Limequat's 4.2l I6 Swap

[SupraN/Asty]

Haven't posted for a while, but it's not for not working on the beast. A lot of fabrication at this point, so the going is slow. Here's what I've been working on:

From left to right they're the downpipe stub and 2 up-pipes. Welding is mostly done need to do some finish grinding and sanding. Hoping to rig up a pressure tester tomorrow to check weld integrity. Then on to get ceramic coated. Here they are getting test fit, please excuse my flux core welding skills:

The above is a good view of my stock manifold, chopped, flanged and ceramic coated by Tom at Stovebolt.

After I get these banged out, I'll start on oil and water lines to the turbo.

what are your plans for the opporation of the vanes on the turbo? is it off a 6.4L, total understanding of its geometry and oppertion can be a VERY strong combination with a 4.2 i6

 

[limequat]

what are your plans for the opporation of the vanes on the turbo? is it off a 6.4L, total understanding of its geometry and oppertion can be a VERY strong combination with a 4.2 i6

It's off a 6.6 (aren't all duramax's 6.6?). 2005-ish. The vanes are controlled via oil pressure. A pulse width modulated signal drives a solenoid that controls vane position. So what required to control it isn't that different from a closed-loop boost controller. Want more boost? Close the vanes to create a smaller A/R, turbo spools faster. Want less boost? Do the opposite. Megasquirt and a few sensor inputs will do this for me. In the short term, I may rig up a pot and a 555 timer circuit for a dial-a-boost type thing.

 

[Bleakvoid]

With a Megasquirt, you can do a lot more than that. You could dial-a-boost depending on throttle position, intake temp, any number of things. Basically, boost pressure would become just another trim adjustment; set a static value, and add to or subtract from it to adjust for motor/environmental conditions. Depending on the language MS uses for custom code, it could be as simple as while/do loops. It's a very interesting prospect; imagine being able to tell your turbo to spool to max pressure early when you're at freeway speed, and slowly at city speeds to keep from burning out all the time at 30psi...I think I could code it using just tach, speed, TPS, and boost signals.

I'm going to be watching this, very closely...you're on the brink of convincing me to go with variable geometry. It just seems like the next logical evolution in turbo technology, to be honest. If you need any help with the controls, let me know...I have ideas, and this is intensely interesting.

 

[SupraN/Asty]

I thought it was a variable geometry turbo off a the newer Ford 6.4, They looked the same. Fords uses a different combination of pids to determine vane position. Youre are building a very inovative project. Im not aware of a vgt on a gas motor making big numbers?, or in a sup, with a chevy motor..good luck..definatly going to volunteer as a consultant i. youve got my support with this thing

 

[limequat]

With a Megasquirt, you can do a lot more than that. You could dial-a-boost depending on throttle position, intake temp, any number of things. Basically, boost pressure would become just another trim adjustment; set a static value, and add to or subtract from it to adjust for motor/environmental conditions. Depending on the language MS uses for custom code, it could be as simple as while/do loops. It's a very interesting prospect; imagine being able to tell your turbo to spool to max pressure early when you're at freeway speed, and slowly at city speeds to keep from burning out all the time at 30psi...I think I could code it using just tach, speed, TPS, and boost signals.

I'm going to be watching this, very closely...you're on the brink of convincing me to go with variable geometry. It just seems like the next logical evolution in turbo technology, to be honest. If you need any help with the controls, let me know...I have ideas, and this is intensely interesting.

I thought it was a variable geometry turbo off a the newer Ford 6.4, They looked the same. Fords uses a different combination of pids to determine vane position. Youre are building a very inovative project. Im not aware of a vgt on a gas motor making big numbers?, or in a sup, with a chevy motor..good luck..definatly going to volunteer as a consultant i. youve got my support with this thing

Thank you both for the offers of help. This will be my first foray into Megasquirt, and I'm sure I could use a hand in the programming.
Bleakvoid, the possibilities are indeed endless. Speed dependent boost? Sure. Exhaust brake? Sure. Once installed, it's just coding to do some wacky stuff.
Check out this guy on youtube. He has an older Subaru Legacy with the Ford version of this turbo (slightly smaller). The link is him doing a 20psi boosted launch, but there are plenty others.
http://www.youtube.com/watch?v=y4lDkm6Xj5g&feature=related

That video completely sold me. If he can do that with a 2.2 liter, imagine an extra 2 liters of displacement and a bigger turbo. The implications are frightening.

 

[Courterman08]

cool build

 

[Bleakvoid]

http://www.ls1tech.com/forums/forced-induction/670015-vnt-turbo-megasquirt.html <- This guy did it. MS + 6.6L duramax VNT, only on an iron 402 LSx motor. Looks like he eventually went with a twin T70 setup, citing that it was too small for the 402; considering that it's basically a GT45 or GT42 (can't quite figure that one out), it should be perfect for the 4.2L, possibly even the 3.0L 7M's and 2JZ's. There's also mention of the Ford version, which is slightly smaller...again, a possible fit for the 3.0L motors.

Just thought I'd share with ya

 

[IJ.]

Gonna be interesting to see how well the VNT copes with the Gas EGT's.

 

[limequat]

Cool, thanks for the link. Hadn't seen that one before. I'm a little skeptical that a vvt "GT45" wouldn't have enough exhaust flow, but whatever works, I guess.
The turbo I have is a GT3788VA. Garrett doesn't make a GT45, so I'm not exactly sure what he had. The ford version is a GT3782VA.

I've seen other builds that use a wastegate actuator for control. It's simple, but it cause the turbo to search for max boost at all times: idle, highway cruise, etc. It would be fine for a 1/4 mile car, but nothing I'd want to drive every day (or buy fuel for). Those that are fortunate enough to have an AEM can use the existing boost control!

---------- Post added at 09:59 PM ---------- Previous post was at 09:57 PM ----------

Gonna be interesting to see how well the VNT copes with the Gas EGT's.

I don't think it will be a problem. I don't plan on running very lean. They sure don't like diesel soot, though. That's why they're so cheap on ebay!

 

[efi-diy]

It will be good to get more 4200 running MS3x code... Later this spring I'll be converting to sequential. Hopefully we can convince the authors to add in sequential support for the GM 7 notch crank wheel. The 7 notch code wheel decoder work ok in batch fire but lacks the cam timing input to support sequential.

On another note the rear sump oil pan is moving from prototype test castings into production version very soon.....

 

[IBoughtASupra]

Forget the 402....454 all the way. That thing would spool a 80MM turbo with no effort and if you bore and stroke it.....or if you wanted to be crazy, a 527 crate motor rated at 620HP on pump gas.

Imagine that with two 70MM or 80MM bolted to it on C16?

 

[87nasupra]

This thread has inspired me to get off my ass and get to work on my car, lol. This is one of the best, if not the best, build thread i've seen on the forums to date. I enjoy looking at all the cool stuff done but this is outstanding! Best of luck, subscribed and will be following it. Good luck !

 

[limequat]

Thanks 87nasupra.

Small updates:

Here's my coolant inlet T-d for the turbo. I had to cut and reweld the nipple on the block to angle it away from the downpipe. Routing all this stuff so that it doesn't melt will be a real challenge

---------- Post added at 09:31 PM ---------- Previous post was at 09:29 PM ----------

Got my up-pipes coated. Supposed to be good for 1800 F. I did have my welds ground down nice, but then they sand blasted them before the coating. Had to redo them. Ugly, but they hold 50 psi and stand up to black magic blasting

---------- Post added at 09:35 PM ---------- Previous post was at 09:31 PM ----------

Mocking up coolant lines. The feed is stock. The drain I lopped off a section and tried to flare a nipple. This may need more work later on. Both hoses will splice into the heater hoses.

 

[kamikazemkiii]

Can I have one of those turbos?

 

[hvyman]

About time this thang got a turbo.

 

[dslocal]

Do you have any examples of where to get these turbo's for cheap? I was looking for one and all I can find are new ones selling for like 1400
And also are these ball bearing?

 

[limequat]

Do you have any examples of where to get these turbo's for cheap? I was looking for one and all I can find are new ones selling for like 1400
And also are these ball bearing?

There's several on car-part.com for under $500. I think I saw one or two for around $100. I scored mine on ebay for $125 and it came with a spare! There's several up right now with buy-it-now prices in the $250 range.

They are all journal bearing. You don't need ball bearing when spool is instant anyway

 

[limequat]

Here's the beginnings of my oil feed line. What a PITA. The diameter of the hose fitting is greater than the width of the Banjo, so I have to fab a little spacer just to bolt it down. Of course that makes it too tall to work with the bolt that Russsell supplies with the fitting. I wish I could have found OE quality stuff for this

 

[supra90turbo]

I can get you a bolt for that banjo fitting if you want to try the GM factory bolt.
It's about $17, though. $12 if you can get it for wholesale price.

Part number is 97365194. Maybe it's shorter, but it doesn't seem so, M12x1.5x24

 

[limequat]

Thank you. I may have to do that, as I think I just stripped the lower one. Seriously, it's made of cheap aluminum and they can only manage 3 threads on it? Might've had better luck making it out of oak.