gtb1749vk vs. gtb1756vk

Hello guys,
I'm new here so I will introduce myself a bit first. I’m a 20 years old student car technologies, next year in my final year of high school I need to do a thesis project. The project that we will do is a dual parallel vnt setup on a 4.2 straight six Nissan patrol engine. The idea is to put a turbo on the first 3 and one on the last 3 cylinders. By that way we create 2 times a displacement of 2.1 liter for each turbo.
Because I know that the VAG group use a lot of vnt turbo’s I registrate here on the forum cause you guys can maybe help me in choosing the right turbo? The idea is to get a flat torque map, ideal would be to get the turbo’s spooling at around 1500rmp until the engine reach 3500rpm so on pretty low rmp’s. I was thinking to use the gtb1749vk that is used on the 4.2TDI VAG engine, that because it has the same displacement as my engine and I can find it pretty easy cause it is used by Mercedes and ford on their 2.2 liter engines. I have read a bit here on the forum and saw that many people use the gtb1756vk on 1.9TDI engines and get the turbo spooling at 1800rpm? So I’m now doubt of what I will do? Can you guys maybe give your ideas of what I can do best? The ideal for me is to get 14psi from 1500 till 3500 rpm and maximum 20psi that the turbo can deliver.
We want also work with electronic actuators, we work with IFM electronic module to write a program that generates a pwm signal to let the actuator react.

Thanks a lot guys!
Grtz Zjuul

Stay with the 1749. Your biggest challenge (and achievement, if you get it to work) is to get the electronic actuator to work with a standard PWM signal. You could make a business selling these.

There are 2 (actually 3) big things you need to think about integrating turbochargers from a Diesel application into a petrol engine:

1. The compressor-side thrust bearing seal may not be suitable for operation under a throttled vacuum; if the engine is an older one with carburetors or throttle-body injection, there's also problem of fuel being drawn-through the compressor (probably needs different seals/thrust bearing);
2. The exhaust gas temperatures of a petrol engine are higher than a Diesel and VNTs don't like high temperatures;
3. The control strategy of a VNT is different from that of a wastegate. The VNT's vanes are constantly moving and modulating, whereas a wastegate simply begins to open once a desired boost pressure is reached.

thanks for the reply,

I don't going to use a diesel turbo in to a petrol engine, the 4.2 Nissan patrol engine is also a diesel you read petrol instead of patrol I guess? Are those gtb1749vk turbo’s easy to find? I think they are used in the 2.2 ford duratorq engine as well in the 2.2 mercedes engines and also in BMW 2 liter engines but they all are gtb1749vk turbo’s but have different garret numbers is that cause the cartridge is the same but the housing is different?

Thanks a lot!
Zjuul

Oh whoops, my bad.
http://www.ebay.de/dsc/i.html?LH_TitleDesc=1&_nkw=057145873&LH_PrefLoc=2

GT1749V is used in many, many applications, as you know:
http://www.turbomaster.info/eng/catalogs/model.php?base=garrett&pagina=GT15V

They are not all exactly the same, however. There are different trims that support different airflow and power levels.
My favourite is a actually a GTC1549VZ. It depends on what your objectives for this project/thesis are. If you want simply to put 2 turbos into an engine, you don't need to be so picky. If your criteria is fastest possible response, something like the aforementioned GTC1549VZ would be good, or a GTC1444VZ or GTD1449VZ. See the thread titled, "Turbo technical database" and also search the forum for more information on these.

I think of starting with used turbos and when the whole thing works well buy new ones. There can be a small fault in the program that blows the turbo out so it would be a shame to break new turbos… so I stay with my first thought and work with the gtb1749vk turbo?

Fine idea.

The 4.2 engine is a conventional diesel engine that has a standard power of 125bhp and a torque of 260Nm. The idea is to give these pretty large engine a modern boost, I don’t want to get 300bhp and 800 Nm of torque out of it but the standard values are very low for that displacement. So my goal is to get 200-250bhp and a flat torque map of something about 400Nm that starts in pretty low rpm’s. That is why I want to work with a dual small turbo setup to get the boost faster and the vnt technology to get the boost over a longer range. I don’t want to use an to small turbo cause it will be blow out of pressure to fast and a large turbo will not spool fast enough. So I guess that the gtb1749vk will be good for my needs. Are the differences between the various models very big?

There is more to is then the turbos for getting a flat torque curve. that part would be done more with the fueling of the motor.

how are you going to control the fuel? by "conventional diesel" do you mean a mechanical pump with a LDO (boost enrichment) or a electric-mechanical pump (Servo, like the ALH)

With the turbo vane control you are going to want to incorporate a PID

yes indeed I know that fueling is also important. the engine has a inline mechanical fuel pump so it is very easy to adjust the volume of fuel that gets injected.

zjuul said:

yes indeed I know that fueling is also important. the engine has a inline mechanical fuel pump so it is very easy to adjust the volume of fuel that gets injected.

Click to expand...

how are you going to control boost enrichment?
the mechanical pumps deliver a base volume as boost build the mixture is adjusted to take advantage of the extra air. On the VW's it is done with the the LDO (ufo looking thing on the pump) also called a ALDA/ADA

this is for a VE pump and not a P-pump
http://www.vwdiesel.net/forum/index.php?topic=30340.0
http://www.cs.rochester.edu/~jag/vw/engine/fi/injpump.html

Benz P-pump
http://mercedesforum.com/forum/dies...t-mw-m-pumps-engine-tuning-maintenance-46097/

don't know what particular injection pump stands on the engine (don't have him already) but if there is no option to get boostcompensation I think to build a merc inline pump on it that have boostcompensation.

doe anyone know what signal that I need as a variable for the controller inputs?

zjuul said:

doe anyone know what signal that I need as a variable for the controller inputs?

Click to expand...

The standard VNT is controlled by injected quantity, desired boost, actual boost and rpm. It uses a base vane map (witch says this vane position for that much injection quantity and rpm) and a PID controller to calculate the final position of the vanes, depending on requested boost and actual boost.

stand alone controller done on a open source platform ( I would get a MAX32 if i were you and you chose that route way better processor)
http://dmn.kuulalaakeri.org/vnt-lda/

There are several ways to control boost enrichment, you can use a servo like above or put a solenoid in between the boost ref and the enrichment part with a small open atmosphere vent hole and duty cycle the valve to get the correct reference. Probably would want a pressure sensor post solenoid to make that part dynamic. it will be a lot of work and tuning to get it right

Tell me about it. I've done about 50 logs using a modified version of that software. Works great now though!

Sent from my HTC One XL using Tapatalk

what do you want to know vanbcguy?

Not so much what I want to know as expressing the many iterations it takes to tune something like this without a dyno. It's hard to find deserted stretches of public road where you can make dozens of runs while logging.

I'm still fighting a bit of oscillation on my setup. But I can definitely break both my front tires free! (yes I have an LSD) I have things more or less under control but I can definitely see why ECU engineers get paid the big bucks, and why tunes cost what they do.

Sent from my HTC One XL using Tapatalk

we have whole load of dyno's at school so tune the system is no problem. I can use the 4 wheel drive dyno as much as needed

I have build and am running a standalone turbo controller using standard pwm works perfectly fine on my gtb2056vk been using it on my mtdi build for over 2 years now. I think the biggest problem most people run into building am electronic controller is trying to over complicate the build, make it simple. I have a tread on here showing my build on the test stand and its simplicity. Only one actual input (EMP not boost its unimportant only a byproduct) to the board the servos position and a rheostat to control my base setpoint for more or less EMP relative to boost, very simple very effective servo reacts perfectly fine and is adjustable on the fly.

I wrote a controller myself a while back, you can overcomplicate it but one of the best control methods I used simulated a theoretical value for torque at the turbine shaft, it was unitless and somewhat arbitrary but served as a scale of what energy we actually had at our disposal (based on IQ and RPM) and meant the setpoint could be quite rigidly controlled to an 'always acheivable' value, which gives a PI controller a reasonably straightforward job to maintain boost in almost all operating conditions.

One thing I wanted to avoid was implausible setpoints otherwise you risk excessive backpressure as it tries to ramp the actuator to try and do what you don't have the ability to do. And no matter how you design a controller, if it uses closed loop control generating achievable setpoints have always been key IME

anywho, from a programming POV it's a lot of fun to have a stab at

Follow the advice of Tylenol, Dakta and Keaton, start off with the controller in http://dmn.kuulalaakeri.org/vnt-lda/ and adapt to your needs. You do not need boost at low load as the engine is throttled anyway. If your Nissan Patrol specimen is a pretty recent model, it will have an e-gas pedal, e-throttle, and knock sensor, so you could modulate the VNT to start boosting only above 50% throttle opening. Keep the absolute pressure conservative to prevent knock - 1.5 bar is likely as far as you will find you can go on 98 RON without knocking. On second thought, a small trim GTB2256V might be better for less EMP - knocking is very sensitive to trapped residual gas and this is a direct function of EMP. Good intercooling is important for sure, water injection can definitely help, all good subjects to research for your studies.

Edit: Wait I'm a little confused this is a Nissan Patrol petrol or diesel? Lol. If diesel ignore my whole last post. 1549, 1749 or 1752 still best turbos to use.