Picked up a large used Garrett VNT turbo. Apparently came off a 14 liter Detroit Diesel Series 60 EGR motor.

There is a Detroit Diesel tag on compressor backing plate with two numbers:

23531102
FKM44621

Tag also says: "Garrett GTA"

Compressor cover says: 0339-4 M24 A/R 0.69
Turbine housing says: A/R 1.72 M3 J282
Center Section says: M52

Anybody have access to a catalog that can shed some light on this turbo?

Popped the compressor cover and took some measurements. 14 blades, 7 tall, 7 short. Inducer looks to be about 70mm. Exducer around 102mm. Works out to about a 47 trim. Looks to be somewhere between a GT4294 (70.3/94) and a GT4202 (74/102). Not much play on the shafts side to side or end to end, but noticed some oil on the compressor blades and the backing plate. Seems like it is seeping from where the compressor wheel is recessed down into the backing plate. Looks like it seeped after it was removed since it is not flung out in a radial pattern, but maybe it was leaking and this is why it was pulled. Any thoughts? How hard is it to put new seals into one of these guys? Can I get the parts? I've seen some DD memos that says they service these as a unit...might be tough to find parts for this turbo.

Anyone know if I can swap in different compressor wheels and covers? The backing plate seals to the compressor cover with a large o-ring and is roughly 21cm/210mm in diameter at the seal. I can't tell you the diameter of the shaft, the nut that holds on the compressor wheel looks to be 18mm but is a little sloppy. An 11/16" socket will almost fit, but some of the shoulders on the corners of the hexnut are a bit rounded so it won't quite slide on.

Disconnected the actuator for the vanes. The lever that is attached to and moves the vanes moves smoothly. Looking into the turbine inlet with a flashlight, I can see many of the vanes and they move fine from almost all the way closed to an "open" position. Was going to pop the turbine housing off to inspect the mechanism and measure the wheel. It is attached to the center section with 8, 12-point bolts. A 5/16" 12 point socket fits on them with a little slop but it feels like these are metric. I only have 6 point metric sockets, so rather that booger them up with the 5/16" socket (they are rusty) I held off. Any tips for removing these rusty bolts after I pick up a set of metric 12-point sockets?

Been researching this turbo a bit, haven't come up with much.

Looking at the Garrett Catalog (pp. 66-67)....

http://www.turbobygarrett.com/turbobygarrett/images/catalog/catalog_common/catalog.pdf

...his compressor wheel looks very similar to a GT4202, although the A/R on his compressor cover could be different. Garrett says this turbo is good for 500 to 1000 horsepower on motors 2.0 to 8.0 liters. The compressor map is on page 51. My thought is his compressor should have been able to deliver more than 13 psi on a 402, so the issue may have been the hotside. Either the vanes wide open may not be big enough, or he couldn't get the vanes adjusted right. The specs on the VNT hotside supposedly are as follows: It can have an A/R of as small as 0.42 in fast spool mode. Then it can open up to provide lower restriction with an A/R as large as 1.72. I think I have the same hotside as his, mine is stamped 1.72. While 1.72 sounds big, maybe the vanes, even when wide open, "clog it up" so it acts like a smaller A/R.

Another guy over on the Corvette Forum (BaldTurboFreak) shared some info on these turbos and said the compressor is made from billet titanium for high pressure ratios. The compressor may be a different design, so even if the inducer and exducer are the same dimensions as the GT4202 in the catalog, it may flow differently due to a different shape and a different cover.

http://forums.corvetteforum.com/showthread.php?t=1413757

I would really like to play around with the VNT aspect of this turbo which seems to be functioning just fine. To control the vanes, i picked up a remanned Detroit Diesel VPOD (variable pressure output device) that is used to supply various air pressures to the VNT actuator. You send the VPOD a PWM signal from a microcontroller to control the pressure output. The pressure works against a spring in the actuator and as it moves in and out, it moves a lever that controls the position of the vanes. I've got a Basic Stamp microcontroller that can send the PWM signal to the VPOD. I'm trying to tap into my ECM to get things like RPM, TPS, MAP, etc to feed into the Basic Stamp real time. The turbo has a built in tach sensor in the center section. If I can get these inputs into the microcontroller, I should be able write a program to control the turbo electronically via the VPOD the same way Detroit Diesel does.

I would be happy to find out I could throw different compressor wheels and covers on this turbo to dial in the cold side for an HT-502 I am planning to use it with. As it sits now, this turbo is probably more suited to a 350, or lower rpms and lower levels of boost on the 502. I'm running an EBL with a 2-bar map sensor with tables up to 200 kPa. I'm only looking for 15 psi max and not planning to take the motor over 5500 rpm. I would be happy for now if I could get the results bigturboz28 got. The problems I see are that my motor is 100 cubic inches larger than his and my compressor inducer is 4mm smaller. If his problem was the hotside simply was not big enough, it will only be worse for me. One guy in his thread mentioned removing some of the vanes to open it up. Sounded easy to do, I don't believe he tried it. If his problem was that he could not get the vanes adjusted right, I may be able to solve this problem by approaching it with a microcontroller, VPOD, and the factory actuator.

Seems like an interesting project, at least to me. Anyone know anything about this turbo? Where can I get seals? Can a bigger compressor be swapped in?

 

cool, sounds v.interesting!

if you can, pics of the unit from various angles would be good.

i presume its undivided t6 flange?

 

I tossed in the obilgatory Cola can for scale, though it might be a bit misleading. On the shot of the turbine exducer, the can looks bigger than the exducer. The can fits into the outlet with about 1/2" of extra room. (The can is closer to the camera than the turbine blades).

The green wire you see in most of the pics leads to the shaft speed sensor.

Pic VNT 3 is the money shot. You can look up the turbine inlet and see the vanes. They are close to "full-open" in that pic.

The last pic shows the VPOD and the microcontroller.

 

I can't get my dial caliper into the turbine outlet, but eyeballing the exducer shows it to be around 75mm.

 

When you get the turbine housing off you should be able to tell what frame size turbo you have. A GT42 should have a 77-78mm inducer turbine wheel and a GT45 will be about 82-83mm. The compressor wheel you have is not a titanium version. The fact that it's a cast wheel and has splitter blades tells me its aluminum. All the titanium Garrett wheels I'm familiar with only have full blades in the cast versions.

The VGT is one of their dual axle versions which is a heavier duty version than the AVNT's like are used on the 6.0L Powerstroke's and the Duramax's. Be carefull when you pull the turbine housing off to keep the Vane assembly in one piece. They can be a pain to reassemble when the pieces fall apart.

I won't tell you it cannot work but be advised that the turbine wheel is Inconel 718 which is only reliable up to about 1600F. The bigger limitation though is the VGT portion. They are designed to live in much cooler exhaust. They are typically are temperature limited to about 1300-1350F. As you push beyond this they may have a tendancy to lock up.

Steve

 

Jay K. Miller mentioned the temp problem with these turbos in his new book that came out a few months ago (pp. 16-17). He didn't provide much information on these beyond that.

Even though the turbo would be mounted underhood, I was planning to route some fairly long tubing before the exhaust enters the VNT, 10 to 12 feet. The hope is that the exhaust tubing can dissipate some heat over this length and eliminate the blowtorch so the VNT mechanism can live. Spool shouldn't be a problem, just tighten the vanes a bit. Hedman Elites (14ga, 3/8 flange, thermal coated), merge behind the transfer case and back up to the engine bay. I've got the room and using off the shelf headers (already installed) really cuts down on the fabrication. Not sure if this will do the trick, especially for an extended load. Perhaps an egt probe at the turbine entrance would be wise. Thanks for the heads up and the temp figures.

BigTurboZ28 kept referring to it as a GT45V. Said he got his from a friend on the inside at Garrett. My turbine exducer is right around 75mm. I can see the larger turbine inducer through the vanes. It looks like it is more than 2 or 3 mm larger than the exducer. If it is a GT45 frame, does this open the door to using the 81.62mm/108mm GT45 compressor? What's involved? Cover, backing plate, and compressor? Would the shaft accept the GT45 compressor?

Titanium Compressor: Mine looks like aluminum to me. Maybe others came titanium.

 

Found this on TiAL Sport.com:

GT4508: direct bolt on upgrade for GT4202
When 1,000HP Isn't Enough.

Already have a 4202R and want more power? What's that you say? A GT45 compressor housing will not fit? Here is your answer. We take a GT4508 and machine a GT42 Compressor housing to accept the GT45 compressor wheel, it looks great too!

http://www.tialsport.com/NewHotness.html

I need to bust open the turbine and find out what I've got...

 

Is it possible the oil leaked out of the center section when it was shipped, but during operation, when the compressor housing is pressurized and the turbo is sitting in a normal position, it doesn't leak?

 

Try to avoid VNT's on peformance apps. Trust me....doesn't work good at all. The vanes will stick with carbon buildup and you will kill the turbo.

 

The Garrett approach to VNT is a lot more complicated and less durable than Holset. I hear what you're saying. I didnt' pay a lot for this turbo, so I would like to give it a whirl. Fall back is I'll have a T6 setup ready to bolt in an S400 down the road. Sinking a bunch of money into the coldside may not make a lot of sense if the hotside won't hack, but I would like to know my options if I can get it to work reliably.

As far as carbon buildup, do you think I could inject water into the exhaust stream periodically to steam clean the vanes? I've seen water injection clean combustion chambers spotless, perhaps it could help here.

 

I disagree on the carbon build up causing problems to a certain extent. I recondition both the garret avnt and the borg warner vnt almost daily. The carbon is not what causes the problem. The problem is moisture. Typically, moisture will rust the turbine housing to the point that the rust bubbles cause the vanes to stick. If you are burning coolant then the coolant will gum up with the carbon and create a mess. A diesel creates a ton more soot/carbon than your turbo race engine does, so carbon build up will/shouldn't be an issue or a turn off. If you suspect the vanes are sticking, then pop the turbo apart and clean up the housing. I wanted to use the borg warner unit on my build but could not convince the boss to let one disappear. My plan was to use a waste gate off an externally mounted wastegated turbo to actuate the shaft and use a exhaust back pressure sensor set up as a safety with a solenoid connected to the turbo waste gate actuator. This would throw the turbo vanes open in the event of an overboost condition.

 

Taking a closer look at this compressor wheel, I'm starting to wonder if it was cast. There are tooling marks all over it.

Miller has a picture of a wheel with similar tooling marks. Here's what he says: While most compressor wheels are a casting, this compressor wheel has been machined from billet. This process is done by a five-axis milling machine and is used in special low-volume, high-durability, or high-performance applications where no production wheel will fit the intended need. These types of wheels are tremendously expensive but are justified in many instances.

Sounds consistent with what BaldTurboFreak said.

How can I tell if this is aluminum or titanium?

 

Can't find a map for this compressor wheel. Trying to get an idea looking at maps of other GT compressors.

IND EXD TRIM HP
Mine 70 102 47 ?
GT4294 70.3 94 56 500-850
GT4202 74.7 102 53 500-1000

Mine has an inducer the size of the 4294 but an exducer the size of the 4202.

This is why my trim [(INDxIND/EXDxEXD) x 100] is the smallest of the 3 at 47. Smaller trim flows less air. So definitely less than the 4202, but more or less than the 4294? Maps below.

 

Followed up on that Tial lead. Basically they are swapping a full GT4508 with a smaller cover in place of the GT4202. Their offering is not a compressor upgrade for the GT42.

 

subscribing.. I am putting a holset VGT on my DD summer car this year.

Cheap and worth a shot ... 60mm variety

 

Sounds like a good project!

How much air do the 60mm Holsets move?

How are you planning to actuate the VGT mechanism?

 

I have many different numbers so I will say 500_hp to be very conservative. The numbers I have seen are mid 60's LBs (ie well over 600hp potential). I plan on manually gating it initially but have some things in the works with my neighbor(also building a VGT) to use and electronic method of control. It will take a ton of R&D and I believe each motor/combo will use a different table for opening input.

 

I wonder how twin 60mm Holset VGTs would spool on a big block if you could get the VGT mechanism dialed in.

 

I like the way you are thinking... Since I am considering twin VGT's on my BBC. I think they would work great. Yes they are heavy.. but for now they are cheap and a little bit of control would really make for a fun ride