Workflow for designing exhaust header

Workflow for designing exhaust header

viperx85
Enthusiast Enthusiast
2,696 Views
25 Replies
Message 1 of 26

Workflow for designing exhaust header

viperx85
Enthusiast
Enthusiast

Ive got a 3D scan of my engine and chassis and i'm trying to design a set of 4-2-1 headers for my KA24DE inline 4. I have the header flange and the collectors in position and just wondering if there is a easier way to layout the tube's?

I need the bends to be a certain radius and I need the individual tube lengths to be equal and a certain length.

So far my workflow has just been trail and error (Alot of trail and error)

Im using a 3D sketch to draw a path with straight lines, then apply a fillet to the corners with the correct radius (using a parameter)

First problem is I cant just draw the straight lines to add up to the length I need because once I add the fillets the length changes and the amount it changes depends on the angle of the bend.

Alot of the time I draw my straight lines but once I apply the fillet's when two bends are too close to each other they overlap. once you have applied fillets they sketch becomes very unruly, By that I mean if you try to move a point or a line alot of the time it either wont move or it will move all other lines randomly so I usually have to ctrl+z to remove all the fillets, adjust the lines, re-apply the fillets, realize I need to move them abit more. rinse repeat then finally use a pipe command to model the tube.

You can see in the pics below Ive managed to get two primary's routed to exact same length and not interfering with each other but that took ALOT of trail and error.

Wondering if anyone has done similar and knows of any tips/tricks on how to improve my workflow?

An AI feature would be lovely. Tell it I need pipes to run from here to here... be this long, have this radius bend and not hit each other or surrounding components... GO


viperx85_0-1727130187991.png

viperx85_1-1727130216870.png

 

0 Likes
Accepted solutions (1)
2,697 Views
25 Replies
Replies (25)
Message 2 of 26

davebYYPCU
Consultant
Consultant

Whilst you know more of the limitations, obstructions (guard bell housing and accessories,)

I would move no.2 closer to the block, (future) pipes 3 and 4, can go to outboard receivers, will get closer to same length.

 

Prepare with construction splines, they can be measured / manipulated easily, and then follow up with the Pipe Path.

 

mfrndb.PNG

 

Might help....

Message 3 of 26

viperx85
Enthusiast
Enthusiast

This is a inline 4 cylinder with firing order 1-3-4-2 therefore cylinder 1-4 need to be be coupled and cylinder 2-3

The primary pipe length i'm working with is just a placeholder for now (567mm from flange to primary collector) as im waiting for a new book to arrive which will hopefully help me work out the primary length. Ive tried a whole bunch of online calculators and none seem to agree. but for now I'm just practicing the workflow as I've never done anything like this before.

Splines can be manipulated easily but since I need to actually fabricate this with off the shelf mandrel bends I need each bend to be a set radius which you cannot do with splines can you?

0 Likes
Message 4 of 26

davebYYPCU
Consultant
Consultant

Prepare with construction splines, they can be measured / manipulated easily, and then follow up with the Pipe Path.

 

Why does the firing order effect which receiver is relevant?

Manipulating a 3 maybe 4 point spline can have very close to correct fillet radius, but if you are chasing 567 mm of pipe, they have to go somewhere.

 

Might help...

0 Likes
Message 5 of 26

Warmingup1953
Advisor
Advisor

I wonder if a library of these (Bespoke Components) would keep you within the realms of what is possible would work? IcengineWorks 

Message 6 of 26

viperx85
Enthusiast
Enthusiast

1-4 / 2-3 are paired together so the exhaust pulses are evenly spaced apart.

Paired that way the exhaust pulses are a spaced a even 180° apart which helps with scavenging and stops pulses from interfering with each other

If you paired say 1 and 3 your pulse timing in crank rotation would be  90° 270° 90° 270° 90° 270° etc etc

0 Likes
Message 7 of 26

davebYYPCU
Consultant
Consultant

Dual exhaust to rear that makes sense, you said 4-2-1.

I like the modules, and had thought of those, in Fusion could be cumbersome.

 

Might help....

0 Likes
Message 8 of 26

viperx85
Enthusiast
Enthusiast

Yes 4-2-1 there is the secondary merge collector slightly further downstream. But If i manage to keep the two primary collectors side by side like that the pipes to the secondary collector will be very easy to position as they will be basically identical.

viperx85_0-1727156361764.png

 

0 Likes
Message 9 of 26

user015M86Y
Advocate
Advocate

Hi,

please share a sample file for demonstration

 

File > export > save as f3d on local drive > attach to post

0 Likes
Message 10 of 26

viperx85
Enthusiast
Enthusiast

@user015M86Yplease see attached.


0 Likes
Message 11 of 26

user015M86Y
Advocate
Advocate

Hi, see attached a proposal

The difference lenght between cyl1 - cyl 3 is 8mm

The difference lenght between cyl2 - cyl 4 is 12mm

If you agree with my proposal I can adjust these difference to almost 0mm

 

0 Likes
Message 12 of 26

davebYYPCU
Consultant
Consultant

Screen shot?  If offering equal, why not?

0 Likes
Message 13 of 26

viperx85
Enthusiast
Enthusiast

@user015M86YI think you may have uploaded the wrong file? its the same one I sent with no pipes in it?

Did you link Cylinder 1 and 4 together into one collector and then 2-3 into the other one as they need to be?

0 Likes
Message 14 of 26

davebYYPCU
Consultant
Consultant

Do you still need 567mm pipe length, my longest simplest pipe in brief test is 449.  No.1.  Even making No.4 that long will be cumbersome.  (WIP)

 

eh449db.PNG

 

Might help....

0 Likes
Message 15 of 26

viperx85
Enthusiast
Enthusiast

567 is just a target im using while practicing, Still waiting on a book to arrive which hopefully will detail selecting the correct length.

So far ive managed to get #1,2,3 to be equal length. working on #4 now.

viperx85_0-1727247386805.png

 

0 Likes
Message 16 of 26

user015M86Y
Advocate
Advocate

user015M86Y_0-1727247861524.png

 

0 Likes
Message 17 of 26

user015M86Y
Advocate
Advocate

Hi,

I upload the only file you attached

I connect 1 and 3 in first collector and the 2 and 4 to the other collector

I correct my error 

0 Likes
Message 18 of 26

viperx85
Enthusiast
Enthusiast

Looks good but 1 and 4 need to connect to one collector and 2-3 to the other.... doesn't matter which collector though and for this exercise I was aiming for each pipe to be 567mm long. The collector location can be changed though, Although I dont have the 3D scans on the attached file as that made the file size much larger.

But the idea isnt for someone to do it for me but hoping to improve how I do it.

What method did you use to create the paths?

0 Likes
Message 19 of 26

user015M86Y
Advocate
Advocate
Accepted solution

Hi,

You will see the procedure I use in my proposal (edit cylinders sketch) :

- sketch with construction line

- dimensions, constraint are applied for all segment and arc

- draw lines between each end of arc, lines to connect for sweep

- sweep with the flange section pipe

- adjust the lenght by editing construction lines : as everything is constraint you should remove some constraint

This is my method but I am sure there is a better way to do that

Message 20 of 26

davebYYPCU
Consultant
Consultant

Your workflow looks ok, once you have a workflow it’s same again for each pipe.

wasn’t going to send my file.

 

Might help…..

0 Likes