Performing Honda K24a3/AST5 6spd Conversion

[dcioccarelli]

Looks like P2185 (sensor 2) is NOT the main sensor in the head - it's a second sensor in the rad. Since I don't have this, I need to figure out where it would be wired & either add one or bypass/fool the system. I don't recall seeing the wiring for that one.

EDIT:

E12 is the ECU signal input/ground for the cooling fan. The manual describes it as a "switch', not a 'sensor'. Not sure what to do with this, and if it could even be related to the rich condition.

View attachment 28761

Hi Hussein,

sorry, I don't understand the link between the P2185 issue and the E12 connection, but in terms of the latter the circuit indicates that

 

[lookforjoe]

OK. So, on my AC box, I seem to have the heater core I/O both on the right side. The feed hose comes up behind, goes through the heater valve, and into the top fitting. The lower fitting feeds into the lower hard pipe & back to the motor. I don't see any connection besides that. Don't see any removable flange that would have a gasket.

I fashioned a "U" pipe to bypass the core for now, so I can figure out the base tuning.

 

[lookforjoe]

Hi Hussein,

sorry, I don't understand the link between the P2185 issue and the E12 connection, but in terms of the latter the circuit indicates that

The code is for this rad sensor - not for the fuel trim ECT. I found there is a way to disable the second ECT in the software, I guess because it's also not used on many swaps. With that done, the code is gone.

Can't figure out the MAP issue yet. Fixed the idle part with the TPS mount mod. Was able to set the reading to 0%.

Still have the idle surge & AFR sweep from 10-22(!), so I'm assuming I can't move forward until I resolve the MAP problem. The P0102 code can be disabled also, but I don't see how that helps. It was suggested to use Speed Density, but in researching that, it uses MAP anyway, not MAF.

screenshots from the logging. On;y ran it for a minute, don't want to screw the motor.

Since I'm now stuck Until I figure this out, I carpeted the trunk panel I made

 

[kmead]

OK. So, on my AC box, I seem to have the heater core I/O both on the right side. The feed hose comes up behind, goes through the heater valve, and into the top fitting. The lower fitting feeds into the lower hard pipe & back to the motor. I don't see any connection besides that. Don't see any removable flange that would have a gasket.

I fashioned a "U" pipe to bypass the core for now, so I can figure out the base tuning.

The hoses connect to these behind the cover: https://www.midwest-bayless.com/p-9...re-elbow-wac-fiat-bertone-x19-1975-88-u8.aspx

On a non AC car the connections are not as hidden. This is why the whole mess has to come out to replace the heater core.

 

[TonyK]

OK. So, on my AC box, I seem to have the heater core I/O both on the right side. The feed hose comes up behind, goes through the heater valve, and into the top fitting. The lower fitting feeds into the lower hard pipe & back to the motor. I don't see any connection besides that. Don't see any removable flange that would have a gasket.

I fashioned a "U" pipe to bypass the core for now, so I can figure out the base tuning.

That is the location, why is it wet there?
Looking at the pictures, it appears that the core must come out to change the gasket under the 90 degree fittings. Try unsnapping the 4 clips on the sides of the heater box, you may be able to see from there the fittings and if it is leaking. I am sure Bob Martin changed the gaskets in the car when he was at my house.

TonyK.

Grimsby Ontario Canada.

 

[lookforjoe]

That is the location, why is it wet there?
Looking at the pictures, it appears that the core must come out to change the gasket under the 90 degree fittings. Try unsnapping the 4 clips on the sides of the heater box, you may be able to see from there the fittings and if it is leaking. I am sure Bob Martin changed the gaskets in the car when he was at my house.
TonyK.
Grimsby Ontario Canada.

That's just silicone spray, to help loosen the hoses off the nipples, there was no coolant up there I'll try doing the lower case snaps as suggested.

 

[lookforjoe]

The code is for this rad sensor - not for the fuel trim ECT. I found there is a way to disable the second ECT in the software, I guess because it's also not used on many swaps. With that done, the code is gone.

Can't figure out the MAP issue yet. Fixed the idle part with the TPS mount mod. Was able to set the reading to 0%.

Still have the idle surge & AFR sweep from 10-22(!), so I'm assuming I can't move forward until I resolve the MAP problem. The P0102 code can be disabled also, but I don't see how that helps. It was suggested to use Speed Density, but in researching that, it uses MAP anyway, not MAF.

screenshots from the logging. On;y ran it for a minute, don't want to screw the motor.

Since I'm now stuck Until I figure this out, I carpeted the trunk panel I made

This Idle surge/AFR sweep is really bugging me. I can't find anything remotely definitive on this, which make sit difficult to know what direction to take. The P0102 is a MAF code, not a MAP code - and my setup doesn't use MAF. so I may be going in the wrong direction trying to figure this out. I'll double check the MAP signal & + - later today. I did also order a new MAP to try. Just frustrating when I don't know where to search for guidance with this

EDIT: just found a post regarding the P0102 code & Ktuner:

So, I do need to disable the MAF setting. Don't undertsand why he didn't just tell me that in my thread I guess that means looking elsewhere, unless having that code causes the EMS to take a brain fart

 

[dcioccarelli]

The code is for this rad sensor - not for the fuel trim ECT. I found there is a way to disable the second ECT in the software, I guess because it's also not used on many swaps. With that done, the code is gone.

Sorry: looks like I accidentally posted before finishing what I was writing (I got distracted!) . I just wanted to point out that based on the circuit diagram you supplied E12 is definitely an output from the ECU and is used to ground the relay operating the radiator fan. It looks as though the radiator fan can either be turned on by the radiator fan switch (in the radiator) or the ECU. Moreover, there are two fan relays being controlled: one for the radiator fan and the other for the A/C condenser fan.

So, just to be clear: there is no radiator sensor in the circuit diagram you provided: just a ECT sensor in the head.

Cheers,
Dom.

 

[lookforjoe]

Sorry: looks like I accidentally posted before finishing what I was writing (I got distracted!) . I just wanted to point out that based on the circuit diagram you supplied E12 is definitely an output from the ECU and is used to ground the relay operating the radiator fan. It looks as though the radiator fan can either be turned on by the radiator fan switch (in the radiator) or the ECU. Moreover, there are two fan relays being controlled: one for the radiator fan and the other for the A/C condenser fan.

So, just to be clear: there is no radiator sensor in the circuit diagram you provided: just a ECT sensor in the head.

Cheers,
Dom.

Hey Dom - the rad sensor is in the bottom left of the diagram, they call it a switch
I called it an input because the ground at E12 is internal to the ECM, not external. I would consider an output to be a + signal, but I guess it's an output whether it's + or - I would defer to you on terminology here - you are far more versed in matters electronic than I am.

It looks like either the switch or the ECM ground at E12 could engage the (pin 4, ground) relays for AC or Rad fan, since they appear to all connect within the relay box circuit? In any event, disabling the E12 signal in the software has resolved that issue, so it is not the cause of the rich condition I am hunting down.

 

[lookforjoe]

The hoses connect to these behind the cover: https://www.midwest-bayless.com/p-9...re-elbow-wac-fiat-bertone-x19-1975-88-u8.aspx
View attachment 28797
On a non AC car the connections are not as hidden. This is why the whole mess has to come out to replace the heater core.

Now I understand Tony's point about the gasket leak vs. core

 

[lookforjoe]

Ongoing discussion with KTuner, not really getting anywhere yet. When I emailed my data to the local KTuner dealer who installed the end-user hardware in the ECU, he was confused about the codebase, as it appears to be for DBW, even though I selected otherwise. Prolly won't get more answers until next week.

You'd have codes for DBW if you loaded a DBW codebase.
Surging could be from not enough air passing, too much air passing, or the need to tune the fuel maps.

I used the 03-05 Accord (K24A4) codebase indicated in my screenshot, which doesn't say DBW like the '06-on.

I'm obviously going to have to research further for the surge, just trying to determine if the surge is causing the AFR sweep, or the other way around

I don't have any codes for idle control (P0511), which were present before I fixed the TPS setup.

The other concern with that is that my motor is the K24A3 that has higher compression, full iVTEC on int/exh like the US K24a2, etc. Maybe I should be using a different codebase altogether? I'm not familiar with Honda product to know which US product listed would have DBC AND the higher compression /iVTEC to use as a base, do you have a suggestion for that?

I'm concerned that the codebase for the accord won't be controlling the iVTEC properly?

I found these descriptions on HybridRacing, so the K24A2 seems a match, but has DBW which doesn't work for my setup:

 

[lookforjoe]

I revisited the (K-Tuned) throttle setup, based on KTuner's suggestions of possible causes. I was able to get the plate better centered in the bore & set to just a crack off closed to prevent the plate sticking when snapped shut. Reset the TPS so it registers 1% as soon as the plate moves off the stop.

Still runs poorly with the Accord base (high idle, rich AFR's), so I tried the custom tune base. The main trim is set at 8 for the 410cc injectors in this tune.

With that, AFR's are better, engine sounds better. Idle is still too high, so I removed the air inlet hose & closed off the IACV port. That dropped the idle to just under 1700, that's an improvement! There are no codes for IACV, so I'm assuming it's just stuck open.

 

[dcioccarelli]

Hey Dom - the rad sensor is in the bottom left of the diagram, they call it a switch
I called it an input because the ground at E12 is internal to the ECM, not external. I would consider an output to be a + signal, but I guess it's an output whether it's + or - I would defer to you on terminology here - you are far more versed in matters electronic than I am.

It looks like either the switch or the ECM ground at E12 could engage the (pin 4, ground) relays for AC or Rad fan, since they appear to all connect within the relay box circuit? In any event, disabling the E12 signal in the software has resolved that issue, so it is not the cause of the rich condition I am hunting down.

Hi Hussein,

They are actually correct about calling it a switch (at least, that is the circuit symbol for a switch). It will be a similar bi-metallic switch as in an X1/9 radiator. The purpose will be to ground one side of the relay coil (the other side of the coil is connected to +ve) thereby turning the relay "on". In terms of nomenclature, E12 is still an output even if it goes low to activate the load: the definition of an output is simply that it is something that performs an action on an external device (like a relay) rather than receiving information from a sensor (like a temperature sensor, e.g. B8). In this case, the E12 output goes low (to ground) to energise the fan relay or, alternatively, the bi-metallic switch in the radiator can also engage the fan.

In any case, certainly not the cause of your issues. This output will only become relevant when you start to look at the overall cooling behaviour and particularly the ECU control of the radiator fan.

Good luck with everything though! I's sure that getting the tuning correct is not going to be a walk in the park: you have so many parameters to deal with and they often interact with each other. Would the settings from the other people who have converted to a Honda motor be relevant to your application or is it significantly different?

Cheers,
Dom.

 

[lookforjoe]

Hi Hussein,
Good luck with everything though! I's sure that getting the tuning correct is not going to be a walk in the park: you have so many parameters to deal with and they often interact with each other. Would the settings from the other people who have converted to a Honda motor be relevant to your application or is it significantly different?
Cheers,
Dom.

Thanks Dom.
Settings from others would only be relevant if it's the same tuning software. I spent more time looking at the intake plumbing today, to see if the remaining high idle is related to intake leaks.

Temps in the high 30's today, so I pushed the car outside to run it. I modded the RSX tune using the TSX VTEC settings & rpm limits & uploaded to the ECU.

After that I ran the motor briefly. Sounds good & AFR's are good. Only obvious problem is high idle now. First up I pulled the IACV & found the crappy seal was cracked, and leaking. With the two ports plugged in the TB, the idle was closer to normal.

after that, I ran the motor to normal operating temp - 180 something for the K24, fan cycled.

then I sprayed around the TB & intake plenum to check for possible air leaks. I did order a smoke machine - been meaning to get one for some time - but it won't be here for close to a week.

I got an idle change when spraying the right side edge of the plenum

unbolted the compressor, undid the plenum

have to remove the support backet also, all easy to get to with the larger opening

Just have to ease the gasket out & new one in

I don't see any wear/uneven compression on the one I removed though

Put it back together, and it sounded worse! So, I took the plenum off entirely, that wasn't to bad either, just have to drop the alternator to allow removal

I found that there is a hole in the welds of the crotch between 1 & 2. I have to take the plenum back to the machine shop, that was one of the tricky areas I had him finish for me. I've also ordered one of the thermal intermediate gaskets (3mm), as I'm not confident that the skinny metal gasket is ever going to seal properly.

 

[kmead]

Bummer on the intake hole. Glad you found it now versus later with a burnt head gasket or worse.

Aaaahh the fun of development.

 

[lookforjoe]

Bummer on the intake hole. Glad you found it now versus later with a burnt head gasket or worse.

Aaaahh the fun of development.

Indeed! I'm just grateful all this can be accessed without any cursing, or the need for drivetrain dropping, now THAT would suck!

 

[lookforjoe]

Dropped off the plenum to get the weld fixed. He gave me a section of quality gasket paper so I could make my own gasket. I may or may not wait for the thermal gasket. I'll have the plenum back tomorrow.

 

[lookforjoe]

First drive Sounds good, no leaks. Brakes suck. Definitely needs to be tuned. Idle is only good with IACV port taped shut. Have to figure that one out.

Also got a P1009 code. That's regarding the VTC control circuit. Assuming it's tune related. All the mechanical elements are new

 

[PaulD]

First drive Sounds good, no leaks. Brakes suck. Definitely needs to be tuned. Idle is only good with IACV port taped shut. Have to figure that one out.

Also got a P1009 code. That's regarding the VTEC control circuit. Assuming it's tune related. All the mechanical elements are new

This is really great to see your car on the road. And it is quick! You have shown great perseverance thee many issues that have arisen. Well done!

In my experience, it took a very experienced tuner on a good dyno to get the tune right. If your engine management system has the capacity to record and play back your data, it is possible to do a "pull" by accelerating on the open road and then stop and read the data, only each variable you can change impacts all of the other variables. My tuner would change one setting, then adjust all of the rest of the settings, then do another pull. And he know what to adjust. I had no idea what to change. My system is an AEM EMS4, a "simple" systems, but not for me. an example: there are over 20 factors that can impact the idle adjustment, all interactive.

Sounds like yours is close though.

Paul

 

[lookforjoe]

This is really great to see your car on the road. And it is quick! You have shown great perseverance thee many issues that have arisen. Well done!

In my experience, it took a very experienced tuner on a good dyno to get the tune right. If your engine management system has the capacity to record and play back your data, it is possible to do a "pull" by accelerating on the open road and then stop and read the data, only each variable you can change impacts all of the other variables. My tuner would change one setting, then adjust all of the rest of the settings, then do another pull. And he know what to adjust. I had no idea what to change. My system is an AEM EMS4, a "simple" systems, but not for me. an example: there are over 20 factors that can impact the idle adjustment, all interactive.
Sounds like yours is close though.
View attachment 28964
Paul

Hey Paul, yes like you I will dyno-tune it. I've already melted enough pistons self-tuning over the years As you say, too many parameters and need to know where to look to make solid tuning decisions. Fortunately for me, the K-Tuner rep is a serious Honda fiend & local to me, and Dyno tunes. I will go to him once I have all the basics sorted out.