Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Topic: Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap (Read 5157 times) previous topic - next topic

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

March 25, 2011, 01:21:29 AM

I've been having problems ever since I performed the HO and T-5 swap. I used an A9L computer (re-manufactured by advance), spliced in a MAF, dropped in the engine and away I went.

I have never been able to get the car to idle correctly. The idle would flux up and down, setting the base idle would lead to stalling or 1000 rpm+ idle speed.

Here are the codes I pulled (from a previous post)
Exhausted over HCI swap problem

Quote from: 88CougarGT;341417

Ok, heres the dirt.

I pulled codes koeo:
51 - ECT sensor signal is greater than the Self-Test maximum of 4.6 volts.
22 - MAP/BP sensor out of self test range.
53 - TPS circuit above maximum 4.5 volts.
54 - ACT sensor signal is greater than the Self-Test maximum of 4.6 volts.
35 - PFE or EVP circuit above the maximum limit of 4.81 volts.
85 - Canister Purge Solenoid circuit failure
95 - Fuel pump secondary circuit failure. The EEC senses infinite resistance to ground from the fuel pump on the Fuel Pump Monitor circuit.
I ignore 95, I did not hook up the fuel pump relay signal during the swap to mass air.
Hmm those sensors share the sig return line. So I added a ground (just a wire clipped on to the ground going to the TPS)

Disconnected the neg cable for 5 minutes to clear the codes.

Re-ran KOEO:
31 - PFE or EVP circuit below minimum voltage of 0.24 volts.
85 - Canister Purge Solenoid circuit failure
95 - Fuel pump secondary circuit failure. The EEC senses infinite resistance to ground from the fuel pump on the Fuel Pump Monitor circuit.
Weird, ok.

Ran KOER:
41 - HEGO sensor circuit indicates system lean (right side).
31 - PFE or EVP circuit below minimum voltage of 0.24 volts.

The car seems to be running rich (sooty exhaust, floods on startup occasionally, dark plugs). Do you suppose my left 02 is bad? They are $40 at autozone. I might try to pick up a few at the junkyard instead. I can't imagine its running lean.
85 perplexes me too because all the emissions is hooked up with good vac lines.

It starts better with the extra ground I added but it still wants to idle really high. If I remove it, it drops down to a nice 900 rpm idle. I'm pretty tired of these gremlins. I have to DD the car too until I get my girlfriends truck fixed (shes driving my normal DD). At 10 mpg its killing me. I will try to shoot the wires this weekend and find out why the sig retn is open.
-Z

*** Note, the Code 41 was from a burned through wire on the o2 sensor - I fixed this before continuting to troubleshoot ***

With the auxiliary ground wire attached, it would run OK. From a cold start, the car would hold a high idle until you cycled the key, then it would drop down nice and slow. I could never get the idle speed into the zone I wanted it. It had a rolling idle, sometimes it would have hard starting issues.

Since then, I have been digging and digging and digging. I finally ran across some posts talking about a burned out trace. There is a subtle difference between the AOD and the T-5 staring circuit.

Corral Forums post about the fried signal ground (46)

It took some more research, but this is what I found.

In a manual, the NDS signal (pin 30) is connected to the clutch neutral safety switch and the trans neutral safety switch in parallel. These, when in the neutral position, provide a ground to the NDS line.

In an automatic, the NDS signal is connected to the neutral safety switch in the trans AND is T-ed off to the ignition switch. The ignition switch provides +12v to the NDS signal when the key is in the Start position.

Now here is the catch, the programming on the A9L computer, expects that line to be neutral. When +12v is supplied, its dumped through the circuit board and though a very small trace to the filtered sensor ground (pin 46). When this happens, the trace fries and you loose your sensor ground.

Here is the circuit from an 88 Tbird/Cougar EVTM

The easiest way to diagnose this is to check for continuity between the TPS ground and the vehicle ground. A high impedance means that the line is probably fried (assuming you have performed a swap similar to mine).

REMEMBER TO FIX THE REASON IT FRIED IN THE FIRST PLACE. Disconnect the NDS (Pin 30) from the HOT IN START circuit. Failure to do this will further damage the computer after the repair. I ran a wire through my clutch neutral safety switch and to ground. You can leave it disconnected (cut and ungrounded) but you will not be able to run KOER codes.

Anyway, enough of the backstory, heres how to repair your fried computer:

1. Assemble the following tools, small wire (I used a strand from an old Ethernet cable, wire strippers, wire cutter, small torx screwdriver, medium torx screwdriver, razor blade, screwdriver, soldering iron, solder, a multi-meter, and toothbrush.

2. Start by removing all the exterior case screws, 10 in total. You will need to cut the tape that covers the programming port and the warranty sticker.

3. Open the case, and remove the screws attaching the PCB to the rear frame.

4. Remove the PCB from the frame and flip it over.

5. Examine the backside. Find pin 40 and follow the trace.

6. If yours looks cooked (like mine) then take your multimeter and check the resistance of the trace.

7. Mine is bad (open loop = infiltrate resistance). If you see less than 1 ohm across the trace, then your sensor ground trace is good. Something else must be wrong.

8. Start by cleaning the conformal coat from the pin contact pad and the nearby through hole pad. Use the razor to carefully se clean the pads. Use the toothbrush to clean up the residue.

9. Strip back a small segment of the wire jacket. Form the end into a small hook.

10. Place the hook around the pin lead and tape the wire to the pcb.

11. Solder the wire onto the pin lead. Remember to heat the joint and apply solder to the joint, not the tip of the solder iron. The components are pretty stout so there shouldn't be too much worry of overheating. (If you have the ability, you may want to remove the old solder first, but as long as you use just a little bit of solder, this is not necessary).

12. Trim the wire to size following the original trace location. Trim the jacket and form the other end into another little hook. Place the wire over the through hole lead.

13. Solder the wire onto the through hole lead.

14. Route the wire cleanly and close to the board. Check for any stray solder or bits of wire. Clean with compressed air and/or the toothbrush.

16. Check the continuity of the trace. Mine is now 0.4 ohms = FIXED. (probably less than 0.4 ohms but its hard to hold the probes and take a picture).

Anyway, I don't know if anyone else will ever find this handy. I hope so. I have spent tons of hours trying to figure it out. This will only happen to people using the A9L (or any other MAF manual computer) with the AOD harness. I guess not all A9L are affected either. It is somewhat common though. There is something to do with a difference in the o2 sensor wiring. Ford put a loop back in the harness to configure the harness for the appropriate trans.

I compared my o2 harness to the images discussed at this link but the cougar's harness is different.

This was so frustrating that I nearly sold the car because of it. Hopefully this will save someone else the same grief I went through.

(I will be annotating the images later. but for now, I'm going to bed)

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #1 – March 28, 2011, 09:59:14 AM

Got it installed and started this weekend. No fluxing idle, no over rich condition, no hard starting, even the check engine light went out! It drives like a whole new beast. Even with my stock granny gears (2.73) I can roast the tires all the way through first. This is what I was expecting when I built this engine. I can't wait to get my new 3.73s w/ locker installed.

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #2 – March 28, 2011, 05:07:22 PM

Nice job, I heard of that issue with the A9L, but never heard of anyone repairing it.

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #3 – March 28, 2011, 05:12:36 PM

Great find!!
The 87 EVTM shows the 2.3 and 3.8 ignition start 12v going to pin 30 but not the 5.0
Pin 30 is not shown at all for the 5.0

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #4 – March 28, 2011, 05:20:20 PM

Quote from: softtouch;356606

Great find!!
The 87 EVTM shows the 2.3 and 3.8 ignition start 12v going to pin 30 but not the 5.0
Pin 30 is not shown at all for the 5.0

Not sure about the 87, it's definitely in my 88 EVTM. Its hidden in the crease almost. Off-page reference "E". Took me forever to find it. I guess this is how the mustangs are wired as well:

I had parked the car for about a year because I hated how down on power it was. I was averaging 10 MPG or worse. I was generally pissed off at it and almost sold it as was. Its like having a new car again. I'm going to go get some lower ball joints and shocks for it next. Might turn it into my DD as I am averaging 16 mpg around town (and climbing) .

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #5 – April 01, 2011, 06:13:54 AM

Really nice work troubleshooting and repairing that. Hope the car runs good for you now.

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #6 – April 01, 2011, 11:59:59 AM

One word: Sticky.

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #7 – April 04, 2011, 10:09:33 PM

In an automatic, the NDS signal is connected to the neutral safety switch in the trans AND is T-ed off to the ignition switch. The ignition switch provides +12v to the NDS signal when the key is in the Start position.

Disconnect the NDS (Pin 30) from the HOT IN START circuit. Failure to do this will further damage the computer after the repair. I ran a wire through my clutch neutral safety switch and to ground. You can leave it disconnected (cut and ungrounded) but you will not be able to run KOER codes.

OK,

So, in a manual, you simply ground it out, but in the auto, you cut the hot feed from the ignition, and do what with it? Your post reads that if i leave it cut, i cant run codes, but it only instructs what to do with a manual...

Im no expert, just trying to figure this out, so forgive me if Im missing the obvious point.

***EDIT***

Jesus, ok, I was complicating this thing way too much. Gotcha.

Im not having problems...yet. Im putting in a new engine, ho, mass air swap, etc.

Fix for poor idle, rich mixture, and poor performance after HO, T-5, mass air swap

Reply #8 – April 04, 2011, 11:21:27 PM

It needs to see ground in order to run KOER if I recall. I need a connector but I am planning to run it to ground though the switch mounted on the clutch (I rigged it up with some spade connectors for testing). With mine floating, I can pull KOEO codes just fine. Haven't tried a KOER yet but I suspect it will complain that I am not in neutral.

What happens when you jump the STI to ground? Any blinks? Are you trying to run KOER or just KOEO?

-DZ