Roman's 4GR V6 Carina discussion thread

[tortron]

I'm gona guess a bit of valve guide, or one of those collets you shot across the room 

[shrike]

Surely someone on here has access to cad software that could analyse your file for stress etc if you sent it to them? 

Have you looked at how the 2GR adapter that MR2 guy uses is designed? I know that's cncd tho

Edit ignore me, not the style I thought it was haha

https://frankensteinmotorworks.squarespace.com/shop/p/2gr-velocity-stack-intake-manifold

[Rhyscar]

You beat me to it. This is my next project to sell a few 2zz manifolds to pay for racecar things. 

 

Great design to minimise the material. Did you draw the port shape as a surface then shell outwards? (Sorry don’t know the f360 version of the features). 

I’ve got the FEA package on 2013? solidworks if you want to come have a play with that? 

[Roman]

This software (like any) can be quite finnicky with order of operations. 

So although shelling outwards sounds like a good plan, and its what I initially tried. 
It ends up having a fit when this collides with other geometry sometimes. 

So I ended up going 2mm oversized on the port shape, and drawing a solid piece of material with a loft. 
Then using a second loft, minus the 2mm perimeter to cut the internal shape.
Then it had a massive fit when I tried to combine this with the flange that already had the cross bracing and fillets on it. 
So I had to undo about 5 million fillets. Then join the parts and re-add them. 

Tedious work! But hopefully good haha.

Thats an awesome idea to make some 2ZZ manifolds, look forward to seeing it! 

[Rhyscar]

Yeah I can understand the painful process! That’s why it’s so important to work with surfaces in solidworks until you’re almost complete. Merging solids gets complicated quickly!

[Roman]

 I absolutely hate using surfaces in either program, always just work from solids. 

I've had nothing but frustration trying to get surfaces to work. 

Although @Stu has said that Rhino has really really good surface modelling features. But I've never tried it.

Not too fussed about FEA stuff for this, I'm not sure if I'd know the correct variables to input anyway. 

[h4nd]

Ahem, @Roman (Maisteer - engine audio nutter) https://www.youtube.com/watch?v=vA-Z3ncTdUY

 

[Roman]

Yeeeaaaahhhh this is his 3rd video though claiming he's cracked the secret this time, but I'm not seeing any results yet...
But hopefully it works out well this time.
One thing that made me chuckle about the first video, is he's absolutely convinced that 6-1 manifold is the reason the Zonda sounds so good. 
But then it shows the zonda exhaust in the video and its a 6-2-1 setup.

[Goat]

That is super cool going for the 3d printed metal approach. Allows some pretty intricate details. Good idea to build balance tube in

What sorta cost are you looking at per manifold?

 

[Rhyscar]

7 hours ago, Roman said:

But then it shows the zonda exhaust in the video and its a 6-2-1 setup.

 

From that vid, I got that it's more about the steps that create the duplication in sound waves closer to a sine wave, than the collector configuration itself. But agree, without results hypothesis is still out. 

 

I've seen dune buggy guys on YT talking about 8-1 collectors making awesome sounds on LS things. 

[Roman]

@Rhyscar his first video that blew up real big, was about making a 12-1 manifold for his v12.

It ended up sounding like garbage.

Its a really great video though if you havent seen it. 

But his first video he was wholly convinced that 6-1 on each bank was why the zonda sounded so good. Compared to the standard configuration mercedes v12 that sounds nothing like it. 

Whixh is why it was funny to see its not actually 6-1 anyway.

He might be onto something about tge stepped headers, i wonder if all of my welding dregs inside the pipes will mean only bats can hear my engine

[Roman]

3 hours ago, Goat said:

That is super cool going for the 3d printed metal approach. Allows some pretty intricate details. Good idea to build balance tube in

What sorta cost are you looking at per manifold?

 

Under $500 for both, delivered. 
However since the shipping cost is fairly fixed I figure I may as well chuck some fuel rails in at the same time.
As my current ones are visually prominent and ended up being a bit of an eye sore.
My manifolds are good to go, but I've been busy last night trying to finalize rail designs.
Looks like it will be around $750 delivered for manifolds and rails. 

[Willdat?]

36 minutes ago, Roman said:

Under $500 for both, delivered. 
However since the shipping cost is fairly fixed I figure I may as well chuck some fuel rails in at the same time.
As my current ones are visually prominent and ended up being a bit of an eye sore.
My manifolds are good to go, but I've been busy last night trying to finalize rail designs.
Looks like it will be around $750 delivered for manifolds and rails. 

I may have missed it, but who is doing the printing?

[Roman]

@Willdat? Craftcloud3d.com 

You can upload a model, and then it spits back out a list of options/materials/prices from a whole bunch of different vendors. 
No need to talk to any sales people and have to politely explain you cant afford to spend 3 million dollars on printed titanium engine block... haha.

[Willdat?]

4 minutes ago, Roman said:

@Willdat? Craftcloud3d.com 

You can upload a model, and then it spits back out a list of options/materials/prices from a whole bunch of different vendors. 
No need to talk to any sales people and have to politely explain you cant afford to spend 3 million dollars on printed titanium engine block... haha.

Sounds perfect, both from the perspective I'm a better CAD user than a fabricator. Annnnnd I do love to a kick a tyre towards a pipe dream...

[Goat]

Wow. Can't believe that its that simple. (and cheap)
What sort of infill are you using for the metal prints? 

[Roman]

There's no infill, they are completely solid. 
They are printed by laying down a layer of metal powder, then a laser zaps it all to shape. 
Then lays down another layer, and so on.
So the part gets bead blasted afterwards to remove excess powder. But there would be no way to get infill powder back out.

However, I'm using 3mm wall thickness for the fuel rail (thicker on the tabs) 
3mm thick flanges on the manifold base, 2mm walls on the runners, and 3/2/1mm thick on the detail areas. 

My throttle pulleys are just way stronger than I would have thought, I thought they would be borderline to thin. But I would happily redo them with even less material.

Due to the process it's probably as if they have been heat treated already. Which is why they feel stronger than expected.

[Goat]

That makes sense.
I was thinking the stl file was out of the slicer. But of course that's the fusion 360 export, so the part is verbatim.

 

Chur

[xsspeed]

whats the finish like on mating surface from a printed product, like for your intake will it need a skim or is it smooth and true enough that a gasket (/goo) is all thats needed

[Roman]

The pulleys that I had printed have a texture similar to casting, but, maybe 10x finer "grain" on it. 
Visually you can see the texture but they feel quite smooth.
I'm sure they will be fine with just a gasket.

The injector holes though, definitely need a really good seal. So I've made the holes 1mm undersized, then will machine them back out to the right measurement on CNC.
I've done the same for the fuel supply hole, 1mm undersize of the drill size before tapping it.
I might need to get that end faced flat to get a good seal with the banjo bolt.