Well appears in my last post I was a bit hasty to reply. I stated the evp was bad (it was) and it was replaced. Code went away. Appears I have a new one. I got a code tp sensor out of limits so I checked the sensor. During my signal voltage from the sensor test the volts actually went on the negative side applied throttle and it went to 1 volt so it was moving over a certain point, just way off (out of limit). It was supposed to be .5 then 5 volts with full throttle. OUt of quriosity, I took the negative probe and touched a metal for ground and the volts went to .5 which it should, I applied throttle and it went to 5 volts, as it should. So from my perspective this either points to something being off in the sensor or something to the sensor. I will replace it monday, but if anyone has any comments/sugestions I will respond and appreciate it, you might save me a buck or two. I would adjust the sensor like the other efi tp sensors but the 4.9 is non adjustable. I am posting because something just doens't seem right with the volt being off with a normal test but right on if I apply a new ground. later

 

Ground problems are common. I am not certain I understand, but it sounds like your initial test was a back probe of the TPS connector to monitor signal voltage through the range. The second test was the same with the meter grounded elsewhere. If that is the case I would first unplug and inspect the contacts at the TPS connector for corrosion or other problems. A little water can over time increase contact resistance and give the problem you described. If the contacts look good I would measure resistance over the ground reference wire from the connector to a known good ground (like the spot you used in the second test) it should be very low, less than an ohm or two. If resistance is higher than this I would start tracing back to ground and wiggle as I went. Sometimes grounds have crimp fly-splices buried in the harness that can also have corrosion problems.

The fact that you got the proper output voltage sweep with the throttle opening and an alternate ground points to the ground circuit. TPS is likely OK.

One other thing, because ground reference circuits are often shared, it might explain the multiple code setting.

Good luck, I would be curious to know what you found.

 

I put the probe on the signal return and a known good ground and the pointer was just past the "0" still touching so little to no resistance.


I put the "+" probe on the tp signal and "-" probe on the sig return (per directions), I get a negative reading as mentioned before, with closed throttle and 1 volt with wide open throttle.

If I put the positive probe on the tp signal and a known good ground I get normal readings. 1 volt closed throttle and 5 volts wide open throttle.


OUt of my own quriosity, this wasn't recommended but I did it anyway, I switched the first test I put the "-" probe on the tp signal and "+" probe on the sig return and the reading went from 5 volts closed throttle to 1 wide open throttle.


Any other thoughts (they are much appreciated) no resistance through the signal return and good ground so I suppose it is grounded well.

 

I think I have this figured out. It looks like you simply have the two outer terminals reversed in your ID of them. They are sig return (ground) and v ref (5 volts to sensor from computer). Try retesting using the opposite outer terminal as ground. This is an easy mistake to make as the diagrams in the manual are not always clear as to which side you are looking at.

I couldn't help but notice from the terminology you use that it seems you are using an analog meter for testing. Analog meters are low impedence and can fry the solid state electronics found in modern EFI (computer especially). A digital multimeter should be used that has a minimum measurement impedence of 10K ohm/volt. Not only is this tool much safer to your system it will be much more accurate at the low voltages you are testing. I got mine on sale at Sears for around $60.

Hope this helps.

I was just re-reading this thread, it would not make sense that the v ref and sig rtn are reversed if you measured no resistance to ground at sig rtn. To be sure you could unplug the sensor and measure for DCV at each of the outer terminals with a DMM, the one with 5 DCV to known good ground is your vref, the other should read 0. DCV between the two should be 5 +or- 1 volt.

If unplugged you establish those two wires are behaving, The TP is giving less than 0.5 DCV throttle closed and 5 DCV at WOT. I would say all is well. Sometimes TPS can get a dead spot. It is hard to be certain without and graphic scope to view a wave form.



I'm wondering about your meter.

 

Well you took a good shot at it. In my last post I explained how I tested first per instructions:

I put the "+" probe on the tp signal and "-" probe on the sig return (per directions), I get a negative reading as mentioned before, with closed throttle and 1 volt with wide open throttle.

Then switched, what you suggested:

OUt of my own quriosity, this wasn't recommended but I did it anyway, I switched the first test I put the "-" probe on the tp signal and "+" probe on the sig return and the reading went from 5 volts closed throttle to 1 wide open throttle.

First test the votage went negative at idle/closed throttle and went to 1 volt at full throttle. In the second test ( hooked up backwards) the voltage started at 5 volts with idle/closed throttle and went to 1 volt at full throttle. Either way, it was way off, as it should have been 1 volt at idle/closed throttle and 5 volts at full throttle.


What throws me is the ground to the sensor is good. When I use an alternate ground during the first test otherwise done by the book. It performs as it should. Where do I look elsewhere? I appreciate your help any thoughts on this?

 

Something is amiss. I don't believe there is any negative voltage available to be measured in this system. V ref is always 5 DCV, Sig rtn is always grounded and TP, by the very nature of a TPS, can only send out something in between. If you actually have a full volt closed throttle, this can set a code. it would be helpful to be able to measure down to 1/100th of a volt to establish the low end. I would expect the computer to be looking for 0.5 volts or less.

What kind of meter are you using?

 

A Car quest special (multi-meter). haha. I agree with you something isn't making sense at all. I am performing the test per instructions but like I mentioned when I used another ground it worked correctly. Could there be something crossed/faulty inside the sensor? It appears to be a bad sensor, without my knowing more at least. The vehicle runs/idles very well though. I would think the volts off the sensor, being off that much, would cause some very noticable problems. Let me know what you think

 

jwtaylor, TPS are actually very simple. I don't think your measurements are correct if it runs/idles very well.

Quite simply, the sensor is a variable resistor just like a fuel level sender. a constant 5 volts to sent to it by the computer (v ref) this voltage flows through a resistor to ground (sig rtn). A moveable contact that travels across the resistor is attached to the throttle opening shaft. The moveable contact is connected electrically to TP terminal. As this contact sweeps across the resistor depending on throttle opening angle, the total resistance and therefore voltage changes between TP and ground just like that fuel level sender. The computer is looking at this changing TP signal. By far the most common failure mode for a TPS is contact problems inside. "Dead spots" can occur where the normal linear increase in voltage from closed to open will have a place where the signal drops completely in one small spot. Problem is it is very dificult to see on a multimeter because of slow sampling rates (cheaper the meter the slower the response). If the drop out is a small enough range your computer could see it before you feel it.

If I were you I would satisfy myself that I have a good connection at the TPS, I have a reliable 5 DCV at the sensor over v ref, I have a good ground over sig rtn (as you have done already, though may want to double check with a wiggle test). last thing I would do is verify I have good solid connections at the computer as well and the TP conductor between the TPS and computer is conducting as it should (wiggle test?). If all of these are a go and the TPS code persists, new sensor time.

As a mechanic I can really appreciate your efforts to understand this problem and try to verify it before going into parts changing mode. Most can change parts, real mechanics understand why they are changing parts. In the end it is usually cheaper, takes less time and is downright satisfying!!!

 

I appreciate your helping me out and explaining things, I can tune a carb (how well remains to be seen) but efi is new to me. Not a mechanic and never had to deal with it, until know. haha I will check the connections again and see what I come up with. I will get back to you in a day or so. Someone mentioned I might be able to do a resistance test on the tps sensor. I cannot find referrence of that in the haynes manual. Is this possible? Again thank you for your attention to my question.

 

Resistance test would quickly tell you if you had total faiure. Unplugged you should have some resistance across any two terminals. Truck still runs OK so you don't have total failure. Measuring the voltage sweep across v ref and TP is basically the same thing only better.

Again the TP is not really a go, no go part. That slight "drop out" I described can be very hard to detect without an o-scope. If all checks out with the wiring external to the TPS, and you don't have a 'scope to check the sensor, I'd install a new TPS. They are fortunately not too expensive.

Any of the other codes come back?

 

I will tell your more tonight if it isn't raining that is. Appreciate your assistance. Later

 

Well Okie Dax

Been a while, but here is where I stand now.

KOEO test

522-vehicle not in park or neutral, its a 5 speed so this doesn't appy either way it was in neutral parking brake set, per instructions. So right now I don't care about this code.


CONTINUOUS CODE

332- insufficient egr detected

RUNNING TEST

332- insufficient egr flow

would appear I have a bad egr valve? As a thought I found the evp sensor to be bad replaced and it worked as the test suggested sensor wise. Being the evp senses the opening of the egr valve and now I get these codes (which I didn't get before) so appears the egr valve is faulty huh? I may get a vacum pump and test it or just leave it alone as mpg is normal I believe, or block it off? Any thoughts?

 

The sensor you replaced reports EGR position to the computer. Fords of that vintage had computer controled EGR. The computer does this via an actuator, the EVR solenoid. Again I would start with a couple of tests.

If you unplug the vacuum source from the EGR and connect a vacuum gauge to it you should have about 1 in. vac at idle. revving the engine should produce at least 10-12 in. vac at this hose.

If vac is behaving you can test the EGR valve with a hand held vacuum pump. It should start to open with approx. 4-5 inches vac and be fully open at around 10 inches vac. It should also hold vac with no leakage. A leaky EGR valve diaphram can cause this code.

If vac is not behaving then there is a problem upstream with either a vacuum line, the vacuum accumulator (basically an empty tank plumbed into the vac line system) or an EVR solenoid problem.

If the vacuum readings are low or non existant I would start with the simple first, check those lines. Check to see if there is manifold vacuum (18-20 in.) going into the EVR. I don't know where it is located on the I-6, but it is mounted on the ignition coil bracket of V8s. If vac is there at the EVR then check the line between the EVR and EGR.

I wish I had a nickel for every rubbed through vacuum line I have found to be causing driveability problems/check engine lights. Anything preventing enough vacuum signal to the EGR will set the code 332. Replacing the sensor on the EGR valve is unfortunately akin to killing the messenger.

If the vac lines are good, unplug the EVR and check resistance between the terminals of the actuator. It should be somewhere between 65 and 100 ohms. Either open circuit or low resistance will require replacement of the actuator.

Sounds complicated, but it isn't too bad. This is a common problem. If you need a new EVR solenoid I would expect to have to buy OEM, I doubt the aftermarket will have this part. Last one I bought ran $25-$30.

 

I will test the vac from the source that supplies the egr as you mention and go from there, again appreciate your assistance. Good news I found the evp to be bad and narrowed the field down so I guess I did learn a little, bad news I still have to look, haha, again appreciate it. I will see what I come up with and let you know the results. later

 

Well seems your on to something. I took the vacum line that runs to the egr valve off at the source not the actual egr valve obviously that would do nothing. So I connected the vac gauge up to the source as you mentioned and it is dead not a thing, no vac with throttle or not. So now I will check the vacum line to the evr if it is okay then I will check the resisitance, per your instruction, where you at? I think you'd make/are a good mechanic. Well I will again get back to you later, thanks again