91NApeewee

Warmup only

Frequency 62

Cranking 12

Duty @ 130 F : 12

Duty @ 180 F : 22


I think this is dependent upon how open the idle air bypass screw is. Mine will not idle with those settings unless the idle screw is opened up a ways

Joe Perez

iTrader: (8)

= not what the OP is after, unfortunately.

The_Pipefather

iTrader: (5)

True, but if it makes any difference, mine was set to idle at 850 rpm with the stock ECU. I didn't touch the screw after that, and only changed settings in MS.

Also Joe, for some reason changing the duty cycles in warmup only mode makes a difference in the idle even after the car is fully warmed up. Any ideas why?

91NApeewee

I was quoting the other dude with his settings but for some reason it did not quote.

I was saying it that his didnt work on my 1.6 so the OP would know my point


Pipefather, I have experienced this to. I believe it is based on the temperature settings your specify, lower and upper limit if I get what you are talking about

Joe Perez

iTrader: (8)

Having not actually analyzed the underlying code, my understanding of the two modes is this.

In "Warmup Only", the system is essentially emulating an electronic choke. It's a misleading name, because the IAC is not only active during warmup, but rather engine temperature is the only variable that it monitors, ignoring actual RPM.

When the engine temperature is below the "Fast Idle (lower) Temp" value, the IAC valve is driven at the value entered in the "Idle duty at lower temp" box. Once the temp rises above that value, MS begins decreasing the duty cycle of the valve linearly until the engine reaches the "Slow Idle (upper) Temp" value, at which point the duty cycle comes to rest at the value entered in "Idle duty at upper temp"

So in this example, when the engine is cold the IAC DC will be 35%. It will stay there until the engine reaches 120°. At that point, the IAC DC will begin to decrease. When the engine is at 140° (halfway between 120 and 160), the IAC DC will be at 29%, halfway between 35 and 23. Once the engine gets up to 160°, the IAC DC will be fixed at 23%, and it will remain there so long as the temperature remains above 160°.

Nothing else will influence the duty cycle. MS is not targeting a specific RPM, it's just blindly spitting out the values you've told it. Therefore, making a change to the "upper temp" duty cycle will affect the idle when the engine is at operating temp- that's exactly what it's for. You tune that value to achieve the desired idle when fully wamed up, just as you tune the "lower" value to achieve the desired idle when cold.





Now, I also find the term "Closed loop only" to be a bit misleading. It seems to imply that the idle control is inactive when the engine is in warmup, coming on only once the engine has entered closed-loop mode. The two have nothing to do with another. "Closed Loop" in this context refers only to closed-loop idle control, meaning that the system is attempting to hold a specific RPM, adjusting idle DC in response to changes in RPM.

Here, the "Minimum" and "Closed" values have been made available, and the specific "upper" and "lower" values have been grayed out. We're no longer giving the MS specific DC targets, instead we're setting up some general boundaries and letting it figure out what to do.

Now, when engine time is below the "Fast Idle (lower) Temp" value, it will attempt to hold RPM at the specific value set in "Fast Idle Speed (RPM)". Once the temperature rises above that, it will begin interpolating the target RPM until engine temp reaches the "Slow Idle (upper) Temp" value, and it will attempt to hold the RPM value entered in "Slow (warm) Idle Speed (RPM)".



Closed-loop is definitely tricker as there are more values in the "Continued" screen which are largely up to guesswork for setting. It can work properly however.

elesjuan

iTrader: (11)

Okay at this point I can't be 100% sure it even worked, but I attempted to press the diode into the rear of the connector to backprobe it. Didn't have my DVM handy so I couldn't see if it was actually making connection, but it changed nothing with my idle. I'll play more later, just don't really have the time to actually install the diode right now.

91NApeewee

that Diode is for flyback purposes. it only helps so that you dont fry your board or anything during a surge

The_Pipefather

iTrader: (5)

Thanks for that explanation Joe. It all fits in with my own observations of the car's behaviour. In light of what you said, I think it would be better to get it right first using warmup mode, then subsequently moving on to closed-loop.

sabretooth

I've done the mod described in the manual for PWM idle users - I have a TIP122, skipped this, jumpered that etc but didn't use the diode across the wires. When I plug it into the car using all of the DIYAT 94-95 base map settings - absolutely nothing happens.

When I put the diode across the plug the idle is about 800RPM higher.. and stays there. It does not change at all - fiddling with the ECU warm/cold RPMs does nothing - and seemingly nor do any of the other settings. The car just acts as if there is no idle control.

I've replaced the IAC valve with another. I've verified the plug is getting power OK. I've verified that the tube going from the intake pipework to the IAC valve is clear. I've replaced the TIP122 already.

Any ideas on what I'm doing wrong, or what could be wrong?

91NApeewee

I know it is probably obvious, but did you check the wire from megasquirt for idle to the stock harness for continuity?

FoundSoul

iTrader: (5)

Is it set to use PWM IAC in the MegaTune configurator?

If you set it to PWM Warmup mode (in MegaTune) does it work any differently? That can be easier to sort out, especially if you are just looking to get your valve working period, than closed loop control. Then once you get your valve working you can go back to CL later.

Joe Perez

iTrader: (8)

In theory, I agree. In practice, I found that when I added the diode it changes the characteristic of my idle considerably. I had to re-tune, but now that I have it idles better than before. It's actually just about 100% like-stock.

The OP however, having a 2.2 board, should already have said diode in place, as it's part of the original schematic and BOM. Why the PWM instructions call for the placement of an external diode (on 2.2 boards) baffles me.

91NApeewee

Joe, I am having idle issues and was wondering if you could PM or post your Idle settings so I could compare them? Also, where is your idle screw in the adjustment range?

Joe Perez

iTrader: (8)

Here are my idle settings. On the 3.0 MSI with the diode installed, these produce an excellent idle. I could probably go just a tiny bit faster on recovery time, but these are damn close to perfect.




As to the idle screw, I have absolutely no idea where it is set. I have never touched it, so I assume it's wherever the factory set it 16 years ago.

91NApeewee

Thanks Joe!

I have to try those later. Are you using High-Res code?

My idle was screwed with by the PO, so I think I will use those as a starting point

Joe Perez

iTrader: (8)

Yes, HR10d2 as it says in the sig.

A tiny update- I've been tinkering with the deadband range. I decreased it to 50 this evening, not sure if there was an improvement or my imagination, but it didn't hurt. (deadband is the null zone around the target- the amount that idle is allowed to deviate without triggering a correction. In theory, smaller is better, with the downside that too small will cause the idle to oscillate or "hunt".)

sabretooth

Yep, checked it last time. I'll check it again this weekend too.

Yep, IDLE_CONTROLLER (group) = PWM_GAUGE

I'll try it and see what happens.

Is it feasible that there could be something wrong with my CPUs because I didn't initially put that diode in?

Could I be missing a ground from the MegaSquirt? Actually I'll look over the diagrams and make sure I have all of the grounds covered.

Joe Perez

iTrader: (8)

Doubtful. It's theoretically possible that the IAC transistor could be damaged, but I doubt it. I drove mine around for months w/o the diode and it seems no worse for it.

Assuming a 3.0 PCA, there is only one ground which is common to the whole Megasquirt. It is presented at multiple pins on the DB37 connector solely because each individual DB37 pin isn't good for terribly much current (it was, IMHO, a damn stupid connector choice) and therefore spreading the ground across multiple pins spreads the load. Personally, I have three ground wires (two factory, one new direct to head) with each wire hitting two pins.

Since you say that "nothing happens at all" when you plug in the IAC, I'd be suspect of the actual IAC mod itself. Have you the ability to monitor individual signals with an oscilloscope while running the MS on a stim?

The_Pipefather

iTrader: (5)

Just reporting back that I baselined with Joe's settings and they work really well! The things I changed to adapt it to my setup and requirements (sorry I am too lazy to post screenshots as its all on my tuning laptop):

Cranking dc = 27

Fast idle speed = 1400

Slow idle speed = 900

TPS threshold = 100

Idle activation adder = 500

Thanks Joe.

sabretooth

Alright, I think I have it working now.

Attempt #1 - when I built the megasquirt I put in a TIP122 without a diode, and attached to the main heatsink using the end where the high-output ignition MOSFET normally went. This did an unspectacular amount of nothing.

Attempt #2 - I replaced the TIP122 itself and used a mica insulator this time. Still nothing.

Attempt #3 - So I then followed this up by putting a diode over the connectors where the plug goes into the idle motor. I also replaced the idle motor with one from another 94. The idle increased by about 700RPM, but the car still wasn't responding to changes such as air conditioning load.

Attempt #4 - this weekend I redid the TIP122 AGAIN - brand new TIP122, new wires. Load compensation is occurring. It works! IT's not going to the exact RPM values - changing the cold/warm RPMs does things but not predictably. I've got a lot of homework to do to sus out what's going on but I think I'm on the right track now. I think the rest is just going to be fine-tuning of settings.

I have the diode over the connectors - and nondestructively too. The plug on the 94 can have its pins poked out. I poked them out, slipped a 1N4004 on and slid the pins back into the plug. Easy! I also redid the TIP122 for the third time using a mica insulator just as I did the second time I did the mod.