When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Several months ago I started a thread titled “Changing hotside from .82 to 1.05”. At that time I was considering upgrading a PT6266 V-band in/out .82A/R to a now available V-band in/out 1.05A/R. During the course of that thread it was suggested that as much of the restriction that we were experiencing was likely to be from the restrictive manifold, as the small turbine housing. After discussion with the car’s owner, we decided to move on to a T4 divided housing and matching manifold.
Having never fabricated a turbo manifold, I spent considerable time on internet research. I watched a lot of videos posted by some very skilled fabricators,,,,,, over and over and over again. I find by repeatedly watching others do a task, I can pick out details that may be missed by just watching once. Over the course of this, I found one or two specific details from each one that the others don’t cover.
To start out, it’s typical to mock fit the turbo to set its exact location. There were several different methods used, including jack stands, and bungee cords. I found one builder that constructed an adjustable jig to hang the turbo from. It seemed to me that this would allow for more precision in the placement, and could be useful for future builds.
My jig attaches to one of the studs on the coil over support and allows for several axis of movement.
First, I attached it to the turbo as it sat in the car. Then I was able to remove the turbo with the jig still attached to the turbo, and swap the turbine housing. Then I removed the manifold and using the jig, I tried to put the turbo back exactly where it was.
Wasn’t happening…., the new turbine housing is just too fat. So then I spent some time moving it around until I found how it does fit, attached the flanges to the engine and the turbo, then tacked a 2 ½” tube between the two, locking their relative locations.
Then I built a fixture from that.
Once the fixture was constructed, I removed the tubing and had the flanges with their relative positions fixed, so I could build the manifold on the bench, knowing that when I finished, it would fit.
Almost universally, other fabricators start with the collector/turbo flange. So that’s what I did.
I used 2” sch10, 321 stainless pipe to construct the manifold runners, and 1 ½” sch10 for the wastegate. The flanges are ½” thick 304 stainless. The circumference of 2” pipe is significantly bigger than one side of the divided T4 flange. Then there’s that round peg - square hole thing… At first I was considering leaving the pipe round cutting away the major overlap and welding it together with a 3/16” divider in the middle. But this makes it so that the transition happens in the ½” thickness of the flange. I just didn’t like that.
So after thinking it over, I made a die to form the end of the pipe to more closely fit the flange.
This makes it so the transition happens more gradually.
The fit up for the wastegate takeoff was the hardest part. Mirror offset angle copes on 2 90deg weld fittings. Of course building it divided all the way to the gate just adds to the difficulty.
Usually people build out the collector and runners weld it to the engine flange, then add the wastegate connection. I was concerned that the distortion that inevitably occurs when welding, would possibly “bend” the manifold slightly, and due to the tight clearances, cause fitment problems. For this reason I elected to construct the wastegate takeoff before finishing the runners, and saved welding the runners to the flange for last.
When I was unbolting the manifold from the fixture, it was apparent that this method was a success.
The downpipe went pretty straight forward
There are a few details to take care of, then before we send the turbine housing and manifold for ceramic coating, I plan to bolt it all up and see if it holds boost.
Looks like a good job. Very close to Coleman's. I am probably going to purchase one from him myself. Looks to be a very efficient design.
Originally Posted by KNONFS
AWESOME work!!
Thank you both.
It hasn't been to the tuner yet due to covid concerns.
So we've limited it to 12psi. it runs like a completely different car. Before it felt like it was straining, but now it feels free.
Thank you both.
It hasn't been to the tuner yet due to covid concerns.
So we've limited it to 12psi. it runs like a completely different car. Before it felt like it was straining, but now it feels free.
What was the original manifold you were using? When do you think it will get tuned and what kind of power did it make?
What was the original manifold you were using? When do you think it will get tuned and what kind of power did it make?
The previous manifold was a shorty Vband flange piece fabricated by A-Spec tuning.
you can see a picture of it on post #10 of THIS THREAD.
The best power was 423 or so.
We want to be there when it's tuned, so as to "participate". And the cars owner and I both have specific concerns about exposure to COVID19. So that's all on hold right now.
This thread and subsequent discussion with the OP inspired me several years ago, so I’m just going to add to it. Which I had some custom turbo flanges CNC machined with radiused inlets and such for this purpose, but decided to head in a different direction recently.
Which is presently using custom laser-cut pie pieces from T321 tubing having a 2” ID and an extra tight bend radius that can’t otherwise be achieved or readily purchased (less than 1D). The manifold is for an RX8 chassis with REW turbo conversion and the space is tight to achieve a low mount turbo position. Building one for a friend with an EFR8374 IWG and one for myself to test various non-EFR turbos. Which is somewhat difficult to achieve with the usual 2” Sch pipe fittings and their slightly oversized ID/OD dimensions.
It proved to be a bit of a challenge initially coordinating the purchase and processing of the tube stock, but the work for getting the pie pieces prepared for TIG welding ended up being more time consuming than I had foreseen. Mostly because these are 9° pie angles trying to make them as smooth as possible. So there are lots of them and being on the tighter radius makes them smaller/thinner and more tricky to handle. Also being picky about detailing them, to ensure successful welds that will stand the test of time.
Still working through prep’ing them all. Fortunately I’m not the welder though, that’a lot of tedious work as well. In addition to the T321 stainless pie cuts, I also had a set laser cut from Inconel 625 for a future hardcore race manifold.
raw laser cut pies
. ready for welding
. example elbow in final form
.
The weld in your example photo is definitely Instagram worthy. But, those spots were they "colorized" it, where it turns gray. That metal is cooked and over time could result in failure.
There is a significant degradation to the base metal anytime you make a weld. The small edge of the pie cuts being so thin, each weld is going to have multiple overlaps in the HAZ, this will have a negative impact on the durability of the piece. If the heat input is properly controlled, like with machine welding or a top tier welder, a pie cut bend like that would be great for a downpipe. I'd be concerned about hanging a turbo from it.
That’s definitely a lot of welding haha. I would have agreed that hanging a turbo from it would be concerning over time. But I’ve recently fallen in love with maven mounts. You can take all the wait off your turbo manifold which will greatly reduce chances of cracking, just run a mounting tube from the maven mount to the block to manifold flange.
