Rewired harness to clean things up and now TPS sensor reads 14% at idle. The car cranked and idled fine with a typical TPS reading of .34%. After the fourth start or so the TPS reading jumped to 14% for no apparent reason.
The only way I can get the sensor to read 0% in the datalog is to unbolt it and turn it clockwise a little. It's as if the throttle blade is cracked open but it's not and the sensor is not adjustable except for what the idle screw can exact.
Any ideas before I start hacking up the harness that took too long to finish in the first place.
3900 V6 aluminum heads. Cadillac throttlebody which is similar to the 3x00 throttlebodies just bigger. I tried a different computer and a different sensor and ended up with the same result.
[This message has been edited by Joseph Upson (edited 03-24-2012).]
IF the sensor reads 0 unbolted and turned to it's full travel, the problem is NOT with the wiring.
I hope you are correct, I had to consider a possible issue with ground arrangement or some kind of induction interference, which in theory would not go away with repositioning of the sensor. Some how I don't think it's that simple.
IF the sensor reads 0 unbolted and turned to it's full travel, the problem is NOT with the wiring.
I agree. The ECM "learns" the throttle-closed voltage from the TPS every time you start the car and then rescales from there. As long as the ECM "knows" that 0.7 volts (14% of 5 volts) currently equals closed throttle, there should be no problem. Properly installed, the TPS voltage with the throttle butterfly fully closed should always be somewhat above zero.
[This message has been edited by Marvin McInnis (edited 03-24-2012).]
Originally posted by Marvin McInnis: Properly installed, the TPS voltage with the throttle butterfly fully closed should always be somewhat above zero.
Yes, it was normally .34% and the ECM's response to TPS input depends on the programming. Remember that the ECM reads throttle position to determine whether it should be adding fuel or pulling fuel at startup as well as state of throttle input such as in clear flood mode use.
What immediately made this a problem is the engine idle, which is around 1300 with the TPS sensor reading that high at startup, so it's a little more complicated than just a position reference.
The weird thing about it is when I offset the sensor to 0 the TPS % stays at 0 until the throttle is opened enough to give the impression that it is overly delayed suggesting it is okay to return the sensor to its proper position, only when I do it reads 14% at closed throttle again. There is also a stumble on rapid acceleration from idle that corasponds with the delayed pump shot that results from the repositioning. I don't want to rescale the table to compensate I'd like to find and fix the problem.
It may be that both sensors are bad since I don't know the history on the other. I just hate the idea of throwing away $40 on something I may not need. Right now it appears the sensor is the problem because when it's unplugged the datalog shows 0%.
Thanks for the input, I'm glad I don't know everything because I'd have no one to go to with my problems if I did.
Also consider that the TPS is only used by the ECM when its value is changing. At constant throttle the TPS has no effect at all on fuel delivery or spark timing.
The reality is that you don’t really know anything unless you know for sure that you’re getting the full 5 vdc and a proper ground. Garbage in garbage out.
I appreciate the input guys but I'm not new to diagnosing problems of this nature and so far the fundamentals have been verified physically or at least are registering in the data as proper.
quote
Originally posted by Marvin McInnis:
Also consider that the TPS is only used by the ECM when its value is changing. At constant throttle the TPS has no effect at all on fuel delivery or spark timing.
This does not seem to be the case at the moment. As I stated earlier, if the sensor is installed in its proper position where it is for some reason registering 14% in the datalink now, I promise you the engine will idle at approximately 1300 rpm and that's without any throttle movement. Unplug the TPS or offset it to 0 and idle returns to normal. I'm running 8F code mask. Your statement is partially correct as there are asynchronus multiplication factors that vary with the position of the throttle and directly affect fueling even if the throttle is at steady state and that's probably not the only table association with that kind of variance.
quote
Originally posted by firejo24:
The reality is that you don’t really know anything unless you know for sure that you’re getting the full 5 vdc and a proper ground. Garbage in garbage out.
Although I have not measured it physically, the datalink consistenly shows 100% with the throttle to the floor, I doubt a voltage measurement is necessary in this case. I installed a new sensor and received the same result and I was expecting it based on what I've done so far.
The only other possibility I can think of now as being a clue is that the vehicle speed signal is jumpy. I use a Dakota Digital and need to go back and twist the sensor wires. If the odometer needle starts to twitch at idle the rpm goes up in that scenario as well. There was no problem with the odometer stability previously.
My plan now if I can find the time before work tomorrow is to pull the harness, go over the grounds again and back probe the harness again to make sure everything is connected where it should be.
[This message has been edited by Joseph Upson (edited 03-24-2012).]
The only time I have ever seen a really wonky TPS voltage is when I had mine mis-wired. FWIW, the 4.9 TPS uses the center pin on the connector as the wiper. The Allante TPS (which looks more like a stock Fiero connector) has the wiper on one end. It seemed to have a stumble when I hammered it, and the tranny didn't have a clue about when to shift.
[This message has been edited by Raydar (edited 03-24-2012).]
Originally posted by Raydar: The only time I have ever seen a really wonky TPS voltage is when I had mine mis-wired. FWIW, the 4.9 TPS uses the center pin on the connector as the wiper. The Allante TPS (which looks more like a stock Fiero connector) has the wiper on one end. It seemed to have a stumble when I hammered it, and the tranny didn't have a clue about when to shift.
That's a possibility, the pin assignment for the Fiero vs. the TGP sensor is different, it's possible I strayed and looked at the wrong diagram since I had them both out. I also recall a misprint in one of the diagrams I have some time ago involving either the TPS or the MAP sensor. The sensor read .34% on the first few starts and then suddenly jumped to 14% and has been there since. I'll do a harness by pass wiring directly into the ECM and see what happens.
[This message has been edited by Joseph Upson (edited 03-24-2012).]
Is the 1300RPM idle higher than usual? Is so, my first thought would be that if the idle is high, and the TPS reading is high, then the throttle body is open a little bit more than usual....
Is the 1300RPM idle higher than usual? Is so, my first thought would be that if the idle is high, and the TPS reading is high, then the throttle body is open a little bit more than usual....
Thanks but that falls under the common sense catagory for those of us that have a good understanding of the sensor function. I mentioned in the first thread that the TPS reading was a sudden and drastic change from normal at idle/ closed throttle.
One thing is for sure, TPS function as a steady state/acceleration monitor is much more sophisticated than the thought that it sets a reference point at power on and then adjusts from there. The unsynchronized voltage output relative to throttle position threw off the entire fuel map. It caused a high idle with the abnormally high reading at closed throttle as well as a rich condition.
When I offset it to restore a 0% reading at closed throttle it was a little too far retarded prolonging the 0% reading, or more clearly, a lower % throttle open reading than what it should which obviously leaned out the fuel map.
The above was observed in the datalog.
3 sensors, and two computers along with checking the harness and making the grounds more direct with fewer combined links all grounding at the same location on the cylinder head and that still didn't correct it. I finally slotted the bolt holes on the sensor to make it adjustable which allowed me to restore a proper reading at closed throttle.
To bring this problem to a close, I never did figure out what caused the high off throttle voltage but it was a problem in the harness that caused it. The problem resolved with a new harness and if I had to take a guess I would say a ground arrangement was the culprit as some grounds go directly to chassis/engine and some go straight to the ECM and I probably crossed one up in the previous harness.
This is the same problem I'm having on my 4.9 w/Allante intake. Except mine reads 66% (1.12v) closed and goes to 4.87v at WOT. The Allante TPS is supposed to be .5v-3v :/ Yes I verified the inputs. Yes I switched the wiring. My idle is also higher than it should be. I was thinking of installing a voltage divider circuit...except I don't know the resistance to the ECM in order to attempt to build it.
