Jump to content

Photo

Installing an AIM XG Log

- - - - -

  • Please log in to reply
7 replies to this topic

#1
PatrickCleary

PatrickCleary

    Member

  • Members
  • PipPipPip
  • 15 posts
  • Location:Santa Rosa, CA
  • Car Year:1995

I'm about to begin the install on a Mychron3 XG Log and I want to make sure it's done correctly. Since there are a number of people with these systems here, I thought I'd check my understanding against the more experienced users. Where possible I'd like to use existing signals/sensors.

 

This will be going on a 1995 slowly winding its way towards ITA and it will replace the dash entirely. For the next track day (my first focus) I'm trying to get speed, RPM, water temperature and oil pressure (dummy light at least).

 

-RPM pickup wire (blue) connects to the dash board/datalink tach connection.

 

-Speed sensor provides +V, GND from the XG Log and a sensor line that varies between +V and GND depending on the hall effect (didn't get these, so I'll need to prep a new set)

 

-Water temp - Here things get a little hairy. The 1995 wiring diagram makes it look like the dash has a signal line (Blk/Blue) that would connect to the analog input. The ground line would run to the dash ground. That leaves +Vbattery and +Vref coming out of the XG Log. If I'm not powering the sensor off of the XG log, do either of these get used? Tie the Vref to the dash +12v line? Both of them?

 

-Same as water temp applies for the pressure switch, although I'm less concerned about wiring that up.

 

Am I even close? Also, does anyone have any idea why the horn ties into the datalink?


Dumb trackday kid.


#2
PatrickCleary

PatrickCleary

    Member

  • Members
  • PipPipPip
  • 15 posts
  • Location:Santa Rosa, CA
  • Car Year:1995

Wow I whiffed on my assumptions above. For future reference, Vbattery is just a 12v output, Vref is a regulated 5V output. The rest I think is pretty close to right.

 

Does anyone know the scaling on a temperature gauge sending unit? It's definitely different from the ECU's sensor (which is documented online) and doesn't match the one the local auto parts store sold me. Maybe it would up at the top end of the range, but definitely not anywhere near room temperature. I can compare values against the ECU's sensor next time I drive it, but known values would be awesome. I'd prefer not to pull it unless necessary.


Dumb trackday kid.


#3
tynor

tynor

    Member

  • Members
  • PipPipPip
  • 54 posts
  • Location:United States
  • Region:Atlanta
  • Car Year:1999

I measured the water temp signal at the dashboard on my 1999 and compared to OBD-II data and got the following chart:

 

 

Volts   Temp(F) - via OBDII

4.31   203

5       190

5.52  181

6.05  172

6.81  159

7.7    145

8.55  125

9.03  111

9.27  100

9.52   86

 

The highlighted rows are relatively linear, so I used that to program my TraqMate.  It reads about 15 degrees hotter than the mechanical guage that reads from the upper radiator hose, but I presume I can trust the OBD data?


Edited by tynor, 11-21-2014 03:16 PM.

Steve Tynor

'99 SM

Atlanta, GA

 

Bona fide - A bonafide Spec Miata driver

#4
PatrickCleary

PatrickCleary

    Member

  • Members
  • PipPipPip
  • 15 posts
  • Location:Santa Rosa, CA
  • Car Year:1995

Whoops, didn't mention the year, I have a 1995, although I'll check that data. Did you leave your dash installed and just tee into the line?

 

Yeah, the ECU data will be correct(ish). Assuming the 1995 and 1999 use the same sensor, they are very non-linear. The gauge senders are temperature dependent resistors (thermistors) that decrease in resistance exponentially. Through some sections of their temperature-resistance curve they're pretty linear, but on a macro scale that's not a fantastic assumption.

 

The 15 Degrees off is likely due to the non-linearity of the system (especially if you're just plotting a line between 125º and 203º). Actually come to think of it, I could probably get close just waiting until the fan turns on, taking a measurement there, and figuring the curve out, maybe thermostat temperature as well.

 

Patrick


Dumb trackday kid.


#5
tynor

tynor

    Member

  • Members
  • PipPipPip
  • 54 posts
  • Location:United States
  • Region:Atlanta
  • Car Year:1999

Yes - I just tee'd the wire on the back of the dash (on mine it was pin 2a (purple/white wire) and added a 470 ohm pull down resistor at the input to the Traqdata.


Steve Tynor

'99 SM

Atlanta, GA

 

Bona fide - A bonafide Spec Miata driver

#6
PatrickCleary

PatrickCleary

    Member

  • Members
  • PipPipPip
  • 15 posts
  • Location:Santa Rosa, CA
  • Car Year:1995

Ok, now you have me, why a 470ohm? Just to gain impedence on the input (sounds like it's been a problem on AIM products)? Also, by pull down do you actually mean a connection to ground, or just that there's a resistor in line?

 

My install will be a bit different as I'm pulling the stock dash (although with your numbers I'm a little tempted to keep it) but again, it helps.


Dumb trackday kid.


#7
TrailBrake

TrailBrake

    Member

  • Members
  • PipPipPip
  • 59 posts
  • Location:Manchester, NH

I'm pretty sure I answered you on apexspeed. If it's a resistance sensor, you need a 1,000 ohm pull up resistor connected from the 5v to the sensor reading. You  are much better off putting in new sensors. Retail on a temp sensor is $45 and it is plug and play.


Matt Romanowski

www.TrailBrake.net


#8
PatrickCleary

PatrickCleary

    Member

  • Members
  • PipPipPip
  • 15 posts
  • Location:Santa Rosa, CA
  • Car Year:1995

Yeah, I read that, and thank you it made the wiring diagram make sense. I think I also explained why I wasn't going with the aftermarket sensor unless I absolutely need to.

 

I get why I need a resistor, but not in Tynor's application. Without a dash in place you need to add a resistor to form the first half of a voltage divider. With a dash still installed, splicing into the harness should give an output voltage without a resistor in place. This comment and a far from complete datasheet don't give me the warm fuzzies. It makes me wonder if the impedence on the AIM inputs is low enough that it could substantially effect shared sensors.


Dumb trackday kid.





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users