eBay 2.7L Mercedes sprinter Turbo (http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&item=8008188608&rd=1&sspagename=STRK%3AMEWA%3AIT&rd=1) :D :cool:

It's from an L-5 2.7L CDI (Dodge/Freightliner/Mercedes Sprinter) so my little 3.0L OM617 should not stress it too much.

Since it is controlled by vacuum, it should be more simple to manage. I'll keep the auto-tranny vacuum valve in the system to keep the vanes open at idle (at least I hope it uses vacuum to open them). This should help exercise the vanes and keep them from carbon sticking.

I can control idle, but how would I control my max boost level (12-13.5psi)? Would a simple electronic boost controller work? I could set a boost switch to trip a vacuum solonoid and let stored vacuum do it.

You have no idea how much excitement I'm trying to hold back. ByeBye KKK Turbo, Hello VNT! :nos:

Looks like it might be a little harder than I thought. I believe the vanes "close" when vacuum is applied.

That might not be as bad as you think. Depending on how quick they move, under partical throttle they open a bit and full open at WOT. It would be self regulating in a way due to the vac bleed for the transmission.

Since this turbo and the VW/chrysler VNT turbos are all garrett, could the actuator pods and brackets swap out? This would give me a dual sided diaphram to control it with.

BTW, anyone have a spare dual-sided pod to spare/sell? :D

will you need different manifolds?

tom w

Someone on this board has already done this. It was posted about 6months ago. He was able to convert the VNT to a standard style wastegate with parts from Turbo City or another turbo supplier. IIRC, it was easy to do. RT

Someone on this board has already done this. It was posted about 6months ago. He was able to convert the VNT to a standard style wastegate with parts from Turbo City or another turbo supplier. IIRC, it was easy to do. RT

I can't find any thread with that kind of info.
That does sound like a good idea. The vanes are normally open with the vacuum pod. If a normal wastegate pod were installed, they would be held normally open until boost kicks in. :D

There is only one problem with that idea, idle. Closed vanes at idle are not very efficient. It could also allow corbon sticking from lack of movement during low load city driving, hence the reason I want the dual port pod.

UPDATE: Mike from Forward Motion has some pods for $50 used and $135 new.

I dont know a lot about turbos but I am always intrigued by adding new technology to enhance where practical. Is turbo need driven by throttle position or RPM or both? could you sense throttle position or RPM and drive the gate position from that?

The electronic controls on modern engines uses several inputs: Engine RPM, Turbo RPM, back pressure, Boost level, Throttle Position...

I have put together a system on my VNT that will use both boost level and engine load (via the stock vacuum control valve) to control the turbo. It is still untested though, the turbo is sitting right here on the floor. :D

why not just flip the vaccume switch for the auto tranny, and flip it upside down?

I don't think it would mount up-side-down due to the shape, plus if you have an auto you would need it to work as it was designed.

I don't think it would mount up-side-down due to the shape, plus if you have an auto you would need it to work as it was designed.

but what if a Manual swap is also in the works? that just might make it easier....



Why Don't you post your control (theory) here so you can have more input on it (more people to pick out the flaws)

I already have a manual transmission. :D

Here is the theory:

The turbo vanes are normally closed for fast spool times. As boost pressure increases a pressure actuator opens the vanes to limit boost (this is how forcedinduction's VNT works), the pressure actuator should limit boost at about 15 psi.

At wide open throttle the Vacuum Control Valve is putting out no vacuum so the turbo will be controlled by the pressure actuator alone. At lower throttle the vacuum from the VCV will open the vanes to further reduce boost via a vacuum actuator mounted opposite of the pressure actuator. This should solve the problem of the turbo putting out full boost no matter what position the throttle is at, which exists when a pressure actuator alone is used.

The vacuum control valve puts out a maximum of 15" or so of vacuum which is around -7.8 psi. My pressure and vacuum actuators are roughly the same diameter so every psi of vacuum applied to the vacuum actuator will be equivalent to a psi applied to the pressure actuator, and should reduce the boost the turbo is producing by the same amount. At low throttle my boost level will theoretically be 15 psi – 7.8 psi = 7.2 psi. The factor of boost reduction from the vacuum signal could be adjusted by increasing the size of the vacuum actuator. I also plan to use a vacuum thumbwheel from a euro headlight setup to reduce or cut off the vacuum signal to the turbo on the fly.

Any input?

Here is my setup installed and working.

Both actuator rods had to be ground thinner in order to fit on the lever at the same time and the intake heater was removed to clear the actuator head.

I used an unmodified automatic tranny vacuum control valve. I am pleasantly surprised at how well it works with the initial settings. All I had to do was shorten the vacuum valve arm a tad so it wouldn't hit the stop peg before WOT. A little more tweaking and smoke tests and I believe it will work out great.

During normal acceleration (Sane, not pedal to the floor) I get 3-5psi and boost response is instant to throttle movement, the same when cruising the highway at 70mph. Flooring it will send it to the normal 11psi right away. It seems to work fine without any vacuum restriction orifice.

The only problem is I can't get more than 11" of vacuum even straight off the main line with nothing but the booster attached. I'll change the pump check valve later but if that doesn't work it may be time for a new pump...

More pictures.

Control valves (Middle picture): The foreground is the constant vacuum supply valve. The background is the vent valve. Vacuum starts at the back valve, passes through it, into the crossover tube, through the fore valve and out the vertical hose in the fore valve.

Dash switch: I hacked the A/C knob light harness for the switch's light power source and use the (missing) cruise control's power supply wire as the switched power source. I also installed a 3amp inline fuse since these small valves draw so little power.
Up=High performance (vanes closed, high boost right away)
Middle=Normal operation
Down=Economy (vanes always open, minimal boost all the time, good for smokescreens)

Glad to see it works. ;)

If you are using the standard Auto Transmission setup there is an orifice between the main vacuum line and the VCV. http://www.mercedesshop.com/shopforum/showthread.php?t=218054&highlight=vacuum

I plan to experiment with orifice size to get a fast boost response when I step on it, but not so fast as to cause a spike in exhaust pressure.

No orifice, just normal tubes and hoses.

After more thinking, adding an orifice the same as used for the auto tranny will cause a spike in exhaust pressure. It will drop the vacuum level to the actuator quickly closing the vanes quicker. You would have to put the orifice after the VCV in the line going to the actuator instead of before the VCV.

I'm not so sure. Now that I think about it I don't think the orifice has any effect on how fast the vacuum bleeds off, only on how fast it rebuilds when the "throttle" is closed. It also should limit the amout of air that is being sucked up the vent when at WOT.

So BGKast is going to a GT2056V
And F.I. has a GT2256V
How would these compare to each other, and for that matter to the stock Garret turbo?
Is the 2056 smaller than the 2256?
once installed how would they differ in performance?

Yes, the GT2056V is smaller than the GT2256V on the turbine side: Link (http://www.turbobygarrett.com/turbobygarrett/tech_center/gt_basics.html)

There is still some question as to if the Liberty turbo is a GT2056V or a GT2260V. This (http://www.turbomaster.info/eng/applications/passenger_cars_model.php?fabricante=JEEP&logo=jeep) site says that a GT2056V with part #s that match my turbo was used in 2004 Jeeps, and a GT2260V was used in 2005 Jeeps. The strange thing is the Liberty CRD was only sold as 2005 and 2006 Models, and they seem to have used the exact same engine.

Regardless, both will offer much faster spool times than the stock T3, and both should be adequate for the 617. At 3000RPM and 14.7 psi of boost the 617 uses about 20 lbs/min of air which is right in the center of the plot.
http://www.turbobygarrett.com/turbobygarrett/images/catalog/Turbochargers/gt20_images/751578-2_e.jpg

I think Forced's turbo has the same AR and trim as my turbo. The 2256 will spool slightly more slowly, but will be less restrictive with the vanes open.

I just called the Jeep dealer and the 2005 and 2006 Libertys used the same turbo accoring to the part number. The GT2056V makes more sense though because the turbine outlet is only 2.25" vs the 3" outlet on Forced's turbo.

This is exciting! I am very tempted to buy another turbo and try to follow along with you guys! I am (just barely) restraining myself from maybe getting in over my head. I don't really know enough about the stock systems that I ought to be allowed to fool with non stock turbos.... yet. Fortunately there is no law against this (even in Canada, and even though it is a LOT of fun....)
sssh don't tell anyone up here or they will pass a safety law of some sort and I will have to take courses, make applications, apply for permits, and probably have to wear a helmet!

That is a very good link you gave me/us and I can see that I am going to do some studying while you guys are doing the trail blazing.
Also I suppose each of us may have some different goals for what we would like to achieve. Me, I would like more power mostly in the mid and upper mid rpms, and of course less lag. Others may be looking for maximum power at WOT. Perhaps the GT2056V would be better for this than the GT2256V...?

The GT2256 is probably best, but the GT2056 is only $100. :D

am i going to have to tear my CRD apart and find out numbers for you guys?
AFAIK the libbys (US models) came with the 2260... i have one guy that has confirmed this for me.

i wouldn't trust any of the jeep dealers for any info on these vehicles. Just my personal experience with them and the 3 CRD's that i have taken care of.

those of you with the libby turbos out let me know exactly where to look and i should be able to take a look at it this weekend.

these are off the one turbo i know is for sale:
3524211sf

S/N-1kg005161& S/N 763380-1

and according to the site that everyone is siting that is the 2260

Just the numbers on the tag on the front of the compressor housing would be great.

to bid that etc etc
There is a GT2056V on Ebay:

http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&item=350052624936&ssPageName=ADME:X:RTQ:MOTORS:1123

I may bid, or not.... If another member PMs me that they are going to bid, I would be less likely to bid against a fellow member....

For $45 I should buy it as a spare. :D

Looks decent. Even at $200 it might be worth the gamble.

some kinda rip-off?
There is another E-bay item with the EXACT same pictures of the 2056 that I missed earlier today (sold for 109.00). It is being sold by another seller in another state.
The first one that I put a 106.00 snipe on and missed is at:
http://cgi.ebay.com/ebaymotors/ws/eBayISAPI.dll?ViewItem&viewitem=&item=350052624936

Now here is someone else selling the same thing with the same pix?:
http://cgi.ebay.com/ebaymotors/2005-Jeep-Liberty-GT2056V-Garrett-turbocharger-turbo_W0QQitemZ270232451178QQcmdZViewItem

-and although he might actually have a good turbo and be using someone else's pix - he also claims that it also has 27,000 miles on it. Something stinks here eh?
To further confuse things the first seller has another 2056 for sale, (although it doesn't look quite the same):
http://cgi.ebay.com/ebaymotors/2007-Jeep-Liberty-GT2056V-Garrett-turbocharger-turbo_W0QQitemZ350055699161QQihZ022QQcategoryZ33742QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

And then there is this fella in China with a short, and sorta shaky feedback, they look like new.... but I wouldn't want Chinese bearings I have had terrible luck with Chinese bearings:
http://cgi.ebay.ca/GT2056V-705954-0013-TURBOCHARGER-S_W0QQitemZ120253850378QQihZ002QQcategoryZ33742QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

Yeah, I saw that one with the same pictures, odd. That third one is NOT a turbo from a Jeep, both years of the Liberty CRD (2005 and 6) used the same turbo. I would also stay away from the chinese one, it is probably a counterfeit Garrett.

Keep your eye out, they come up quite often.

Yeah, I saw that one with the same pictures, odd. That third one is NOT a turbo from a Jeep, both years of the Liberty CRD (2005 and 6) used the same turbo. I would also stay away from the chinese one, it is probably a counterfeit Garrett.
Keep your eye out, they come up quite often.

I shouldn't but might be tempted by the Chinese counterfeit, if I hadn't had just appalling experiences with Chinese bearings in the past. (It does look very good, but maybe TOO good.:rolleyes:)
I think I may just lay back a while and keep checking the wreckers now & then.

I have been thinking about a way to control the VNT without two actuators, and I have drawn up a schematic that shows how it could work. The only problem is I can't find a needle valve that is pressure actuated. Anyone know of such a thing? Basically the vacuum signal sent to the actuator would be a function of the pedal position and the manifold pressure. What I need is a piston that is pressurized by the manifold pressure connected to a needle valve. As the pressure increases it pushes the valve shut which increases vacuum thus actuating the vanes. Make sense?

https://engineering.purdue.edu/people/brett.j.bell.1/300TD/VNTcontrol.png

It could work. The hard part would be calibrating the pressure actuated valve. You would also still have to make a custom bracket for the vacuum actuator to get it to work opposite of how it works on the stock turbo.

I guess I had it backwards. I was under the impression that the vanes directed the exhaust to the turbine in the default position. But, now I see they actually function more or less as a wastegate.

In the stock arrangement the vanes default open and the computer supplies vacuum to close them to spool the turbo. In our case we want the vanes to open with the application of vacuum so we can use the VCV for the transmission.

After thinking about it some more, that would actually make it easier to control. You guys probably already discussed this, but you could just reverse the action on the VCV valve, and use a boost controller with a very small pin hole to dump vacuum at a given pressure. I think that I will play around with this idea a little to see if I can make anything productive happen.

That is what I originally thought of doing but the problem with it is unless you isolate the control system with a separate pump, the door locks, climate control, shutoff and brake booster will only be charged with vacuum at WOT.

That is what I originally thought of doing but the problem with it is unless you isolate the control system with a separate pump, the door locks, climate control, shutoff and brake booster will only be charged with vacuum at WOT.


what about a separate anti-air chamber with a couple of check valves?

Hey Guys,

I've been following these posts rather intently for a few months now with the intention of building my own diesel vnt project. I don't have a Mercedes though but a 2.8l Toyota Hilux which I have turbo charged using the GT2256v. I have also implemented an electronic programmable boost control system where I can specify the vane position by RPM and throttle position.

I have as instruments a pyrometer and boost gauge.

With this system I can post 5 psi of boost at 1500rpm climbing to a maximum of 12psi by 1800 in a full throttle accel. It will easily get to 20 if you let it but for engine preservation, 12 is good.

EGT's are around 8 - 900 F while doing these pulls.

My problem is that at part throttle, around 3 -5 psi, the turbo sounds like it is surging.

Has anyone else noted similar results?

If I try to push the turbo above 10psi before 1500rpm it will surge. Above that speed it will do any pressure up to 25psi.

Even at 2-2500 rpm part throttle it still makes the surging chatter. If I accelerate it goes into the smooth whine. I seems to more depend on load than rpm. If I climb a steep hill it will never surge.

My engine starts at 87hp @ 4000 stock aspirated. I was wondering if maybe at part throttle the turbines torque is insufficient to drive the compressor against surge.

Is it possible that the vanes are opening and closing rapidly? Perhaps the loop rate on the controller is too slow? It sounds like a controller thing to me.

I assume you have the model 3L engine.

Put your specs into this forum and it will let you play with RPM, pressure, intercooler, altitude, etc and plot it on the compressor map. I already put the 3L's 96x96 bore/stroke into the form- http://www.not2fast.com/turbo/glossary/turbo_calc.shtml?FeetASL=5280&Tamb=30&Bore=96&Stroke=96&nCyl=4&RPM=4000&VE=100&Boost=10&Ec=75&Eic=0&PdropIC=0&TambIC=30&wiPercentMethanol=0&wiRate=0&wiTemp=0&SFC=0&AFR=0&maxInjectorDutyCycle=0

Its not perfect but it will give you an idea of where you are hitting the surge line and how much boost you can run before it chokes.

Thanks for the leads guys. This phenomenon has been occuring before the controller was built and I had the vanes in a fixed position right through the range.

I tried the Not2Fast link but I could not get it to work.

I have a theory on what is happening but I guess the link could help solidify this.

Assume vane position fully open is 0 and vane position fully closed (resting on the stop screw) is 100.

The tables in my controller only allow me to build boost at values of 50 or higher. Below that the vane position is too wide open to allow boost greater than say 5lbs to be built.

I am thinking from this that my turbo is too big and I should be running perhaps a turbo with a smaller back end so that the full 100 percent of vane motion is usable.

Let me know what you think about this.

(PS Should this be in a new thread or is this ok?)

You don't want the vanes to see 100% open in normal use except maybe at WOT at max engine RPM. If the vanes get all the way open you loose the ability to limit boost pressure.

The surge may be due to cam profile. Your camshaft is setup for a non-turbo engine so if it has any valve overlap (scavenging) in the profile the higher exhaust backpressure pressure will not let the cylinder exhaust completely (backflow into the cylinder/intake), especially since VNT turbos have very high backpressure at low speeds. If there is a turbo version of your engine it would be a good idea to find out if it's camshaft is different.

(This is on topic, I'm fine with it here.)

Never thought of that. Finding a turbo cam may be a long shot.
I was thinking a while ago of using a cam gear. I could advance the cam timing to at least get the exhaust valve closing event in sync at the expense of early intake closing.
I know on gasoline this makes much more low end torque but kills the upstairs.
I will see which of these I can accomplish. Should be interesting to note the results.
Thanks for the suggestion.

Today I added a jumper wire and diode from the cruise control harness to the vacuum venting solenoid valve. Now when I use the brakes the vanes will automatically close to provide an exhaust brake effect. The 1N4001 diode is to prevent the brake lights from coming on when I close the vanes with the switch on the dash.

It works great and it does provide significant braking effort above 2500rpm.

does it make the jake-brake sound?

Great Idea!

Today I added a jumper wire and diode from the cruise control harness to the vacuum venting solenoid valve. Now when I use the brakes the vanes will automatically close to provide an exhaust brake effect. The 1N4001 diode is to prevent the brake lights from coming on when I close the vanes with the switch on the dash.

It works great and it does provide significant braking effort above 2500rpm.

I can't imagine why you would want this on a passenger car even with a stick. Sounds like it make for some jerky driving.

I agree, it isn't immediately apparent that this would be useful on a car, but I can imagine that going down long mountain passes, this will keep the heat out of your brakes, not to mention increasing the life. I mostly think its a good idea to utilize what's already there to your advantage even if its not necessary.

I agree, it isn't immediately apparent that this would be useful on a car, but I can imagine that going down long mountain passes, this will keep the heat out of your brakes, not to mention increasing the life. I mostly think its a good idea to utilize what's already there to your advantage even if its not necessary.

Vented brakes will take care of that. Also I have to wonder about EGP when the vanes slam shut, particularly as stated in your example, after coming up a large hill.

Of course EGP will be high, thats the point. Letting the engine cool more efficiently is another issue.

Of course EGP will be high, thats the point. Letting the engine cool more efficiently is another issue.

I dunno I would just be concerned with eating a bunch of food and then having my hole slam shut. Particularly after the engine got heat soaked.

I can't imagine why you would want this on a passenger car even with a stick.

I live next to the Rocky Mountains.

As long as EGP is under 30psi, it doesn't matter how quickly it goes up.