STI TMIC Testing Discussion

Summary:

 

I'm testing my 08+ STI TMIC to characterize the thermal efficiency and pressure drop. Will use these results to determine what improvements are needed, if any. See mod list for details of my setup.

 

 

Equipment:

 

Logging

- AEM AQ1 Datalogger

- Tactrix 2.0 w/ Romraider Logger

 

Sensors

- Setra 230 5PSI Differential Pressure Sensor

- Omega Pt100 Class B RTDs (2x)

- Stock STI MAF/IAT

- Stock STI MAP

- Stock Ambient Temp Sensor

 

 

Method:

 

1. Use Setra to monitor ΔPintake for given MAF g/s. This will give me an idea of what the pressure is at the compressor inlet.

 

2. Log ΔPIC, T1, T2, Tambient, and MAP via AQ-1 and RPM, MAF, IAT, MAP via Romraider.

 

3. Combine datalogs and run through spreadsheet to get compressor efficiency, intercooler efficiency, and IC pressure drop.

 

 

Configuration:

 

For ΔPintake testing the Setra high-pressure port will be plumbed into the Cobb airbox. The low-pressure port will be plumbed to a nipple on my Perrin turbo inlet (where the EVAP used to connect).

 

For the IC testing, RPM, MAF, IAT, and MAP all feed direct to the ECU already.

 

To log the MAP via the AQ-1 I will tap into the signal wire via an extension harness Ballenger Motorsports made to my specs. Logging MAP on the AQ-1 and via Romraider will give me the ability to overlay the logs.

 

To log Tambient via AQ-1 I will simply tap into the signal wire for the ambient temp sensor located in front of the A/C condenser.

 

The Setra high-pressure port will be plumbed into the hose coming off turbo, going to the GS EBCS. The low-pressure port is TBD (see below).

 

The RTDs will be potted into 1/8NPT plugs. T1 will be tapped directly into the Y-pipe, before the split. The location of T2 is TBD (see below).

 

 

*** So at this point the only thing I still need to figure out is where I want to plumb in the Setra low-pressure port and T2.

 

My initial thought was to use two meth nozzle adapters (like this one) in the TMIC/TB coupler.

 

I also like the ATP quick-tap, as the E-clip and attached hose would prevent the fitting from falling into the engine.

 

Then I thought about the possibility of using a post-TB meth injection plate, like this one from Snow Performance.

 

I have also considered tapping directly into the TMIC end tanks, or having bungs welded in.

 

All of these options have pros and cons. The modified coupler (with either type of adapter) and the injection plate are both easily 100% reversible and similar price points. Modifying the TMIC is not totally reversible - if I plug the holes they will always be potential leak points - but it is free if there is enough wall thickness to tap directly. It seems like many people on NASIOC tap their meth nozzles right into the TMIC end tank, but I'm skeptical there is enough metal. I also see a bunch of people having bungs welded in (usually for the 3/8NPT GM IAT sensor, which I am also testing), but I don't want to sink money into the TMIC that may make it harder to sell later.

 

Anyways, just trying to get my thought process down so I can free up some brain space. I'm still playing around with the AQ-1 configuration and some bench testing.

 

 

 

 

Ok, getting all that out reminded me that I'd rather not tap the IC end tank. If I end up swapping ICs then I want to duplicate the testing without tapping the new IC.

 

That leaves me with the nozzle adapters in a silicone TB coupler, or the meth injection plate. Both options are in the same ballpark price-wise and will have the same fitment challenges (moving the TMIC towards the firewall).

 

Snow Performance actually offers a nozzle adapter where the female NPT threads are on the piece that goes inside the coupler, which means with a sensor or fitting installed it can't fall into the intake manifold. This alleviates my concern of a catastrophic failure. USRT offers a similar adapter made from aluminum, at a competitive price.

 

Otherwise I'm just debating the usefulness of the meth injection plate in case I decide to run meth in the future. That being said I can't find anyone injecting post-TB, except for port-injection systems. Not trying to reinvent the wheel, but wondering if it's simply a matter of convenience that I can't find any mention of someone using a TB injector plate (most people tap right next to IC outlet).

 

I also went back and found this thread from NASIOC that I had stumbled upon years ago, which probably planted the seed in my brain to try the STI TMIC before determining if an upgrade was needed.

Post TB has the issue noted on the plate page: "*Note- For safety, a solenoid (part #40060) must be used when injecting post throttle body to prevent vacuum siphoning."

 

If you used the plate for a sensor you'd have the same problem as running an IAT sensor in the IM if it's aluminum (like an STI mani), the sensor will get heat soaked. That's why it's generally best to put the sensors in the silicone couplers or very thin FMIC aluminum pipe.

Thanks for the feedback.

 

Regarding heat soaking of the sensor, that's something I have considered with the sensor selection, but a valid point for sure. I chose the exposed element thin film RTD in part because it has an exceptionally low dissipation constant (and therefore fast response). It's basically a 2mm x 2mm element at the end of PFA-insulated flying leads, encased in epoxy. This will get potted into a 1/8NPT plug, and I can control how far into the air stream (and away from the metal NPT plug) the element ends up.

 

All else being equal, at this point I'm leaning towards bungs in the TB coupler. I had started to fixate on the solution before characterizing the hardware I already have; a habit I fall into frequently. Once I stopped worrying about where/if I would inject meth the coupler location started to make more sense than the TB injection plate.

 

Next step is wiring up my GM IAT to the AQ-1. I want to see just how quick this thing responds, and what temp it reads up to. It is less expensive than the RTDs (which won't arrive for a couple weeks at least) and if the response time looks good I may just buy another and cancel the RTD order.

Is there a place where you could use something like this joiner?

 

http://www.anfittingsdirect.com/exhaust-pipe/turbo-piping/iat-installation-p-1231.html

Unfortunately not with the stock TMIC & y-pipe.

Subbed, im a geek for tech related posts....

 

 

Dave

Hit a few snags yesterday. The AQ-1 stopped recognizing the SD card suddenly, and all attempts to correct the issue have failed, rendering the AQ-1 inoperable. AEM wants me to send it in for service, so we'll see what develops.

 

It failed while I was attempting to datalog the GM fast response IAT sensor (thermistor style) against a bare element 100 ohm RTD. As soon as I hooked up the RTD I found that the temperature output was jumping around +/- 10F; It seems that the AQ-1 is not very sensitive to small resistance changes. The GM IAT sensor signal is very stable, and at first blush looks like it will be sufficiently fast to respond when air speeds are high; by my estimate the air going through the TMIC inlet and outlet will range from ~60-420 ft/s over the engine's operating range.

 

The good news is that the GM IAT is very inexpensive and easy to use. The bad news is that its listed operating range is up to 135C (calibration curve goes up to 150C), so I'm a little wary of putting one directly post-turbo.

AEM has offered to fix the AQ-1 if I ship it to them on my dime. Frustrated with the expectation that I should pay additional money to have a defective item repaired coupled with the clumsiness of the interface, poor sensitivity for the RTD I've selected, and blatantly contradictory statements between the manual and tech. support, has me stuck with whether to pursue repair or return for a refund and find another datalogger. Either way I have to eat the cost of shipping.

I'm curious why you didn't use the AEM IAT sensor with the AQ-1?

 

30-2014 - Air Temp Sensor Kit

Accuracy: < 1.5°C / 2.7°F

Operating Temp: -20C to 150C / -4F to 302F

Resistive Range (Ω): See Instructions

Thermal Time Constant: < 15 sec. The time required for the sensor to achieve 63.2% of its steady state value when subjected to a step change in ambient temperature [Tc=(Tf-Ti)*63.2%+Ti]. Test medium: silicone oil

Sensor Body Material: Brass

Thread: 1/8" NPT

Hex Size: 1/2"

Installation Torque: 13.5Nm / 10 ft-lbs

Weight: 10g

Electrical Termination: Weatherproof High Performance DTM-Style connector. Includes Mating Connector, Pins & Wire Seals

Includes: Air Temp Sensor High Performance DTM-Style, 1/8" NPT Aluminum Bung, Mating Connector, Pins & Wire Seals

Considered it but essentialy identical to GM IAT at a far higher price point. Only real difference is 1/8 vs 3/8 NPT threads. The temp limit is symbollic, both sensors have calibration curves up to 150C.

As far as using a different data logger: I had the AIM MXL Pista Dash Logger in the race wagon so am familiar with them now. They also offer standalone loggers (the EVO3, EVO4, EVO4S, and the EVO5). I've tried finding ones that do everything the AIM loggers can do but have yet to beat their total package. If you buy their own sensors though they are very expensive, but pretty much plug & play.