When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Who knows AC regarding the correct settings in an Elite on an FD? I know some settings will be user preference, but I'm wondering if I have anything incorrect in here that would keep the compressor from getting power at idle at least. I can do the grunt work on the troubleshooting and wiring and all that, but I'm not very good with things in the ECU, or at least being able to identify a problem in the calibration.
When swapping to the new ECU this winter, I bought a less used front harness and checked it over before installation (I was having weird issues with the last front harness and/or ECU), and we started with a base map on the Elite so I assumed the settings would be based on the stock ECU parameters, but I want to verify.
This is the last function I'm tackling on this install, everything else works (aside from the stock fuel pump wiring in the front harness, but that's another story). I got the AC charged today but they said the compressor isn't getting power. I verified tonight that the issue is there. I have the AC request switch input (O/G) going to 1E (GY/R) on the stock connector and AC output (G/R) going to 1L (Y/B) on the stock connector.
Once I posted this last night, I noticed the button mode box and realized that's probably not right. Once I selected 'push on, push off', another box came up for what it should do when starting the car and I selected 'last state'. I haven't tried them on the car, but I would assume those are the correct settings. However, I'm not sure about the switch on and switch off voltages, whether pull up enable is needed, or what to select for active state or drive type.
For the Air Con settings, I would think they all look descent, but some of them I'm not sure what they're for.
Since my AC request switch input is around 3.1-3.2v, I changed on to 3.5v and off is still 5v. I had to up the idle increase from 70 to about 200RPM because the engine kept dying with only 70RPM increase. I eventually changed it to 100RPM.
And of course button mode is now push on, push off, and startup state is last state.
My problem now is when I push off the AC button to turn it off, AC stays on. When I press it again, the button lights back up but the compressor disengages, and the fans stay on. Then when I press it again, it turns AC off fully, compressor stays off and the fans turn off.
Does it let you record the A/C input and output states in a log? It could be helpful to log those while you troubleshoot what's happening with each button press. I'm assuming you're using stock fan wiring (thermoswitch input, A/C input, ECU input) and relays?
per Haltech instructions (Air Conditioner Control (haltech.com) , appears to be out of date) example "The switch is at 5V when it is Off and is grounded to 0V when the switch is On. Correct settings with tolerance would be to set On voltage to 0.5V and the Off Voltage to 4.5V. i.e. when the voltage falls below 0.5V it is considered to be ON, and when it rises back above 4.5V it is considered to be OFF."
However in the above service manual pin 1E, the ON state for input is below 3 volts and the off state is B+ (i.e ~12V). If you look at the diagrams, the signal is passing from the HVAC panel to a thermoswitch and A/C pressure switch. See what happens if you set your A/C on voltage to 3 volts or less (2.9?). Maybe raise the 5 volt switch off as well? Basically you need to check if following the voltage spec for pin 1E makes a difference.
I think your output setting is correct. It's just sending a ground to the compressor relay.
Does it let you record the A/C input and output states in a log? It could be helpful to log those while you troubleshoot what's happening with each button press. I'm assuming you're using stock fan wiring (thermoswitch input, A/C input, ECU input) and relays?
It probably does, but I'm not very good with all this and have no idea how to get it to do that lol. I'll look into it. But basically it displays on the calibration as follows...
First button push- ~3.18V ON (AC button is lit, compressor is engaged and fans are on)
Second button push- ~3.18V OFF (AC button is not lit, compressor is engaged and fans are on)
Third button push- 5.00V ON (AC button is lit, compressor is off, fans are on)
Fourth button push- 5.00V OFF (AC button is off, compressor disengaged and fans are off)
Correct, I'm using all stock wiring and components aside from the Haltech harness.
Originally Posted by arghx
per Haltech instructions (Air Conditioner Control (haltech.com) , appears to be out of date) example "The switch is at 5V when it is Off and is grounded to 0V when the switch is On. Correct settings with tolerance would be to set On voltage to 0.5V and the Off Voltage to 4.5V. i.e. when the voltage falls below 0.5V it is considered to be ON, and when it rises back above 4.5V it is considered to be OFF."
However in the above service manual pin 1E, the ON state for input is below 3 volts and the off state is B+ (i.e ~12V). If you look at the diagrams, the signal is passing from the HVAC panel to a thermoswitch and A/C pressure switch. See what happens if you set your A/C on voltage to 3 volts or less (2.9?). Maybe raise the 5 volt switch off as well? Basically you need to check if following the voltage spec for pin 1E makes a difference.
I think your output setting is correct. It's just sending a ground to the compressor relay.
Last night I did try the 'on' voltage at below 3v, but it wouldn't turn on at all. I also tried it at 0.8v based on another person's settings, and it would only turn on if 'on' was above the voltage it was displaying (~3.2v). I tried 3.5v and 4v and both worked. And I also tried 'off' voltage at above 5v but, but 5v is the max setting it allows apparently. I tried 4.5v and that worked as well. Well, it eventually turns off anyway haha. I'm thinking it's not due to the voltage settings though. Maybe the enable pull up? IDK.
I was also trying to enable minimum pressure settings but it won't let me. I would think that's info that would come from the pressure switch but maybe it's in the wrong part of the sequence for the ECU to control. Mine is a 94 so the placement of the pressure switch in the system is a bit different. The idle time on delay is also a little weird because it still bogs down no matter what amount of time you set, but if it's lower than 0.4s and around 100RPM offset or less, it will kill the car every time, and if you have the time set to 4s for example, it will take 4 seconds to turn on but still bog down. It's like it's not taking that time to get the idle up, only delaying the AC turning on for no reason. But I'll address that with my tuner. He doesn't really mess with AC so I'm trying to learn that part on my own.
All in all, it's not the worst problem to have lol. At least AC works, I'd just have to remember to press the button an extra two times to shut it off. But I'd rather have it working properly.
I'm not sure I have anything else to contribute regarding the button pressing concern. I can't find anything in the service manuals floating around online about the A/C system. It was probably some separate guide that Mazda published that I don't have. As far as pressure goes, I think it's a pretty dumb system. The pressure switch is in-line with the A/C input and while cycle the compressor on and off. Maybe @DaleClark can chime in? He's smart with this stuff.
That being said, as far as idle up control goes - I can tell you as someone who had a stock 1995 car with factory R134a A/C , it wasn't very good from the factory; it would surge and dip noticeably. In fact A/C and electrical load control on old manual transmission cars was generally not good back then. My 2002 Miata I had for a while was the same way. You have a low torque engine, and without electronic throttle it's difficult to get a smooth transition.
That being said, you need to try bump up the A/C base offset. That's additional duty cycle to the ISC valve to compensate for the drag on the engine (you do have a working ISC, right?). It doesn't look like the Haltech can do spark control to compensate.
The concept is shown above. Basically the ECU should add a burst of idle air, wait for idle to come up before turning on the compressor, and taper off a lot of the air when the compressor kicks on. It's sort of an inverse when it cycles off.
I'm not sure I have anything else to contribute regarding the button pressing concern. I can't find anything in the service manuals floating around online about the A/C system. It was probably some separate guide that Mazda published that I don't have.
Haha understood. I appreciate your help! I never had this issue before, so I think it's most likely an incorrect setting.
Originally Posted by arghx
As far as pressure goes, I think it's a pretty dumb system. The pressure switch is in-line with the A/C input and while cycle the compressor on and off. Maybe @DaleClark can chime in? He's smart with this stuff.
Now that you mention it, I guess the pressure setting in the ECU would only apply if I had the pressure switch wired directly to the ECU, so that makes sense. It's just there to complete or not complete the circuit, making it a dumb system. I hadn't thought of that.
Originally Posted by arghx
That being said, as far as idle up control goes - I can tell you as someone who had a stock 1995 car with factory R134a A/C , it wasn't very good from the factory; it would surge and dip noticeably. In fact A/C and electrical load control on old manual transmission cars was generally not good back then. My 2002 Miata I had for a while was the same way. You have a low torque engine, and without electronic throttle it's difficult to get a smooth transition.
Ok well that's good to know it's not just my car or ECU settings. I had AC when I first bought it in 04, but all I remember is that I wasn't impressed with the output. I certainly don't remember anything regarding whether it was smooth when turning on AC.
Originally Posted by arghx
That being said, you need to try bump up the A/C base offset. That's additional duty cycle to the ISC valve to compensate for the drag on the engine (you do have a working ISC, right?). It doesn't look like the Haltech can do spark control to compensate.
The concept is shown above. Basically the ECU should add a burst of idle air, wait for idle to come up before turning on the compressor, and taper off a lot of the air when the compressor kicks on. It's sort of an inverse when it cycles off.
Yeah I have a working IAC, I didn't even think of that as an option. I'll look into that, thanks!
Tagging in @estevan62274 as he has setup AC on the Haltech.
Yeah the stock system is pretty basic, it's mainly a "chain" - if any link in the chain is broken (AC switch off, fan switch off, pressure too high/too low) the AC will not engage.
I think at this point it's just dialing it in with the Haltech.
And I knew that from doing all the troubleshooting with the Syvecs the last couple years, but for some reason I didn't think of it that way when I saw the pressure settings.
I agree, I think it's just going to take some tinkering with the calibration.
Oh, @arghx I realize now what you're talking about with AC base offset. I did adjust it from 3% to 4% to see if it made a difference, but I'll add more and see if that helps.
And it totally makes sense now why it was acting the same, bogging down, but delayed by however many seconds. I had the base offset the same each time and only changing the idle time on delay.
1. Set a fixed rpm offset
2. try a high value for the base offset, like 30%. Make sure it's actually doing something. The first thing you want to do is solve the engine near-stalling behavior, then dial back.
3. Observe what the various offset settings do. You would think 30% would make a big idle surge, so you may have to drop it down. The big thing is -- your RPM offset is 200, but if 30% gets you to like 220... it's actually not so wrong. Every rpm offset is going to need a specific duty cycle offset to get there. If 30% gets you stuck at 400 rpm idle bump, you know it's too high for the rpm offset.
4. Try different delay timer settings. A shorter delay will kick the compressor on earlier and interact with the idle surge created by the base offset.
Do this in a methodical way, don't just randomly pick a bunch of numbers. You need to get the offset rpm and offset base % figured out. The delay is probably less important. See below example of what I'm talking about. You don't have to do every one of these combinations, but understand the method here. Don't go chasing your tail. This also assumes your idle AFR and spark are stable, and environmental conditions are stable as well.
Yeah my plan was to drop the idle offset back down to 70RPM, leave the delay set to 0.4s, and only adjust the base %. I don't really need it 200RPM over, I was only doing that because I thought it helped. It did help keep the car running, but it didn't fix the problem. And as long as I get the offset correct to where it doesn't bog down or kill it, I'm good with the delay at 0.4s. So I'm going to play with the base % first, then combine delay if I need to, but try to keep the RPM the same. And I won't go ***** nilly changing settings with no rhyme or reason lol.
Or since the chart starts at 150RPM, maybe I'll try that lol.
Thanks for the chart, where did you find that?
Oh and my AFR seem to be rock solid. I believe spark is too but I have nothing to back that up lol.
I'm seriously going to light this car on fire lol. Now it's doing the same thing it was doing with the Syvecs. The AC is on all the time now no matter what I do with the button. Although back then that was only with the system not charged.
The only thing I did was change the base offset to 5%, and I may have changed the switch off voltage to 4.5v, I don't remember. But I uploaded that, and I don't remember if the AC was on immediately or only after I pushed the button the first time, but it wouldn't turn off after that. So I'm thinking now maybe it's the AC button itself.
EDIT: my buddy logged in and now it's back to the 4 push situation.
Ok everything is working great now, thank you to everyone for your help! I played with the base offset, RPM offset, time on delay and even time off delay, and this was what worked best for me. These are the final settings I came up with overall. Ryan Heinrich pointed out that the button mode should be momentary. It doesn't seem right, but it is.
Tagging in @estevan62274 as he has setup AC on the Haltech.
Yeah the stock system is pretty basic, it's mainly a "chain" - if any link in the chain is broken (AC switch off, fan switch off, pressure too high/too low) the AC will not engage.
I think at this point it's just dialing it in with the Haltech.
Dale
@DaleClark
Apologize, for not responding been super busy at work.
@speedjunkie My setting are very similar to yours in the last post.
Glad to hear that it looks like you got it dialed in now.
Thanks Steve! I did increase the RPM offset from 100 to 200 last night, after I was apparently too cocky using AC on the way home and the car died in rush hour traffic and then didn't want to start and tripped the circuit breaker so I couldn't start the car for another minute haha. It was much hotter in the car after that with all the embarrassment. And I adjusted cranking fuel of course.
All the other settings are the same though, at least for now.
I have a different ECU, but here are my car's AC / Idle settings in case it helps:
AC Idle RPM offset, 150 RPM
AC Idle base offset, 12%
AC Clutch enable delay, 0.75 seconds
AC Maximum Throttle, 35%
AC Maximum RPM, 4000 RPM
AC request switch voltages: on below 2.4V, off above 2.5V.
I find that the important part of getting the car to behave with the AC on is to carefully watch the cells in the fuel map when the AC turns on. On my car, engine load increases pretty significantly when the AC compressor is running and it's important not to let the engine run lean when that happens.
Thanks! I hadn't thought about asking people with other ECUs, yours aren't far from where I'd tried when I was messing with it. My max RPM was originally 4500 but it was 5500 on my old ECU so I figured it would work well enough.
That's a good idea. I'm going to look into that tomorrow. My AFR is pretty steady while idling, around 13.5. Actually I think I noticed today that it was idling around 1200 and AFR was about 13.9 with AC on.
hopefully this will help someone else but these were the settings that were needed to get this particular car to work. long story short, the compressor was not engaging after servicing. pressure switch, relay and fuse all checked out. brand new denso rx8 compressor checked out as well. you can get the clutch to engage by changing the active state to high so that alone ruled out a wiring issue.
the individual components and wiring all checked out but the ECU was not completing the chain for some reason. with "pull up" disabled, the voltage would should 0.5v with the ac button off and the same with it on. with it enabled, it shows 4.95v with it off and 1v with it on. with the correct values in for the on and off trigger and the pull up set to enable, the compressor kicked on and it works as it should.
unlike in the version that speedjunkie has, this one only gives you digital switched and CAN as options. the button "mode drop" down also does not exist. could also be that this car is on a 1500 and maybe those are features of a different version. the information in this thread helped tremendously in helping us figure out what settings were needed.
Thanks for your input! I'm hopeful it will help others in the future as well. I'm confused though, as mine is also a 1500. So I'm not sure why there are differences, unless it's a feature of an update. I'll have to see if mine has changed too.
08-22-23, 11:47 PM
