2270cc type 4 EFI Turbo: my first engine build

[buguy]

Very nice. I never thought about the stainless being difficult to cut.

[split1950]

Very nice. I never thought about the stainless being difficult to cut.

I'm finding that my air saw blades hardly touch it.
I'm using a cutting wheel on an air die grinder type thing (with a guard around the wheel) that they use for cutting body panels. It takes about 15 minutes to make the cut and dress it up with a file (again which doesn't cut much off). The cuts are coming out quite nice, so welding shoudn't be an issue.

[buguy]

Wow! It's not easy to get a straight cut with a cut-off wheel! My next one will be stainless. I'm gunna have to learn to weld stainless, never done it before.

[split1950]

Wow! It's not easy to get a straight cut with a cut-off wheel! My next one will be stainless. I'm gunna have to learn to weld stainless, never done it before.

I found an effective if not weird way.....
I use a toilet roll inner tube, slipped over the tube, then draw the cut line around it with a Sharpie (other brands of pen are available lol).
Then use the cut off wheel to carefully cut around on the line bit by bit. File is only used to deburr. As you can see the results are good.
As for (Mig) welding, I found it no different to mild steel, but I'm told back gassing is essential to get a good bead on the inside. My main concern is bits off the inside of the welds dropping off and destroying my new turbo. I might try and borrow a Tig plant for the seam welds.

[buguy]

Good idea. When I did mine I used a zip tie around the pipe and it makes a straight line for marking. It worked really well and I was able to get nice lines even on the radius.
That's interesting. I don't think I've seen anyone use a mig welder on stainless before. I guess as long as you have the right wire it would work fine. I was also worried about pieces of the wire destroying my turbo after welding mine.

[split1950]

Made some progress over the weekend with my exhaust header system. #1,2 & 3 done, only #4 to do (should be the easiest pipe I hope).

20190813_083748.jpg

20190813_083804.jpg

It's been a satisfying process so far!

[andy198712]

Wow! Looking amazing!

[rubenski]

X2!

[split1950]

Thanks guys!
I think that what really paid off was doing the "design" on CAD first.
I only used a rubbish online thing "Tinkercad" made for 3D printers etc. I had to scale down to 1:10 but it hasn't hurt.
So far not a single bend has been wasted. I'm sure if I'd winged it that would be a different story!
I will say again though cutting the SS 304 pipe is the worst part by far.

[Wally]

I will say again though cutting the SS 304 pipe is the worst part by far.

Thats why its the only material we should be making exhausts from

Curieus how you will go from here? Seems you run out of space from the 4-2 section onwards?

Does the CAD program also tells you how 'equal length' the 2 headers are per stroll?

[split1950]

I will say again though cutting the SS 304 pipe is the worst part by far.

Thats why its the only material we should be making exhausts from

Curieus how you will go from here? Seems you run out of space from the 4-2 section onwards?

Does the CAD program also tells you how 'equal length' the 2 headers are per stroll?

The plan is to bend ~90 degrees as tight to the #1 & #2 stubs as possible and run forwards up to / near to the starter which is roughly where the turbo will be.
I need to find a good solution for the air inlet pipe routing into the turbo. I need the car back here.....

No, sadly the CAD is just a space model, I measured the lengths with a digital ruler function. They are +/- a few cm equal into the Y collectors, it's not exact. It was helpful for working out the routing, bend angles, lengths etc. Also very easy to use.

[split1950]

So last night I finished making up the tube for #4.
I've also loosely put on the Turbo slip joints so I don't weld it all up and then remember when it's too late. The idea is to have the joint at the inlet to the collectors to allow for expansion and also removal of the system later on. I have some "ears" and bolts to use on the slip joints and hold them together. Once I've done the test fit on the car providing nothing needs changing I'll get it Tig'd up by my buddy.
As for the original goal of equal lengths, I have 2 lengths @ 870mm, 1@ 890mm & 1@ 850mm, so not as equal length as I would have liked but I had to use the bends I could buy not the bends I thought I could buy.
Now I've made this first one and adjusted the CAD model I presume in the future I could get it bent up out of single pieces for each cylinder rather than welding loads of pieces together. Hopefully this will not be necessary!

20190816_073922.jpg

20190816_073934.jpg

20190816_073946.jpg

20190816_073958.jpg

Hopefully it's all goood from here forwards to the Turbo.......

[Wally]

Awesome!

[split1950]

I'm just ordering some fittings for my turbo and have come up on a few issues, anyone got any type 4 experience here?;

1. The -AN10 turbo oil drain I had originally planned on using was by feeding back to the oil filler block off plate, but I think the pipe will go slightly uphill. I assume this is no good as this is not gravity? I'm running dry sump so I'm wondering if I can get away with that as there is some suction in the crank case?
2. I thought about going in somehow at the oil strainer cover as I'm running dry sump the scavenge is higher volume than the oil feed, so will I get away with the turbo oil return being below the oil level (I assume dry sump isn't really dry and still has some oil in the bottom of it) helped by the increased pick up suction from the scavenge?
3. My EFR7168 has an Aluminium centre housing and ceramic ball bearings, Borg says it must be water cooled. Bugger. Using the water system from my chargecooler is dumb and going to ruin the performance of it by heating the water? I think I probably should have a separate water cooling system / radiator / pump for cooling the turbo centre housing?

What do you know?! Thanks!

[Wally]

I'm just ordering some fittings for my turbo and have come up on a few issues, anyone got any type 4 experience here?;

1. The -AN10 turbo oil drain I had originally planned on using was by feeding back to the oil filler block off plate, but I think the pipe will go slightly uphill. I assume this is no good as this is not gravity? I'm running dry sump so I'm wondering if I can get away with that as there is some suction in the crank case?
2. I thought about going in somehow at the oil strainer cover as I'm running dry sump the scavenge is higher volume than the oil feed, so will I get away with the turbo oil return being below the oil level (I assume dry sump isn't really dry and still has some oil in the bottom of it) helped by the increased pick up suction from the scavenge?
3. My EFR7168 has an Aluminium centre housing and ceramic ball bearings, Borg says it must be water cooled. Bugger. Using the water system from my chargecooler is dumb and going to ruin the performance of it by heating the water? I think I probably should have a separate water cooling system / radiator / pump for cooling the turbo centre housing?

What do you know?! Thanks!

Uphill is no good of course, I used the (old BAS) elongated oil strainer ("tuna can") for return as I run dry sump as well. Its the lowest part of the engine by far. Worked very well. I did use AN12, so if you can...bigger IS better as far as oil return lines go.

Watercooling the center section is overrated imo
But seriously, I run the EFR7670 version with cast iron center and same ceramic ball bearings and mine has the same 'advise' from Borg. I discarded it (I have with all other turbo's as well, inclusing Garrett's GT-3076R): all worked fine without it. In fact, floating bearing turbo's don't need it from the factory and do not even have a provision for it, like the EFR7670''s predecessor: the S200-7670 I ran before.
My thinking is that the ball bearing versions produce even less heat in the bearing section and the alu centers cool themselves (since alu) even better too. So why would they need it now?
YMMV however