[HugoW]

Hi,

it seems more people than me alone have trouble with stable idle, aspecially when using more things that take energy at the same time, like headlights, AC and cooling fans. I can use Idle Up for one of those, not all three. If someone could allow for a PWM duty adder in the Fan Settings, percentage or hard number, we have two idle adders. And most cars will benefit from idle up with the fan on.

Cheers,

Hugo

[PSIG]

A few thoughts, Hugo. One is that you can input multiple signals to an idle-up input, as an option. Use diodes on each signal line so you don't back-feed. Only one bump for any and all devices, but it might fit your issue as one option.

Second, you might consider using Idle Advance Settings idle control as a faster and effective additional stability tool than Idle-Up. Previously, issues such as multiple loads were handled with a simple ignition table "timing trough"; a valley of normal idle timing with considerably increased timing immediately below idle range. If anything (any load) pulled idle down, the timing would instantly advance for increased torque to recover.

If your idle target is 800 rpm, your idle timing cells in the 800 rpm column would be perhaps 10°, but the next-lower column would be 700 rpm with double the timing from minimum to 70 or 80kPa, for example. You might want to try that as a well-proven and simple alternative, and the torque-reserve theory Idle Advance Settings was based-on.

And finally, a good tune should not struggle with added loads. Sure, some minor drop in rpm is expected with accessory loads, but not anything concerning. Think back to carburetor days, where the carb and distributor tune had to run with the same occasional loads you have now, without stalling or stumbling. They had to do it without any help from controlled idle valves or other cool features. Get as close as you can with basic stability tuning to make it solid, and then add the electronic magic to make it really nice. TL;DR — You may want to revisit your balanced (fuel & spark) tune for adjustments towards stability and efficiency, which directly affect the idle. If you have not developed that skill yet, it's a good time to start practicing.

[HugoW]

Thanks for the reply. I am working with those suggestions but I have some issues with them (I call them carburator fixes).
The first is using multiple inputs via diodes, the system will treat them as one input. So it doesn't react when a second user is switched on (standing at the trafic light at night, lights are on, fans kick in).
The second is MOT. Having the idle not set at optimum but with room to spare gives exhaust gass readings that don't pass. I need lamba 1, highly efficient (no unburned carbons).
The third is, it seems standard practice on many if not all OEM ECUs and I think it is for a reason. The ECUs know when airco kicks in, fans switch on, lights are switched on, etc. And they react. They might have an added PID loop to fine tune, but the basics react to input and settings.
In Arduino, I would be able to write the code. In this complicated Speeduino environment I wouldn't know where to start. For the moment, I will hardwire pin 47 (fan out) to 41 (idle up in) and get it going. But with that done, I cannot hook my lights up. If this feature is added, I can use pin 41 for the lights. I would need a second Idle Up in if I had airco, but I don't.
BTW, I read everywhere how fantastic people's idle is, but in my Fakebook topic, when push comes to shove, no real good idles are offered in the MX-5 NB VVT engine I have. All have high idle, not very stable, running rich to compensate with low idle advance, etc.
So, I would like this feature. And I think it would be beneficial to others, too.

[EDIT]I just did a little test with the family car, 2009 Peugeot 308 SW, bone stock and dirty. Car idling, I turn on the headlights, no change in idle. Then I disconnected the headlights (I didn't bother with the rears, they use low current and you need to remove the clusters to disconnect them). Fired the car up again, stable idle. Then I switched on the headlights and the RPM went up (reacting to switching on the lights, not to the load, as there hardly was any!) and then it compensated back to idle quickly (closed loop compensation). No fault codes on the dash, though. I plugged the lights back in so my wife can go to work now.

Looking at the wiring of the MX-5, the ECU controls the alternator field. So, it knows load and can compensate open loop for it, before closed loop correction if any. The ECU also has a signal wire from the AC coming in, I suspect to do the same. Of course it also runs the fan, so it knows that as well. [/EDIT]

Cheers,

Hugo

[theonewithin]

On modern OEM ECU's it will have those functions.

Speeduino is not a replacement for a modern OEM ECU.

It is for "off-road" use.

[PSIG]

Aw, damn. You had to half-read it and respond negatively while missing some points. Solve the issues as you see fit or require, but there is some good stuff hiding in Speeduino if you use it.

Thanks for the reply. I am working with those suggestions but I have some issues with them (I call them carburator fixes).
The first is using multiple inputs via diodes, the system will treat them as one input. So it doesn't react when a second user is switched on (standing at the trafic light at night, lights are on, fans kick in).

Yes, I said that, and with some loads quite low, only one IAC idle-up for any or all of them may be enough.

The second is MOT. Having the idle not set at optimum but with room to spare gives exhaust gass readings that don't pass. I need lamba 1, highly efficient (no unburned carbons).

Unfortunately, the concept is backwards from a torque, stability and economy perspective. Running an engine. Lambda 1.0 or stoich is very inefficient except for creating exhaust products that work best with older catalytic converters. It is neither powerful nor economical, and so is rarely seen in tunes—except emissions tunes—where it is not desired, but required. That doesn't mean you can't have a good running engine and good idle, but it does mean you'll have to work harder to get it. See more below.

The third is, it seems standard practice on many if not all OEM ECUs and I think it is for a reason. The ECUs know when airco kicks in, fans switch on, lights are switched on, etc. And they react. …

They are for reasons. As you fully grasp the reasons, the solutions will become more clear. Hang in there.

BTW, I read everywhere how fantastic people's idle is, but in my Fakebook topic, when push comes to shove, no real good idles are offered in the MX-5 NB VVT engine I have. …

Big red flag. Assume nothing 'offered' is usable. For something like this your idle will have to be created, by you, for that specific engine and fuel, to your requirements. That's diagnostic tuning. A shared tune, nor Lambda 1/14.7 everywhere, won't get the job done for the sniffing machines or your idle. If you are not willing to dedicate to ongoing learning and tuning, Speeduino (or other aftermarket ECM) may not be a good option for you. They have the ability to pass emissions tests, but you have to find the secret combination to unlock it. Tuning.

An awesome advantage you have is the ability to load tunes when you like. So you can get your fingers wet creating a killer performance and economy tune for off-road, non-emissions, with peak performance, great idle and economy. Then store it while making a MOT tune. Load whichever one you need, when you need it, from your phone or tablet. This allows you to find what it wants to run well, before fighting it to run "OK" with what it doesn't.

[EDIT]I just did a little test with the family car …

True, but that's a given, that you can take the signal from the source or the result (ECM process or end device). Nobody is limiting you to one or another, any more than requiring you to get TPS signal from the throttle body instead of the throttle pedal, or in-between. Get it where you can or where it makes functional sense.

Finally, I noticed you said nothing about the largest section of my post, using ignition advance for idle stability. I'll leave you to do your thing, but your vision appears somewhat restricted, limiting your options. In any case, have fun!

"People who say it cannot be done should not interrupt those who are doing it." -Puck Magazine, 1902

[JHolland]

On modern OEM ECU's it will have those functions.
Speeduino is not a replacement for a modern OEM ECU.
It is for "off-road" use.

Its not a 'modern' thing, almost every ECU that I've worked on, back to around 1987, has inputs for AC and power steering and a large number also have inputs for electrical load.
For AC the ECU usually just switches to a different target idle table.
Its an interesting issue, I may spend some time looking more closely at how an older 8-bit ECU handles this.

[HugoW]

I will only react shortly to a few points, not to contaminate this feature request any more. Change of request at the bottom, BTW.

... It is for "off-road" use.

I don’t use the car on public roads, of course, I wouldn’t dare. Nobody does, right? I sometimes do illegally to test something, but most of the time I cross the border to Mexico to do that.
https://www.youtube.com/watch?v=MgxeThyjf4I

Big red flag. Assume nothing 'offered' is usable. For something like this your idle will have to be created, by you, for that specific engine and fuel, to your requirements. That's diagnostic tuning. A shared tune, nor Lambda 1/14.7 everywhere, won't get the job done for the sniffing machines or your idle. If you are not willing to dedicate to ongoing learning and tuning, Speeduino (or other aftermarket ECM) may not be a good option for you. They have the ability to pass emissions tests, but you have to find the secret combination to unlock it. Tuning.

First of all, you missed the point or I didn't make it clear enough. I have 5 people telling me their idle is stable, they even sent me their tune. But when I ask, they all don’t have stable idle. On their cars. One idling at 1300 rpm, and dying when the airco kicks in. One idling at 1000 rpm, but with hunting and significant rpm drop when lights or fans are activated. So, I was disappointed that people tell me they have beautifully stable idle, making me think I am failing, while they actually don’t.
Secondly, I don’t copy other people’s tunes. I save them for reference and analyse them for my education. I use parts of it, compare to my tune, see what the difference is and what the result is. The only tune I copied so far was the MX-5 base tune, when starting the project.

Finally, I noticed you said nothing about the largest section of my post, using ignition advance for idle stability. I'll leave you to do your thing, but your vision appears somewhat restricted, limiting your options. In any case, have fun!

Ah, yes, that part. I *think* it was also you suggesting that on Facebook, I thanked the poster there and will give it a shot, certainly. And for my vision being restricted; I am not the sharpest tool in the shed, I know, I learn every day but I will never come close to most of you here. I appreciate all help, and I think adding idle up features will help others. And therefor my suggestion here might help others.

And now for the change in feature request, maybe simplifying it;
Can we add some idle up inputs. Not fan or anything related. If I want to hook up the fan to one, headlights to the second, I can. If someone wants to hook up the airco, steering pump, stereo, or whatever, they can. Just allow for five idle up inputs, instead of the current one. The mega has enough ports. No need to add it into the hardware for me.

I have wired the fan pin to the single idle up pin yesterday evening and will tune it today, so it idles up when the fans kick in. The headlights-dip I will try to deal with, with the ignition advance.

Cheers,

Hugo

[LPG2CV]

Hi Hugo

It was I, that posted on FB and directed you to the @psig post.

You need to do some work on you tune (as posted on FB) at idle before it will behave as you want. Extra features will not fix a poor tune.

There is not one fix for a tune, all areas need to be configured correctly for it to work.

Key areas to understand are:

VE table
The values in the VE, AFR and Advance tables interpolate. What that means is a value is calculated based on the current cell and the eight cells surrounding it. So if at idle, a value to the left of the current idle is less than the current VE, then, if the idle rpm drops, then the AFR will run leaner. However, the table and the interpolation is two dimensional. The effect of the leaner burn will result in the map changing and result in an upward or downward cell in the table.

Voltage Correction
Injectors
Generally speaking (and for our purposes), the opening time of an injector, is proportional to the voltage applied to it. And, to clarify, the opening time is, the time it takes to open an injector, and there is not a significant amount of fuel being passed.
So, looking at the injector voltage correction table, the idea is to add or decrease pw (as a %), as the voltage changes.
If for example, the voltage at idle is 12v, and the correction is 100%. That is, no correction, Then as the voltage rises (from the alternator) and the injector opens faster, then the correction will be to reduce the %.
Working the other way down the scale, if we add a load to the charging system, by way of lights, heater fan and cooling fan, then we not only reduce the voltage at the injector, but we add extra load on the engine by way of the alternator trying to keep up with the voltage and adding load to the engine.
So, to enable the injectors to maintain the correct amount of fuel. (in effect, the correct AFR), extra % needs to be added, when electrical load is applied. A good way to determine the %, is to turn on the headlights and watch the AFR. If it runs leaner, look at the voltage, and add some correction %. Then turn on the heater blower and repeat and rinse.
Generally, the cooling fan is the same principle. However, be cautious, because the fan has activated, as the engine is hot and the IAT may also affect the AFR.
Cranking. Obviously, cranking a cold engine on a cold day is going to push the system voltage down even further. So even more correction % is applied for cranking.

Whilst aiming for an acceptable fuel setting based on system voltage, it won't work well without a a decent spark.
Which brings us to the dwell correction table. Unlike fuel correction, there is a cranking dwell. Unlike fuel, the is a fixed value for running dwell and then the correction table is applied to it. Whereas with fuel correction, you chose where to put you 100% in the table.
So same approach to dwell as to fuel, if you add electrical load, then the voltage will drop, and needs correction.
To little dwell results in a poor spark, to much can cause the coils to overheat and possibly fail.

Happy tunes

[HugoW]

OK, I'll give up. Please strike my request.

I have, like many others I've met via Fakebook, added the extra idle ups in the firmware. It seems semi-common practice. We'll keep doing this 'underground style' and have a nicely behaving car, like factory solution. Too bad these extra idle ups cannot be tuned via Tuner Studion, I'm not that competent of a programmer. It does feel good not to be the only idiot using these features.

Happy tuning,

Hugo

[nm145]

Hugo,

Can you point me in the direction of the code for these extra idle ups? I have a miata engine in my exocar that has no shrouds and a huge intercooler in front of the radiator so I have 2 very powerful fans to keep it cool. Even with a idle advance and a timing trough I can't get my car to stay consistently running when the fans kick on. I even have the idle up triggered by the fan output but it is too late. The stock ECU would idle up then turn on the fans and it was never an issue.