VNT or Twin Scroll
Turbo Mini Forums

While looking for a turbo for the Formula Student car we did consider a VNT turbo, and some other teams have used them in the past.

The link below shows how one of the teams have actuated their turbo (Page 70ish). Using a 2 port actuator (Bosch
204103 Dual Port), and I imagine some sort of smart piping with check valves relative to atmosphere so that the vanes act 'open' when off throttle and at high flow rates.
This aids fuel economy by providing little back pressure when off throttle, and limiting the maximum boost created at high rpm.

Unfortunately it was found that they also needed a wastegate as the diesel turbine could operate over a wide enough range at high rpm.

http://digitalcommons.calpoly.edu/cgi/view...55&context=mesp

That's an interesting read. Thanks.

I's got me thinking about different methods of controlling the vanes. How about hydraulic using oil pressure or pneumatic using a small electric vacuum pump??? Obviously in combination with a PWM output from the ECU and a fast acting solenoid.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I started to look for a turbo for the engine I'm building. Originally I wanted to use GT1549 but then I found out that the GT1549V version is one of the most popular turbo available here in Poland - we do love the 1,9 tdi.

Anyway the whole idea of the VNT sound really interesting but since I'm new to the topic I thought I'd better ask for your opinion.

GT1549V has flange simular to T3, so maybe I could get away with using stock Metro turbo manifold with an adapter plate. I will try to sort out a damaged unit to check it. I was going to need a custom manifold for the non VNT variant so if it doesn't fit it won't be a problem. However I'm going to use carb and this may be the main limiting factor.

So correct if I'm wrong:

GT1549V should spool up even better than non VNT GT1549 from the lower revs, but the issue might be controling the boost. VW uses N75 valve to do exactly that. Basicly 3 vacuum hoses connected to actuator, servo, or air filter. There is the an electric connector as well and here is my first concern, what is it for exactly, does ECU influences the valve and therefore turbo or maybe it's just pickin up some signals to adjust other things? I'm planing to use Megajolt and it does allow me to wire other output/input signals so could it be used to control the vanes somehow? External wastegate might help here as well, that would require using a custom manifold but again I was planing to use one anyway.

I'm sure it can be done but if it's to much hustle I'll stick with a normal turbo. I don't want to complicate things to much.

Made in Poland build thread:
http://www.turbominis.co.uk/forums/index.p...tid=570363&fr=0

Personally I would not use the GT1549V.

Because you don't use a wastegate with a VNT, the turbine has to pass all the exhaust gas. I don't think that the GT15 turbine would be big enough.

The GT1749V has a much larger turbine and a large A/R ratio, which on a fixed nozzle turbo would be slow to spool. However the variable vanes can be controlled to effectively reduce the A/R and spool it up quicker.

The N75 valve is controlled by the ECU with a PWM output. That varies the vacuum seen by the actuator and hence controls the vane position. I can map this with the MS3, but I'm not sure how you would do this otherwise.

I've been searching for a solenoid to do this job and had not seen the N75. It does just what I want. Thanks for that.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I'm only aiming for 120HP, so I think GT15 should be good enough but I'm exploring all the options to get it working as low as possible that's why I thought about the variable geometry variant.

How about not using a control valve? I could connect the actuator directly to vacuum and use external wastegate to prevent overboost. Or that won't work as well.

Made in Poland build thread:
http://www.turbominis.co.uk/forums/index.p...tid=570363&fr=0

You wont have vacuum at wide open throttle when you need to control the vanes.

Unless you put in a separate vacuum pump

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Schrödinger's cat - so which one am I ???

Schrödinger's cat - so which one am I ???

There is a huge amount of those electronic actuators on ebay. That says a lot in itself.

Yes, trying to size up a 12v vacuum pump

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Well, the plan is to control the turbo with the MS3 controlling a N75 valve (maybe the Saab equiv.) on a vacuum supplied by a small vacuum pump like this:

https://www.coolcomponents.co.uk/vacuum-pum...CFafnwgod9yYA0g

It should be big enough to move the actuator at an acceptable speed. The maximum vacuum will provide nearly 15kg force on the actuator arm which is plenty to overcome the force of the spring.

I plan on using 2 step control from the MS3. 1st step will be to switch on the vacuum pump at, say, 80% TPS. 2nd step would then be to control the N75 with a PWM signal. The N75 will be configured to bleed air into the system, thus reducing the vacuum at the actuator.

I'm going to use the spare injector outputs on the main board for both the pump on-off and the valve PWM. They are unused at present (if they are soldered in - must check and add if necessary). No need for relays if I use the injector outputs.

Edited by Paul S on 17th Oct, 2014.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Why the added complication of something else to go wrong? Can't you just have a vacuum chamber? Perhaps a roll-cage cross-member or something? That has to be a lot more reliable surely; or are you never planning on letting off? LOL

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I'm using an N75 pwm valve to control boost via MS2 boost control, it's a rpm vs tps table, it works really well. Interesting to see how it will work with vacuum. I can't see why the nozzle actuation has to be controlled with vac, why not just a setup like a waste gate, pressure vs spring, open to closed.

Joe C had issues with that method.

Best explained here:

http://www.msextra.com/forums/viewtopic.ph...lit=vnt#p299935

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I see from that. So it's seems, as the car manufacturers have gone with a full servo motor control system I would guess this is the only way to fully get the results needed.
Looks like the control for this is going to need to be mapped as the fuelling VE table is done 16x16 cells to make it work nicely. Engine load vs rpm. Could it be possible to run a dual table all the time and use the injector pw to control the servo motor or even an N75 via the vac pump you've been discussing?

Not sure what you mean by dual table.

As I use the injector outputs on the MS3X, the 2 injector outputs on the MS main board are available to be assigned to a mapped PWM output.

I did think about just outputting a PWM to drive the pump, but as it is a single acting piston pump, I'm not sure how it would react. So, as I plan with the Water Injection on my car, I'll run the pump at 12v and control with the PWM on the valve.

As for the table, I think it will have to be TPS vs RPM. You can't use engine load (MAP) as that is what you are trying to control.

The standard PWM maps are just 6*6 cells. As it is the control is likely to be linear and I'm only looking at TPS from, say, 75% upwards and rpm from 2000 upwards, 6*6 will be adequate and a lot easier to set up.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I was thinking of the secondary fuel load, VE table 2, I'm not sure on how that works I've not needed it, I thought the injector pulse width output could be utilised. That is if it is an extra injector output.
Although as you say your trying to control MAP so I guess tps vs rpm is what's needed.
That's back to the boost control table, good thing there is all the variable settings for different pwm devices. Be it a motor or valve n pump setup.
Thinking along the lines of using boost pressure again. If the turbo where to run away with out any control, then it will always give pressure for use to control.
I'm also assuming you are including a blow off valve as pressure bleed/relief while closing and off throttle, as an extra basic control.

Edited by Nick king on 18th Oct, 2014.