Honda D Series Forum banner

1 - 19 of 19 Posts

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #1 (Edited)
91 CRX Si running turboedit (chipped PM6) with RC550 injectors. At cold idle, engine runs at ~13.5:1. After warming up to about 180-190F coolant temps, the AFR will settle around 14.7:1 which is ideal. However, after reaching full operating temp (several rad fan cycles, oil pressure at it’s lowest levels) the idle AFR drifts between 16:1 - 16.8:1 and the engine will run noticeably rougher.

Previously I was using Injector Nation 600cc injectors with a bypassed resister box. After speaking with HiProfile and The_Acid_Beaver, I came to the conclusion that the PM6 ECU couldn’t compensate for the significantly slower saturated injectors during lower voltage scenario. I assumed that this was the cause and ordered some P&H RC550s and installed an unmodified resistor box. Everything seemed fine after I installed the new injectors and I assumed the issue was resolved.

That brings us to today. Another member messaged me saying he ran IN600s with turboedit and had no issues. That prompted me to do some additional testing, at which point I found that the issue still occurred with the RC injectors.

Soooo...I’m back at square one trying to figure this out.

-I’ve already tested the ELD per the FSM and it checked out.
-Battery voltage during both normal and lean idle conditions is 14.2-14.4v.
-Fuel pressure during both normal and lean idle conditions is ~40 PSI.
-Alternator was replaced a few years back with a reman Denso unit. I’d guess less than 100 miles and 10 hours of run time with this alternator.
-Engine wiring harness is new and unmodified.
-I’ve installed new OEM battery cables (positive and negative) and a new OEM valve cover ground wire.
-Datalogging shows similar injector pulse width during both conditions.

I’m really at a loss. The only thing I can think of is that the voltage regulator on the alternator isn’t functioning properly after it has heat soaked.

Looking for any ideas, suggestions, or outside the box thinking.

Is there a way to directly test the alternator output voltage?

Thanks guys
 

·
Registered
Joined
·
416 Posts
What are your IATs when it is heat soaked and idling vs. just after it has initially warmed up? I recall a friend of mine with a JRSC that had to setup his with IAT correction? Going to have to go find my copy of TE and recall the settings it had...
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #3
I start pulling timing after 110 degrees on the IAT.

When this issue happens, the IATs are between 80-95F so no ignition adjustments are performed.
 

·
Registered
Joined
·
830 Posts
does this use iat fuel corr?

plastic or metal iat?
iat location.

alternator direct testing? multimeter on alt stud can't get much more direct...
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #5
Plastic OEM IAT mounted to the manifold. No IAT fuel corrections.

Unfortunately I can’t reach the alternator stud due to the supercharger. Maybe I can trace it up further and test at the under hood fuse box.
 

·
Registered
Joined
·
830 Posts
so no iat fuel correction happens.?

I would say try moving the iat sensor to the intake tube but probably a waste of time..
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #7 (Edited)
Correct, no IAT fuel compensation. I only pull spark with IATs above 110F. IATs are never above 100 when the issue occurs.

I’m going to go outside and do some datalogging /testing right now.

And I can’t move the IAT. Engine is supercharged so it has to be in the manifold to read boost.

Looking over some previous data logs, it looks like the ELD was reading a high load at points with a voltage of .75v. Seems awfully high...like something is pulling a lot of amps. I’m doing a 1 sec datalog session with a cold start all the way until the issue occurs. Will see if there’s any noticeable difference.
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #8 (Edited)
Over the weekend I datalogged in 1 second intervals for about 30 minutes. The intercooler pump and fan were off during this time.

-I created a new turboedit map from NG63_final_(5Q).BIN with the latest ET Handler script. I removed ALL compensations from ET so there's absolutely no ignition or fuel corrections occurring.





-My idle fuel/spark map:





-Here are some datalog snippits from 3 different points while warming up:

Code:
Time	RPM	MAP	A/F	TPS	VSS	IAT	ECT	ELD	Baro	Row	Col	Gear	Pulsewidth	Ign. Adv.
30.156	889	-18.07	13.51	1	0	96	158	2.75	14.45	3	3	0	3.21	18
31.171	876	-18.07	13.44	1	0	96	158	2.75	14.45	3	3	0	3.17	18
32.171	873	-18.07	13.52	1	0	96	158	2.67	14.45	3	3	0	3.22	18
33.171	896	-18.07	13.48	1	0	96	159	2.59	14.45	3	3	0	3.16	18
34.171	874	-18.07	13.33	1	0	96	159	2.73	14.45	3	3	0	3.21	18
35.187	882	-18.07	13.27	1	0	96	159	2.73	14.45	3	3	0	3.21	18
														
300.343	886	-18.26	14.27	1	0	98	184	2.71	14.45	3	3	0	3.09	18
301.343	880	-18.26	14.46	1	0	98	184	2.75	14.45	3	3	0	3.1	18
302.343	869	-18.26	14.35	1	0	98	184	2.69	14.45	3	3	0	3.11	18
303.343	872	-18.26	14.26	1	0	98	184	2.76	14.45	3	3	0	3.1	18
304.343	880	-18.26	14.35	1	0	98	184	2.75	14.45	3	3	0	3.09	18
305.343	879	-18.26	14.42	1	0	98	184	2.73	14.45	3	3	0	3.11	18
														
1270	872	-18.07	15.26	1	0	115	205	2	14.45	3	3	0	3.12	18
1271	899	-18.26	15.36	1	0	115	203	2.75	14.45	3	3	0	3.08	18
1272	898	-18.46	15.64	1	0	115	203	2.67	14.45	3	3	0	3.06	18
1273	885	-18.46	15.6	1	0	115	203	2.67	14.45	3	3	0	3.09	18
1274	880	-18.46	15.54	1	0	115	203	2.76	14.45	3	3	0	3.03	18
1275	869	-18.46	15.43	1	0	115	203	2.65	14.45	3	3	0	3.04	18

As you can see, during three different intervals, the AFR varies between 13.27:1 - 15.64:1, despite the same row/column being referenced (3,3).

If you look at the injector pulse widths, you can see that there's a decrease of 3-6% despite the same row/column being referenced.

I'm really not sure what to make of it. Obviously the ECU is sending shorter pulse widths and I feel that 3-6% reduction in fuel is enough to account for the difference in AFR. I just can't understand why the ECU would be sending shorter pulse widths. I even made sure that the adjacent cells used the same spark/fuel values.

I may reach out to Joe and see if he has any input but honestly I don't want to bother the guy over software that he stopped developing years ago.

Any thoughts/ideas? Am I missing anything?

I may just try bumping the fuel up 3-6% in 2/2-4/4 and deal with the overly rich idle. Maybe there’s something hard-coded into the software that is still adjusting fuel based upon ECT.
 

·
Registered
Joined
·
463 Posts
Do you have a spare ECT sensor? I'd be interested if you still have the leanout issue if you disconnected the electrical connector from the block and connected it to the spare sensor which you just keep hanging in the air reading ambient temps.

The fan switch will keep the temperature modulated and you can see if the ECT reading is the source of your problems.

If all else fails.... Can you just run a wideband sensor? Bit of a band-aid solution...
 

·
Brokedick Millionaire
Joined
·
40,228 Posts
Turn the fan on and watch ECT vs AF as it cools down....or better yet, just use the interior heat to modulate different temps.

I know Blundell found a coding error in OEM Honda setup which might explained why I was always getting stupid great gas mileage. Basically if TPS voltage doesn't change, but is in the good range, the ECU keeps trying to lean everything out while cruising.

I was always one to not move the gas pedal as much as possible when driving. But once the TPS would change a little (slight lift for traffic) the ECU would go back to base tuning.

One trick I want to try before this was known was using a servo or cruise control unit that didn't adjust based on MPH setting. You would set it by percentage of throttle opening you wanted. Blundell thought it would be worthless.

Then again he didn't think shutters in the grill opening mattered much either. How many cars have them now from the factory?

https://www.greencarreports.com/news/1079678_active-shutter-grille-vents-how-they-help-improve-mpg
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #11
I don’t have a spare ECT unfortunately. I am already running only a wideband. The OEM 02 sensor is disabled in TE so its only running open loop.

Bone, I like the idea of modulating the ECT with the heater. I don’t have the rad fan wired to a switch (prob a future mod I’ll do if I can find your wiring diagram again). I bumped the fuel in those low cells and I’m letting it idle outside right now. Once it’s completey warmed up, I’ll turn the heater on full blast and see how much I can get it to drop the ECT.

Thanks for the quick feedback guys. This issue has been driving me nuts!
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #12 (Edited)
OK so I wasn't able to really make any drastic impact to the ECTs with the heater on full blast but I don't think that was the issue.

I ended up adding some fuel (around 6%) to the idle cells which seems to have had a positive effect on the AFR.

BEFORE:

Code:
Time	RPM	MAP	A/F	TPS	VSS	IAT	ECT	ELD	Baro	Row	Col	Gear	Pulsewidth	Ign. Adv.
11624	869	-18.07	15.52	1	0	94	203	1.92	14.54	3	3	0	3.1	18
1625	869	-18.07	15.42	1	0	94	203	1.92	14.54	3	3	0	3.1	18
1626	869	-18.07	15.55	1	0	94	203	1.92	14.54	3	3	0	3.1	18
1627	869	-18.07	15.6	1	0	94	203	1.92	14.54	3	3	0	3.1	18
1628	869	-18.07	15.23	1	0	94	203	1.92	14.54	3	3	0	3.1	18
1629	869	-18.07	15.32	1	0	94	203	1.92	14.54	3	3	0	3.1	18

AFTER:

Code:
Time	RPM	MAP	A/F	TPS	VSS	IAT	ECT	ELD	Baro	Row	Col	Gear	Pulsewidth	Ign. Adv.
563.64	875	-18.85	13.94	1	0	104	205	2.65	14.54	3	3	0	3.09	18
564.828	875	-18.85	14.1	1	0	104	205	2.65	14.54	3	3	0	3.09	18
566.015	875	-18.85	13.91	1	0	104	205	2.65	14.54	3	3	0	3.09	18
567.187	875	-18.85	13.91	1	0	104	205	2.65	14.54	3	3	0	3.09	18
568.375	875	-18.85	13.95	1	0	104	205	2.65	14.54	3	3	0	3.09	18
569.562	875	-18.85	14.02	1	0	104	205	2.65	14.54	3	3	0	3.09	18

It seems like there's definitely some behind-the-scenes stuff going on in TE. The pulse widths in the datalogs never seem to directly correlate to the values from the fuel map. There must be some sort of base pulse width or TE can't accurately log the actual pulse widths. Case in point, the pulse widths from both of the above datalog snippits are nearly identical but there is 4% fuel added in the 2nd snippit and a noticeable change in AFR.




So I'm not entirely sure what is going on but for now, the addition of fuel in those cells seems to have fixed it. I'll do a complete test tomorrow to confirm.
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #15 (Edited)
I just like the idea of the engine running off of a set of static variables. I don’t want the ECU try to compensate for things, thinking it knows best.

Also, there’s only one bung in my header and running the LC1’s narrowband signal to the ECU means more wiring.

I don’t care about gas mileage and I don’t see any other benefit to running closed loop
 

·
Registered
Joined
·
463 Posts
I just like the idea of the engine running off of a set of static variables. I don’t want the ECU try to compensate for things, thinking it knows best.

Also, there’s only one bung in my header and running the LC1’s narrowband signal to the ECU means more wiring.

I don’t care about gas mileage and I don’t see any other benefit to running closed loop
At least you've got options if you can't get this one figured out.
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #17 (Edited)
Thought I had figured this out but no dice. I added some fuel in these two idle cells andI got the car to end up at 14.7:1 after like 40 minutes of idling so I figured the issue was resolved. However, turning on the headlights and heater threw everything out of whack and the AFR drifted to like 16:1 again.

I checked the injector voltage at the resistor box and it’s 14.1v so I’m confident that there’s not a voltage issue that’s causing this.

Beginning to think this is something that turboedit is doing. I’m emailing with Joe to try to get a license for tuner express. Looks like you can enable and disable the ELD with tuner express so I may play with that when I get the license.

I’m also going to see if I can find another under hood fuse box to test with. The ELD voltages I see at the ECU always look correct and the ELD tests fine but I suppose it’s worth checking.
 

·
Registered
Joined
·
830 Posts
sounds like your having a good time with this..

I don't use obd0 and never played with TE ..
try a previous version of the code?

fuel pressure change ,funny regulator?

like you said probably ecu enter problem is to blame..
 

·
Meat Popsicle
Joined
·
2,922 Posts
Discussion Starter #19
The problem is, older versions of the NG63 bin have their own bugs, one of which was a false rev limiter that I encountered. However, you do make a good point which I haven’t considered. When I was running the DSM 450s, I think I was still using an older bin file that has the false rev limiter. I’m actually waiting to hear back from another member who says he’s running turboedit with the same injectors and is not having this problem.

If he sends me his bin file, I’ll try copying my tables into it to see if it’s related to the NG version I’m using.

Thanks! Hadn’t even considered that since Joe wasn’t aware of any issues that would cause this.

Regarding fuel pressure, I do have a gauge and the values are always consistent. I’ve even clamped the fuel return to up the pressure and it didn’t have an effect on the issue I was seeing.
 
1 - 19 of 19 Posts
Top