So enriching or leaning out the mixture take this for example - I was running lean at top, and much better at lower rpms IF i tightened the bypass, because it has a great affect at idle and has less effect for the more air that is actually being pushed through and metered at high RPMS with high air flow. So all i had to do was turn the spring CCW inside the afm about 4 teeth or so, then open my bypass more for a leaner mixture at idle and up.

(BTW, do not close your bypass screw all the way doing this, you very well may damage your afm and flap because the bypass screw acts as an escape at rapid deceleration... or your flap goes FLAP FLAP FLAP and bounces around, which is a no-no)

richer all the way up to the top, smoothly*

Ok, I did a little testing: I let the engine cool after work for a good 2 hours..

fuel pressure : 40 psi (not running)

Fuel pressure: 36 psi (running at 800 rmps)

I let the engine warm up for about 25 min, normal running temp

Fuel pressure: 36 psi (running at 800 rpms)

disconnected the vacuum line to FPR: 42 psi

applied close to 20 in Hg to get the fuel pressure to drop to about 28 psi.

So I do note that vacuum is having an effect on fuel pressure, I am not certain that my fuel gauge is accurate, if it is, fuel pressure seems high. I am not sure this is where I need to look though.

I tested the vacuum at the brake booster hook up with the mityvac it read 14 in. Hg this seems low. I will test the vacuum advance with the mityvac and the timing light tomorrow but everything seems to be operating ok, smooth running engine, stable idle ect. I guess I need to test the connections at the ecu using the EFI bible but how would I know or adjust mixture. I am weary of messing with the AFM at this stage. I think I would do that as a last resort.

Edited June 29, 201114 yr by grantf

Well, your fuel pressure does seem a bit high, but as you say it could be inaccuracy in the gauge. If it's really high, the effect would be to richen your mixture.

It would be good to have a static fuel pressure reading, just to be certain your pump is 100%. You can do this by plugging the return line. Alternately, you can just pull your supply line off of the fuel rail and connect it to the gauge.

The only mixture adjustment on an unmodified EFI system is the idle mixture screw just below the outlet of the AFM. In theory, you screw it in (clockwise) to richen the idle mix. In practice the adjustment doesn't do much at all, at least on my engine.

You can seek out some useful articles on the Atlantic Z website (google it) that discuss adjustments to the AFM. This is one approach to adjusting the fuel/air ratio. The other approach is to add a potentiometer (maybe 5 or 10 kOhms) in series with the coolant temp sensor. This is only useful if you want to richen the mixture. You could also add resistance in parallel to the CTS to lean out the mixture, but none of us have tried that yet -- haven't had to.

Don't worry too much about messing with the AFM. If you're careful about what you're doing, you won't mess anything up. Just be sure to mark all your positions before moving/adjusting anything, so that you can return back to factory settings. (You may find that your AFM has already been violated -- broken glue blobs.) You can get into the potentiometer housing by prying off that black plastic cover on the side. It's very easy to tell whether your mix is too lean or too rich by gently tweaking the vane with your finger to find the fastest idle. CW leans out the mix, and CCW richens it. By doing this, you'll get a pretty good preview of how your engine can run with the correct mixture, and you'll also confirm which direction your mix needs to go (richer or leaner).

The vacuum is really low. You need to find and fix that issue, or at least rule out leaks and other causes of low vacuum. That will lower the fuel pressure some, probably not enough, but some. Are you 100% certain you don't have vacuum leaks ? Going by your comments, I would assume that if you had a proper 18" of vacuum, you would be right around 30-32psi on the fuel pressure at idle which should be about right.

Eric, if his non-running pressure is really 40, the fuel pressure will always be too high, irrespective of engine vacuum. However, depending on what his engine wants (and if he has yet another lean-running condition that seems to be so common), that might not actually be a problem.

FAIW, the conversion between psi and in Hg is very close to 1:2. So it would take approx. 20" vacuum to drop the fuel pressure from 40 to 30 psi. (His gauges are a bit inconsistant, perhaps.) If he's getting 34 psi at 14" vacuum, that would correspond to 41 psi with no vacuum. An additional 4" vacuum (total 18" vacuum) would give him a pressure of 32 psi, when he should really be getting more like 36.3 - 8.8 = 27.5 psi.

Again, it's not a problem if the engine has a lean-running condition anyway. Fixing the problem would make the engine run leaner/worse.

Edited June 29, 201114 yr by FastWoman

Oh, getting back to Eric's question about remaining vacuum leaks, have you done the "yogurt cup test" (aHEM!... intake/crankcase leakdown test) yet? Oddly, you can have a vacuum leak and still have a high vacuum, as long as the resulting mixture is correct. In that case if you were to fix the leak, the mixture would be too rich, and engine vacuum would actually be lower (after you opened up the throttle plate a bit to compensate for the drop in RPM, of course). All the vacuum reflects is engine running efficiency, although a leak will certainly throw your mixture off.

The key test at the ECU connection is for the water temperature sensor. My engine loses 1-2 mpg during cold weather just because the engine never really gets hot. At the ECU, measure resistance between pin 13 and ground. Compare the value you get to the table attached. Notice that as the resistance goes up, the ECU thinks the engine is cold and adds more fuel. Corrosion or oxidation at any of the connections between the ECU and the sensor can increase the resistance.

The air temperature sensor does not have as much effect but you might as well measure it also. Measure between pins 6 and 27. It's the same type of sensor so the same table can be used.

If you measure before you drive the car for the day, then the sensor should be at ambient temperature. If one reading is correct, the odds are the other temperatures will be also.

Eric, if his non-running pressure is really 40, the fuel pressure will always be too high, irrespective of engine vacuum. However, depending on what his engine wants (and if he has yet another lean-running condition that seems to be so common), that might not actually be a problem.

FAIW, the conversion between psi and in Hg is very close to 1:2. So it would take approx. 20" vacuum to drop the fuel pressure from 40 to 30 psi. (His gauges are a bit inconsistant, perhaps.) If he's getting 34 psi at 14" vacuum, that would correspond to 41 psi with no vacuum. An additional 4" vacuum (total 18" vacuum) would give him a pressure of 32 psi, when he should really be getting more like 36.3 - 8.8 = 27.5 psi.

Again, it's not a problem if the engine has a lean-running condition anyway. Fixing the problem would make the engine run leaner/worse.

I can't remember exactly what PSI it was, it's been a while and I've forgot. He does say in his original post though that when he applies 20" of vacuum with the mighty vac, it drops the pressure to 28. Something still doesn't sound right though, even with my FPR disconnected, my fuel pressure isn't that high (40psi) I know. I'm also wondering how the non running fuel pressure was read, since the fuel pump shouldn't be running when the car isn't running.

The main thing is that the higher the vacuum, the lower the fuel pressure. More vacuum opens the FPR up to drop the pressure. He's got to get the vacuum up.

Zed is right too, the WTS is a good place to check, as well as other tests on the EFI system. I figured ruling out vacuum leaks would be the first place I would start before moving on.

Edited June 29, 201114 yr by cozye

My fuel pump always runs with the key in the on position. I know it shouldn't but it does. Also I do have an aftermarket fuel pump. I am still looking for vacuum leaks, perhaps It's time for yogurt (I don't really like yogurt).

measuring the temp sensor:

I get 2.27 k at the ecu connector it is about 65 f outside my car is a little warm been cooling for about an hour and a half. I will check again before I head out later tonight but I would think that it is at least working and I have a signal.

So at this stage I think tracking down vacuum leaks seams like the best thing to do, I feel that adjusting the AFM before doing this would be working backward, meaning that I would be using those adjustments to potentially compensate for inefficiencies elsewhere in the system. Does this make sense?

measuring the temp sensor:

I get 2.27 k at the ecu connector it is about 65 f outside my car is a little warm been cooling for about an hour and a half. I will check again before I head out later tonight but I would think that it is at least working and I have a signal.

So at this stage I think tracking down vacuum leaks seams like the best thing to do, I feel that adjusting the AFM before doing this would be working backward, meaning that I would be using those adjustments to potentially compensate for inefficiencies elsewhere in the system. Does this make sense?

Yes, it makes perfect sense. I wouldn't mess with the AFM or anything on fuel metering until I've thoroughly checked out everything else. On the WTS, just because you are getting a signal doesn't mean that its working. 300 ohm difference can make a huge change in your fuel metering. You should check it and make sure it's giving an accurate resistance for a given temp. I would get the car hot, up to full temp, then take the measurement. When they go bad, they don't break completely, they start giving higher resistance which makes the car run rich.

Do Sarah's yogurt cup test. Also you can hook a propane torch up to a small piece of hose, and blow propane all around the intake and vacuum lines and listen for a rise in rpm. The yogurt cup test will find small leaks pretty well though I think.

Are you sure no one has messed with the AFM in the past ?

Are you sure no one has messed with the AFM in the past ?

No I am not sure, but it is very possible.