• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

LH 2.4 swap idle issue

paulcurran

Member
Joined
Feb 27, 2005
Just finished conversion, having trouble with idle after a speedy drive. Decided to bypass the idle motor. The lh 2.2 throttle body I picked up doesn't have the bypass hole drilled to use the black knob to adjust idle. Anyone know how big a hole to drill?
 
You have no speedo speed signal going to ecu? It’s a well documented gripe that lh2.4 has. Speed signal goes from diff to speedo and is transformed into signal for ecu. There are ways around it.
 
Wait. First... what is the idle doing? 2.4 should sort idle out or you fudged something up.

Also, what do you mean your 2.2 tb doesn't have the hole drilled for the black knob? How is it a 2.2 TB then?
 
idler

If I drive the car a while fast, when I come to a stop the idle hunts in the 2 to 3k range. I know this is lack of speedometer input, in looking for an easy way out, I read that some people have disconnected the idle motor & blocked off the pipes, then used the 2.2 lh throttle body with the black tapered knob to set the idle. I'm trying a 2.2 throttle body but twisting the tapered knob does nothing because the passage that allows air through is sealed like it was never drilled at the factory. Put the throttle body in a vise & put a punch in there, hit it lightly with a hammer & it easily made a hole (thin metal) so I drill it closed to the i.d. of the pipe. See what happens when I get it back on.
I drove it with the 2.4 throttle body & the pipes blocked, but it idles about 600 to 750 rpms & a little rough. I thought about turning the throttle stop screw in a bit to raise the idle (then reset the TPS) but something makes me remember the 2.4 is supposed to have the throttle closed at idle?
 
It may be wiser to cure the disease rather than treat the symptoms. What's the car?

LH2.4 base idle is set with the idle screw. You block the idle air tube with a pair of tubing pliers and adjust to around 500 rpm, if I remember correctly.
 
Trying to get lh2.4 to run well without the iac is not worth it. You can get it to work in the environment you set it up in well, but once the weather/temperature/humidity changes it'll run like crap again. Trust me on that, I tried to bypass a dead idle valve for too long before giving up and finding good one in a junk yard.

Fully functioning lh2.4 is a great system, you should complete the swap and make it work as intended.

Like shoestring said, fix the problem, not the symptoms
 
2.4 speedo signal missing, choices to make a signal

So I took the road I headed down to the end, & I'm not happy with the results. I removed the idle motor & hoses, blocked off the pipes, installed a 2.2 throttle body, set the TPS , then adjusted the idle with the black knob. Now the idle is low & rough when it starts & runs cold, and it settles in about 1200 rpm's warm. But the rest of the 2.4 swap went so well it just seems crude & there must be a better solution.
On the back of the 1984 242 turbo's speedo head is the little plate with 2 screws and 2 male connectors that was used for cruise control those years. Is there any way to use that as a speedometer reference signal? If so, how? If not what other choices are there short of changing the rear axle & using the late dash cluster? I'm supposed to deliver this car to my nephew in a week & I'd love to have this resolved. Help appreciated, thanks.
 
Assuming your lack of speedometer is due to the rear harness to the sensor in the diff, it?s quite easy to repair. Unplug the connector from the sensor, cut the ?hinges? off the black plastic box on the other end of the connector and remove it. Once out, cut each connector off and splice in some new wires. I used some 12 gauge wire with mine, and has been going strong for about 500 miles or so now. I?m not sure if it makes a difference, but I was extremely anal about making sure the wires were oriented correctly, so most of my time was making sure the colored wires on each connector were attached to each other. I think it?s one brown and one green wire originally. After some spirited driving my idle would hunt from about 900-1600 rpms continuously. That problem is long gone now. Good luck!
 
On LH2.4 cars, the VR sensor on the rear differential goes to the speedometer. On the speedo is a circuit board that both drives the needle and generates the Vss signal used by the LH2.4 ECU. See http://cleanflametrap.com/speedo.html

The 3pin connector in the first picture is ground, +12volts, and VR sensor input (the other VR pin goes to ground, polarity doesn't matter). The Vss output to the ECU is the outer-most of the 3 silver barrel pin connectors on the right side of the picture.

On cars with early cable-driven speedos, there is a VR sensor for cruise control on the back of the speedo (optional, may not be present if no cruise control). This VR sensor generates a signal that is about half the voltage and half the rate of the rear differential VR sensor used on LH2.4 cars.

I've been curious what would happen if you wired the early speedo cruise VR sensor into a scrap circuit board from a late-80s speedo (pre-ABS), and then used the Vss output from the circuit board to drive the ECU. It might work OK, or the voltage might not be enough, or the pulse rate might not be enough. I have the parts to try part of this experiment - if I have time this weekend, I'll try it and post a follow-up.

Otherwise, you can use a front wheel ABS hub and sensor with an ABS speedometer circuit board to generate the Vss signal for the ECU, or swap to a differential with a VR sensor and use the matching speedo circuit board.

Am I explaining this OK? If not, ask some questions and I'll try again.

-Bob
 
Quick Update:
I tried wiring the cruise VR sensor from an early 80s cable speedo to a 91 (pre-ABS) speedo. It seems to be sensitive enough to detect the lower signal voltage from the cruise VR (versus the normal rear differential VR sensor).

When the cable speedo is at ~30 mph, the 91 electronic speedo reads ~12 mph (~2.5x slow), and the VSS signal looks fine. I don't know what the LH2.4 ECU does with the VSS signal for idle control. Even though the pulse rate from the cable speedo is ~2.5x slow, it might work fine for idle control -- if idle simply uses "car is moving" versus "car is stopped", the difference should be OK.

If you'd like to try this, you need an electronic speedo for a pre-ABS differential. It should say something like K9800 on the face. If it says K39200, or similar, it's from an ABS car and needs 4x faster VR tooth rate. If you can, get the 3-pin connector with the speedo too.

I'll post pictures and wiring details this weekend.
 
The ECU will cut fuel to the motor at deceleration when you take your foot off the gas(TPS clicks shut) above a certain speed. Not sure above what speed. Hope this helps.

Simon
 
Generating LH2.4 VSS from Cable Speedometer Cruise Sensor

The early 240s use a cable driven speedometer. When cruise control is installed, a VR sensor in the back of the cable driven speedometer is used. (click on pics for full size)



The later 240s use an electronic speedometer with a VR sensor in the rear differential. The electronic speedos also supply the VSS vehicle speed signal to the LH2.4 ECU. It's possible to use the circuit board from an electronic speedo with the cruise control VR sensor on an early cable speedo to generate a VSS signal for the LH2.4 ECU.

Here's a picture of the circuit board on an electronic speedometer (a 91 in this case), showing the main 3 pin connector in the center and the 3 single pin connector sleeves on the right. +12v power comes in the main connector and goes directly to the leftmost single pin connector. This needs to be jumpered to the center single pin connector (or you can just wire +12v directly to the center single pin connector). One side of the VR sensor goes to ground, the other side to the leftmost pin on the center 3-pin connector. Polarity of the VR sensor doesn't matter. The VSS signal to the LH2.4 ECU is the rightmost single pin connector. In the center of the circuit board is the stepper motor that drives the odometer gears. The blue and brown wires at the top of the board go to the speedometer needle galvanometer.



Here are pictures showing a hand drill driving a cable speedo, with the cruise VR sensor connected to an electronic speedo. Notice that the electronic speedo reads ~2.5x low. I don't know if this will work OK with the LH2.4 ECU for idle control, or if it needs to have a higher rate. (The electronic speedos use either 12 tooth differential sensors, or 48 tooth in the later ABS 240s.) If this doesn't work, it's possible to use an ABS front wheel hub sensor (48 tooth) with the speedo circuit board from a ABS speedo.



There are 2 screws holding the circuit board to the speedomter, and a couple pairs of wires. I unsoldered them but you could just clip them off. The rotor for the odometer stepper motor just falls out.



By itself, the circuit board draws ~1.4 watts of power. Most of this goes into the odometer stator coils. I think that you'd be fine wiring up the circuit board as shown above and wrapping it in electrical tape. The stator housing does get hot from the power (~115 ?F after an hour on the benchtop at 60MPH).

I went further on mine and removed the stator - clip the 4 plastic heat stake pins flush and unsolder the 3 winding pins. Without the stator, power was only ~0.4 watts. I also had a scrap of big heat shrink tubing but it wasn't quite big enough so I trimmed off the edge connector and wired directly to the circuit board.



If anyone tries this, please post an update on how the LH2.4 behaves -- does it have good idle even though the VSS rate is ~2.5x slower than standard? Does the Shift light (if equipped) do anything weird, or does it just not work?
 
The ECU will cut fuel to the motor at deceleration when you take your foot off the gas(TPS clicks shut) above a certain speed. Not sure above what speed. Hope this helps.

Engine speed >1500rpm (roughly)+ TPS switch closed=no injector pulse. If you do a higher gear cruisedown, you can feel it when the injectors are activated.
 
Followup:
I've been running my LH2.4 swapped '85 GLT AW-71 K-Jet for a while now with no weird idle issues. Generating the LH2.4 VSS signal from the Cable Speedometer Cruise Sensor seems to work just fine.

Here's a picture of the cluster, with speedo Vss PCB hidden in black heatshrink, and mods to use the Lambda Sond Service light for ECU/EZK check engine light. Note: I clipped out the diode on the lambda sond light -- it was just above the 2-pin connector with the added yellow CEL wire.



Caveat: I haven't tried unplugging the VSS connector to see how idle behaves without the VSS signal, but I'm assuming it will be erratic like all the other reports.
 
Engine speed >1500rpm (roughly)+ TPS switch closed=no injector pulse. If you do a higher gear cruisedown, you can feel it when the injectors are activated.
Agreed. If you raise the idle speed settings in the ECU tuning, it actually raises the rpm that the injectors kick back on at the same time(which I don?t like). There may be some additional tables in the newest XDFs that give even more control but I haven?t experimented with them.
Followup:
I've been running my LH2.4 swapped '85 GLT AW-71 K-Jet for a while now with no weird idle issues. Generating the LH2.4 VSS signal from the Cable Speedometer Cruise Sensor seems to work just fine. .
Nice work!
 
Caveat: I haven't tried unplugging the VSS connector to see how idle behaves without the VSS signal, but I'm assuming it will be erratic like all the other reports.

After some side discussions, and more reading, my curiosity won out and I unplugged the VSS wire last night, then went for a leisurely 20 minute neighborhood and highway drive. Surprisingly, I didn't notice any idle issues. I checked for codes (ECU hole #2), and there were none (1-1-1). Remember, this is a AW71 automatic, not a manual. If it matters, I'm running a stock '937 ECU and a stock '148 EZK.

Tonight, I went for a slightly more spirited drive (I still have less than 500 miles on the rebuilt engine, so when I say spirited, think granny's a little late to church and may not get her favorite pew spirited). I also clicked off the overdrive when on the highway for a few minutes -- cruise RPMs went up from ~2K to ~3K.

With this, I did notice a high idle that wouldn't go away until the engine was turned off. The idle was ~1000 to ~1100 RPM in Drive, and went up to ~1400 RPM in Neutral. Normally, it idles a bit under 800rpm. This time, I got the expected 3-1-1 diag code for missing VSS signal. I was also watching the WB02 gauge to see if deceleration fuel cut changed at all, but didn't notice anything obvious.

I'm assuming the automatic tranny masks the high idle as compared to a manual tranny, and that you need higher RPMs plus some load to trigger the issue in the first place.

Earlier this week, during our late season snow storm, I put together another VSS board for smoothdurban. It will be interesting to see if it works for him, and if he sees the same issues.



 
Back
Top