Problem Roush 427IR on the way to Roush

[olddog]

I would assume it ran great for Roush. You are at a different altitude and have different gas. How far would you have to drive to get down to the altitude Roush is at? If easy to do, that would tell you if it was the gas or the altitude.

I had a 1987 Mini Van that use MAP (no Mass flow sensor). Every time I drove it to Florida it would trip a code. The Map sensor had a problem reading the higher pressure at sea level, but it worked fine at the lower range.

From what you said it reminded me of this and makes me suspect a bad MAP sensor, however it doesn't make logical sense. Higher altitude is lower pressure. An engine always runs through the low pressure range, as the throttle closes.

Ask around and see if anyone can test or verify your MAP sensor.

[twobjshelbys]

No trouble code.

The altitude (barometric) table is only available in the pro key version. It's in the ignition /steady state menu.

I have data logged it with the built-in CalMap and we did a dyno run last summer with a monitor and they matched pretty well.

[twobjshelbys]

Originally Posted by olddog

I would assume it ran great for Roush. You are at a different altitude and have different gas. How far would you have to drive to get down to the altitude Roush is at? If easy to do, that would tell you if it was the gas or the altitude.

I had a 1987 Mini Van that use MAP (no Mass flow sensor). Every time I drove it to Florida it would trip a code. The Map sensor had a problem reading the higher pressure at sea level, but it worked fine at the lower range.

From what you said it reminded me of this and makes me suspect a bad MAP sensor, however it doesn't make logical sense. Higher altitude is lower pressure. An engine always runs through the low pressure range, as the throttle closes.

Ask around and see if anyone can test or verify your MAP sensor.

Roush is in Livonia (Detroit) is about 650'. We are about 5400. It's a ways to get lower. The lowest point in Colorado is 3200' and that's at the Kansas border

The Roush setup does use a MAP sensor to detect base pressure (altitude) but the compensation table that it drives is "null".

[vector1]

your v/e table looks pretty messed up from 3400+.

the a/f base map is pretty rich also and the spark advance is pretty conservative at the lower rpm and cruising.

look at the 427 chevy sb 48# injector prom ram 544 hp setup for comparison if you want something close.

if you're datalogging you should be able to come up with a better v/e table and smooth it.

i think i would find somebody else to tune it or do some learning.

when your v/e table jumps around like in your table from 3400+ i don't think it is going to run well. it needs to be smooth.

[twobjshelbys]

Originally Posted by vector1

your v/e table looks pretty messed up from 3400+.

the a/f base map is pretty rich also and the spark advance is pretty conservative at the lower rpm and cruising.

look at the 427 chevy sb 48# injector prom ram 544 hp setup for comparison if you want something close.

if you're datalogging you should be able to come up with a better v/e table and smooth it.

i think i would find somebody else to tune it or do some learning.

when your v/e table jumps around like in your table from 3400+ i don't think it is going to run well. it needs to be smooth.

Standard rules may not apply. This is a faux 8-stack.

I don't have the base tune you mentioned. Can you email it back to me?

[olddog]

This is my gut feeling. I only know about half as much as I need to know to be of much help.

If the map sensor calibration is skewed at the lower pressure range, and someone has tuned it to work, the tables might look strange.

As I understand it, when the key is first turned on the MAP reading is saved before the engine is cranked. This lets the ECU know what the atmospheric pressure is. Even though you could tune an engine to run correctly with a skewed MAP, that tuning would only be good at that elevation. When you moved to radically different elevation, I think the tune would fall apart.

[vector1]

bring up the table you have in question and then hit f1, that will give you the information you need.

[twobjshelbys]

Update: While I was looking around to get some more info for vector1, I found that there are two altitude compensation tables in CalMap, one in the Ignition and one in the Fuel. The ignition one (which adds or removes advance) is indeed null (0). But the Fuel one is populated. It is a multiplier, and most of the terms are >1.0, especially in the range where I am at. It would seem to me that they should be <1.0 at altitude - that is, less air => less fuel to maintain a constant AFR. Do I have this wrong?

I may have some failing memory cells but I have a vague memory that this table might in fact be "inverted" in that it was treated as a "divide by" instead of a "multiply by" so that the numbers should all be calculated as 1/x... Might be worth a try...

I'm going to load some of the sample tunes from Accel into the ECU to see what they look like.

[vector1]

you don't have to load them into the ecu, bring the program up on the laptop and 'go offline" and there should be a pretty good selection of programs to load and look at on the laptop. the accel website also has some programs if you can find them. welcome, welcome to accels world.

[twobjshelbys]

Originally Posted by vector1

you don't have to load them into the ecu, bring the program up on the laptop and 'go offline" and there should be a pretty good selection of programs to load and look at on the laptop. the accel website also has some programs if you can find them. welcome, welcome to accels world.

I was using off line mode with yours and the others. But there are some items that are only available with the Pro key, and having it attached is not enough. It has to be connected to the ECU. The altitude tables are Pro-key only items.

[olddog]

Some EFI system use lamda (?) instead of AFR, which is AFR/14.7 so that 1.0 is the ideal AFR. In this case, 1.1 is lean and 0.9 is rich.

Others use the inverse, so that an increasing number is richer. You need to get specific information from the manual of your ECU. Then you need to read some books on tuning EFI to try to understand what the manual is talking about.

[vector1]

well tony, how did it go? you still need the dvd?

[twobjshelbys]

Originally Posted by vector1

well tony, how did it go? you still need the dvd?

I listened to the two sets of things on YouTube. It was some good info. I imagine the DVD has the same info - right? If so then I don't think I need it.

I'd been searching for a Roush guy but then I had the revelation of looking for an Accel EMIC. The place I tried last fall is one but didn't seem to be very scientific about it. I found another guy that was at the update class in the past few weeks and has done a few of the installs. I'm going to connect with him this week.

[twobjshelbys]

And then a miracle happened!

When it came back from Roush they said it worked at their location. This left the altitude and some other differences. So I tracked down an Accel EMIC in Ft. Collins, described the problems I was seeing, and sent them some data logs. They confirmed that the issues were probably altitude related and I brought it up to them.

Today I got it back. We did as spirited a drive as we could up to Cameron pass (10K+ ft), I put it through the spirited acceleration test several times along the way, and it works flawlessly! Wow, what a difference.

The major changes were done to the altitude table from 5000+ up, with a few low RPM changes to the base VE table.

Now it runs like a BOH.

Now we're not sure what it will do going up to from lower altitude to the 5000' region. It's hard to take it down in altitude but I may take it to Vegas in September. They said they can blend the two tables and that the stuff he did from 5000 to 10000 was pretty linear so it shouldn't be too hard to interpolate.

Kudos to Garry and Jordan at Lundstedt Automotive in Ft. Collins.

[vector1]

good job........

[twobjshelbys]

Today I got a call from Roush. There is some debate over whether the tune in the engine as delivered is appropriate for a Cobra at this altitude, so they offered to send someone out to tune it. I told them what I'd done and said I would send them the current tune for their comments, but that I believed that we were 90% there. I'm satisfied it is fully operational, I just think a few tweeks wouldn't hurt. First is that the mileage has gone way down. On trips with highway miles (say 80% cruising in 5th) I was getting 20-22MPG. Now I'm getting 12-13MPG. He indicated they could lean it out a little. I need to do a few more controlled tests to verify this.

Other than that, it is running great now. All of the other problems are gone.

I will be towing it to Vegas for the BJ auction (I got a slot in their parade down the strip.) I'll have it out there for the week and should be able to do some low altitude verification.

[twobjshelbys]

Just a side note - the engine has nothing to do with the pending sale which I will describe in another topic (since I've gotten about 5 queries)

[Got the Bug]

I'm not sure I could ever pull the trigger on selling a car at a BJ auction. Maybe as a buyer, but aren't you worried about the downside risk as a seller?

[twobjshelbys]

Originally Posted by Got the Bug

I'm not sure I could ever pull the trigger on selling a car at a BJ auction. Maybe as a buyer, but aren't you worried about the downside risk as a seller?

I'm not selling at the auction - only bringing it out there to drive in the parade and do some monitoring of the engine at lower altitudes. Sale happens at the snail's pace of all the others.

[ekrupa2]

Two...with my 402 SR I have seen 11-14 MPG...never any better. 20-22MPG sounds great but seems out of the norm. Doug - what do you see for MPG?