I'm testing my EGR valve to eliminate it as a possible cause of my MPG issues.
First questions:
(1) What is "herky-jerky"? I experience a very mild jerkiness in lean mode.
Basically it feels like the engine looses power for an instant and then runs normally for an instant. Hence the nickname "herky-jerky".
The engine needs to be thoroughly warmed up to have a smooth lean burn. My car won't go into lean burn until I see at least 154° F coolant temperature with my OBDIIc&c with the ambient air temperature above ~68°. At that point I may feel the slightest amount of "hesitation". A few more degrees of coolant temperature rise and the engine smooths out.
(2) Have some people with "herky-jerky" cured it with an EGR valve replacement?
I found a new looking 1997 Acura 3.0 CL V-6 Siemens EGR valve at my favorite junk yard for $5. It cured my herky-jerky.
I measured the sensor voltage while driving with my OBDIIc&c. While in lean burn the sensor measured a constant 0.9 volts.
Unfortunately I don't remember what the max sensor voltage was when I was checking the valve[edit] OUT of[/edit] lean burn. I seem to remember something like 3 volts. But that wouldn't be the applied voltage to the solenoid.
Do your tests show that the valve is open at all with a 0.9v sensor reading? I haven't tested my original EGR valve to determine that.
I bought used 2001 Accord valve (ebay, $18 ) to experiment with. Same connector, same mating surface, but wider top and it did not fit without some "encouragement" (I have not tried to fully install it yet).
I found a way to bench test the Accord EGR valve by holding it open with 6 V, then pushing it closed with my finger while monitoring the potentiometer leads (from the connected, EGR not opened). (It takes 0.6 volts at 6 V and 1.8 amps at 13.8 volts but I have a feeling it is not designed for constant operation at the higher voltage, so I would not advise extended operation at a high voltage.)
I would guess that the voltage to the solenoid is a pulse width modulated X volt signal. Your 6 volt DC applied voltage would amount to 50% PWM drive if the peak voltage is 12 volts DC.
Also since the EGR plate has a restriction plug in the passage to each cylinder, the EGR valve probably has very little movement to reach the maximum flow rate that the restrictions would allow.
Please explain what "It takes 0.6 volts at 6 V ..." means. Is 0.6v the sensor output voltage at 6 volts or perhaps you mean the solenoid current is 0.6A at 6 volts?
I found that the valve's resistance changed linearly through most of the travel, but when partially open, it deviated from this straight line, indicating that the wipers may have cut through the material in this area, where the valve most likely operates most.
This means that the Accord EGR, a junkyard special, is No Good. It gives erroneous position readings in the range it would be used most. The ECU would then command it to reposition. Could this cause "herky-jerky"? Thus the questions above.
Sure sounds like it could. But that assumes that the bad area is used mostly in lean burn. That raises the question of which wears out the sensor's resistance the most, moving to the 0.9 volt lean burn spot or the normal variation of the sensor voltage when not in lean burn.
You could answer that question by varying the solenoid voltage to give a ~0.9 volt sensor reading and see if that is in the bad area of output voltage.
The next step is to run this test on the Insight valve in the car. I have a feeling it will behave similarly just from having read about potentiometer wiper problems being common. I am working out a way to test it without removing it from the car.
