Electrical Guru's - need help before replacing ECM!

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, Skimlinks, and others.

MountNGoat

KK6WRY. Always Learning
Joined
Dec 12, 2013
Threads
15
Messages
545
Location
Northern NV
My '97 LX450 has been having idling problems for awhile now. I'll skip all the things I've checked (vacuum hoses, EGR, exhaust, main air intake hose - all good). The truck started throwing codes relating to the throttle position sensor, P0120, P0510. I purchased a new TPS and installed it, cleared the codes, and drove it this weekend. The rough idle is gone, but it idles high now, and while their is one more system that needs to reset and no CEL light on yet, there is a temporary code of P0510 again in the system.

I have a '96 EWD and '97 FSM. For a P0510 code the FSM says that with the key turned to the 'on' position I should be reading 9-14V between terminals E2 and IDL of the wiring harness at the TPS. But there's nothing. To check if the wiring is bad (I've already opened the harness at the EGR - didn't find anything, and inside near the glove box also no problems) I decided to measure the voltage coming out of the ESM. The FSM shows which terminals at the ECM are the E2 and IDL (ground side), and the EWD shows the same. So, I unplugged the E5 harness at the ECM and checked pin 11 (+) and pin 9 (-) and there's nothing. Pin 1 is the VCC, it was supposed to have 5 V, and it did.

So I'm inclined to believe that my only diagnosis here is that the ECM is bad since it's not putting out any voltage at the E5 connector, pin 11 with the key on - and I am really hoping someone can tell me I'm wrong (@TrickyT )!
 
You need to know something about the logic of the circuit. Does it depend on the state of some inputs (like the speed sensor) before it sets the voltage high? The diagnostics section of the FSM is pretty good at pointing out all the possibilities in order of likelihood for most of the error codes. No doubt the ECM is one of the possibilities, so if you have one laying around, give it a try.
 
The ECM's on these vehicles are very robust and failures rare, so I would be looking to other causes before replacing the ECM. That said, I do think you should see some voltage on ECM E5.11 when the harness connector is unplugged. Are you absolutely sure you are testing the correct pin number in the ECM and making good contact with the pin? (Note that if you are trying to test for the called for 9-14V reading then either the E5 connector needs to be unplugged or the TPS in a non-idle position or else the idle switch in the TPS will pull pin E5.11 to ground.) Also, the 9-14V reading can be taken between E5.11 and any vehcile ground, not just to E5.9.

The EWD shows partial circuitry of the ECM and the E5.11 input is pretty simple (couple of resistors, a diode and an implied transistor), so if a component has failed it should the possible to replace it. Is the truck driveable? Any chance you couple come by my house on Saturday or Sunday?

EDIT: Lastly, note that TPS is a standard OBD II sensor output, so you can also check for correct TPS operation by attaching a Scan Gauge or other OBD reader and looking for the TPS reading. I vaguely recall the numbers vary from ~10 to ~40, at least on my Scan Gauge. I don't recall if the idle switch is specifically checked though, and that's what's in question here.
 
Last edited:
I think I got the pin numbering down, Tom. Looking at the FSM and EWD, I see the following:

upload_2018-5-29_15-29-4.webp


At the connector, I definitely do not have 9-14V between E2 ad IDL. Those are labeled as pins 9 and 11 at connector E5. Then there's this diagram of where those are on the connector:

upload_2018-5-29_15-31-13.webp


Which, since when you pull the connector off you're looking at the male side of the connection, matches what the EWD says for pin order according to this diagram:

upload_2018-5-29_15-32-29.webp


So I take that to mean that when I'm upside down and looking up at the bottom of the ECM with E5 removed, pins 9 and 11 are on the back row, with 9 being farthest to the right. With the truck not running and the key in the ignition turned to on, I have no voltage at 11 whether I try grounding it to 9 or to another ground. And 9 shows continuity with other grounds.

It may be driveable by this weekend!
 
Looks to me like you've got the correct contact positions.
Can you see the part number on the ECM? Likely a 89661-60650, but I need to confirm when I get home.
 
Coming back to wrap up this thread in case someone has a similar problem in the future. @MountNGoat and I moved our exchanges to PMs after my last post above and ultimately we met up today to definitively sort out the problem. What we discovered was...

1. His Fluke 113 multimeter was giving wacko readings at TPS connector terminal #2, while my Fluke 115 showed a steady +12v which is exactly what the FSM calls normal. The 113 is sold as a "utility" multimeter and has limited selectable ranges, but I have no idea why it could not reliably measure a simple DC voltage coming from the ECM. (The 113 did fine measuring voltage at the battery and resistance values. Maybe the diode in the TPS circuit caused it problems for some reason. If I figure this out I'll post a follow-up.) At any rate, this bad reading meter reading lead @MountNGoat down the proverbial rabbit hole and into thinking the ECM was faulty.

2. A loaner ECM was swapped in and exhibited exactly the same symptoms as the old ECM, thereby definitively ruling out the original ECM as being faulty.

3. The rabbit hole mentioned in #1 above lead to the TPS being replaced prior to today. The TPS uses slotted screw holes and must be carefully adjusted using the procedure described in the FSM. This was not done initially and lead the the idle position switch inside the TPS never closing and therefore the ECM never really knowing that the throttle was truly closed and that the engine should be considered to be "at idle." So we adjusted the TPS in situ using the FSM procedure described in the FSM. (Requires applying vacuum to the throttle opener at the same time as measuring the TPS idle switch contacts at the same as adjusting the TPS. In other words you need six hands. Fortunately @MountNGoat and I had the very welcome and capable assistance of @justintpryor, making this step a breeze.)

4. Because some of the previous stored OBD-II codes implicated bad ground connections, we unbolted and wire-brushed pretty much all of the grounds inside the engine compartment. None looked bad and I can't say doing this made any real difference, but it only took 1/2 hr and now we know that all the grounds are good.

The truck now seems to run fine, did great on our freeway and slow speed uphill seat-of-the-pants engine power tests, and shows no codes and all OBD-II emissions tests as "completed." So we are declaring victory and that this vehicle Rubithon ready. :bounce:
 
Coming back to wrap up this thread in case someone has a similar problem in the future. @MountNGoat and I moved our exchanges to PMs after my last post above and ultimately we met up today to definitively sort out the problem. What we discovered was...

1. His Fluke 113 multimeter was giving wacko readings at TPS connector terminal #2, while my Fluke 115 showed a steady +12v which is exactly what the FSM calls normal. The 113 is sold as a "utility" multimeter and has limited selectable ranges, but I have no idea why it could not reliably measure a simple DC voltage coming from the ECM. (The 113 did fine measuring voltage at the battery and resistance values. Maybe the diode in the TPS circuit caused it problems for some reason. If I figure this out I'll post a follow-up.) At any rate, this bad reading meter reading lead @MountNGoat down the proverbial rabbit hole and into thinking the ECM was faulty.

2. A loaner ECM was swapped in and exhibited exactly the same symptoms as the old ECM, thereby definitively ruling out the original ECM as being faulty.

3. The rabbit hole mentioned in #1 above lead to the TPS being replaced prior to today. The TPS uses slotted screw holes and must be carefully adjusted using the procedure described in the FSM. This was not done initially and lead the the idle position switch inside the TPS never closing and therefore the ECM never really knowing that the throttle was truly closed and that the engine should be considered to be "at idle." So we adjusted the TPS in situ using the FSM procedure described in the FSM. (Requires applying vacuum to the throttle opener at the same time as measuring the TPS idle switch contacts at the same as adjusting the TPS. In other words you need six hands. Fortunately @MountNGoat and I had the very welcome and capable assistance of @justintpryor, making this step a breeze.)

4. Because some of the previous stored OBD-II codes implicated bad ground connections, we unbolted and wire-brushed pretty much all of the grounds inside the engine compartment. None looked bad and I can't say doing this made any real difference, but it only took 1/2 hr and now we know that all the grounds are good.

The truck now seems to run fine, did great on our freeway and slow speed uphill seat-of-the-pants engine power tests, and shows no codes and all OBD-II emissions tests as "completed." So we are declaring victory and that this vehicle Rubithon ready. :bounce:

Glad to know my ride is ready for the con @TrickyT
 
Okay, here's the deal with the Fluke 113. This meter has a low input impedance (LoZ), about 3k ohms according to the specs. Contrast this to most multimeters that have a very high input impedance. (The input impedance on my Fluke 115 is >11 megohms.) Fluke did this so that it can "simultaneously test for voltage or continuity" without moving the function switch. In other words, the 3k input impedance (= just resistance when dealing with DC circuits) is just like a load on the circuit under test and must be taken into account when making measurements. For many measurements, like measuring battery voltage, the 3k ohms will have very little or no effect and the meter reading will be accurate. But for the TPS idle switch circuit under consideration here, the 3k ohm meter resistance is in series with a resistor and diode that are internal to the ECM and all three will have a voltage drop across them. I'll guess that the internal ECM resistor is 10k ohms, maybe even more. So most of the voltage drop will be across this resistor and thus the meter will measure a much lower voltage than it should. As the Fluke 113 manual says "This load [i.e., the 3k input resistance] can alter the voltages in electronic control circuits." This is exactly what was going on with Steve and what led him to believe the ECM was faulty. This doesn't mean the Fluke 113 is a bad meter. Indeed it is highly rated and comes from a very respectable manufacturer. It just means you have to be aware of its LoZ input impedance when making measurements, especially in electronic circuits involving other components like resistors.

EDIT: Here's a situation where the LoZ input on the 113 is advantageous: Often times electrical connections on vehicles get corroded and develop high resistance, say a few thousand ohms. And say that corroded connection is in a voltage supply circuit. A regular multimeter with high impedance input will not detect any problem when measuring the voltage, since it will draw very little current itself and thus there will be very little voltage drop across the corroded connection. The LoZ on the 113, on the other hand, will draw current (12v / 3k = 4ma) and this will result in a voltage drop across the corroded connection and a lower meter reading than expected, so you know you have a problem.
 
Last edited:
@TrickyT , can’t thank you enough for your help!!

Here’s some more background info for you on the Throttle Position Sensor replacement. The TPS replacement was triggered by both P0510 and P0120 codes being thrown initially, although the 510 did go away. Both relate to the TPS. The TPS has a very narrow range (rotationally) to sense when it is closed at idle and when it is open above idle. There is an internal rotary sensor that is engaged by the back side of the rotating assembly at the throttle cable. With the engine off there is a plunger extending off a a vacuum operated diaphragm, known as the throttle opener, that keeps the sensor rotated just far enough off of closed to cause pins 1 and 2 to not connect internally (pin 1 is ground and 2 is ~12V). So, when you start the engine the ECM sees ~12V at pin 2 but after a second the vacuum builds up and pulls the throttle opener away and allow the throttle assembly to rotate a bit more toward less throttle, allowing the sensor inside to make a connection between pin 2 and ground (pin1), sending 0V as the pin 2 signal to the ECM. As you apply throttle the connection quickly opens and pin 2 is no longer grounded, sending out the ~12V signal instead.

When the TPS is out of the vehicle and nothing prevents the rotary switch from closing all the way, there should be zero ohms resistance between pins 1 and 2. I measured my old switch just now and instead of zero ohms it is 8, which was causing the codes as the ECM never got the correct signal that it was idling. So the TPS was bad. Tom already went through the issues with my meter (thank you Tom!), but not properly adjusting the new TPS after installing it was a big mistake by me.

The other odd thing is that this improper adjustment somehow also led to P0110 and P0115 (intake air temp and engine coolant temp) codes to manifest. The FSM literally says “hint: if you get those codes along with P0120 then the ground may open (they all share the same ground, so makes sense). That’s why we chased grounds. But really everything was probably good enough (Like Tom says, though, now we KNOW they're good). Somehow not having the TPS properly adjusted caused those codes.

Two days of driving and all is good!!!
 
Last edited:
So, was the original problem actually the TPS and everything else the result of the multimeter not reading.
 
So, was the original problem actually the TPS and everything else the result of the multimeter not reading.


And/or possibly a marginal or intermittent ground, since that's the other area we addressed and there were stored codes that had nothing to do with the TPS and where the FSM advised to consider a faulty ground connection.
 
During TPS adjustment how did you pull vacuum during calibration?
 
Back
Top Bottom