possible with this design. It's an annoying limitation, but using the E86 braces wouldn't be possible otherwise.

I still need to modify the design slightly so that it's actually manufacturable, but no big deal. Just need to add some extensions so that the smaller flanges can be bent correctly.

I did also quickly run it through FEA using that same 21 kN figure that should buckle the E86 braces:



As you can tell, the analysis says that the welds will fail before the E86 braces. There's also some areas in the part where the safety factor is worryingly close to 1. Not ideal, but I'm gonna go ahead with this design anyway for the following reasons:

• There is no space for another design. The engine harness box gets in the way on one side and the positive terminal gets in the way on the other. A beefier design will require modifications to both of those.
• Simulating welds is not straight forward. There's a bunch of spots that will be welded together in real life that the simulation software didn't let me mark as welded.
• The real loads that this will be seeing will likely be below 21kN. This is more of a max value and not a typical load value. If I had a way to know what loads the braces will typically be seeing, I would have ran the simulations with those.
• The cost to get both brackets manufactured is less than $200 and they take 5 min to install. These are not like the windshield mount where I only get one shot, I can easily swap in updated parts if needed.

Additionally, seeing how thin the aluminum end pieces on the Slon brace are makes me confident that my design will survive. I have yet to hear of one of those failing. Pic for reference:



Anyway, lots of coming up with excuses for lazy engineering and very little doing. I'll get myself in the machine shop this week and finally trim back the stock brackets. Once I have confirmed the final dimensions of those, I'll order my brackets.

Also, simulation seems to work out if I make these out of Titanium...



Potentially a future upgrade?

[heinzboehmer]

I just use the crank lock timing pin and break the 4 bolts loose.

Yeah that is an option, but I've always been worried about cracking the timing chain cover by doing that. Especially when torquing the new bolts.

I bought the tool for like 40 bucks last year I think?

Edit: 83300493843

Okay well, turns out this tool wasn't as hard to find as I thought. Thanks for that info.

Looks like there's no stock of them in the US, so at least the laser cut part will get to me sooner.