Holset VGT 5VZ-FE turbo build

I had the TRD supercharger on my 260,000 mile '01 TacoTurd, but got bored with that and decided at VGT Holset sounded like fun to experiment controlling boost with.

I have a sweet excel file with all the parts I've purchased if anyone wants to duplicate this build. I'll post that up when it's complete, but one should be able to do this whole setup for under $2k.

Main parts:

- 2007(?) HE351VE Holset variable geometry turbo
- CX racing manifold/downpipe kit
- Keating machining downpipe adapter (allows the Holset to bolt straight to the CX downpipe)
- LBB (Lil' Black Box) arduino-based VGT controller

Update: project complete-ish!



Current mockup: I'll probably have to do a body lift or move the turbo down.




First off, this tool is awesome and saved hours of time. Seriously worth the $150. Can't believe I never bought one before. Such tool.

I started by pulling the motor because I wanted to do the timing belt and replace all the seals on the motor while I was in there:


Booger-welded a -10an fitting on the pan:


Oil fitting into block: 3/8" NPT to -4AN:


Oil line (48", -4AN)





More to come in the next few weeks.

 

That Tacoma hood is going to have a bulge in it.

@Blackdawg

 

Hmmm

 

Yea...way to big of a turbo for this motor. Not worth it.

 

Holy big fucking turbo. Going to be a tight squeeze. Maybe recommend a hx35

 

The turbo is actually pretty appropriately-sized for this engine, but I sure won't be running it very fast. Remember that the 6.7L diesel this comes from only needs to rev to 3000 RPM, and makes 260HP. The 3.4 is basically half the displacement, but revs to 6000 RPM, so it will be just about right for required airflow.

In the turbo world, you can get the same airflow from a large turbo at low shaft speeds as you can with a small turbo at really high shaft speeds. People want to run the slowest shaft speeds possible while still having acceptable transient response, since bearing losses increase rapidly with your typical hydrodynamic fluid film bearings.

Here's a HE351/HX40 (they're the same) compressor map. I'd like to get the full-load, full-rpm operating point to sit right on the ~76% efficiency island, which should work out to about 30 lb/min @ a pressure ratio of 2, so ~12.5 psi of boost at my altitude. Remember that boost and airflow are not totally dependant on each other. AKA 6 psi of manifold pressure from the TRD supercharger does not create the same mass flow of air as 5 psi from a larger turbocharger.

Edit: here's a better version of the map:

 

Sorry for my ignorance, I can't figure out what that chart represents. I am interested in learning more about forced induction.

 

I am well aware of everything you are talking about. I had a supra for a few years and am very familar with turbo systems.

While yes you are right about the shaft speed and efficiency and all that. You are not right that that turbo will spool fast.

the 6.7l diesal it came off moves a LOT more exhaust volume then the 3.4l gas motor ever will. That is why it works so well. The 5vz will not produce enough exhaust to cope with it. And since our motors actually only rev to 5.5k rpms, you'll be lucky to get 2k rpms of full boost from that thing. Id want more usable boost personally.

From my research and looking at compressor maps, our motors would be perfect with a 55mm Compressor inducer and 68-71mm turbine with .58-.68 A/R. Full boost would be about 2.5k rpms.
Yea you'd be dealing with higher shaft speeds at that point, but thats where getting a high quality ball bearing turbo pays off.

Best place to learn how to read a compressor map is Garrets website. Go there and start reading!

 

Well, we'll see with the VGT functionality! It basically works to vary the exhaust scroll A/R ratio. This is more of an experiment than it is trying to squeeze the best overall performance. You're right - this isn't the ideal turbo to go with if that were the goal, but based on a little figuring, I don't think it'll be too bad. Lots of DSM guys run HX35's on their 2.0 4G63 motors, and the HX40 wheel in the 351VE isn't much bigger.

Remember that air mass flow in = exhaust mass flow out + fuel mass. The volume flow rate is really a secondary factor and doesn't play when calculating turbine power. The important parameters for turbine power out put is the enthalpy difference of the inlet and outlet of the turbine, and the velocity difference across the turbine. For a similar power output of around 290 HP, the 6.7 will use about 25% more air mass/rev than the 3.4, but the catch is that the diesel's exhaust enthalpy will be lower overall since more energy gets extracted in the cylinder, due to the more efficient diesel cycle. So really turbine power produced shouldn't be waay different. It's really hard to predict without really expensive software (GTpower) and lots of dollarbucks worth of pressure and temp sensors. I'm no scientist, just a lazy engineer that takes a more "cowboy" approach to these sort of things - which is to just try it first and see what happens.

Here's the conservation of energy equation for a turbine operating at steady-state. You can see that the work extracted by the turbine is only a function of mass flow rate, enthapy change, and velocity differential (the potential energy term, g*z is almost always ignored in compressible flow problems, and Q is taken as zero assuming the turbine housing is perfectly insulated). The cool thing about the VGT turbo is that you can basically control the velocity term (capital V) of the equation. There are other implications to doing this, but it's fun "knob" to turn.

 

Ok figured the turbo wasn't going to sit up like that. Behind the headlight seems to be the easiest location without going too stupid on the body lift.

You gots your mathz in place. A bit under budget. You'll probably be over $3k unless you got some good hook ups and have access to stuff use garage guys don't have. The stupid little parts add up quick. What are you doing to support the turbo? For example meth, injectors, piggyback, afr gauge, data logging, intercooler

 

Well i guess the only thing is to fire it up and try now..

 

I've already ordered everything actually, it's sitting right around 2k. And I'm using the CX manifold kit, so that's where the turbo sits.

Tuning will probably be something simple since running low power levels. Probably an Apexi S-AFC just to up the high load MAF voltages a bit.

 

Did some measuring. I think the turbo may barely clear the hood. I will probably have to cut away some of the hood support underneath the skin.


Other updates:

Test-fitting the IC: I had to move my transmission cooler up to sneak the IC in there but it should fit behind the stock bumper cover really nicely. I'll probably trim some of the bumper cover for more airflow.


The 2.5 - 2.75" elbows line up perfectly with where I want to punch holes through the radiator support.



Painted the manifolds and downpipe with this Helix high-temp paint. It's the only stuff I could find that was supposedly rated to 2000F. Everything else is only rated to 1400F, which is right about what the max EGT's of a turbo-gasser will put out. We'll see how it holds up.



Decided to replace the knock sensor harness and found the lower intake manifold gaskets had disintegrated. Replaced those.



New OEM exhaust manifold gaskets


Coolant and oil lines all mocked up and/or cut to length:




The oil drain is funky. The way the 351 has to be clocked makes a standard drain fitting interfere, so I got one of these things on order. I'm thinking it will barely fit.

 

Should be a fun build to watch..

I was actually just finishing up some practice problems on conservation of mass and energy for my thermo class when I stumbled on this thread Kinda nice seeing actual applications of what I'm learning.

 

Intercooler mounting is finalized and am awaiting a co-worker to weld so brackets to the intercooler.

Aw yiss

I might trim more back for more airflow, but boy, the bumper cover fits so nicely without any mods.

I wanted the core to fit behind the bumper beam, but the radiator stay piece was protruding too far. I just cut the bottom part off, welded a new piece to the core and drilled a bolt hole though so it can be removed like normal.



Used a combination of a 3" hole saw, and lots of dremel cutting discs to cut the holes for the IC pipes. A quick coat of paint for rust-resistance followed.



The A/C drier thing had to be moved up a tiny bit (you can see where the clamp used to sit). I just had to trim down the plastic headlight bracket to get it to clear.


I had a really hard time finding an appropriate turbine outlet adapter to go from Holset to CX downpipe, so I just cut the v-band off the stock elbow, cleaned it up with a wire wheel, then welded another V-band that is close the CX in dimensions. I clamped the two to their mating parts during welded to prevent warping.


Finalized oil drain routing. I added foil tape insulation after the pictures were taken.




Clocked compressor housing:

 

I hated doing the hole saw through my core support

 

Yeah it was not really until I switched to a corded dremel with some mini cutoff wheels. Then the whole process took all of 20 minutes.

 

Here's the turbine adapter plate from Custom Fab Shop. I had to do some customization to get it to work with the weird clocking of the VGT:

** You may not have to do some of these steps - I'm pretty sure someone re-tapped the threads on this particular turbo - I found that my turbine housing had a mix of 3/8-16 and 1/2-13 thread. It should be an M10x1.5. So check before ordering stuff. ***

1. Drill out one of the holes to clear a 12mm bolt
2. Tap cooresponding hole on the other plate with an m12x1.75 thread
3. Grind relief for coolant banjo bolt
4. Drill out one of the countersunk holes on turbo's adapter to thread into the bigger hole on the turbo.
5. Drill out matching hole in turbo-side gasket
6. Grind down the head of the 1/2-13 low-head screw for the turbo-side of the plate.





There's a lot of weight sitting on the relatively flimsy manifold, so I made a quick support bracket and welded it to the charge pipe. It isn't ideally placed but was the easiest way to get some useful support structure made.


Engine goes back in tomorrow!