To be honest Im not expecting too much below 4000rpm... I was hoping that due to the extra cc's I can get away with a bigger turbo than your average 2ltr..

Im also helping that the heads, cams, headers, and mapping will sort out the bottom end.. I also have the option of antilag and the nitrous..



David

What was it mark said?

Built by Wallis, Rebuilt by Lateral Performance

 

I also only find lag and issue in first and second.. ie, thats where Id rather have no lag..

David

 

The dump valve will open with boost pressure on one side and atmo on the actuator side no problem.
To test this, just remove the signal line to the DV and see how much boost you get at WOT

 

Andy,

Is that with an OE DV? I'm about to fit one that's a little less reluctant to open

Richard

 

I'm sure a Forge DV with a red spring is easier to open than a 1 bar actuator.

 

Can anybody answer my question here ? I am interested, especially as Pavlo says it's some months away, as to what exactly he's talking about.

 

making the turbo work like a gas turbine.. spinning the turbo producing air...

do a search for "jet engines car turbos" or similar.

Very interested in this aswell.. deserves its own thread

David

 

I think Flat Four is talking about the same method as Paul was some time ago. I assume that by keeping the mix rich the temperature is kept within the capabilities of the turbine.

I posted a link in drivetrain to 'home made gas turbines' some time ago. In fact, David it was you I recommended to fit one in the back of your wagon

Richard - Not sure of 'all' the aftermarket ones but any that I have tried will not hold more than 1 bar without pressure behind the diaphragm. Generally they feel stiffer to push open but this is offset by the larger surface area.

Andy

 

I guess a cracking pressure of 1 bar helps the cause

Richard

 

Yes, your right. As the vent duty cycle will be irrespective of boost pressure (I think...JB ?) the last thing we need happening is the DV opening at a low pressure during spool up in a lower gear !

 

Sounds to me like you could use a preset frequencied multivibrator circuit driving a solenoid in the feed line to get the boost you want (what do you need? 1.3 upto 4k?).

Only question then is when to switch it on and off - AH!, but, now I know why you were talking of using an MBC........you *cannot* use factory boost control with such a system, it's MAP based error correction wouldn't be too happy with what you're planning and turbo would probably self destruct quickly

Richard

 

Turbo dynamics to zero and max/min duty cycles the same

= no closed loop boost control/ECU interference, just a dumb TPS vs RPM duty cycle

 

I still don't want to use an MBC though - call it pride

Richard

 

Pat,

If you need a 2.0 litre Impreza still running a standard manifold to try your exhaust out you can borrow my purple WRX. It's running a hybrid, front mount etc. I'm just trying too get hold of some 550cc injectors and changing to front entry turbo in the next few weeks for the G force rolling road day.

Colin.

 

Andy - problem is, the MAP will see the dropped pressure ('cos dv is cracked) and raise the duty, working the turbo even harder. MBC is the only way to go with such a solution.

Richard

 

Thought that was the idea, more PR and flow, but higher efficiency?

 

Yes, that's correct, the pressure should not drop (if it does then the duty cycle needs reduced) The pressure should basically rise to hit set point, then a fine balancing act between WG duty cycle (or Dawes) and recirc vent duty cycle to maintain boost with minimum recirc to avoid surge.

 

bttt

 

Using a Dawes for boost control and the solenoid for the DV would work. But I reckon trying to maintain use of the OE solenoid wastegate control plus a second circuit to control the DV would give stability issues.

In the very early days I tried using a Dawes with the solenoid to remove creep but maintain factory control. It worked up to a certain point in spool when the dawes was cracking open - the solenoid went completely, and unpredictably, nuts. My boost gauge (manifold pressure) didn't show the cause, just the effect. I didn't have the ability to monitor MAP pressure at the time, but assume the instability started here.

Maybe with the DV being so far from the MAP things would be more stable? Only one way to find out

Richard

 

Hope I'm not being stupid here, I know very little about compressor maps & surge lines etc but how about adapting some kind of variable nozzle turbo, like the Garrett units used on the VAG PD diesels?

I don't know what the flow capacity of these is, but with the VNT you SHOULD be able to get good boost at the top end and quick spool up, with manageable boost in the middle. Don't know how you would manage the duty cycle of the VNT controlling solenoid but you guys are doing some very complex stuff with ECU maps, it shouldn't be that hard ...

Just 10p worth -- and probably worth about the same too

 

Thanks for the suggestion BB We already have the good top end boost and quick spool up with the TD05/06-20g It is achieving the high boost at relativly low flow in the midrange that we are trying to overcome.
The VNT would not change this as it is effective on the exhaust turbine, the midrange surge is a function of the compressor flow map. What we really need is a variable compressor trim !

Andy

 

Andy,

The boost threshold on that particular turbocharger I posted the map for should be somewhere below 2000 RPM, but without trying one it'de be difficult to know exactly where. Just need a slightly bigger version of it and one that will run slightly higher PR.

All,

with regard to the exhausts, they are now being made, and I have no problem in getting more made. I also do not believe that they will be noisy on a car with headers.... both the development cars (EJ20 and EJ22) were running headers and the exhausts were really quiet. My issue at the moment is that I later fitted one of the development ones to my Legacy Estate, and it's somewhat louder on that, probably same sort of level as the Scoobysport or Hayward & Scott 2.5" item, but deeper sound and quieter on boost. Now I don't know if this is to do with the stock headers, the fact that the Legacy Estate is a big open car (people generally complain of noise more in 5 door cars), whether it's downpipe related, whether it's turbo related (most of the early TD05 or VF12 cars are just plain loud anyway) etc. I am therefore uncomfortable selling them on the premise that they will be "quiet" when they may not be much quieter, depending on the remainder of the exhaust system.

I have noticed an improvement in flow and in turbo response when going from the 2.5" to a 3" system (with the addition of a splitter). My next step was to get it onto a 4 door car with stock headers, to get an idea about how loud (or not) it will be on that. I believe that I have a volunteer lined up for that, fortunately the car also has a Link ECU I should be able to alter fuelling if required. I also have one going on a car with HKS headers, so it will be very interesting to compare

Cheers,

Pat.

 

Pat, I am currently skulking around looking for a set of headers for my UK turbo,

I'm currently running a prodrive box, straight through centre, scoobysport down pipe, h&s up pipe and headers are standard cast items with a little grinding work done.
turbo is standard TD04
running link ECU
any use for you in "testing"

where in the country are you ?

{cheeky}
maybe some sort of deal on headers I want as mentioned above
{/cheeky}

Andy

 

pat is near herts..

 

theres been a huge wave of really really cheap headers floating around the US in the past few months. Borla recently released one and its selling for $397us shipped and they are offering international shippin for a few more dollars. Who knows how well it performs but it is damn cheap!

http://forums.nasioc.com/forums/showthread.php?s=&threadid=318810

 

Those Borla headers look quite horrible.

The Gruppe-S ones which are similar to about half a dozen others, seem much better for about $500US

 

Could always retain ECU control of boost and fit an aftermarket controller for the progressive recirc control. Or JB could make one like once before ?

Using a MBC to test the theory will suffice in the short term, then we can look at a tarted up 'bells n whistles' option

[Edited by Andy.F - 3/4/2003 2:31:12 PM]