The Quarky Engineer

The engine is looking good. Precision 5858s, ID1050Xs primary, ID1300s secondary. English Racing doing kick ass work as usual.

With just a little bit of time in the garage this afternoon, I did manage to get the vacuum/boost block assembled and a bracket made. Simon Ko stopped by to pick up an intake for Tony Ng, so I had a chance to see his awesome GTR. 315s on all 4 corners! It is a beautiful wide body with a trick mid bumper output exhaust setup.

The weekend gets better when a box arrives from these guys! (Thanks Tony Palo)

I really dislike vacuum/boost sense hoses with Ts in them, especially when you have multiple a in row. I am guilty of that travesty on my GTO. In an effort to redeem myself, I got a manifold vacuum/boost block, which I will use AN fittings on to connect to all the things that need references. I even made a little bracket to hold it.

Of course I looked and while I have some NPT to 4AN fittings, they are blue, and that just isn’t going to fly… Thanks Amazon for having them Prime.

Temp sensors ready to go. They look ok. Myles Kerr I even put them in little bags so they feel official. 😉 Oh an each is labeled for the location it goes it.

I also made an exposed tip one to try out back to back (actually in series) with an epoxy one, so we can see the difference in response time.

Time to update the calibration in my 06 TT GTO. Since I changed the fuel system to a front regulator return-style with a manifold referenced regulator I need to tweak a few things. I still have the Deka 60lb injectors I started with back in 07. This PCM (E40) has a few interesting internal limits. One of them is the internal scaling of the injector flow table. It does not allow entries above 63.5 lb/hr, even with the use of the additional multiplication maps. As a result you have to scale the injectors a different way. Back when I first did the turbo setup in 2007 the common way was to scale all of the airflow tables and effectively cut them in half. (Note I’m running the HPTuners SD OS with no MAF).

Fortunately someone noticed that you could just change the ‘Stoich AFR’ table by doubling all of those values and then cutting the injector flow tables in half. That gives you an injector flow range up to 127lb/hr. Thanks to Dave Steck for pointing out this idea many times on many forums.

The car stock does not reference the fuel pressure to manifold pressure, so the flow table is not a single value but a table indexed by manifold vacuum. Since the fuel pressure is now referenced by manifold pressure this table is flat.

The injector offset by voltage is a 3D table in the GM world, indexed by both voltage and manifold vacuum. I had a good reference table for these injectors that I need to dig up, as this one is flat across VAC. I need to think a little about this table as a result of the change to a manifold referenced fuel pressure system. It would seem that since differental pressure across the injector is now constant that this table should be flat across VAC.

Most tests of the 60lb injectors seem to show around 62-63 lb/hr(660cc/min) at 43.5psi, and 72-73lb/min (760cc/min) at 58psi (the default GM fuel pressure on this car), thus I have the IFR set to 72.6/2 = 36.3.

My small offset table is almost certainly wrong. I have a graph of flow percentage difference I need to use to make a better table.

What all this really means is that I need to put some ID1050Xs in this car. 8x 1050Xs at 58psi base will be plenty of fuel.

There is still a second limit, the airflow limit, that I need to work around. As it sits now once you are in boost you are in the last row of a number of tables including the ignition timing. This can be fixed by doing a rescale of all of the airflow related tables to that airflow in the ECU represents say 1/2 of the real airflow. A hack, but a hack that works.

Does this scream out ‘Fantastic evening on the way!’? Yea, I thought so. Lots-o-DTMs plus lots-o-sensors and lots-o-MIL-25579/16 wire equals kick ass Motec harness!

I pulled out my ID2000s from my 08STI after 5 years of use. Almost every tank of gas over that period has either been E85 or E98, with more E98 in the last year. They still function great, but I suspect they could use a good cleaning. The upper o-rings that go into the fuel rail leak a little bit on one of them, I assume from continuous ethanol exposure and heat. The tips look pretty dirty, although I didn’t notice any difference in idle quality. This really has been a pretty worse case use, as I would typically run the car on E98, drive it hard, park it and let it sit there for weeks at a time.

Paul Yaw can probably tell me when I bought them (SN 61636, probably sold through Cobb). Those little hats have small filters in them as well. Can those filters be changed Paul/Tony Palo?

I know the ID1700Xs and ID1050Xs are the stars of the show now, but I like these little guys. They have been flawless for 5 years of shenanigans. I think I’ll see about getting them cleaned and put back in. Or perhaps I should just get a set of 1700Xs. Hmm.

Speaking of sensors, I need lots of air temp sensors in my GTR setup. I could buy them, but I thought it would be more fun to assembly some. I really like the glass silicon diode based NTC sensors because the have very small thermal mass and thus very fast response time.

Back in the day I used a few Autronic air temp sensors that relied on the KTY/83 glass silicon sensor, and I have made a few using that same sensor. That part is now deprecated, but there are lots of similar sensors. For this particular application I picked the US Sensor 103JG1F, which is a 10k@25C J curve NTC sensor. They are $1 each, so cheap. The wire is M25579/16, 20 gauge. I should have used 22.

To make them usable I put them into a 1/8 NPT brass fitting and used epoxy to seal them up. 99.999% of you guys should just buy the air temp sensors from Tony Palo at T1, as he does a much better job, and uses a kick ass 1/16 NPT stainless steel fitting.

I’ll shrink wrap the wire side, put a DTM connector on each one and be good to go. Most of them have the temp sensor covered in epoxy, but I did make a couple with exposed tips. I’ll put one of each in my Subaru test rig and post up data comparing how they react. In my limited testing previously the epoxy made very little difference.

It is amazing how fast these respond to temperature changes compared to the standard GM AIT sensor (which is a ceramic NTC element). I’ll post up comparison data once I get my Subaru back on the road.

Good things coming to an R35 GTR near me. 😉