ive just finished installing my new LINK ecu in my jdm s6 only to find that i cant get the A/C request to work with the computer. i have simply cut the cable that went from the stock computer to the ac button (and then on through the a/c t/stat then grounds out through the fan speed selector switch) and connected it to a digital input on my computer but cant get it to recognise when the circuit is complete to ground giving me no ac input so the computer doesnt pull in the ac compressor relay.

i have tried searching for things like air con wiring ecu etc, but nothing really comes up. so im just chasing how others with standalone ecus have wired their ac request signal?

thanks guys
Brad.

 

So, what happens if you just send that pin on the ECU straight to ground? Does it then recognize that it is grounded?

There is an issue with North American cars equipped with the MANA system where the ground signal going to the ECU is a bit "fuzzy" (1-2 volts instead of 0) and aftermarket ECUs don't recognize it as being grounded....but you have a JDM car, so this is likely not your issue.

On North American cars, the ECU has the last "say" in whether or not the compressor will run. The wiring sequence is: A/C button -> Thermoswitch -> Pressure Switch -> ECU -> Clutch relay Are you saying you've bypassed the pressure switch?

 

I would think that you should be able to use a test light or a multimeter with a "beep" tester to see if any kind of ground is coming along that wire at all. It's basically a 20 year old harness and even though you don't have a North American market car I wouldn't be surprised if you are still having that issue.

If you use a jumper wire to bypass the ECU (mostly as a test), in my limited experience it will work.

 

my car is a 93 model so the hp cutout is in series with the compressor clutch. i did get this to work briefly by turning the key from acc to when everything powered up but it didnt click to "ON" and when i pressed the ac button, the input would show as active, the blower fan wasnt running with the key in this position and my sensors on my ecu software were reading all out of cal. as soon as i turned the key fully to on, the sensors were readign correctly and the fan was working but hte ac signal in no longer worked.

i think i may have to go through and clean up some ground connections as im also having another strange electrical issue in the cab, where when i press my brakes my odometer dims as if i've turned on my headlights, and the lights on my defi gauges come on as well but thats another story and im trying to search for answers for that as well.

long story short, if i cannot fix this by cleaning up ground connections, i believe i am going to have to find another way to ground the ac request signal but i still need an interlock with the blower motor, any ideas?

Thanks,
Brad.

 

Today I resolved this issue that I’d been having with the air-conditioning not working with my new ECU. What I did was made a patch loom that plugs in between the connector G-08 which is the thermo switch one. I’ve attached an image at the bottom of this post and I will reference the coloured dots on it in relation to what I did with the patch loom. Firstly the loom basically just extended all the connections but broke one; the one between the red and the yellow marks. The wire with the red mark was connected to the common pin of a STD Bosch relay and the NO pin of said relay was connected to ground. The green mark is the connector to the blower motor itself and this is supplied the power to the coil of the Bosch relay. The yellow mark indicates where the coil of the relay was grounded (to ensure that there was some sort of fan related interlock so the compressor couldn’t continue to run without the fan causing icing up issues etc.). I did measure the differential between (battery voltage and earth) and (battery voltage to the resistor earth) and the results varied by roughly .6V. So this setup i have now, ensures that the ecu digital output/input and the coolant fan relay coil both get a solid ground to earth and the relay coil has an earth that is only active when the blower fan is on.

 

This is some good information and I might have to do the same mod that Brad did.

I took my AC out in 2008 because I put in my VMIC and didn't figure I'd need it, but black car with black interior almost makes it a requirement. I put my original MANA system back in (my car is a 94 Touring btw) aside from using a universal condenser and having to get a Denso dryer since they don't make the MANA one anymore, both of which in turn required having lines modified. The system is charged but my compressor clutch won't engage unless power is applied directly to the clutch and/or HP cutout. I've jumped both the LP and HP cutouts, and the AC button and fan speed switch both seem to be working properly (the fan speed switch changes the blower motor speed and the AC switch turns on the radiator fans as long as the fan speed switch is on). That made me think it was the thermoswitch but I tried jumping it and it didn't change anything. At least I'm pretty sure I did it correctly. I jumped VP and V.

So now I'm wondering since my system is MANA maybe the ground isn't being read properly by my Syvecs ECU. Neither 1E or 1L are seeing ground. I might have to try the relay like Brad did.

 

Bumping this old thread to get an understanding. I have a microtech lt10 I have been told that I can no longer use the oem ac switch to make the ac work. I have modded it to work but I am not thrilled with the results. This is what I have done. I taped into the compressor wires and ran two wires. One wire runs from the compressor wire to the evaporator. I e wire from the evaporator two a toggle switch and one wire from the fuse box to the toggle switch. It works but I am trying to understand if there is a way for it to run with the microtech and the bac. Please help and explain as though I was in kindergarten please.

 

Don't know if this will help with a Microtech ECU, but here's how I got it to work with my Link G4+ ECU. For the Link, you need to sense if the A/C switch is on or not via one of the Link's digital inputs (DI's). The switching thresholds for all the DI's on a Link is based on voltage, so anything below about 1.2V = binary 0 or off state, and anything above ~ 1.2V = binary 1, or on state. To do this on an FD (93 USDM in my case), I tapped the violet (V) wire from the B1-01(F) harness and wired it to DI#6 on my Link ECU, thru a pack of 10 forward biased 1N4007 diodes wired in series with DI#6, and it worked perfectly with the Link, sensing A/C on or off state regardless of the fan blower speed. If the DI's on a Microtech have similar switching thresholds, the same setup should work.

The reason you need the diodes is because the voltage on the V wire is not a simple binary 12V "on", 0V "off" affair. If the A/C switch is OFF, and the ignition switch is ON, you'll always see a reliable 12VDC on the V wire, but if the A/C switch is ON, the voltage will vary depending on the blower fan speed setting - so you might have 1V at speed #1, 2V at speed #2, 2.7V at speed #3, 3.5V at speed #4 for example. Actual voltage will depend somewhat on how worn out the fan switch contacts are, but the point being is most/all of these voltages are likely above the switching threshold for the DI on your ECU, so it won't be able to correctly sense the A/C on/off state change. For my FD, without the diodes (directly wired to the DI), it would work fine for fan speed #1, but none of the others. Putting in a bunch of series diodes results in a reliable voltage drop, and conditions the input so the ECU can correctly detect the state change. So now when the A/C is OFF, instead of seeing 12VDC at DI#6, it sees about 7VDC, which is still above the 1.2V switching threshold, and when the A/C is ON at any fan speed, it will see way less than 1V, well below the threshold.

Attached is my FD's schematic diagram for the Link installation, refer to sheet #4 for the details on the diode wiring.