This is a 10:1 motor, 222/230 cam at .050", tfs alum heads and miniram.
Small blower. $8D AXCN, open loop.


I can't get the thing to idle and its driving me crazy. I used to have the comp version of this cam and it idled just great.

I'm seeing around 60kpa at idle, around 11hg on my vac gauge. I have ALL the idle based sa stuff zero'd out so I'm running on just the closed tps SA.

Reset min air a dozen times. I've tried leaning it out, richening it up, idling at 20*, idling at 22*, idling at 24*.

I've even lowered fuel pressure to 40psi idle 45 wot to make sure I wasn't fighting an erratic pulsewidth.


It will idle fine at the commanded rpm of 900 for a few seconds at about a 1.7msec pulse width, with lets say 24 degrees total sa.

All of a sudden RPMS will clime and hold at about 1200-1300, then shortly after fall down. It appears just before RPMs clime the pulse width goes down to about 1.5, then when idle comes back down pulsewidth climbs back up to around 1.7..

Spark doesn't change - I zero'd out all the RPM error and multiplier tables because I didn't want any spark correction. The only thing that appears to change is the pulsewidth and RPM (target idle says 900)

Checking plugs, they are tan and look clean, though 02 mv is in the 700-800 range. I'd imagine if it was rich the plugs would be wet or sooty.

IAC is 160 steps parked, and idles around 140-150 steps.

I'm out of ideas. Naturally when idle goes up to 1200-1300 vac increases to about 13-15hg. When idling at the commanded 900rpm vac is around 11hg on the gauge, and 60kpa in the log.

I did a compression test sunday and saw 165-170psi in every cylinder, so I'm at a loss. New plugs, wires all seem fine.

-- Joe

 

Why so many steps? That's WAAAAAY too many IMO. I run 0 steps in park and about 25 in drive. Try opening the throttle blades until you get between 0 and 10 steps. In my experience, the more I took the ECM out of idle control, the better the results I got.

 

I actually backed off the throttle blades because it would idle higher and the ECM would try to command it down with RPM error. I zerod out the RPM error spark retard table first, but then I was still left with the "it idles too high".

I've always gone by this procedure:

Ground the ALDL, key on. (CE flashes, fans enable)
Disconnect iac
Key off, remove jumper
Start engine and adjust blades until engine is idling around desired idle.
Shut off engine and reconnect IAC.

This starts me off with 160 steps, and idles around 150.


-- Joe

 

Hmmm.... all I did was set my desired idle speed in the calibration, then open the throttle blades, start the engine and see what the IAC step count was. If the steps were too high, I shut the engine down, open the blades a little more. I repeated the process until I got it where I wanted it.

I think the process you're describing is for people who can't reprogram their own chips and want to adjust their idle speed.

 

Joe, FWIW, I idle around 900RPMS and my commanded steps are only like 35'sh...I can get them to go down to almost nothing around town, cruise. Stupid question but how's the min air (idle) screw on your throttle body? Are the blades mistakenly closed all the way?

 

I would recheck the wiring to the IAC. Compare the year of the harness against the year of the IAC connector. Something seems fishy there.
I had mine wired incorrectly after blindly reconnecting the 95' style TB to my older style wiring of the 87'. Connections are not A=A, etc and could be inverted on one coil or both.
Drastic measure would be to take it out and confirm the direction is correct.

 

Joe,

I agree with what others have posted. Get the counts down first by opening the throttle and resetting the tps, then work on the mixture. Spark and the PID control gains also come into play.

With my present settings, I'm idling now at 800 rpm with 26 degrees advance at about 0-5 counts warm with the fans off, and about 10-20 counts with both fans running at an 80% duty cycle.

I can also idle up to about 1800 rpm at 145 counts if desired, or anywhere in between really. The IAC has quite a bit of authority.

By opening the throttle, you can minimize the undershoot and unless you've cranked up the underspeed gains it shoudn't overshoot too much. This should help.

It sounds like you're somehow setting the min air with the IAC fully open instead of closed.

Also sounds like you have too much fuel. With 23# injectors at 54 psi, I used to idle at about 1.6 ms. Now tuned for my 42# injectors at 46 psi, I only idle around 0.8-0.9ms. 30# injectors should fall somewhere in between.

 

I'll give this a whirl tomorrow and see where I'm at.
----------
Harness is '87, IAC came off the original motor. It's repinned to run a '730 ECM so I can't see why there would be an issue. ?

 

Hrmm. So you pretty much have your advance locked in from idle to wot, with higher advance just in the cruise and decel areas huh?

Why do the counts go up when the fans come on?

I think a lot of the problem is the lack of an idle circuit in the miniram. I can try opening the blades up again and see what happens.

Isn't that the factory procedure? They call it setting "minimum air", so when you ground the aldl it should fully extend the pintle blocking the IAC ports, then you set the blades to get your desired idle - so the iac could deliver more air if needed ?

How is that possible? I've had some really long conversations the past two days (like, 2 hours on the phone each) with Rich at Cruzin Performance and he doesn't know of an injector that is reliable under about a 1.4 msec pulsewidth. ???

-- Joe

 

Because there's a higher load on the engine coming from the alternator. You can look at it this way: The fans draw say, 30 amps. At say, 13 volts, Power= current x voltage = 30 x 13 = 390 watts = .52 hp. So when the fans come on, the alternator is sucking up 1/2 hp off the engine, which would drop the idle speed if it weren't for the ECM attempting to keep it constant.

EDIT: This is assuming a 100% efficient alternator (which is impossible). I reality, you're losing more hp due to losses within the alternator itself as it converts mechanical energy to electrical energy. I.e., if it were only 85% efficient, you'd actually be losing more like .52/.85 = .62 hp.

It's similar to the A/C clutch kicking in.


I solved that problem.

https://www.thirdgen.org/forums/diy-...tml?highlight=

 

I see the point. I'll re-adjust things and see where I'm at.



I read that. Interesting. I'm actually selling the miniram. I have a holley singleplane manifold, and will be running a sideways doghouse/elbow with an LS1 throttle body hanging over my valve cover. That will reduce the amount of 90* bends from 5 down to 2 in my plumbing from the blower to the intake.

Just trying to find the right elbow. Edelbrock makes one but it's only good for 400hp. I contacted "intakeelblows.com" but the owner wants me to sketch out like a cad drawing of what I want. I told him 4.5" max height, squarebore, ls1 throttle body, and no more than 8" center to TB flange. Guess that wasn't enough info.

-- Joe

 

Here's yesterday's spark table (I just bumped it to 26 degrees for high load at 4800 rpm from a previous 24:

Spark Advance Table
RPM Load Variable LV8
32.00 48.00 64.00 80.00 96.00 112.00 128.00 144.00 160.00 176.00 192.00 208.00
4800.00 40.08 40.08 40.08 48.16 48.16 48.16 47.11 42.89 39.02 35.86 35.86 26.02
4400.00 40.08 40.08 40.08 48.16 48.16 45.00 43.95 41.84 39.02 37.97 33.05 22.85
4000.00 40.08 40.08 40.08 48.16 48.16 45.00 43.95 41.84 39.02 37.97 31.99 20.04
3600.00 40.08 40.08 40.08 48.16 46.06 43.95 41.84 40.08 37.97 35.16 31.99 20.04
3200.00 40.08 40.08 40.08 46.06 45.00 42.89 41.84 41.13 36.91 34.10 31.99 20.04
2800.00 40.08 40.08 40.08 48.16 46.06 43.95 43.95 42.89 39.02 35.86 33.05 20.04
2400.00 40.08 40.08 40.08 48.16 46.06 45.00 43.95 42.89 39.02 35.86 31.99 22.15
2200.00 40.08 40.08 40.08 48.16 47.11 43.95 41.84 40.08 40.08 35.86 30.94 23.91
2000.00 35.16 40.08 40.08 46.06 46.06 41.84 40.08 40.08 39.02 35.86 30.94 23.91
1800.00 31.99 37.97 37.97 46.06 41.84 40.08 37.97 35.86 35.86 34.10 28.83 22.85
1600.00 29.88 35.86 35.86 43.95 41.84 41.13 37.97 34.10 29.88 26.02 23.91 22.15
1400.00 26.02 34.10 34.10 41.13 39.02 39.02 35.86 33.05 29.88 23.91 20.04 20.04
1200.00 26.02 31.99 34.10 40.08 40.08 37.97 35.86 31.99 29.88 22.85 16.88 16.88
1000.00 26.02 31.99 31.99 35.86 35.86 34.10 34.10 28.13 26.02 20.04 14.06 14.06
800.00 26.02 26.02 26.02 26.02 26.02 26.02 26.02 26.02 26.02 17.93 14.06 14.06
600.00 26.02 26.02 26.02 26.02 26.02 26.02 26.02 26.02 26.02 16.17 14.06 14.06
400.00 26.02 26.02 26.02 26.02 26.02 26.02 26.02 26.02 26.02 16.17 14.06 14.06

Increased alternator load requires more air for same rpm.

Yes, but your high counts suggest something is wrong.

They're Lucas 42s the so called "009" injectors. All I had to do was reduce the voltage offsets and short pulse width offsets to get them to work well at the low pulse widths. Here's what I have now for settings:

Injector Pulse Width Correction vs Battery Voltage
VDC usec added
(null)
25.6 0.00
24.0 0.00
22.4 0.00
20.8 0.00
19.2 0.00
17.6 0.00
16.0 0.00
14.4 0.00
12.8 0.00
11.2 91.55
9.6 213.63
8.0 305.18
6.4 396.73
4.8 396.73
3.2 91.55
1.6 91.55
0.0 91.55

Low Pulse Width Injector Offset
msec msec added
(null)
3.90 0.000
3.66 0.000
3.41 0.000
3.17 0.000
2.929 0.000
2.685 0.000
2.44 0.000
2.197 0.000
1.95 0.076
1.708 0.122
1.46 0.183
1.22 0.229
0.976 0.275
0.732 0.305
0.488 0.336

 

Maybe I'll change the IAC or test the wiring. 150 counts should be what, damn near fully extended right?

Yeah but... Your final pulsewidth is actually below 1.4... That isn't supposed to work.. I've been told, on a flow bench, they get all wacky.

-- Joe

 

Maybe I will have time later to read this in full to see what is going on.

1) The 730 ECM V8 code/bin doesn't report the real BPW sent to the injector. It reports a shorter BPW than what goes to the injector. That is why an ALDL BPW of 0.9ms seems to run an engine OK..........because it probably is 1.4 - 1.7ms depending on the tables.
I find it interesting when 730 ECM users report injector duty cycle when it seems high like in the 90-ish % range.........because in reality it is actually much higher than that.

2) Lots of MiniRam users have had idle problems. Your problem may just be a VE table problem. Flatten the VE table out in idle MAP and RPM areas over a 20 KPA and 600 RPM range to see if it helps. It sounds like it hits a leaner map/rpm VE area and the idle goes up because it is lean.

EDIT: One other thing. JP86SS mentioned the wiring differences between IAC (stepper motor) pin-out. That is a real thing and confuses a lot of 730 / 749 swap users. I run a 1995 LT1 TB with the 730 ECM and AUJP code.

 

Ohh.. That's great. So we're actually going static way before that. Is there any way to see the actual PW through aldl?


I'll try try that tomorrow when I get home. I flattened everything out between 50-70kpa, from 400-1400RPM and it didn't seem to help but I can give it a stab again.

I dunno. I repinned the '165 ecm connector to the '730... How would I even go about testing if it's going in the right direction..

-- Joe

 

I would think doing your mim idle setting but them pull the iac out before starting and see if its all the way out or all the way in. should be moving out of the iac as for the 730 to 749. since some seem to need changed and soom don't. your being backwards would be an easy first check. as I understand it. higher numbers means more air passing by because the plunger is moving into the iac. less counts is moving out to seal off air. so less should idle down. 150ish is all the way open, passing lots of air.

 

I can change your latest AXCN_V3 bin to report the actual BPW in the ALDL at the stock ALDL BPW location if you like. Send me your latest bin and I will modify it to report the BPW or at another ALDL word location. Putting it at a new location will show you how different what you would normally see reported vs. what the actual BPW is.

If you want the BPW reported at a new location you will need to be able to modify the ALDL definition (ADS file using TunerPro).

 

If you set the 2-D commanded idle table higher and the idle indeed does go higher then the IAC would be going in the right direction. If you open up the throttle blades for a given commanded idle RPM, then you should see the IAC counts go down AND the idle stay the same. If the idle increases then it would be wire wrong.
You could check that the IAC is completely closing off the air passage after doing the IAC reset procedure, but I think you would need to remove the TB and/or IAC.

Don't try resetting the IAC when it is not in the TB.

 

I think I'll take you up on that first offer. I'd rather know the "true final, at the injector" pulsewidth. That way I'll know if my idle pulsewidths are reasonable, and then when I get some boost testing I'll know if I'm out of injector up top.

I'll play with the commanded idle + with TB blades too and see if I can verify a broken idle.

I've got a lunch from 3-4:30pm, then I gotta go back in til late tonight but I might get 20 minutes to play with the car. I'll email you my latest bin either way.

-- Joe

 

The 0.80-0.90 ms pulse width that I reported is from a 165 ecm/$32B mask, (byte 45) running in double fire mode.

This value does not include the AE pulse width, but should include the other trim and offset corrections.

How does the 730 reported BPW differ, or are you just adding in the AE pulse to show the total?

At idle, with no AE present, I think it is representative of the actual pulse. Hard to verify though, without an oscilloscope.

 

I don't know how the 165 $6E or $32_B code reports.

The V8 730 code doesn't send the low BPW correction value x 2 or the BPW battery correction x 2.

Suppose the ECM is at idle with a low BPW near .75 ms. The ECM will add 2 x .14ms for low BPW and 2 x .793 ms for a battery at 12.8 volts:
The ALDL will report .75ms for the BPW.
The ECM will send .75ms + (2 * .14ms) + (2 * .793) = 2.6ms to the injector driver (all sync).

For high RPM, the low BPW correction will be zero so only add in the 2 * batt_correction.
Suppose the engine is at 6000 rpm (10ms inject time):
If the ALDL reports 9ms then that would be 90% duty cycle, but the ECM sends 9 + (2 * .793) = 10.59ms to the injector driver............over 100% duty cycle.

 

Are you sure that the $32B mask includes the low BPW and battery correction values? It looks like a lot goes on after the staa 0x00D5.
----------
While we are talking about injector BPW. I think a good code mod. is to go 100% DC on the injector when the ECM tries to go over 90% DC. This would cure the sometimes seen lean condition due to injector flutter. Yes, it would make the AFR too rich, but that is better than too lean........especially on boosted and/or N2O engines.
I am surprised the Megasquirt code doesn't already do this.

 

Here's what I can see from a better commented $32 hac:

STX L00D5 ; Base PW, Last Inj
LDD L011A ; ACCEL ENR PW
ADDD L011C ;
STD L011A ; ACCEL ENR PW
BRCLR L0034,$80,LF748 ; BR IF NOT b7
; ... else
INC L018D ;
BNE LF748 ;
DEC L018D ;
LF748: BRSET L003E,$20,LF750 ; BR IF b5
; ... else
BRCLR L0038,$40,LF758 ; BR IF NOT b6
; ... else
LF750: LDAB L00E0 ; Num of Inj,s simce Start Of Acel Enr
INCB ; INC NUM INJ'S
CMPB LC314 ; Injects in fuel limiting
BCS LF768 ; BR IF NUM INJ'S LT LMT THRESH (2)
; ... else
LF758: BRSET L003E,$80,LF76A ; BR IF b7
; ... else
BCLR L0038,$40 ; CLR b6
BCLR L003E,$20 ; CLR b5
CLRB ; ZERO OUT NUM ASYN PULSES
STAB L00DF ; NUM OF ASYNC PULSES SINCE ACEL ENR
CLRA ;
STD L00D9 ; ACCEL ENRICH, ACUM FUEL
LF768: STAB L00E0 ; NUM OF INJ,S SINCE START of ACCEL ENR

It looks to me like after the staa 0x00D5, there is only the AE pulse and decisions whether to include it in the current pulse or next pulse and limits etc. The other blm, low pulse and vbatt corrections occur prior to the staa 0x00D5.

 

And now I have an oil leak...

I'm gonna push this car into the river...

-- Joe

 

I can see a bat voltage PW correction table, but where does it add steps?
See, it's increasing my idle TOO much. It's adding like 50 steps which is ramping me up to like 1100-1200 RPM.

I ditched the miniram btw, and am on a singleplane now.

-- Joe

 

Have you been able to get your idle steps down from 140-150?

I don't see anything in the calibration that says "add steps" per se. Though the ECM will kick the idle speed up (by adding steps) if it thinks you fall too far below the commanded idle speed. Again, try to have the engine idle completely from the throttle blades (take the IAC out of the equation).

Have you zeroed out your stall savers? Typically these values are down in the 500 rpm range, but all the same...

You said you're seeing a steady .700-.800V on the O2 sensor. That's certainly peculiar. Should be bouncing around. Have you tried a different sensor (verify you don't have a bum sensor). Also, check the voltage drop from the O2 sensor body to the negative battery terminal. If it's too high, that'll screw up what the ECM sees as O2 voltage.

 

Yep, idle at 0 steps now. Perfect smooth idle. I actually got the miniram all set too, but then decided to swap to my singleplane and sold the miniram.

But in any event, yep it idles at 0 steps now.


All the RPM error stuff is zero'd FYI. If I recall, those are all SA tables though. And really, that was a huge problem. The ECM was fighting too hard to get the idle right that it was making it worse. Now it idles great at 0 steps. But when the fans kick on, it adds about 50 steps and raises the idle too high..

The A/C table is zerod out too.


It is.

No. I figured with the manual car the stall saver might be a good idea to leave in..

02 swings from .100 - .700 now rapidly so I think the fuel is dead on. I have a little bit of a rich tip in on take off, but as far as idle goes I think it's perfect.

Thanks!

-- Joe

 

Next time I run my car (which probably won't be until this weekend unfortunately), I'll look to see how many steps get added when the fan comes on. I know my idle stays constant and the fans switching on and off is pretty much invisible to me. So I never really paid much attention to the number of steps being added. I can't really say 50 steps is too much until I have some data...

 

Ok, I ran the car today. I basically get 1 or 2 steps more IAC count with the fan on.