55 replies to this topic

[speedster]

Anybody have any experience with VGT's. I believe the 2006 Porsche 911 Turbo has this technology installed.

[siztenboots]

Anybody have any experience with VGT's. I believe the 2006 Porsche 911 Turbo has this technology installed.

Nothing special, even my Mondeo TDCI has one, widens the powerband plus you get an overboost when you really clog it.

[slindborg]

work with VGT quite a lot on the new transit and MY07 defender It allows for better boost control etc BUT its not really that well used on gas engines as the exhust temps are much much higher causing reliability issues. Plus the software to control the bendy turbo is rather complex so its not as simple to get good aftermarket gains. give it a couple of years and bendy turbos will be common place but at the moment they are a "new" technology for gas engines

[Winstar]

yes, as i design them for big diesel engines,

This is our version, clicky, that uses a sliding vane mechanism to change the inlet critical area.

The Pass Cars normally use the a swing vane mechanism that changes the angle of incidence of the flow into the turbine wheel. but these are alot less durable

These systems require alot more ecu calibration to make use of the performance.

Anything in particular you wanted to know.

[speedster]

yes, as i design them for big diesel engines,

These systems require alot more ecu calibration to make use of the performance.

Anything in particular you wanted to know.

Cool!
Here's a few Q's for starters...


Whats the power rating on the unit your using?
What ECU manufacturer is providing you with control?
Is there a standalone VGT control system you can use in conjunction with a normal ECU?
How do you calibrate? Is it with specific software or via a dyno?
Compared to a standard turbo what is the % increase in generated heat?
Can you use pre-compressor injection (injection of a % of fuel into the intake prior to the turbo)?
What physical size is a VGT, compared to a standard turbo rated for the same power output?
Does a VGT make much of a difference on a engine pushing say 400-500 HP?
Do you know what unit the Porsche 911 is using?
Any idea how this would operate with Bioethanol?

[slindborg]

i can answer some lol. The VGT we run is controlled via a Hella/Garret actuator that is electronically driven by a PWM signal with a hight and low side. The software/calibartion for the actuator is done on a dyno and vehicle to iron out driveability etc. Its an emmence software feature that has taken many many many months to get into its current state.... now wehter the software is over complex or not i have no idea as we only implement what the customer asks for

[ronbot]

If you are thinking of fitting one to a gasoline car, I don't think you would be able to get hold of one suitable for gasoline temperatures as only the porsche one is currently available in mass production. If you could get a porsche one, i suppose you may be able to mod it to suit your application, but you would still have to sort out some software to control it. I know several OEMs are working on High temp VGT's for gasoline engines but I think they will all be at prototype stage so even if you contacted garret, 3K MHI, IHI etc I don't suspect they would sell you one;-) Might have to wait a few more years yet.

Edited by ronbot, 01 January 2007 - 09:25 PM.

[speedster]

Thanks ppl! Been doing a bit more digging myself. The 2006 Porsche 911 VGT's were developed by Borg Warner. To get over the temperature issues the unit is using special materials from the aerospace industry so, yeah, looks like this technology is still a bit out of reach... Couldn't locate a breaking 2006 turbo on ebay... what a pity Though it looks like the Mitsubishi Lancer Evolution 4G63 uses a VGT.......

Edited by speedster, 02 January 2007 - 12:21 AM.

[paulf-cam]

The software/calibartion for the actuator is done on a dyno and vehicle to iron out driveability etc.
Its an emmence software feature that has taken many many many months to get into its current state.... now wehter the software is over complex or not i have no idea as we only implement what the customer asks for

What makes the software so compilicated? Surely its just a case of changing the geometry based on the revs?

[slindborg]

add in driver demand (pedal position and rate of change), engine temp and ambiant air temp compensation, air density compensation, emissions controls, self testing, etc etc Its never as simple as people think lol. Just like aftermarket ecus that simply run fuel maps based on throttle position and revs, whereas the OEM stuff is full of various compensations for emissions regs/driveability etc.

[speedster]

add in driver demand (pedal position and rate of change), engine temp and ambiant air temp compensation, air density compensation, emissions controls, self testing, etc etc

Its never as simple as people think lol.

Just like aftermarket ecus that simply run fuel maps based on throttle position and revs, whereas the OEM stuff is full of various compensations for emissions regs/driveability etc.

OK, we have access to all that sensor data and ECU's like my Emerald ECU have aux outputs that can be configured for specific use. So, in general is there some mathematical formula that can be applied to kick off a base VGT tune?

[slindborg]

add in driver demand (pedal position and rate of change), engine temp and ambiant air temp compensation, air density compensation, emissions controls, self testing, etc etc

Its never as simple as people think lol.

Just like aftermarket ecus that simply run fuel maps based on throttle position and revs, whereas the OEM stuff is full of various compensations for emissions regs/driveability etc.

OK, we have access to all that sensor data and ECU's like my Emerald ECU have aux outputs that can be configured for specific use. So, in general is there some mathematical formula that can be applied to kick off a base VGT tune?

from a basic principle its not "that" far from a normal boost control type of arrangement if you only want to have X boost at X rpm. but it depends what the aux output can drive and how configurable it is in the software.

plus youve got the ballache of getting a VGT..... there are lots of FGT turbos about that are more than capable of being driven normally on a big power engine. have a look on the vauxhallsport.com forum for turbo info

[Winstar]

I work in the Aero-Mechanical design of the turbos not the calibration side of thing as the OEMS do that, but slindborg seems to have that side covered. Compared to a standard turbo what is the % increase in generated heat? The increase in heat generated depends on the operating point of the turbo as this changes then the efficiency and the part of the turbo where the expansion of the gas changes. on our VG turbos we use water cooling in the bearing housing. Can you use pre-compressor injection (injection of a % of fuel into the intake prior to the turbo)? The compressor end of a VGT is the same as that in a standard turbo, I'd think post compression plenum injection is a better route, as Lotus did with the Esprit V8, as the energy required to vapourise the fuel is take out of the gas and acts as a form intercooler. As long as you don't need to pass Hydrocarbons emmisions. What physical size is a VGT, compared to a standard turbo rated for the same power output? The actual sizes of the VGT compared to a standard then they're not much bigger, generally longer as the mehcanism fits within the scroll of the trubine housing. However you have to accomodate an actuator to move the mechanism which can be quite bulky and also sensative to heat. Does a VGT make much of a difference on a engine pushing say 400-500 HP? Yes the VGT should make the engine more responsive at lower engine speeds as it allows the best efficiency can be obtained at low swallowing conditions and then still allow high efficiency at a higher swalloing conditions. Do you know what unit the Porsche 911 is using? i don't know what the porsche is using and as some have said the gas temps are higher with petrol, the main problem with this is pass car turbos are made as cheap as possible, about $20-50, and churned out in high volume and car manufactures generally have to make the engine fit whats available. So very few manufactuers are prepared to pay for or accomodate whats required for the durability. Whereas in Heavy Duty Deisel our turbos are matched/designed to the engine application and have more expensive material and watercooling so they will last over 1/4 of a million miles. Rob

[paulf-cam]

add in driver demand (pedal position and rate of change), engine temp and ambiant air temp compensation, air density compensation, emissions controls, self testing, etc etc

Its never as simple as people think lol.

Just like aftermarket ecus that simply run fuel maps based on throttle position and revs, whereas the OEM stuff is full of various compensations for emissions regs/driveability etc.

Yeah, i can see how all that stuff would make developement take a little longer...

[ronbot]

As someone who works for one of the passenger car OEMs and spent a couple of years on turbo system design, I would say the OEM's these days cannot afford to ignore the durability aspect. Competition is so tight these days that the customer will not tollerate a product that will not last at least 100k. Typically we design engines to do at least 100k miles even under very hard use. The high speed maifold crack/turbo tests run on dyno are pretty harsh, ocillating between peak power and peak torgque for a few hundred hours for example. This means the majority of cars thesedays (other than maybe those that spend there lives as track cars) will vast exceed 100k, even turbocharged ones. In addition, the package constraints that make it ever harder to fit all the components under the bonnet, mean a lot of the turbos are bespoke for a given powertrain configuration and sometimes even vehicle specific as well. Although a given powertrain might use basically the same turbo in terms of frame size, often inlet/outlet flanges on compressor/turbine connections have to be different to package intake/exhaust routings and to minimise pressure drops for best efficiency.

[Winstar]

I wasn't saying that pass cars don't take durability into account, the last pass car engine i worked on the design target was 150k miles/10 years. However ask the shopping people how much they're willing to pay for a turbo then it will far less than what it would cost to make one for a gasoline engine.

[slindborg]

yeah you spec up something that will do the job that your required to do, then the bean counters will be offered one at 1/4 of the price that is jsut as shiney but lasts 1/10th as long and they buy that lol

[speedster]

BorgWarner sent me the following details when I inquired about the unit

Hello,

we are not allowed to sell these turbos in the Aftermarket. The turbos are
only available
from Porsche dealers. The part numbers are 5304 970 0060 (OE part no.
997.123.014.72) and
5304 970 0061 (OE part no. 997.123.013.72).

Kind regards,

BorgWarner Turbo Systems GmbH
Sales IDS/Technical Support

Anybody know anybody in a Porsche dealership with a parts department? Be interesting to see what they sell for.

Edited by speedster, 03 January 2007 - 12:27 AM.

[Winstar]

My bet is around £1500 - £2000 if our aftermarket prices are anything to go by. The Borg switched system from the BMW 535d cost about £2500 from our local dealership, I wonder how i know that . I'll ask if anyone at work knows how much they are. Rob

[Mike9009]

Rob I work at Trucast and supply Holset with the turbine wheel! Small world ehh! In my limited experience the design of the Holset (Cummins) VG seems much simpler (cheaper??) than the BorgWarner/ Honeywell design. How do the two differing designs compare in terms of performance? Is the design difference due to the exhaust temperature differences? Mike