Air temp ignition compensation

[nurv2600]

Bump, it's been a month and with warmer weather on the horizon, I'd love to get some basic failsafes and protection that run fine when the car is stock. Not being able to compensate for IAT/ECT is one big reason I'm holding off on the CT-e blower; tracking the car would be a little dicey if I was to rely on knock to bring timing back where it's supposed to be!

[Hondata]

Read my previous reply. The ECU will compensate ignition for IAT as long as you set the knock ignition limit tables. You don't need knock for it to compensate.

[Six_speed_demon]

Read my previous reply. The ECU will compensate ignition for IAT as long as you set the knock ignition limit tables. You don't need knock for it to compensate.

So what's being said is IAT comp to run against MBT table will not happen... Also tuning a IAT sensor post blower will not be able to work either...... Meaning limiting the power these boosted 9thgen will make....and it won't be safe seeing that there is no FAILSAFE Capablities!


This is awesome! Hopefully you guys will figure out how to get this done one day..... But from the sounds of it doesn't look like that's going to happen!

[Hondata]

Sarcasm aside, you are making many false assumptions and are panicking (failsafe? - it's a compensation). You can get it to work given the existing behavior of IAT compensation. In fact, the stock behavior is better than the altered behavior for the 2006 Civic - do you want the ECU to retard with temperature under conditions where it is close to knock, or over the whole table, even if the ignition is nowhere near knock? The first type of behavior is what the stock ECU uses. The second behavior has been requested since people don't understand or wish to set up the knock tables.

[Six_speed_demon]

Sarcasm aside, you are making many false assumptions and are panicking (failsafe? - it's a compensation). You can get it to work given the existing behavior of IAT compensation. In fact, the stock behavior is better than the altered behavior for the 2006 Civic - do you want the ECU to retard with temperature under conditions where it is close to knock, or over the whole table, even if the ignition is nowhere near knock? The first type of behavior is what the stock ECU uses. The second behavior has been requested since people don't understand or wish to set up the knock tables.

Ok I get what your saying. And we all understand that. But there is still things like being able to run a IAT sensor post blower so that I can see what the actually temps are that are going into the motor. That is a nice option to have and to me give the option for tuners to go more aggressive with the tune. I'm trying to be a asshole or a dick but I'm just saying, some of nice features that the 8thgen have that I thought the 9thgen would have had by now that's all I'm saying. This platform is really beginning to take off and as more ppl beginning to the push the motor harder harder we are going to need those small option and other compensation that's all I'm saying. I know this stuff takes time and testing and what a lot would like to know is it even being worked on. That's all that's it. That's for the above answer.

[VitViper]

Read my previous reply. The ECU will compensate ignition for IAT as long as you set the knock ignition limit tables. You don't need knock for it to compensate.

Really? I have our S/C 9th gen here, and K.Control sits at 23%. Under every situation I had it under: 92 octane, E85, no knock, lots of knock (confirmed to be knocking on the dyno even). K.Control sits at 23%. Even under heavy detonation during a WOT pull K.Control simply did not budge. A lot of good that does me in real time -- the IAT can soar and with no K.Control response so much for the KIL tables.

And yes, even on the 9th gen the blowers still do generate some false knock, I have no interest in the ECU fighting my settings based on blower noise.

Or the flip side: such noisy motors that I've seen KIL sitting at 150% (think internals, cams, valvetrain). Really useful knock feedback there, again.

or wish to set up the knock tables.

You're right, I have no interest whatsoever in setting up knock tables for heavily modified cars, where in many situations the knock sensor is totally useless. You know how many cars I've heard audible knock on with not even a blip from the knock sensor? With OEM knock sensitivity settings even. Once the motor moves away from the OEM parameters the knock sensor was calibrated for, it's usefulness is almost zero (or just zero). Stroker AP2 on FlashPro that I tuned while in Arizona -- audible knock, knock sensor doesn't see it. Yes I even spent time calibrating the knock sensitivity tables and KIL to see if it would register any knock -- not even a blip.

Just because Honda set it up one way for an OEM engine does NOT in any way imply it is "superior". Honda by far has the crappiest knock system out of any OEM ECU I've worked with. For example, Ford & Mazda's knock systems are infinitely superior, and actively hunting for the real ignition limit, adding or removing timing actively as necessary. The knock sensor calibrations in the ECU are actually worth the time invested to calibrate. But even in situations with upgraded internals even their knock systems start to lose value unless you can tune the knock window and frequency (and Ford's ECU even have some of that available to me).

And guess what... they still do IAT compensations against the tuned ignition tables. So do most OEM's... so does every aftermarket ECU I've run into, to date.

Example: http://www.youtube.com/watch?v=-RmMrtS2sZY Nice detonation on the "final stage 1 basemap tune". So much for those KIL tables, made my ears hurt listening to that car cry.

[nurv2600]

Really? I have our S/C 9th gen here, and K.Control sits at 23%. Under every situation I had it under: 92 octane, E85, no knock, lots of knock (confirmed to be knocking on the dyno even). K.Control sits at 23%. Even under heavy detonation during a WOT pull K.Control simply did not budge. A lot of good that does me in real time -- the IAT can soar and with no K.Control response so much for the KIL tables.

And yes, even on the 9th gen the blowers still do generate some false knock, I have no interest in the ECU fighting my settings based on blower noise.

Or the flip side: such noisy motors that I've seen KIL sitting at 150% (think internals, cams, valvetrain). Really useful knock feedback there, again.

Indeed! I've noticed since...version 1.5.x maybe?...that my K.control also just sits at a random value determined when I start the engine. It's usually around 50%, but either way, going from a cold start and idling for a while, all the way through a 30-minute track session, it sure as hell shouldn't sit still!

Plus, if you guys are so adamant about using these KIL tables (of questionable usefulness), why is it that you saw fit to implement the IAT comp tables in the 8th gens?? I want to get the CT-e S/C, but with the current ECU logic, it seems that could be a downright DANGEROUS idea for somebody who tracks the car! If Vit's getting major detonation issues with just single WOT pulls, I can't image what might happen doing WOT after WOT for hours on end, with no way to get live feedback. And NO, I'm not just going to de-tune the car to the point where it simply CAN'T knock because the timing has been retarded all to hell.

[VitViper]

Really? I have our S/C 9th gen here, and K.Control sits at 23%. Under every situation I had it under: 92 octane, E85, no knock, lots of knock (confirmed to be knocking on the dyno even). K.Control sits at 23%. Even under heavy detonation during a WOT pull K.Control simply did not budge. A lot of good that does me in real time -- the IAT can soar and with no K.Control response so much for the KIL tables.

And yes, even on the 9th gen the blowers still do generate some false knock, I have no interest in the ECU fighting my settings based on blower noise.

Or the flip side: such noisy motors that I've seen KIL sitting at 150% (think internals, cams, valvetrain). Really useful knock feedback there, again.

Indeed! I've noticed since...version 1.5.x maybe?...that my K.control also just sits at a random value determined when I start the engine. It's usually around 50%, but either way, going from a cold start and idling for a while, all the way through a 30-minute track session, it sure as hell shouldn't sit still!

The ECU determines what fuel you are using, that, in summary, is the extent of the OEM Honda knock logic. On other vehicles I have noted as the K.Control value moves up and down depending on noise detected by the sensor (heavy valve train, aftermarket forged internals, etc, will all make the knock sensor angry).

Plus, if you guys are so adamant about using these KIL tables (of questionable usefulness), why is it that you saw fit to implement the IAT comp tables in the 8th gens??

Because we asked for it, and yes, it is superior to the KIL table method. There are dozens upon dozens of 8th gens I have tuned where the KIL logic is useless due to the modifications done to the motor. We have cars with cams and upgraded valvetrain that just makes the knock feedback go bonkers. I even have had wastegate chatter on my car during onset of boost control set off my knock sensor.

I want to get the CT-e S/C, but with the current ECU logic, it seems that could be a downright DANGEROUS idea for somebody who tracks the car! If Vit's getting major detonation issues with just single WOT pulls, I can't image what might happen doing WOT after WOT for hours on end, with no way to get live feedback. And NO, I'm not just going to de-tune the car to the point where it simply CAN'T knock because the timing has been retarded all to hell.

I am not getting major detonation issues, if you want to go S/C we won't have a problem doing a reliable map for the car. Our 9th gen is our R&D car, it's a car I can do things to that I cannot/will not do to customer's cars. As such I have intentionally pushed it into detonation in a controlled environment to see how the ECU parameters respond. The lack of this feature shouldn't prevent you from enjoying your car. My goal is to bring more features/options to how we can use the software (this is an OPTION on the 8th gen, you can use both methods if you so chose) so as to make the software BETTER. I cannot think of a single feature request I have ever put in with Hondata that has not been met with some kind of friction. I completely understand that they are busy, but this gives me the sense that no one is interested in the requests or improving the software with more features. It is actually exhausting to even bother asking for anything after years of this kind of back and forth and I really am getting to the point where I simply won't care if the software gets anything new/useful.

Every time I think about putting in a feature request I stop myself and ask "what kind of excuse am I going to hear this time? Or is it going to be the typical excuse of Honda did it this way so we think it's superior". Sorry if this offends anyone at Hondata, but in the years we've been going back and forth on feature requests, you know it's true. You guys have come through on quite a few things I've asked for, and it's appreciated, but it really has been met with some sort of friction in one way or another before we saw any progress.

[Hondata]

It pains me to see a post like this when we have generally given people what they asked for if it was at all possible. We are not ignoring you - the 2012 ECU is an order of magnitude more time consuming to work with than the 2006 and the there is a long lead time in the development process. I'll explain.

1. Development time. Currently there are two changes in the development pipeline which have been there for over a year, and they will have to be completed first. This may give you an indication how long it takes to develop, test and roll out a change. I would not say that the two changes I made as per Vit's requests were done because of the yelling and screaming - that was more an indication of not understanding how long it can take to make a change and not giving us a chance to implement it before making a lot of noise.

2. Development effort. The 2012 ECU is about an order of magnitude more time consuming to make a change than the 2006 ECU (and that is an order of magnitude higher than the KPro). Just finding out if the change is possible can take a hundred hours or so. At some point the change may not be economical to make at all, so I can't even guarantee that a enhancement to the 2006 ECU is possible on the 2012 without sinking in that 100 hours first. And it's not like we're here twiddling our thumbs for work to do.

I often point out if the ECU uses factory so that you can distinguish between a bug and an enchantment. It's not an excuse.

At this point I'm going to close the thread as non-constructive. If Vit wishes to throw more brickbats this way you can email me.