SW20 MR2 AC clutch control

[MGV101]

After installation of the MR2LINK V3 pnp ECU, the AC on my MR2 had stopped working. After spending a long time troubleshooting and looking up factory electrical diagrams and research on how the AC circuit works we noted the AC clutch is controlled directly by the AC amplifier unit self and for the output from the ECU, an ON position actually tells the AC amplifier to disengage the clutch (presumably at high RPM to protect the compressor) rather than sending out a signal to engage the clutch.

The map was initially locked by the tuner so I cannot simply disable the AC clutch output in the map. Instead, I simply removed the AC clutch output pin on the ECU connector and the AC immediately started to work again. I had been running my car the whole summer with the pin removed and manually switching off the AC when I am had on the throttle.

Now that the map has been unlocked and I have finally found some time to work on the car, I intend to fix the AC issue the properly way with ECU control. See the attached screen shot for my solution to the problem. I am still in the office so have not tested it yet but I think it should work.

However, I do find it strange that such consideration had not been made in the base map or hear this problem being mentioned by anyone else. This makes me fear  hat there is actually something wrong with my car rather than the map itself. Can anyone help me confirm?

[Adamw]

I dont remember anyone ever reporting that our base map AC settings dont work correctly on an MR2.

But if all you are trying to do is "invert" the AC clutch output so that it works opposite of what it does now then all you need to do is just change the polarity setting on the aux output from low to high.

[MGV101]

2 hours ago, Adamw said:

I dont remember anyone ever reporting that our base map AC settings dont work correctly on an MR2.

But if all you are trying to do is "invert" the AC clutch output so that it works opposite of what it does now then all you need to do is just change the polarity setting on the aux output from low to high.

Yes that is what worries me, seems strange that no one else is having this problem. Maybe this is specific to a JDM MR2.

If I only revert the AC clutch output, when AC request is off the output would still be at the "on" position wouldn't it? Meaning the AC amplifier would still be receiving a signal from the ECU to disengage the clutch even though the AC is already switch off. I doubt that would cause any problem but the OCD in me wants it to behave exactly like stock where no signal will be given to the AC amplifier when the AC is off and hence I was proposing to use the method above.

[cj]

Is this a 3sgte powered mr2? rev1/2 or rev3 engine? What colour wire is connected to the AC pin on the link?

The service manual and wiring diagram for all 3sgte's doesnt even show an AC control circuit coming out of the ECU. It shows pin A10 (B/W) as input to the ECU so it knows when the AC clutch is engaged so it can idle up and thats it. All other signals about water temp, AC pressure, etc all go direct to the AC amp and never touch the ECU. The one bit thing that _might_ be relevant is that the AC amp receives a tacho input signal (black wire from IG- on the coil, same as the one to the tacho), and apparently uses some internal logic to disable the AC clutch at high RPM. Unless you've done a COP conversion or otheriwse chaneged the tacho signals this should still be in place.

From the 2nd gen service manual:

Second gen is a little different on the pinouts for 5sfe's but 3s's have the same AC control regardless of 2 vs 3 generation.

On the 3rd gen engines, the same thing applies to 3sgte's but 5sfe's are a little different and the service manual states:

3SGTE

 

5SFE

Things that are the same on both

 

End result of all of this, is that the ACT pin was never used on 3sgte's from the factory, so you dont need it. If you want to set some logic to control the AC activation (eg above 75% throttle) then you can wire the ACT (Pink-Green) wire to an output that you set Low Polarity (ground) = "triggered" to disable the air con compressor (the air con amp outputs a 12v signal so floating or 12out on the link output will do nothing).

If you have a black-white wire hooked up to a link _output_, you should really disconnect it as any result you are seeing from placing 12v/ground signals on this wire are because you are providing current to what should be the output side of the aircon activation relay and you are bypassing the aircon amp or possibly feeding current into the wrong side of it. b-w wires in all the factory manuals are some form of input signal (as far as aircon is concerned)

[MGV101]

Wow... thank you for such a detailed answer cj.

Yes, one of my main source of information were the factory wiring diagrams too. Somehow my version did not cover the 5SFE specific part through, and it had no mentioning of what the ACT pin does. Only with trial and error that I have figured out and grounding this would cause the AC shut down. The information you found affirmed my suspicion.

It's a Gen 3 engine and the car itself is a JDM Gen 3 MR2. Seems like although this input is not used on the 3SGTE, the wiring is still there on the stock ECU connector and engine loom. Although the base map on the Link G4+ notes presence of this output, it was never configured properly and treated it as a ordinary AC signal used to switch on the AC even though it was a signal to stop the AC instead. (So yes, it is the pink green wire going to the G4+ AC output)

I am actually now running a full coil on plug system and replaced the distributor with crank and cam sensors so there is no tach signal going to the ECU. MAYBE the AC amplifier is configured to ignore the ACT signal with a tach signal is detected and that is why no one had previously reported this problem. I actually had the car running briefly with a distributor system on the Link G4+ but it was only for a matter of days before I dropped the engine to carry out works and never refilled the refrigerant a year later and I had no recollection of whether the AC was working back then.

I got a friend running a stock ignition system on a link G4+ too and he was having the same problem until we disconnected the ACT. However, his MR2 was one of the few rare ones which had a stock TRC system and it's own separate TRC ECU which further complicate things to draw a conclusion whether the AC works with the base map on a MR2 with a stock ignition system.

Without a rpm signal for the AC amplifier to determine when to shut down the AC pump under high loads. Yes, it is a good idea to make use of the ACT input on the AC amp to give the compressor some protection.

[Aqmar]

Hi @MGV101,

I've been reading your thread with alot of interest. I am currently using the 3SGTE ST246 (Gen5) engine into my MR2. This engine originally comes with COP setup. In that case it is as similar with your issue.

So what exactly your solution? What did u do to makes the A/C runs same as stock?

[ellisd1984]

I know this is an old thread but thought I should post my solution to this problem.  What I've noticed is the ECU doesn't have any control over the AC clutch, it just reacts to the AC amplifier activating the clutch.  This means if you've done away with the standard Idle up valve (as I have) the idle will stumble when the clutch comes in becase theres no delay between the AC request and the clutch being energised,  What I did was disconnected the ACT and MGC from the AC amplifier connector and wired them together.  This means the AC amplifier does it's thing but the Link actually activates the clutch and fans in reaction the the AC request.  Meaning you can program the required delay for the idle to adjust and stop it stumbling

https://www.dropbox.com/preview/Screenshot 2020-08-08 at 14.52.31.png

[Island_Racer]

Lets revive this one last time. So i installed a simple push button on my center console and wired it to the DI2 pin on my link g4x (ac request pin). I also created a GP output and assigned it to activate the pin that is connected to my ac idle-up valve and is triggered by the DI2 pin (which is wired to the button)…..and i used two conditions…1st- the DI2 pin must be in active state (aka the button is pushed) 2nd- the TPS position is less than 50% (aka when i press beyond 50% throttle, the ac idle-up valve is deactivated)

So when i push the button the ac idle-up activates and my car’s rpm rises….i then push the “on button” on my ac control and the ac starts, clutch engages and the rpm drops back to “normal” levels…and the car cools nicely...

now i personally have no issue with now having to push two buttons to get my car’s ac going….the car cools well, the car behaves great….i just need to know and confirm one thing….

what tells the clutch when to disengage? I kno and see the pressure switch on the ac pipline…but does that trigger the ac amplifier and the amplifier disengages the clutch? Or is the ecu supposed to get the pressure reading and then it tells the amplifier to disengage the clutch??

i want to ensure i dont damage the compressor because it doesnt disengage when/if the pressure gets too high…so i need to know if there is one more step i need to take.

it would help to say that i tried using the ac clutch control in the chassis and body section of the link g4x but it doesnt disengage the clutch…doesnt do anything from what i observed….i used “basic” mode, used the DI2 pin as ac request (push button), set the aux6 pin (ac output) as the output…and when i push the button. The ac request lights up and says active, the ac idle-up valve output lights up and says active the valve activates, the ac clutch control (aux 6) lights up and says active…i noted that the setting by default had disengage clutch at 90% tps….i changed that to 1% to test if the clutch disengaged when i held my pedal higher than 1% tps…and when i hold above the 1% the light (aux 6) turns off and says inactive, but the clutch did not disengage…so i dont believe that it controls it…the clutch only disengages when i turn off the ac controls...i saw one person above say switch the driver from low to high and that will do it….but i need confirmation that that is in fact exactly what i am missing….i wouldnt want to try that randomly without some kind of verification by a pro….

 

[Island_Racer]

Also….on the link g4x the pin labeled DI2 (ac request)  when compared to the st205 manuals corresponds to a wire going from the ecu TO the amplifier whereas the pin labeled Aux 6 (ac output) when compared to the st205 manuals corresponds to a wire going FROM the amplifier to the ecu….is it possible that Link accidentally swapped the two pins when they fabricated their version of the ecu???

so where it says “from ecu” wouldnt that be an output from ecu??

 

and where it says “to ecu” wouldnt that be an input to the ecu??

 

 

[Vaughan]

https://www.mr2.com/files/mr2/techinfo/Wiring Diagrams/1993 MR2 Wiring Diagram.pdf

PDF viewer page 177 or page 185 based on the numbers on the pages.

AC magnetic clutch relay is controlled by the ac amplifier not the ECU. AC pin into the ECU lets it know the whether or not the clutch is engaged.

[Island_Racer]

Vaughan,

I see your point and thanks for that….the ST205 also has the dual pressure coming into the amplifier and the line from the amplifier going out to the magnetic relay….

 

so essentially, what i did with my button, input and output assignments is ok…the amplifier will disengage the clutch for pressure safety….

But then that leaves me with the one remaining question….what disengages the clutch when im making a hard pull?

[Vaughan]

38 minutes ago, Island_Racer said:

what disengages the clutch when im making a hard pull?

maybe investigate ACT and AC1 to see what they do. The PSCT ECU pin is used by the ECU to turn off power steering when so desired so maybe one of those two pins does something similar.

[DerekAE86]

Toyota changed the names of ACA, AC1, ACT, etc around a fair bit. Depending on car/engine/year things could be slightly different.

But typically there'll be a pin from the Amp to the ECU to let the ECU know the AC is on. This can sometimes be the same pin that engages the actual clutch.

And then there'll be a pin from the ECU to the Amp to override/cut off the AC. Could be because youre RPM drops too low to maintain idle or you're accelerating fast and want that extra power.

The Amp itself will control the pressure/temp of the system by cycling the clutch on and off though. So no need to worry about accidentally overriding that function.

[Island_Racer]

31 minutes ago, DerekAE86 said:

Toyota changed the names of ACA, AC1, ACT, etc around a fair bit. Depending on car/engine/year things could be slightly different.

But typically there'll be a pin from the Amp to the ECU to let the ECU know the AC is on. This can sometimes be the same pin that engages the actual clutch.

And then there'll be a pin from the ECU to the Amp to override/cut off the AC. Could be because youre RPM drops too low to maintain idle or you're accelerating fast and want that extra power.

The Amp itself will control the pressure/temp of the system by cycling the clutch on and off though. So no need to worry about accidentally overriding that function.

Thanks for that info….that was the picture i had shown…that was also what made me curious. The oem manual said the pin going from AMP to ECU is the Link G4X’s Auxilary 8 pin and i was wondering if it is going “to” ecu shudnt that be an input…..

and then vice versa, for the oem manual saying the the pin going from ECU to AMP is Link G4X’s Digital Input 2 pin and i was wondering if it is coming “from” ecu shudnt that be an output.

hence me asking it maybe the pins were accidentally swapped…remember the ST205 3rd gen is still different from the SW20 3rd gen…

based on what uve said, i will try to swap those two wires and see if the system operates how it should.