Jump to content

MAF Translator Gen II install on a 89 Conquest TSI


kf4eok
 Share

Recommended Posts

Here is some notes of a install of a MAF translator Gen II I did.

This is how I installed my mass airflow translator generation 2 on a 89 conquest TSI or starion. First I removed the kick panel on the passenger side and unpluged the two connectors use from the ECU. I have included pictures of all the connectors involved with this install. You will need to run a new 14 to 18gauge wire from the distributor harness plug wire-to-wire on the hardware side to the inside of the passenger compartment. I ran mine to be ECU location. I have a wideband O2 sensor I will be using during this install. I will also use a 3" GM MAF sensor and a GM map or barometer sensor and a IAT sensor.

 

wiring from the translator:

1. connect the pink wire to the smaller ECU connector (A36) pin 1 which should be a red wire

2. connecting black wire to connector pin 2

3. connect the blue wire to the larger ECU connector (A37) pin 15

4. connect the orange wire to the 0-5 V output wire from your wideband controller

5. connect the purple wire to connector A36 pin 2

6. connect the white wire to the wire you run from the distributor connector

7. connect the green wire to connector is A37 pin 2 in this step you will physically cut the white wire with black stripe and connect the green wire to the wire from the connector to the ECU.

8. connect the yellow wire to the wire you just cut in the previous step that runs to the MAF sensor

9. brown wire not used

10. grey wire not used

wire of the old MAF plug to the new barometer, IAT and MAF:

cut the wires one at a time because there is 2 blk-wht wires and 2 grn-red wires (look at tht pictures)

1. take the red wire going to the old MAF sensor and cut it and hook it to the 12 V in of the GM's mass airflow sensor

2. take wire-to-wire with black stripe and cut it and hook it to the output wire from the GM's mass airflow sensor

3. hook the Gm MAF sensor ground wire to sensor ground wire fron the old MAF sensor wiring harnes

4. Hook the IAT sensor to the sensor ground wire from the okd harness

5. hook the IAT other wire to the green- red wire going to the ECU pin 5

6. hook the MAP or barometer sensor output wire to the blk-wht wire from the old wiring harness

7. hook the MAP or barometer sensor ground wire to the sensor ground wire from the old wiring harness

8. hook the MAP or barometer sensor 5 volt supply to the grn-red sensor power wire from the old wiring harness

doing this will return all the sensor inputs that was lost in removing the orignal MAF sensor. Because the old MAF has the baro. and IAT sensor built into it. It is a good idea to install the new IAT in the intake or just before the throtle body, or anywhere after the intercooler. This way you will be gatting a truer temp reading of the incoming air to the engine.

http://thestarquestclub.com/images/ecu_connector_a36.JPG

 

http://thestarquestclub.com/images/ecu_connector_a37.JPG

 

 

http://thestarquestclub.com/images/maf_connector_a67.JPG

I hope this will help you during your install.

I set the V1 and V2 to setpoint on thg Gen 2 Translator

I am still working on ALL the other settings

Let me know if this helps you or if you hane any questions.

http://handymandan.us

Link to comment
Share on other sites

  • 2 weeks later...

Hey, great write-up. This is exactly what I was thinking of doing with my MAFT Gen2--using actual sensors instead of just feeding the ECU fixed (wrong) values from the Gen2's aux outputs. I think this is the right way to do it. Edit: Turns out I was wrong, see post #8.

 

I'm assuming you used the standard GM 1-bar MAP and IAT sensors? There are some specs in the FSM regarding what the stock Mitsu IAT and baro sensors should output at certain temperature/altitude. Have you done some research to verify that the GM sensors have compatible outputs? That's something I still need to look into.

Edited by obsolete
Link to comment
Share on other sites

Update: I did some research on Mitsu & GM sensors and here's what I found.

 

According to my '87 FSM, here are the specs for the Mitsu baro sensor:

101 kPa (14.7 psi) = 4 V (sea level)

88 kPa (12.8 psi) = 3.5 V (~3800 feet above sea level)

 

According to my '87 FSM, here are the specs for the Mitsu IAT sensor:

20 C (68 F) = 2.7 kΩ

Unfortunately, they only specify that one point.

 

According to the internet, here are the specs for a GM baro (or 1-bar MAP) sensor:

100 Kpa = 4.9V

90 Kpa = 4.4V

80 Kpa = 3.8V

etc.

 

Sources:

http://www.diy-efi.o...fo/sensors.html

http://www.robiether...m/mapsensor.htm

 

According to the internet, here are the specs for a GM IAT sensor:

20 C (68 F) = 3.457 kΩ

25 C (77 F) = 2.752 kΩ

30 C (86 F) = 2.205 kΩ

 

Sources:

 

http://www.diy-efi.o...fo/sensors.html

http://www.bmotorspo...products_id/400

 

As you can see, neither GM sensor matches the Mitsu specs.

 

Good news if you like to solder things (which I do), here's an absolute pressure sensor that appears to match Mitsu's spec: MPXA6115A, MPXH6115A Series $9.66 at Digi-Key. Still looking for an IAT that will work--maybe just gutting a junk/spare Mitsu MAF is the best way to get one. In fact, I have a spare MAF around here--maybe I can do some testing with it and get enough data to find an aftermarket IAT replacement.

Link to comment
Share on other sites

Okay, I kept researching, and learned that what I was planning on doing (and what it looks like Dan is planning on as well) is wrong. You can't use anything but the fixed IAT and baro setpoints that the MAFT outputs to the Mitsu ECU if you want it to work "correctly". This has to do with the fundamental difference in how the Mitsu MAF and GM MAF measure air, and knowing the relationships between temperature, pressure, mass, and volume.

 

http://i48.tinypic.com/2nu0heg.jpg

 

The Mitsu MAF isn't really a mass airflow sensor. It doesn't measure the mass of the air flowing through it. What it does is measure the volume of air flowing through it. A given volume of air can have a different mass (and a proportionally different amount of oxygen) depending on its temperature and pressure. As most of us already know, a given volume of cold air at low altitude has more mass, and contains more oxygen than the same volume of warm air at high altitude. Since the amount of fuel the ECU commands from the injectors must directly correspond to the amount of oxygen in the air, the ECU needs to know the mass of the air, not just the volume. So, to calculate the mass, it needs three things: the volume, the pressure, and the temperature. That's why a Mitsu MAF is actually 3 sensors in one: a Karman vortex sensor delivers the volume, a baro sensor delivers the pressure, and an IAT sensor delivers the temperature.

 

The GM MAF actually measures mass airflow. As the density and temperature of outside air change, the mass airflow reading of the GM MAF changes proportionally. Great technology, right? Just input that to our ECU and we're done. Well, not exactly. Our ECU still expects three parameters from which to calculate mass airflow. So, in order for it to work, we need to give it all three: volume airflow, baro and IAT. The problem is, we no longer have a volume airflow reading, we have a mass airflow reading. So, by setting the baro and IAT readings to known constants, we can effectively eliminate them from the airflow calculation. That's the simplest way to do it, and that's how the MAFT works. If instead, you gave the ECU the real IAT and baro from actual sensors, it would double-compensate, which would make the car hard to tune consistently and very sensitive to weather and altitude changes.

 

The airflow calculation with a MAFT, as many of you have experienced, is very accurate, but the unfortunate side effect is that your ECU now thinks you're always at sea level at 80F. So, anything else that the ECU uses temperature and pressure for, besides just airflow calculation, is going to be wrong. What are those things? Probably not much on a simple old computer like ours. The only thing I could find in the FSM is that above 3,937ft altitude, the computer advances ignition timing by ~5 deg. That doesn't really mean much to those of us who spend most of our time east of the Rockies.

 

Hope this was helpful.

Edited by obsolete
Link to comment
Share on other sites

Okay, I kept researching, and learned that what I was planning on doing (and what it looks like Dan is planning on as well) is wrong. You can't use anything but the fixed IAT and baro setpoints that the MAFT outputs to the Mitsu ECU if you want it to work "correctly". This has to do with the fundamental difference in how the Mitsu MAF and GM MAF measure air, and knowing the relationships between temperature, pressure, mass, and volume.http://i48.tinypic.com/2nu0heg.jpgThe Mitsu MAF isn't really a mass airflow sensor. It doesn't measure the mass of the air flowing through it. What it does is measure the volume of air flowing through it. A given volume of air can have a different mass (and a proportionally different amount of oxygen) depending on its temperature and pressure. As most of us already know, a given volume of cold air at low altitude has more mass, and contains more oxygen than the same volume of warm air at high altitude. Since the amount of fuel the ECU commands from the injectors must directly correspond to the amount of oxygen in the air, the ECU needs to know the mass of the air, not just the volume. So, to calculate the mass, it needs three things: the volume, the pressure, and the temperature. That's why a Mitsu MAF is actually 3 sensors in one: a Karman vortex sensor delivers the volume, a baro sensor delivers the pressure, and an IAT sensor delivers the temperature.The GM MAF actually measures mass airflow. As the density and temperature of outside air change, the mass airflow reading of the GM MAF changes proportionally. Great technology, right? Just input that to our ECU and we're done. Well, not exactly. Our ECU still expects three parameters from which to calculate mass airflow. So, in order for it to work, we need to give it all three: volume airflow, baro and IAT. The problem is, we no longer have a volume airflow reading, we have a mass airflow reading. So, by setting the baro and IAT readings to known constants, we can effectively eliminate them from the airflow calculation. That's the simplest way to do it, and that's how the MAFT works. If instead, you gave the ECU the real IAT and baro from actual sensors, it would double-compensate, which would make the car hard to tune consistently and very sensitive to weather and altitude changes.The airflow calculation with a MAFT, as many of you have experienced, is very accurate, but the unfortunate side effect is that your ECU now thinks you're always at sea level at 80F. So, anything else that the ECU uses temperature and pressure for, besides just airflow calculation, is going to be wrong. What are those things? Probably not much on a simple old computer like ours. The only thing I could find in the FSM is that above 3,937ft altitude, the computer advances ignition timing by ~5 deg. That doesn't really mean much to those of us who spend most of our time east of the Rockies.Hope this was helpful.
If you notice I did not use (did not hook up to ECU) the grey and brown wire from the Gen II. I did however hooked up a BARO. Sensor (I used a Toyota vapor sensor) back to the ECU so that the original ECU would do its normal calculations with the Baro. Input. As for the IAT sensor I did use a GM open air IAT sensor and wired it to the orignal input to the ECU so that the Original ECU can do its original calculations with the original IAT input. The Gen II takes the reading from the GM MAF sensor and converts (translates) it to (in our case) DSM signal so that the orignal ECU sees a signal that looks like the original MAF signal.Whith that being said, the original Baro sensor was built into the original MAF sensor and it was installed WHERE? inside of the air filter. I could be wrong on this but I do not think so, the barometer reads atmospheric pressure pretty accuracy right. Well what happens inside of the air filter when you start up the airflow through the filter, the pressure will drop inside the air filter and the sensor will read a inaccurate reading, even worse when the air filter is dirty. I will have to say that the original Baro. should have never been located there. Yes there might be slight differences in manufactures sensor readings but I do not think that it makes much difference on these old ECU in our cars. How did the ECU comspate for a clogged or dirty air filter? It did not but the cars still run. If you put in a K&N air filter there would be a difference in the original Baro reading wouldn’t there, but the car still runs without any noticeable changes.The IAT is similar in the same regards. The IAT should not have been installed before the turbo, or the intercooler. We want the temp. of the air going through the intake to make A/F ratio calculations right. Do you think that the ECU known how much temp. rise the turbo is adding to the air temp, how about how much the intercooler is reducing the air temp at different outside air temps.. These old ECU systems are just that, old and outdated by today’s standards.I am using set points with the Gen II for the IAT and BARO. to match my location. Is this perfect? no but I would guess that it is pretty close compared to the old sensor readings with all of their design and location flaws. The way I hooked everything up the baro and ECU will calculate the altitude close to the original setup if not better and the IAT should have much more accurate readings if installed in the intake or close to the throttle body. Using set points with the Gen II and not using the grey and brown will not give double compensation.This is just my opion. Edited by kf4eok
Link to comment
Share on other sites

Using set points with the Gen II and not using the grey and brown will not give double compensation.This is just my opion.

 

Okay. I have no firsthand experience, but I think your tune is going to drift. Florida is pretty flat and temperate, so who knows, you might be fine. Please report back when you get it running and let us know how it goes.

Link to comment
Share on other sites

I set mine up on and rpm and tps trigger. Once those triggers are hit the Gen 2 adjust fuel from the settings I program for each 1k rpm.

 

My car ATM is 1800 miles away so I am of little help.

Link to comment
Share on other sites

Its my car, and working on it since its only been a week. Having a learning curve with the GEN2 but I'm getting there. Hope I can share with others some of their/my challenges or not with the GEN2 so hopefully the dialogue continues.

 

Daniel rebuilt the engine for me and his knowledge and efforts have been impressive. Mods are Chad's equal length tubular header, rebuilt M7 head Lshores Perf. (all the goodies). Trilogy injectors (delphi?), 20g td06h 8CM^ , 35mm tial external wgate, HKS compressor bypass, HKS blow off valve, HKS EVC, RRFPR. (all the other periphals, HD sbars/lowering springs/camber plates, HF exauhst dynomax muf, upgraded alt & rad etc...)

 

I've been tuning it as I go, but I need someone else to drive while I or others, well tune. Currently I'm close good Low load / Mid Load 12-14.7 just goes pig rich on WOT sometimes even when I raise the boost to 18-20lbs. My mainscale is 19%, and well I think I need to get that number as low as I can while continuing with the low/mid/high AFR % -/+ changes to acheive the desired results. WOT its connected to an AEM 5v Wideband so I would think that should hold AFR at my settings per the rpm range.

Edited by ptracy26
Link to comment
Share on other sites

There is no MAP sensor with this setup.

 

According to your install instructions you are.

 

Last sentence in first paragraph. Also, under wiring instructions 6,7 and 8.

 

 

Common GM map sensor.

 

http://cdn.maperformance.com/images/P/gm3bar.jpg

Needs a manifold source on that nipple.

Edited by BuGG
Link to comment
Share on other sites

Great, any input is better then none. Great work.

 

Ironside, can you post your settings for the GEN2 related to you TPS/RPM setup? It may be an effort, but figured I would ask. I'm curious to see the results. Currently I feel like I'm chasing a tune with the MAFT/RPM setup. (you can probably skip your 1k rpm settings if it makes it easier, thanks)

 

Also, after warmup and at Idle I would be curious to know what your Hz In/Out readings are at the sensor monitor page if available to share,

 

regards

Link to comment
Share on other sites

You may find some thing use full in here. http://www.starquest...ic=106923&st=40

 

On my neo afc i have my tps low throttle point at 22% and the high at 35%. You have 3 points to work with. It might help to have the mid point set richer than highest throttle point. You will likely find as the car come into boost you need to pull fuel. That is to say negative correction of about 15%-20% from your cruise setting.

 

What size injectors are in this car?

Edited by StarquestRescue
Link to comment
Share on other sites

Ptracy26, I would if I could but my car is 1800 miles away. Starquest rescue is very good with tuning. If I remember right set your main scale to negative 10 to start.

Good luck.

Link to comment
Share on other sites

Ironside, I thought you were both 1800 miles away therefore couldn't help locally, now I get it. lol

 

As an update I did switch to TPS/RPM and moved from the LT1 3.5 to LSI ALuminum MAF, and switch from "set point" to "MAF Volt" on the V1/V2. Its been two days driving to/from work and I must admit I think I found its sweet spot for cruise and low/mid throttle. She stays steady at stoich. Yesterday I had the mainscale at 00.00, but today I moved it to -8. Not much difference yet, but its a little premature at this point.

 

I will add that I did a pull turning my EVC up a little (mainscale at 00) and with the 20g flowing to almost 25lbs (i know too much), I was still pig rich 10.5 and <, almost chocked it will too much fuel (thats with a rebuilt/hiflow M7 head, s/b more cfm's then stock which should require more fuel). So at this point, I'm going to give a couple of days of normal driving to make sure low/mid throttle stays constant, then I'll start leaning out the High loads slightly to see if that helps under WOT/ to the point of the correct mixture. ( I have the GEN2 for AF Tracking WOT to the wideband, surprised that didn't lock in those settings under WOT)

 

StarquestRescue, I have Trilogy's first run of injectors offered, which I believe are the Delphi's 650/950 (maybe off a little) they have the Trilogy stickers on them, but only have about 2k miles with them in there.

Link to comment
Share on other sites

You have 3 points to work with. It might help to have the mid point set richer than highest throttle point. You will likely find as the car come into boost you need to pull fuel. That is to say negative correction of about 15%-20% from your cruise setting. (startquestrescue advice, didn't qoute correctly haha)

 

 

Understood, though I will add under slow/moderate pedal pressure (moving towards WOT) going into boost, with my car it starts to lean out (1-3lbs of boost ~) 15-16 range, then immediately corrects itself richening up to 11.7, then til full pedal pressure/full boost, 10.5<. Again I have all my low/mid/high points still at zero, except for the lowest rpms to assist in a "Tip In". Hence, now I can start tuning,

Edited by ptracy26
Link to comment
Share on other sites

You may find some thing use full in here. http://www.starquest...ic=106923&st=40

 

Read that one about 10x's in the last 3 months, ok i'm slow, and even tried his latest setup. The car was terrible to drive, so I had to immediately reset the GEN2 and start over with the MAF/RPM setup, with a mainscale at -25%, that showed some signs of improvement for me. But ultimately after a week or two, it seems I was always chasing my tail to get the mixture correct throughout the low/mid/high

 

It just maybe the issue with Double Compensation since I still have hooked up to the ECU the BAR/IAT (see Dan's reply's on setup), maybe why I'm seeing better consistent readings using the TPS/RPM & Volt MAF setup throughout low/mid, just rich under WOT.

Link to comment
Share on other sites

I would start with tps as the load source. It is scaled in #'s with 320= to 5 volts per the online instruction. Try and figure out what # = 1.5 volts and set that as your low tune point say 2.5 volt for mid and 3.5 for high.

 

Find a main scale setting that the car idles at. The set your low load correction at each rpm point up to about 3200-3500 so the car cruises around 14.7 Cruze is level road steady throttle position. Above that rpm set it the same or a little richer than the 3200-3500 point, richer at higher rpm. You will not really run at the higher rpm at 1.5 volts of tps any way.

 

Now work with the mid at high load settings. Forget part throttle boost for now, that's the hardest to get right. Pick a modest boost say 15-16 psi. If you start to make boost at say 2500 rpm and reach full 15-16 psi of boost at say 3500 you will have to pull fuel in this area. Example 0 correction at 2500 on the mid and high load point -7 at 3000 rpm and -15 at 3500 rpm. All ways start pig rich and lean it down slowly. Set your mid and high settings the same at each rpm point at this time. After you get it to run good at wot and get a better understanding of the unit, Later you can try and adjust the mid to improve part throttle.

 

It will likely run richer the more boost you run. (up until the point it runs out of injector and leans out, or fuel cuts.) That is why you need to run a consistent boost pressure and also why part throttle boost is hard to get right. Example If you tune the car to run wot at 15 psi, it may run richer at 18 and leaner at 10. At part throttle you may not be at full boost so it may run lean.

 

At this point i wound wire up the unit the way the manufacture intended. Using the set point voltages for baro and air temp. You can always try the baro and temp senders much later. And turn off all the advanced features like afr wide band tracking, Keep it simple till you get a handle on the basics. You will not achieve a the optimal tune in a week or 2.

Edited by StarquestRescue
Link to comment
Share on other sites

  • 3 weeks later...
 Share

×
×
  • Create New...