QuestionI tested O2 Sensor Output Voltage with a DMM and an OScope in parallel. O2 Sensor output was still connected to engine management computer. Measurement leads spliced into O2 output node and grounded on the engine block.
Turned ignition on (engine not started): 0.5 volts.
Started engine cold: for the first few mintues, voltage remained steady at 0.5 volts.
As engine warmed, voltage slightly rose ... 0.52. Revving engine intermittently. After several minutes, voltage began oscillating. The DMM was now jumping around quickly, indicating 0.44, 0.6, 0.4, 0.6, 0.8, 0.7, 0.6, 0.3 ... Wile revving, voltage seemed to average around 0.6 - 0.7. While idling, voltage also averaged around 0.6.
I was unable to get an exact measurement of frequency/period - the o-scope was forty years old (no kidding). It was indicating around a 200 ms period, whichi corresponds to 5 Hz, while your number of 4 ms corresponds to a much higher frequency of 250 Hz. (Are you sticking by this value? :) The DMM, meanwhile, indicated a varying frequency, ranging between 30-60 Hz, but I don't know if that result is valid or not [not sure if my DMM is designed to measure the frequency of this type of signal (oscillating DC)]. In any case the 02 sensor's output frequency was far less than your 4 ms figure.
My mixture seems slightly rich overall, since my estimate of the average O2 voltage is 0.6 rather than the ideal 0.45. This is true for idling AND revving. I know how to adjust the idle mixture. How can I adjust the part throttle mixture? Or is the 0.6 average an acceptable figure, if slightly rich?
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Is it easy to detect whether an O2 sensor is operating correctly by evaluating the engine's running characteristics?
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
This would be funny if it wasn't so sad. A few weeks ago I cleaned out my Alfa Spider's Aux Air Valve with carb cleaner, rinsed it out, and then had the bright idea to lube it by spraying some Silcone Spray Lube in there. Didn't change the valve's behavior at all, but...today I read that silicone in any form--even in the tiniest amount--is instant death for O2 Sensors. I hadn't heard this. I drove around for a few weeks after spraying it with silicone, then last weekend by dumb luck I put in a new O2 sensor, which is probably toast now too. Do you concur? What I want to know is, how would I check to see if the sensor is working without disconnecting it? (It's a splice, not a connector.) In other words, how do I tell if I'm in open-loop? Recommendations? This sucks.
Also, could my car want to die after startup due to a clogged catalytic converter? It was literally red hot a few months ago due to a missing cylinder...did I mention this sucks? Gotta love it though...
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Do you know a way to check if my O2 sensor is working (i.e. verify my wiring splice is ok) without disconnecting it? It didn't seem to run much differnt when in open loop... Also, when I was adjusting my idle mixture with the sensor disconnected, I reved it up a few times - then realized, doesn't this diconnected condition cause it to run extremely rich and risk fouling the plugs (and indeed, the O2 Sensor)?
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Great information. Again, what is an acceptable power supply to the fuel pump? My measurement of 11.88 VDC at the pump terminals was taken at idle - am I correct in assuming that voltage would nudge up if the engine (Alternator) were running faster?
How do you mean about the positive bypass? Connect 12V directly from the battery with engine on? Would I disconnect the existing positive conmnection at the pump for this test? Finally, how would I go about reducing the positive line voltage drop?
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Hello Mr. Sokol - Regarding my Alfa Spider's noisy fuel pump, and startup-then stall-if-it's-been-sitting-awhile syndrome, I followed your electrical measurement suggestions - will you please help me interpret them?
First, with the engine cold and not having been started, I set up test leads from my voltmeter so as to measure resistance from the negative battery terminal to the negative lead of the fuel pump, having measured the inherent resistance in my test setup's wiring (associated with the test leads daisy-chained as necessary) as 0.2-Ohm. I found that the measured resistance between the negative terminals of the battery and fuel pump also measured 0.2-Ohm, indicating a near-zero resistance, or, in my interpretation (being aware that zero electrical resistance is highly unlikely here), at the very least, a good ground.
I then started up the car and took additional measurements. Under running conditions, the voltage drop between the negative terminals of the pump and battery was 0.2-VDC, significatnly higher than your number of 0.03-VDC being the upper limit of acceptable.
The above result seems to contradict my measurement of the resistance of the ground path; as you know, in general, the lower the resistance, the lower the associated voltage drop/division. Although I took these measurements with care, I cannot make these numbers "add up" per the formulas I used when I was earning my electrical engineering degree. Perhaps the resistance in the path to ground increased due to heat when the pump started to run? By the way, the battery voltage (engine running) measured 13.8-VDC. Furthermore, the voltage between the positive and negative terminals of the pump (engine running, pump buzzing away) measured 11.88 VDC. (1) Is this an acceptable value? And (2), isn't this what we're REALLY interested in, in the final analysis (i.e. pump's supply voltage)? In light of above, do you still recommened I provide another ground for my fuel pump?
Turning my attention to the O2 sensor, I measured its output voltage while idling and adjusted the AFM screw so as to obtain 0.8-VDC, which is about where it already was. Just curious - what is this really adjusting and why does it only affect the idle mixture? Also, why didn't the car run any different when the wire was disconnected?
Thanks
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Could a voltage measurement at the main fuel pump terminals, measured while the engine/pump are running, tell me if I need to bother with splicing in another path to ground? (E.g. if I see 12V under load vs., say, 5V?)
I question that I need an in-tank pump upgrade due to an inherent insufficiency in that system. The car hasn't always had this little problem. And as I noted the in-tank pump was replaced just a few days ago.
It starts and idles fine immeaitely once it's been running. Could this indiacte something as easy to fix as idle mixture adjustment? Maybe fuel pressure is good but since the engine has more friction, it struggles...I know the idle speed is high if anything now, so it could only be mixture now, for an EASY fix...
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
My in-tank pump (and sock) was recently replaced with a larger Delco unit for a Volvo, so probably ok there. Q1: If my Alfa were losing resisual fuel pressure when parked, would this explain the "starts instantly-->nearly stalls-->runs perfect..." syndrome? Q2: How do I adjust the valve flap tension? [With these idle adjustments I've been making, I think I'm addressing the symptoms but not the cause.] Q3: What does a buzzing main fuel pump usually indicate? It's always buzzed steadily when the car is running and is only a few years old. Drivability is perfect except for the start-then-stall bug.
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Great info. My Spider's in great shape and running quite well (never quite well enough for me though...) but I'll keep bugging you, for my education.. :) Now that the CSI is working, it starts so fast when I turn the key that I wonder if fuel pressure isn't quite up to speed immediately after startup, causing it to stumble in the first few seconds. Next, I'll fiddle with the screw on the AFM.
I should clarify my description of the first item below - the strange, off-and-on "mooing" noise I hear, often in parking lots in the minutes after startup - it's not the differential, because it happens moving or not, often just idling. It almost sounds like it's coming from the Engine Management Computer (not likey I assume). Definitely somewhere aft of the seats. An odd noise. Doesn't affect running. I'd think it was a fuel pump, but I can hear them too and it sure doesn't sound like it.
When I asked you about the rather loud buzzing my new(ish) main fuel pump has always made, you recommended grounding the fuel pump to the chassis - why? - you must suspect a supply voltage issue. Please note the fuel filter and in-tank pump were recently replaced, so should be ok there. Replacing those parts did not change the buzzing noise.
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
Interesting ideas. I appreciate your expertise. I couldn't adjust the timing to save my life. However, I bet they checked it when it was in the shop recently.
I've got the warm idle at a little over 1000 rpm. I adjusted it with the "idle adjustment" nut over by the intake manifold. Also put in NGK plugs. And an MSD coil. It still stumbles and almost stalls on startup. Fuel pressure was checked good recently. It's running GREAT besides this idle issue. Always something! Haven't touched the nut on the AFM yet - afraid I'll screw up the mixture.
You mentioned the fuel pump. I repalced my Bosch pump with a new one just a few years ago, and it has always made a rather loud buzzing sound, all the time. I can easily hear it over the engine when outside the car. Is this normal?
By the way, what is that intermittent "howling" / "mooing" noise I hear over my shoulder when driving it? It doesn't sound anything like the fuel pump buzz, although it seems to be coming from the rear.
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
At what temp is the aux air valve supposed to close off completely? Wondering if it's supposed to be a factor when starting up in warm ambient temps (e.g. 75 degrees outside). Should I try messing with the adjustment screw?
-------------------------------------------
The text above is a follow-up to ...
-----Question-----
I'm shining a flashlight into my 1984 Alfa Spider's Auxiliary Air Valve. How wide should the opening be at 30 degrees-F? Mine opens like a peice of pie, and less than a quarter of the area of the circle opens at that freezing temp. When hot, it closes completely. Is this how they're supposed to operate?
Reason I'm looking at is, I've been having idle problems. If my Alfa sits out all day in the sun, it starts up instantly, then dies. Over and over. In that case, once I get it running by applying throttle while turning the key, it's totally fine the whole drive home. Starting in a COLD morning, it starts instantly, then ALMOST stalls, then is fine...
-----Answer-----
About half the area should be open around 30F. Any Bosch air valve will do so get the best price. You could also try advancing the timing 4 degrees, and make sure the warm idle is 1000rpm. Readjust the idle co screw on the air flow meter to slightly richer (turn in--clockwise).
-----Answer-----
You have two problems which is why I gave you the tuning parameters. I have had hit and miss luck messing with the adj. screw especially on units more than 2 years old. The aux. air valve should not be open at 70F. Do the tuning stuff and then you might try loosening the spring tension on the air valve flap, but, only after trying the tuning stuff. Of course. concerning the hot start problem: it might be a fuel pump starting to fail. Bosch pumps seem to not start rather than quit while running.
-----Answer-----
Re: the noise. The differential is limited slip type and needs a 75w-90 synthetic oil that needs to be renewed from time to time--now is the time.
Re: the pump. Ground the pump to the chassis and replace the intank pump with a BBK unit for an 86 and above Mustang. (Summit Racing has it).
Re: the pressure check. This is a common error that mechanics make--volume must also be checked.
Re: the mixture. You will only affect the idle mixture.
-----Answer-----
I am not there to hear the noise but it must be fuel pump related and there are two ways that noise occurs with a new pump. 1) the ground is faulty so splice a wire into the negative lead of the pump and attach it to the chassis nearby.
2) the intank pump is not sufficient as originally designed and that is why I recommend the higher pressure BBK pump be installed--it always brings an improvement.
-----Answer-----
1)Now is not the time to adjust the tension of the air flow sensor.
2)The check valve on the output side of the main fuel pump is required to hold and gradually lose residual pressure for at least half an hour.
3)Volvos with two pumps quite often suffer this malady also because they use a low pressure in tank pump--increasing its volume does not always cure the problem but a higher pressure pump does.
4)The noise of the pump I have always found associated with one or both of the issues I mentioned.
5)Most of the time the need for an intank pump is because of excessive heating of the returning fuel...
-----Answer-----
In an earlier exchange I said to turn in the screw on the air flow sensor (under the plug that should be there) and to do it while measuring the voltage from the O2 sensor--you want to see .8 volts at idle from the disconnected O2 sensor. This will richen the idle mixture only--remember, use a digital voltmeter.
RE: testing the pump. You do a volt drop test by connecting the (digital) voltmeter positive lead to the negative lead of the pump and connect the negative lead (you will have to add cable)to the negative terminal of the battery. If all is good you will see a reading of 0.03 volts, any higher means the ground is poor.
-----Answer-----
I did not catch the misprint. I meant to write 0.3 volts. Your ground system is fine, it is the positive that has an unacceptable volt drop, a maximum of a 1 volt drop is considered acceptable in automobiles. By the way, I have worked on Italian cars since 1957 and the electrics have always been interesting. Run a bypass to the positive side of the pump and see if it improves performance of the pump.
Hot/cold resistence does change. take a reading of a cold light bulb and do the math. The engine runs fine when you disconnect the O2 sensor because it reverts to the open loop program. The screw in the airflow sensor controls bypass air, the less bypass air the more the flap moves, the more bypass air the less the flap moves. If you experience a lean surge at a steady 70mph then you weaken the spring tension of the flap to richen the mixture. Above a 66% throttle opening open loop programming is again engaged.
-----Answer-----
Yes, run an additional wire directly from the battery while engine is running and see if it helps the pump. However, even though automotive electrics must have a wide voltage tolerance the real concern is the excessive volt drop--it indicates a potential complete failure of delivery due to a burned/corroded connection that will get worse over time. The only fix is to check all connections in the system including the relay and its connections (a common place for the fault. Over time spade connectors lose their tension).
-----Answer-----
In open loop the computer uses a prepared map to run much like a carburetor. If you did not see a voltage change while revving the engine and turning the screw then the sensor is faulty. As you opened the throttle the voltage should have dropped (lean) then recovered. (down to maybe .2 and then up to .6). Now to test the entire system as well as the sensor. With the sensor lead disconnected read the voltage coming from the computer--it is a reference voltage and should be as close to 0.5 volts as possible. If not there is a ground problem at the computer. Now, reconnect the lead and attach the volt meters lead there also. With a warmed up engine you should see a voltage swing above and below 0.5 volts. The more rapid the oscillations the better. However, a DMM cannot record all the swings of a fresh sensor. You can with a DSO (digital storage oscilloscope). A fresh sensor will have a 4ms phase. As the sensor ages it will slow down. When it gets lazy (hopefully after 2 years) fuel mileage suffers and it is time to change it out. If you get no voltage change with the lead connected but do with it disconnected then the computer has a problem.
Also, with the lead disconnected, touch the computer side of the lead to ground and see that the voltage rises (driving the system rich) then touch the battery positive with one hand and with the other hand touch the lead from the computer and the voltage should drop (driving lean). Do not apply 12 volts directly to the lead and do not ground the sensor lead ever. And all of this must be done with a DMM not an analog voltmeter.
-----Answer-----
The answer to how to test is in my last post to you. Now..about the silicone. That statement is partially true and only applies to silicone gasket maker not silicone lubricant and what causes the failure is a product released during the curing time. That is why most currant gasket making silicone glues state safe for O2 sensors on the package.
-----Answer-----
Were you able to read my previous 2 posts... The answer is: not without the tools I wrote about.
AnswerThe DMM cannot read the cycle of the O2 sensor. All the readings are correct and the system is functioning as designed. The perfect combustion point for efficiency is not the perfect combustion point for power. The idle reading is what I would want... If that was an analog oscilloscope it might not have had the impedance needed to safely read this stuff.
