Nice solution to the oil drain. Nothing holds my turbo in place other than the exhaust plumbing, but its old style 5 bolt flange and very substantial.

Wow just WoW !!

The 3D printed stainless exhaust pieces came in today. Can’t wait to get this all welded up. These were not cheap, and there is some warping. It will require a couple hours of fitting, but they will be perfect with some work.

I’m also getting through the intake manifold design. It’s such a challenging assembly but it’s getting there.

http://forums.pelicanparts.com/uploa...1756185078.jpg
http://forums.pelicanparts.com/uploa...1756185078.jpg
http://forums.pelicanparts.com/uploa...1756185078.jpg
http://forums.pelicanparts.com/uploa...1756185078.jpg
http://forums.pelicanparts.com/uploa...1756185078.jpg

What is this, an intake manifold for ants?!

Making progress on the intake manifold design. I ended up printing a few 50% scale versions to get the joining features right (it’s four prints that will be glued together).

I’m really excited about ‘cartridge’ intercooler. I had to buy new cores but is so nice and compact. Will share more on that once I have all the parts.

The full scale one in the PPA-CF will be about $150 in plastic. I’d really like to find all the design screw ups before I print it for real.

http://forums.pelicanparts.com/uploa...1756525665.jpg
http://forums.pelicanparts.com/uploa...1756525665.jpg
http://forums.pelicanparts.com/uploa...1756525665.jpg
http://forums.pelicanparts.com/uploa...1756525665.jpg
http://forums.pelicanparts.com/uploa...1756525665.jpg
http://forums.pelicanparts.com/uploa...1756525665.jpg

That is coming along really nicely Have you checked in to the cost of having it 'professionally' printed in one piece in PPA-CF or a similar filament? Just curious.

Looking forward to seeing the final design and the intercooler that you have selected.

Good call, just checked. It's $5,400 to it printed in SLS in Hong Kong...the only place I saw that could do a part this large. If I had the 4 components printed in China it's $1,600. Good news it that it's only $11k in aluminum. Ha.

I also think the PPA CF material I'm using is far superior to SLS PA12.

Gluing nylon is not straightforward. I will flame treat and then use Plexus MA310. I will also have the joins extensively tabbed for additional strength. I'd like to use fasteners but there's not really enough room.

Lastly, there's the open question of whether or not I can print parts this large without warping. I'm printing on a H2D which will help with it's heated build chamber....but this is still pushing pretty close to the edge of what's possible. We will see.

WOW! Those prices seem really excessive. Your initial plan is looking better and better.

One other thought (also not cheap) if you haven't looked into it is to have it printed with/by Markforged performance filaments which have the option of a continuous carbon fiber thread inserted in the stress areas that require it. I would be curious as to the estimate you receive for a one piece build from Stratosys (I think that is the name of one of their vendors).

While I’m finishing the intake design I’ve been testing various adhesives. This started with some basic tests but quickly escalated to more precise heated shear testing of the bond.

I’m not done with the testing yet but a few things have become apparent:

- Nylon is tough to bond to and there’s only a handful of industrial adhesives that are made for it.
- Flame treating dramatically improves bonding and it’s easy to do.
- As temps go up the adhesives weaken. At 80 C they have lost 30-50% of their room temp strength.
- JB Weld PlasticWeld is no joke. For $5 a tube this stuff is crazy strong and even bonds to Nylon. That said, it loses a lot of strength at 80 C.
- The front runners are Plexus MA8110 and 3M DP8910DS.
- Testing so far is showing something like a 15X factor of safely on the intake manifold joints. This stuff is strong.



http://forums.pelicanparts.com/uploa...1757183831.jpg
http://forums.pelicanparts.com/uploa...1757183831.jpg
http://forums.pelicanparts.com/uploa...1757183831.jpg
http://forums.pelicanparts.com/uploa...1757183831.jpg

You should make a youtube channel. I'd watch it.

agreed

It will be 10000 times better than the clown car mechanics we see there.

I’ve wanted to do YouTube for a while. Appreciate the vote of confidence!

The prototype PLA manifold is printed and assembled. Leaned a lot while printing and assembling it. One funny thing, the manifold is about 2mm too narrow. Not a huge deal. It’s very stiff but can flex that much without issue.

But the cause? I ran 1.2mm base shims which pushed each cylinder bank out by 0.7mm and the manifold is based on the 3D scan of the engine before the rebuild. Forgot to account for the thicker base shims. Guess it’s pretty accurate…


Most importantly, I really like how it looks. Ha. Looks muscular and to me the aesthetic feels like it fits the car. The final manifold will be finished to remove seams and painted in a textured black. The logo plate will be 3D printed in aluminum.


http://forums.pelicanparts.com/uploa...1757555413.jpg
http://forums.pelicanparts.com/uploa...1757555413.jpg
http://forums.pelicanparts.com/uploa...1757555413.jpg
http://forums.pelicanparts.com/uploa...1757555413.jpg
http://forums.pelicanparts.com/uploa...1757555413.jpg
http://forums.pelicanparts.com/uploa...1757555413.jpg

I also wrapped up the glue testing. I created a setup to get shear testing data using Knipex pliers (20:1 lever ratio). The handle is pulled with a scale and the force on sample should be 20X the force at the end of the handle.

There’s plenty of room for error but the end result was that I have a 50X factor of safety at 100C (100X at room temp). So yeah, this stuff is strong. Still some risk with heat cycling and unexpected loads, but I’d be comfortable with a 10X factor of safely….nice to be 5X that.

The Plexus MA8110 and 3M DP8910 were both similar in strength at room temp, 80C, and 100C. They both lose about half their strength at 100C vs room temp. I’m going with the Plexus bc I have used their stuff with success before.

For reference, the glued area is 8mm X 8mm and it held roughly 400lb at room temp and 200lb at 80C for both glues.

For kicks I tested crazy glue on PLA plastic. I have lots of hands on experience for how strong that is from prototyping, making 3D printed RC planes, etc. It’s a really strong bond but these adhesives on nylon were 4X stronger. Nuts.

http://forums.pelicanparts.com/uploa...1757556347.jpg
http://forums.pelicanparts.com/uploa...1757556347.jpg
http://forums.pelicanparts.com/uploa...1757556347.jpg
http://forums.pelicanparts.com/uploa...1757556347.jpg
http://forums.pelicanparts.com/uploa...1757556347.jpg

Looking good. I am interested to see the final version with the intercooler installed.

You should rename this thread......

Finally finished the manifold 3D print in PPA CF. Total print time was 150 hours, about $300 worth of fancy plastic, and the total weight is about 7.5 pounds. It is stout!

I ended up needing to reprint one section. Thankfully I realized right away that the prints shrink in the z axis by 0.5%. It’s a small amount but it would add up to 2mm across the manifold.

Next up is bonding it together and then finishing and painting it. The intercooler ‘cartridge’ is all designed and now I need to order the 3D printed aluminum parts.

I’m a little worried about the size of the Intercooler cores. They just feel small but they should be good for 500hp (~150hp more than I’ll run). I’ll only really know how they perform when the car is on the road…

http://forums.pelicanparts.com/uploa...1759449047.jpg
http://forums.pelicanparts.com/uploa...1759449047.jpg
http://forums.pelicanparts.com/uploa...1759449047.jpg
http://forums.pelicanparts.com/uploa...1759449047.jpg
http://forums.pelicanparts.com/uploa...1759449047.jpg
http://forums.pelicanparts.com/uploa...1759449047.jpg

sick

You'll have access to the triangle area

while you've got your scanner out, I wonder how difficult it would be, to scan and design an airbox centre section, to convert 3.0SC to efi "on a budget"...

Even if it is just to use something like a Holley Sniper efi unit..

http://forums.pelicanparts.com/uploa...1759457134.jpg
http://forums.pelicanparts.com/uploa...1759457134.jpg

The SC is begging for a better intake manifold. Well, honestly, I don't really know if the stock one performs poorly. It could look cooler at least!

The CIS system is still port injection and I'd be hesitant to move to a sniper or similar EFI throttle body. I'm not sure it would be better as I'd guess you would have issues with fuel condensation on the runners. But, who knows, maybe it would work great.

It's really a shame the EFI systems are so expensive. I just bought the ECUMaster Pro-8 for this build and my wallet is still smoking...

The more I think about it, it would be really cool to make a Speeduino based drop in solution for the SC. If you really tried to keep the budget in check it could be done for pretty cheap. Just managing fuel and leaving the ignition system alone would keep cost down. CIS works well but it doesn’t last forever and it would be nice to get rid of the restrictive air flow meter. I can’t imagine there would be much of a HP boost, but probably a little!

The idea is probably so we can run a decent camshaft and not have to rely on a fuel metering plate. I have tried bigger cams with CIS, and at idle and low rpm, with the overlap it can make the plate flutter.. which will alter the fuel delivery continuously.

My thoughts were, an airbox that disposes of the original airbox, and which can adapt a throttle body with standalone ecu OR a holley sniper. I've run a sniper on the engine stand and it seems to work fine. There are others who are running them in 914 and work well.

ignition can be taken care of by the original distributor and CDI.

If you went standalone, you could get fuel rails and injector adaptors from BITZ.

Someone made this manifold to run a holley carb and a turbo: i bought the intakes but not sure how I would go to get this replicated.
https://youtube.com/shorts/IkXNI71DzGo?si=EwuPdlZLQ9bf-6Pf

And here it is running (this had a BIG camshaft).
https://youtu.be/rLlUsFeiOVw?si=XJMg14Jgdiu4xf6s

I have been fiddling with different stand alone ECUs with my 930 during the past few years and I will tell you that this industry has come to a point where its almost a plug-n-play to convert any internal combustion engine to the most advanced ECU technology. What you need to do is basically 3 things, injectors, ignitions, and sensors. The cleanest and most user-friendly setup I have ever done is to have a DBW, that makes your engine very clean and neat. I have come across an ECU called ECU Master EMU Black, and that will support EVERYTHING that any 2025 engine have, DBW, closed loop fueling, Knock control, traction control .. etc, and all that for just $1100, add $300 for a set of generic DBW throttle and pedal, and another $1500 for injectors, fuel lines, ignition, sensors .. etc and you will end up with a very modern and state of the art engine control for under $3000 USD.

Yes, I'm using an emu black... compared to motec, link, haltech, emtron etc.. it is "cheap"... and has everything on board.. no more extra modules or licences etc..

This is really super cool, that you designed and 3D printed your own intake manifold at home! This was unthinkable just a few years ago when Stratasys kept the industry locked up with their patents.

Which PPA did you end up using, and how are you going to seal it up? Is the nylon glue going to be enough or did you design the interlocking pieces to use some form of gaskets?

Following closely as you are blazing this trail, as a fellow printing nerd. :D

Thanks! I used the "Siraya Tech Fibreheart PPA-CF Core" filament. It has the carbon concentrated in core so the layer bonder is improved. I think this is the strongest PPA right now and it's pretty cheap too.

The goal with the bonded seams was to have the glue in shear as much as possible. The two halves of the main body have a huge overlapping joint and should be crazy strong. The runner to main body joints incorporate tabs that go across the bonded joint. This gives me some good shear surface for the glue, but honestly, the flat faces of the joint would probably be strong enough.

I will share photos of it being glued together. I first need to print TPU scrapers that I will pull through the runners to clear up and glue the oozes inside the runner. It would be impossible to remove once cured.

Sharing an intermediate’ update. Not too much newness but some progress.

The manifold is coming together with brass inserts and mounting of the MAP sensor and IAT sensor. The IAT is from an Audi 1.8T…a good sensor.

Also, after three tries I finally successfully printed the back cover. This one is tricky bc there’s a lot of support material that was impossible to remove on my first tries. Even the Bambu PA support material which is made for this was impossible to remove. I ended up using ASA on the support interface and it worked perfectly. A huge unlock for printing in nylon.

The throttle body brass inserts mount from the back. They are crazy strong, but this is a spot where I wanted to make sure they couldn’t pull out.

http://forums.pelicanparts.com/uploa...1759878264.jpg
http://forums.pelicanparts.com/uploa...1759878264.jpg
http://forums.pelicanparts.com/uploa...1759878264.jpg
http://forums.pelicanparts.com/uploa...1759878264.jpg
http://forums.pelicanparts.com/uploa...1759878264.jpg

More photos.

http://forums.pelicanparts.com/uploa...1759878532.jpg
http://forums.pelicanparts.com/uploa...1759878532.jpg
http://forums.pelicanparts.com/uploa...1759878532.jpg
http://forums.pelicanparts.com/uploa...1759878532.jpg
http://forums.pelicanparts.com/uploa...1759878532.jpg
http://forums.pelicanparts.com/uploa...1759878532.jpg

Looking very nice. Is that a 74mm throttle body?

Yes, it's a 74mm. Way bigger than I need but also very cheap...

This is very cool and I have been wanting to design a Nylon intake myself.

Great work!

Here are the ITB adapters I designed years ago that were physically machined in China from a fellow member.

Looks like you have the same interface:


http://forums.pelicanparts.com/uploa...1759950444.jpg
http://forums.pelicanparts.com/uploa...1759950460.jpg

Nice! I reverse engineered mine from a 3D scan of the stock manifold. The injector angle was a tricky part.

A bit of an update. I decided I should do a test torque of the runners before gluing everything together. The result, total failure. Crapola...

I torqued them early on but this time I used the proper plastic intake spacers. I think what happens is that the intake spacer is soft and deforms under the 3,000 lbs of clamping force for each of the mounting bolts. The compliance of the spacer transfers a huge amount of load across the full face rather than just below the bolts. The plastic flange has basically zero flex so it cracks.

The plastic flange would work on a metal surface but it's not going to work on the soft plastic spacer.

So...the plan is to have the lower sections 3D printed from Aluminum and then bond them to the nylon intake. It's an extra $350 and a little annoying.

I think I'll take a break from this and get the turbo plumbing in order.

Yes, the injector angle ate my lunch. I was just learning CAD, so it took me weeks of messing around before it was right and mimicked the Carrera 3.2 perfectly.

This is a super cool project!

I hit the same point when doing my 3.0 EFI conversion, the plastic adapters just weren't up to it so had them 3D printed in aluminum. Pretty easy for my application since I was already using the 964 manifold and factory gaskets so I didn't have to make many design changes to just bolt on the new ones.

The amount of material does significantly impact the price though, so for mine I found I could get away with a much thinner flange overall with just adding more thickness at the bolt hole with a healthy chamfer for support.

http://forums.pelicanparts.com/uploa...1760019773.PNG

Beautiful part. Love it. My adapters are much heavier weight unfortunately bc I designed them to be plastic. My CAD model is such a mess and Fusion 360 is quite weak in terms of parametric referencing. Changing the flanges (the first features in the model) isn't an option anymore!

Yeah I use Fusion too so I feel your pain, keeping things neatly parametric is pretty much impossible for anything other than super basic parts. But yeah, juice probably isn't worth the squeeze on that.

I'll be following along though, what you've got so far is super exciting!