Wtf?

[Bill Blue]

Forgive me for chiming in - you guys are both light years ahead of me on this subject, but something Bill said earlier in the thread struck me.

Bill has had basically the same problem (unpredictable performance) with two different engines (Pinto & Zetec) and two different fuel delivery systems (carb & EFI). So let's identify what these two setups have in common, and look for a problem there.

Bill will have to answer that question, but I am thinking fuel tank, possibly fuel filter, fuel lines, fuel pump. Could we have some debris in the fuel system randomly influencing the results?

I hope this helps. Hang in there Bill!

Todd, believe me, I've given this considerable thought.

Here is the background. During the initial rebuild of the car, the tanks were cleaned, stripped and recoated with POR 15 tank sealer, which seems to be holding up. The tanks and plumbing are dead stock. There is a fuel pump mounted under the trunk floor. Initially, it was a Bosch, intended for use in a Volvo 240. It is now an OEM Ranger/Mustang replacement. The two are quite similar, the Ford unit has about twice the flow. I've measured the flow at zero pressure, it is very impressive. Two small steel bodied, 5/16" filters are used, one before and one after the pump. Return flow has always been to the passengers side tank. 5/16" O.D. steel tubing was used for the supply and return line. Initially, 1/4" O.D. was used for return. The 1/4" line was presenting considerable back pressure, the 5/16" line stopped that.

In short, the only common components are the supply fuel line, which is new and was plugged for installation and the fuel filters. They have been replaced with new ones of the same description. I don't see how the problem could be with the filters as the issue presents itself across the rev range, with idle and off idle performance a biggie.

Jarrid has inspected my tune and found a few problems. He also suggested a minor wiring change, jumping between too computer terminals, which should help stabilize the tune. Well, nothing is ever that easy around here. Way back when I wired the computer, I screwed up a couple of wire terminals in the computer connector. So a couple of then unused wires were left out. Now that they are needed, I had to order some terminals in order to finish wiring the connector, they should be received tomorrow.

I am interested in determining the problem, so I am holding off on the tuning change until I can implement the jumper mod. The bulk of Jarrid's tuning changes are designed to trouble shoot the system and if the jumper works, I'd rather skip that exercise and return to tuning. But those changes may represent the quickest way to a good tune. I'll just have to wait and see.

Bill

 

[Bill Blue]

While waiting for the wire terminals (last Friday I received an email they were ready for shipment, this morning I received notification from UPS they have been shipped) I have continued to drive the car, mostly short drives. It seems the variation in tune gets smaller the more I drive the car, but it is settling down on the wrong side. Has picked up a "hole" from 1K to 2K, especially when cold. Hiccups during moderate acceleration, the "good" off idle response is not as good as it was (and is getting rare) and it has picked up a soft popping in the exhaust between shifts.

Interesting, to say the least.

Can't wait to install the jumper to see what effect it will have. Will resume data logging at that time.

Bill

 

[RootesRacer]

While waiting for the wire terminals (last Friday I received an email they were ready for shipment, this morning I received notification from UPS they have been shipped) I have continued to drive the car, mostly short drives. It seems the variation in tune gets smaller the more I drive the car, but it is settling down on the wrong side. Has picked up a "hole" from 1K to 2K, especially when cold. Hiccups during moderate acceleration, the "good" off idle response is not as good as it was (and is getting rare) and it has picked up a soft popping in the exhaust between shifts.

Interesting, to say the least.

Can't wait to install the jumper to see what effect it will have. Will resume data logging at that time.

Bill

The popping while shifting is a no brainer, your decel fuel cut is not enabled.
I noticed this when I was looking over your map.
This wont effect driveability but likely rumbles when your lightly coasting too.
Your decel fuel cut is currently set to 10kpa. Setting it to a higher value but lower than your idle manifold pressure will "cut" the injectors on an overrun.
No fuel, no cackle/popping.
You also will want to setup your MAX TPS FOR FUEL CUT parameter to prevent a fuel cut while you are free revving the engine. Setting it just 2% above the idle TPS value is usually good.

One thing that has bothered me on your setup is the mismatch you have on the fuel delivery side.
Recall at one time your fuel return line was undersized and the fuel pressure was all over the place as a result.
I am not convinced this is still the case.
Keep in mind that if you have an oversized fuel pump, ALL extra fuel not consumed by the engine must return to the tank by the dumping fuel pressure regulator.
If say when the gas is cold, the pump is able to produce a higher flow at its control pressure, and if the return line is restrictive, fuel temperature and fuel viscosity will affect the rail pressure.

Having the fuel pressure vary by 2 or 3 PSI in a non repeatable way will prevent you from having a reproducible tune for the engine.
One important reason for having matched fuel system parts.

Is your FPR an unmodified (pressure modified/boosted unit)?
Whatever your pressure is now, based on your logs the injectors are either a bit large or the fuel pressure is higher than it ought to be.
This is not necessarily a problem as having large injectors should not create
a non-reproducible tune. The condition does make one want to come up with an explanation as to why the injector duty cycle is low as it may point out the source of a problem.

 

[Bill Blue]

It is interesting to note the decel fuel cut is not enabled, as it certainly was when I first fired up this baby lo those many years ago. It is something I never messed with as I was very happy with the way it worked. In fact, I don't believe I have ever looked at it.

The FPR is a stock Isuzu 43 psi unit, the injectors are 34 lb Bosch. I have no idea what the stock injectors are, I had read they were 27 lb at 51(or so) psi, but the change I had to make in the tune when I went to bigger ones make me think they were closer to 20 lb. The per cent injector duty went from about 94 to 60. When I swapped injectors, I also dropped rail pressure.

As to the potential changes, I have resolved to change nothing until I have installed the jumper, determined its effect and we can proceed from there.

These last few posts are more or less idle musings.

Bill

 

[RootesRacer]

It is interesting to note the decel fuel cut is not enabled, as it certainly was when I first fired up this baby lo those many years ago. It is something I never messed with as I was very happy with the way it worked. In fact, I don't believe I have ever looked at it.

Bill

When I shipped the the ECU to you it was disabled, it still is and unless you tuned it and it changed back to how it shipped, it is what it was.
I always ship with fuel cut disabled becuase it can create a show stopper at startup if I know nothing about how the engine idles ETC.

If you had ever complained about popping, I would have then told you how to tune it. Many engines will run fine without having fuel cut, but this depends on the exhaust system and how lean you like to run the engine.

 

[Bill Blue]

Hmm, good things to know. My memory is playing tricks on me.

Something you mentioned that has me somewhat puzzled.

"One thing that has bothered me on your setup is the mismatch you have on the fuel delivery side."

What is this mismatch?

Bill

 

[RootesRacer]

Something you mentioned that has me somewhat puzzled.

"One thing that has bothered me on your setup is the mismatch you have on the fuel delivery side."

What is this mismatch?

Bill

Your mismatch is on your fuel pumps capacity and the size of your return line.
You made it better by going to a larger diameter return line.
If memory serves you ended at 5/16 ID as it was the largest you could run or something like that.
OEM setups usually have 3/8 or so (only the smallest systems use 5/16 line).
3/8 will have 44% lower pressure drop than 5/16.

Keep in mind again that a 3psi variation in non reproducible injector pressure differential will result in a 10% shift in AFR (14.5 could shift to 13 or 16 AFR).


The second mismatch is on the size of your injectors or fuel pressure for same with respect to your engines needs.

For your engines RPM reality, you should ideally have max pulse widths of around 15 to 16ms. This would show that the injector sizing is optimal for your engines HP and RPM requirements.

I'd like to see 80% duty cycle at for worst case fuel requirements which for you is going to be at around 6500 RPM.

The equation for this is:
Max PW = .8 * (2 * (1/(RPM/60)))
Where the .8 is 80%
RPM is the max horsepower RPM
2 is the revs per injection cycle
and 60 is converting RPM to revs per second.

 

[Bill Blue]

Previously, I had used a 100 psi gauge to measure rail and return pressure. All I could tell about return pressure was it would move the needle a barely perceptible amount. So I dug out a turbo pressure/vacuum dash gauge. It reads to 15 psi over a 150 degree sweep. Zeroed it and hooked it up with 6 ft of 3/16" hose that will reach to the drivers seat.

Initial pump run gives about .8 psi. Running at idle, it is 1.0 psi. With the revs up and throttle closed, it goes to about 1.5. I have yet to drive it on the road, but on several start ups and a little first gear driving from garage to basement to garage, it is very predictable.

I see that Dorman makes an affordable nylon fuel line. I am going to check into that tomorrow as I think I could insert a larger line that was flexible. I am maxed out on the rigid stuff.

Bill

 

[socorob]

I picked up some of this but haven't gotten round to running it yet. It's a good price for the stainless Teflon stuff.
http://aeroquip.cc/s-6twfc807-06aeroquipptfehosefc807-6.aspx

The alpine isn't a very long car so if you ran it right you would only need a few feet and a couple fittings.

 

[RootesRacer]

Previously, I had used a 100 psi gauge to measure rail and return pressure. All I could tell about return pressure was it would move the needle a barely perceptible amount. So I dug out a turbo pressure/vacuum dash gauge. It reads to 15 psi over a 150 degree sweep. Zeroed it and hooked it up with 6 ft of 3/16" hose that will reach to the drivers seat.

Initial pump run gives about .8 psi. Running at idle, it is 1.0 psi. With the revs up and throttle closed, it goes to about 1.5. I have yet to drive it on the road, but on several start ups and a little first gear driving from garage to basement to garage, it is very predictable.

I see that Dorman makes an affordable nylon fuel line. I am going to check into that tomorrow as I think I could insert a larger line that was flexible. I am maxed out on the rigid stuff.

Bill

That it runs even at 1.5psi is not a good sign.
Any pressure variation at the dump side of the regulator means the rail pressure will be affected the same.
That pressure will be affected by many things, the greatest of which is the pumps efficiency which will wain with time.

What is the ID and length of your return line right now as it stands?

IIRC you were using hard brake line or something?

 

[Bill Blue]

Currently it is 1/4" I.D. hard line, about 15 feet long.

 

[RootesRacer]

Currently it is 1/4" I.D. hard line, about 15 feet long.

3/8 ID would yeild about 1/3 (44%) the pressure drop compared to 1/4.
How you bend the tubing (its bend radius) is very important too.

If possible reduce the length too.

If I saw a worst case 1/4psi pressure at the regulator, I think I could live with that.
I would probably want a wideband O2 sensor and closed loop lambda control though to allow for variability in all components so the inevitable change in parameters and the effect on tune doesnt effect your tune enough to spoil the driving experience.

 

[Bill Blue]

WOW. The narrative goes from 3 lbs unreproducible variation to 1/4 lb total back pressure. My head spins.

As to hose diameter and length. This is going to take some evaluation. Increased flexibility creates options, 1/2" O.D. eliminates options. At the moment, I have no idea what is possible. I do know that everything has to remain tucked in above the bottom of the cruciform. This is not a show car or boulevard cruiser.

Bill

 

[Ken Ellis]

Gentlemen, I have a question to pose as I'm watching this thread.
It's a plumbing question: Does the fuel return line tap in right after the fuel pump and regulator, or "after" the fuel rail and injectors?

And if it's after pump and regulator, can/should these two items be at the tank, rather than up front? (I know many modern cars have the pump in or at the tank.)

 

[RootesRacer]

WOW. The narrative goes from 3 lbs unreproducible variation to 1/4 lb total back pressure. My head spins.

Bill

Bill you need to read closer.

The 3 psi was to show you the effect that 3psi would have on your AFR.

I can live with 1/4 psi because then that source of error becomes in the same order of magnitude of other uncontrollable sources of error.

 

[RootesRacer]

Gentlemen, I have a question to pose as I'm watching this thread.
It's a plumbing question: Does the fuel return line tap in right after the fuel pump and regulator, or "after" the fuel rail and injectors?

And if it's after pump and regulator, can/should these two items be at the tank, rather than up front? (I know many modern cars have the pump in or at the tank.)

There are two ways to do it.
The most common is to have a fuel pressure regulator which allows flow through the fuel rail and excess pressure is dumped to the tank.

The other is to put a pressure transducer on the fuel rail and control the pump via PWM to maintain a controlled rail pressure.

Each has advantages, the second being cost. The former is best IMO in that bubbles can return to the tank if they get into the fuel lines, whereas with the latter the bubbles have to be ejected through the injectors which is slow and creates short term loss in fuel control.

 

[Ken Ellis]

Got it. Thanks... I'll get back on the sidewalk now!

 

[Bill Blue]

Bill you need to read closer.

The 3 psi was to show you the effect that 3psi would have on your AFR.

Right. At the present, it appears I have .2 psi of unreproducible variation. How much impact does that have on the tune?

BTW, it appears fuel line is sized by O.D. The largest nylon I can find is 5/16" I.D. x 3/8" O.D. If I install the 5/16" x 3/8" line, wouldn't that decrease the psi of the variation by 44%?

Bill

 

[Bill Blue]

Just looked up my fuel pump. Ford used it on the 1985 2.3 Ranger. That would not that be considered a large system by today's standard, would it?

Bill

 

[RootesRacer]

Right. At the present, it appears I have .2 psi of unreproducible variation. How much impact does that have on the tune?

BTW, it appears fuel line is sized by O.D. The largest nylon I can find is 5/16" I.D. x 3/8" O.D. If I install the 5/16" x 3/8" line, wouldn't that decrease the psi of the variation by 44%?

Bill

Bill,

How can you say what the unpredictable pressure variation is?
If you are seeing 1.5psi static error, the error is going to be at least that since the pumping variation and fluid viscosity changes will only make it worse at lower temps. The back pressure will also drop as the gasoline heats up.

The back pressure will be cross-sectional area an length dependent barring kinks and bad bends.

The pressure is going to be low (by EFI standards) so it doesnt much matter what the material is so long as it is compatible with the fuel.

Thinwall stainless is often used, I've seen thinwall aluminum too but I would worry about corrosion with all the ethanol they are putting in the fuel.

EDIT:
Looked back at the thread and saw that Bill was seeing 1.5psi static error not 1.75.
Even still that is a huge back pressure compared to what it should be.