New Turbo for project GSi

[exist3nce]

I'd like to revert back to allowing the Nubira ECU to run a set of Nubira injectors for situations where I'm not in boost.

Currently the big injectors that are in there right now are causing the car to be very rich at idle... which fouls the plugs quite quickly.

I think I may pull out the Greddy E-manage and drop in a Megasquirt.

Then it will be a hybrid.... the Nubira ECU will still control basic functions when the engine is making less than 0psi of boost.

Sounds alot like my setup, it works quite well!

[greenbluewoo]

what are you doing with the old manifold

[PrecisionBoost]

Good question.... well.... I don't know yet.... odds are I will have two or three manifolds left over ( still have a cast manifold kicking around )

Last night Tom and I took the car apart and checked the fitment of the turbo ... looks good but the offset S14 manifold didn't work all that well for the position of exhaust (too close to bellhousing)

Tom and I fit the old manifold back onto the car and to our surprize the new turbo acutally sits pretty decently if the flange is angled a bit.

Turns out there were a number of cracks in the old "XS POWER" manifold so I finished fixing that allready this morning

So here is the car...



Even though the old manifold works decently I thought I might as well screw around with the S14 manifold since I allready cut it apart.

Oddly enough I shifted the runners over one to the left and it looked pretty good.

I figured that if I cut the farthest left pipe it could be used for the wastegate.... then I could insert that pipe over on the right hand side.



I did some quick positioning estimates and it looked good so I figured what the hell.... might as well try it and see.



I lightly tack welded the manifold onto the flange, moved the pipe to the right and drew where I needed to make the hole for the new pipe location



Then I broke the tack welds and drilled out as much of the material as possible.

Started with an 11/64" bit and then used a step drill to increase each hole size

Lastly I cut inbetween the holes with a dremel tool to remove most of the bulk of material



Next I finished off the hole with a die grinder and carbide bit, tacked the manifold back on the flange, inserted the new tube into the manifold and welded it all together.



That took me about an hour and a half to get to that point.... a good portion of that was trying to get the exact right position for the new runner and about 1/2 hr of welding

Oh... forgot to mention one important step.... at one point before tacking the manifold for fitment I heated up two of the pipes red hot and separated them slightly to get the alignment right.

[PrecisionBoost]

The next "issue" is that the ports are really large ( even larger than my ported head ).... that would be perfect if you could move the pipe inside the flange and weld it but the odd shape of it makes it hard to do that.

Needless to say the ports require a 2" pipe be used just to cover the circumference of the port.... and since 99.5% of all headers are 1.5" to 1.6" in diameter there are giant holes all around the pipe.

So.... I need to take some 2" pipe and cut a wedge shape out of the side.... then crush the sides together and weld it back together to make a cone shape to go from 2" down to 1.5"

Unfortunatly I don't have any 2" pipe here and it's a holiday monday for us up here ( thanksgiving in Canada ) so the supply shops are all closed.

So... I'm not sure if I will be able to try out this S14 manifold.... I will probably just finish the other one and make it work so I can get the car back togther by the end of the weekend.

Who knows... I might find something somewhere in the shop that will work.

Chris

[gse_turbo]

hey chris, it's a bit late now but here is link to what I posted a while back on migweb. CLICK HERE

...there are some decent pics on there of how I used my nissan header.

[PrecisionBoost]

I ended up completly redoing the old manifold..... as the turbo had to be in one very specific spot ( +/- 1/4" in three dimensions ) in the engine bay thanks to my giant pain in the ass 4" inlet.... good thing I will be needing the Anti-surge machining ( which only comes in the 4" inlet )

The 45 deg Silicone hose alone was almost $70 locally ( that's with my discount )

Anyways I have some pics but I'm tired right now so I will upload them tommorow.

I basicly had to cut off the flange move it down and reposition the angles.... which sounds easy but it wasn't.

The old manifold was set up with a very large collector... because it fit both T3 and T4 (dual flange).

So I had to weld in individual pieces of stainless plate at various angles so that it would clear the bolt holes for the T3 flange and funnel the walls to the flange correctly.

Needless to say it's finished now.... but it took me a good 4 hours of cutting, shapeing, bending, welding to get it to where it is now.

It's too bad the local supplier was closed today.... it would have been far quicker for me to finish off that offset S14 manifold I converted to a "non offset" design.

I swear I went through about 30 feet of stainless rod !!!

Probably added a good pound of weight... but at least it's heavy duty and the ports match reasonably well.

[PrecisionBoost]

well... progress is slow.... mostly due to the fact that I didn't like the wastegate setup.... so yet again I cut up the manifold and modified it.

I think I finally have it to where I want it..... so the car should be up and running by the end of the weekend if all goes well.

here I'm chopping up the manifold yet again...



here is the manifold getting it's position checked.... I simply tack welded a chunk of stainless to hold the T3 flange in relation to the exit point of the manifold runners.



here is what the manifold looked like after I finished relocating the wastegate port out front.

Orginally I had way too much length to the wastegate and the position was less than ideal.... so I was worried about boost spikes (since the wastegate was too far from the manifold ) and I was worried about the position since it wasn't as "vertical" as I wanted it to be..... in fact it was at about a 30 deg angle from the horizontal.... so practically sideways.... still would have worked that way but the "prefered" position is vertical or as near to vertical as possible

I also enlarged the port area going to the wastegate by about 40%.... giving it roughly equal flow capabilities as the turbine.

[PrecisionBoost]

Here is the thermostat..... I drilled and tapped a hole inside for the coolant feed to the turbo.

It ended up almost every fitting I tried was too big for this location ( due to the shape of casting ) so in the end I made my own out of a stainless 10mm bolt.

Basicly I drilled out a 1/4" center hole and machined the head into a neck to grab the hose ( which has a 9.5mm ID )

I also had to remove some of the aluminum casting near the fitting to allow the hose to fit all the way down.

In the end it should work very nicely.

[PrecisionBoost]

This is the final fitment..... the synapse wastegate fits really nicely.... the hood goes upwards about an inch prior to the wastegate and there is lots of clearance on top.... so even if the engine manages to rotate 30 degrees it will still clear everything ( looks tight on the picture but it's really not that close... it's just the camera angle and the fact that the radiator extends outwards about 3" away from the wastegate )





So at this point it's primarily a matter of hooking up the coolant and oil lines.... I still need to make a 0.040" restrictor for the oil feed but that shouldn't take long.

The giant 4" diam by 10.5" long intake filter location may give me some problems.... originally I was going to relocate the battery.... but after going through all the NHRA rules I'm not sure if I feel like doing that right now.

I will definitly get her back on the road before snow hits....... and I'm keeping the car at my shop this year.... so I should get quite a bit done this year.... especially if Tom helps out ( thanks Tom.... you definitly helped with those situations where I needed 3 or 4 hands to get the material in just the right spot when tacking things together)

[daewoomofo]

Basicly I drilled out a 1/4" center hole and machined the head into a neck to grab the hose ( which has a 9.5mm ID )

I also had to remove some of the aluminum casting near the fitting to allow the hose to fit all the way down.

In the end it should work very nicely.

now thats being enginuitive, very nice work!

[gse_turbo]

Basicly I drilled out a 1/4" center hole and machined the head into a neck to grab the hose ( which has a 9.5mm ID )

I also had to remove some of the aluminum casting near the fitting to allow the hose to fit all the way down.

In the end it should work very nicely.

now thats being enginuitive, very nice work!

enginuitive...

here is where the water supply for the turbo should be tapped, there is more pressure here than the back side o fthe head where it was before.

I tapped two 1/4" pipe thread holes, both spots are just as good but I like the one by the bolt hole becase there is more material fo rthe threads.

Garrett

hahah just bustin your balls.

...looks like things are coming together nicely.

[PrecisionBoost]

hahah just bustin your balls.

Sorry man.... I don't read most of what you post ...... just joking

I used the same exact location as the Z20LET... since it's the natural choice.....the casting even has a circular dimple where your supposed to drill through.

I am suprised you managed to get a 1/4" pipe thread in there.... I was going to do that but figured I wouldn't be able to get it all the way down due to interference with the hex area of the fitting and the side of the casting.

Then I was going to use a 1/8" NPT but figured the hole would be a little on the small size..... so I settled on this design

For sealing the thread I didn't run the tap all the way through the hole.... so that the end of the threaded portion tapered like an NPT.... which made for a very nice snug fit

The 10mm bolt is actually quite long.... it goes all the way through the casting and sticks out the other side by 0.25mm

In the end it works.... but I think next time I do one I will give it a 3 deg angle to give more clearance to the casting.


Hey... I noticed you have like 500 posts on MIGWEB..... I had no idea you were that involved there..... I use to be on there all the time but as usual my business and family has sucked out most of my free time..... so now I only skim Daewootech once every few days and post here.

[PrecisionBoost]

Well it's finally on the road.... I had a death in the family ( close younger cousin ) so I haven't exactly had a pile of time to work on the car.

So a couple hours here a couple hours there and it's running again with the new Garrett GT3071R

I ended up making a pile of fittings and added a bunch of other things I didn't need to... just to make sure it works well

In a word..... wow..... this turbo and wastegate made a significant difference to the way the car pulls.

I haven't had much time to fool with it but at just 9.5PSI the car is way faster.... scary fast with heavy torque steer in all gears.

The new wastegate is definitely helping to build boost quick despite the fact that this turbo is larger than the T3 super 60

I also installed an oil pressure sensor into the fuel line (since I didn't have a fuel pressure sender and figured it should work the same )

I am getting dangerously high fuel pressure levels at high load ( 90 PSI ) which is what I feared.... it's even higher on the road than the 80PSI we saw on the dyno.

The Nubira ECU is still fighting hard for some reason.... even at that level the duty cycle is just 45% meaning the Nubira ECU is cutting fuel.

So I'm left wondering what to do now.... since the Greddy Emanage can only adjust the duty cycle coming from the Nubira ECU so much.

In theory if I can get the injector duty cycle up to where it supposed to be I can bring the pressure back to 3 bar.

My goal right now was just to get the car put together and on the road for some quick testing before winter hits.

Given most nights it's already below freezing it's just a matter of time before we see snow.

Over the winter I may completely redo the tuning system and go standalone.... but the more I get into this the more I'm thinking about preparing the white "polar express" car to take this engine.

I'm thinking of going completely standalone with the Polar express ( probably megasquirt ) and starting everything from scratch.

I will still keep the engine in the silver GSi but the idea is to complete everything ( wiring harness ) so that it's a simple matter of pulling the engine and dropping it into place on the white car.

Then I would transplant the C20GET from my other Passport Optima into the Silver GSI ( using C20GET ECU and Greddy components )

The Silver GSi would be more of a show car/ fun drive around car...... the white Polar Express would become an all out race car.

Personally I thing this would be the best mix and match since the C20GET is a little more streetable.... and it would allow me to make radical changes to the white car to allow for the useage of all that power.

I'm also tempted to look into the idea of installing a C20LET Coscast head ( one german supplier can supply a Coscast head good for 10,000 RPM )

[exist3nce]

Sorry to hear about the death, family comes before cars.

Sounds like its a fun car albeit hard to manage. Hows does the GT30 sound when its spooling up? I bet its sweeeeeeeet!


I do agree 90psi fuel pressure is on the high side......... do you know what is causing the stock ECU to fight the changes? Is the OEM O2 sensor hooked up with an unmodified signal to the ECU? I am wondering what is the ECU seeing that it is trying to change?

[PrecisionBoost]

I have an innovative motorsports LC-1 wideband

Basically the issue stems from the fuel maps stored under open loop operation.

The O2 sensor is completely ignored under full throttle acceleration.

The O2 sensor is used in Closed loop during partial throttle (regular driving)

So.... what I'm saying is that under the conditions at WOT the factory Nubira ECU is only giving a 40% to 45% duty cycle

The ECU is looking at throttle position, MAP sensor, IAT sensor and probably engine coolant temp sensor.

You can fool the ECU a bit by changing the MAP and IAT ( which the Greddy Emanage does ) but you can only go so far with those numbers

That is to say you have a maximum of 50% change available.... and that comes at a price.... the ignition timing is changed when the MAP and IAT are changed.

So yes.... I could go back and do further tuning to decrease the fuel pressure and fool with the Greddy unit but at maximum I'm still only going to see 60% to 65% duty cycle (with changes to ignition timing I don't want )

Basically the issue is the size of the factory injectors..... I'm guessing that they are a little larger than required.... thus at the maximum power that engine can make the duty cycle of the factory injectors would only be 40% to 45%

If instead of 21lb/hr injectors the Nubira system had been designed to use say 19lb/hr injectors then a higher duty cycle would be required.... probably closer to 50% or more to get the same amount of fuel.

So I hope you get my little bit of my point..... I can't force the Nubira ECU to get up to 75% duty cycle no matter what I do.... my only choice is to use a secondary injection system, a different piggyback with more capabilities or a standalone system.

Tom and I were talking and now I'm contemplating the idea of using jets like nitrous controlled by the Greddy Emanage and a rising rate fuel pressure regulator.

I'd get a dual port system.... feed fuel in one and water/methanol into the other.... then inject them into the runners of the intake.

The idea being to replace the 48lb/hr injectors with the stock nubira injectors to get functionality at low rpm, idle and partial throttle.

Then as boost comes on a solinoid trips and fuel starts being injected (with water and methanol ) into the intake system.

I can preset the RRFPR quite low if I want... 10 PSI if I want.... and then increase to 100 PSI if I need to.

I can then use the Greddy to do fine trim adjustments on the factory injectors ( which would have the factory regulator.... not a rising rate unit )

I was going to go with 8 injectors.... and I still might.... I haven't decided.

I could also use a megasquirt to control the fueling.... but if I'm going to do that I might as well rip out the Greddy unit.

Too many choices..... all I know is that the factory ECU is a poor match to the factory injectors and that is the cause of my problems.

If the factory ECU managed to get up to 60% duty cycle ( which would require smaller factory injectors) then at least I'd have something to work with.

Who knows.... perhaps I can figure out a way to design or buy a simple latch system to extend the duty cycle without modifying the MAP or IAT signals.

The idea being that when the factory ECU cuts the signal to the injectors you simply delay that signal slightly.... increasing duty cycle.

That could be done via a WOT switch and a latch system that extends duty cycle based upon boost or RPM.