-
Welcome to the new SAOCA website. Already a member? Simply click Log In/Sign Up up and to the right and use your same username and password from the old site. If you've forgotten your password, please send an email to membership@sunbeamalpine.org for assistance.
If you're new here, click Log In/Sign Up and enter your information. We'll approve your account as quickly as possible, typically in about 24 hours. If it takes longer, you were probably caught in our spam/scam filter.
Enjoy.
Duratec Performance
- Thread starter Bill Blue
- Start date
You know a little fly cutter on your mill would cut those O-ring grooves just like store-bought...
Yeah, and I have a fly cutter. But that would require at least a 3/8" flange and this one is 1/4". If I went with 3/8", the weight gets stupid and it really should be aluminum. I can't weld aluminum and don't know where to source aluminum tubing shapes. Cost would get out of hand.
Life is way too complicated and this works fine. Did I mention cost would sky rocket like a Congressional Investigation?
Bill
Robbie, I have no direct knowledge on the subject. What I THINK I have read:
Externally the 2.3 and 2.5 are the same.
The 2.5 "goodies" will swap into the 2.3 block.
Megasquirt is capable of controlling VVT.
I know nothing of the history of the 2.5 and direct injection. But it should be possible to mate a 2.5 block with a 2.3 head which, at least in the earlier years, are port injection and do not use VVT.
I will venture to bet no more than a cold beer on any of this info.
Bill
Externally the 2.3 and 2.5 are the same.
The 2.5 "goodies" will swap into the 2.3 block.
Megasquirt is capable of controlling VVT.
I know nothing of the history of the 2.5 and direct injection. But it should be possible to mate a 2.5 block with a 2.3 head which, at least in the earlier years, are port injection and do not use VVT.
I will venture to bet no more than a cold beer on any of this info.
Bill
socorob
Donation Time
I think the 2.5 has 175hp out the box. It shouldn't take much to get that to 200. If its direct injection, I'm not sure what that changes. I think Ive heard direct injection engines use mechanical fuel pumps? Besides the intake, what is the hardest things you had to deal with installing a duratec, and how much ground clearance do you have? How long is the duratec from the bell to the most forward thing on Front of the engine, not including the fan?
The problems involved with installing a 2.3 into an Alpine:
1. The crossmember has to be notched.
2. The starter has to be modified.
3. Intake has to be fabbed.
4. Exhaust has to be fabbed.
5. Exhaust is on the wrong side.
6. Alternator mount has to be fabbed.
7. A mechanical fan is pretty much a nightmare, but not needed.
8. A water rail (from rear of head to rad) has to be fabbed.
9. The stock Ranger oil pan is pretty deep, but accommodates the exhaust crossover very nicely. The Focus pan might be a better bet, I don't know. A Ranger pan that has been shortened about an inch would provide adequate clearance.
10. I elected to install air conditioning which complicated things considerably. Ended up caring some stuff off the block to have room for the alternator.
Which item would be the biggest problem depends on the person doing the swap, his abilities and equipment. The starter would have been beyond most. To me, it was mostly a technical challenge. Its not that I am that good, but that is the area of my interest and tools. From what I've seen, I think you are up to the challenge.
I do not know how much my ground clearance is. The bell housing is the lowest point. The engine is 21" long, including the water outlet.
If you decide to make the swap, I'll help as much as possible. If a 2.5 would not put a smile on your face, check for granite.
Bill
1. The crossmember has to be notched.
2. The starter has to be modified.
3. Intake has to be fabbed.
4. Exhaust has to be fabbed.
5. Exhaust is on the wrong side.
6. Alternator mount has to be fabbed.
7. A mechanical fan is pretty much a nightmare, but not needed.
8. A water rail (from rear of head to rad) has to be fabbed.
9. The stock Ranger oil pan is pretty deep, but accommodates the exhaust crossover very nicely. The Focus pan might be a better bet, I don't know. A Ranger pan that has been shortened about an inch would provide adequate clearance.
10. I elected to install air conditioning which complicated things considerably. Ended up caring some stuff off the block to have room for the alternator.
Which item would be the biggest problem depends on the person doing the swap, his abilities and equipment. The starter would have been beyond most. To me, it was mostly a technical challenge. Its not that I am that good, but that is the area of my interest and tools. From what I've seen, I think you are up to the challenge.
I do not know how much my ground clearance is. The bell housing is the lowest point. The engine is 21" long, including the water outlet.
If you decide to make the swap, I'll help as much as possible. If a 2.5 would not put a smile on your face, check for granite.
Bill
socorob
Donation Time
Im not going to do anything until this heat is over. We went to a u pick it blueberry farm with some friends of ours from Arizona, and I didnt think they were going to make it. Its awful out there now. Im still trying to decide what direction I want to go in. I dynoed my car, and it had 110 rwhp. Its pretty quick with that, so I know if I get the 175hp or so a Cologne is good for, it will be really quick, but I weighed my car and it is front heavy. If I could get the same power out of an engine that weighs 100 pounds less, the car will be balanced better.I just dont know yet if I want to go thru all the work of an engine swap. It seems ill never get done with this car if I keep changing things, but I know I dont want to end up with any less than 175 rwhp when im done.
Robbie, if you decide to go with the 2.5, I'll say this. Do not pull your old engine and start from scratch. There are a lot of things you can do before ever dropping the 2.5 in for a trial fit.
The intake manifold, which consumed an incredible amount of time for me (YMMV) can be made, as well as the alternator mount, AC (if you choose) water rail, oil pan, starter mod, all can all be done based on my experience. I would not recommend making up the header before hand, it was a very tricky fit for me.
I gave you some bad info yesterday. The engine length, from bellhousing mounting surface to the front edge of the serpentine belt, is 19.5", as close as I can measure with the engine in the car. HOWEVER, the water rail bolts onto the rear of the head, which is on the same plane as the rear of the engine block. So in order to get overall engine length, you must add the thickness of the water rail, in my case, 1.5". 19.5 + 1.5 = 21. The rail is located high enough it will interfere with the firewall, I have about 1/4" clearance. Yes, its tight in there. You might have to notch the firewall in order to retain your transmission location.
Looking at my engine, I cannot see any problems that would be created with the engine set back a couple of inches. You would have to run the exhaust outlet to the rear of the steering arm, but I see nothing preventing that. The upside, the notch in the crossmember gets considerably smaller.
I just looked up the "numbers" for the I4 Duratec 2.5. 172 ft lbs of torque. Based on my experience, I'd guess that it has 150 or so from 1500 revs on up. I don't think you could ever match that with a 2.8 unless it has been turbo or surpercharged.
Bill
The intake manifold, which consumed an incredible amount of time for me (YMMV) can be made, as well as the alternator mount, AC (if you choose) water rail, oil pan, starter mod, all can all be done based on my experience. I would not recommend making up the header before hand, it was a very tricky fit for me.
I gave you some bad info yesterday. The engine length, from bellhousing mounting surface to the front edge of the serpentine belt, is 19.5", as close as I can measure with the engine in the car. HOWEVER, the water rail bolts onto the rear of the head, which is on the same plane as the rear of the engine block. So in order to get overall engine length, you must add the thickness of the water rail, in my case, 1.5". 19.5 + 1.5 = 21. The rail is located high enough it will interfere with the firewall, I have about 1/4" clearance. Yes, its tight in there. You might have to notch the firewall in order to retain your transmission location.
Looking at my engine, I cannot see any problems that would be created with the engine set back a couple of inches. You would have to run the exhaust outlet to the rear of the steering arm, but I see nothing preventing that. The upside, the notch in the crossmember gets considerably smaller.
I just looked up the "numbers" for the I4 Duratec 2.5. 172 ft lbs of torque. Based on my experience, I'd guess that it has 150 or so from 1500 revs on up. I don't think you could ever match that with a 2.8 unless it has been turbo or surpercharged.
Bill
I think I have solved the variable performance issue. Over attenuation of the vacuum signal to the ECM. Drivability is much improved as well. The off idle problem seems to be solved.
Anyway, we took it to Kansas City. Mileage varied from 27 @ 75 mph with air on about 50% of the time, to 30 @ 65 with 100% air. This is with the big tires, which cut rpm's about 7%. To me, the taller overall gearing does not appear to have helped mileage. The 3.55's are noisy and need replacement, so I've purchased a set of 3.80's.
I've decided to collect some data to evaluate the difference in performance between the 3.55 and 3.80 gears. So I did a little third gear 20(1518 rpm) -80(6074 rpm) mph run. In summary, the e.t. was 14.5 seconds, but some caveats are in order.
The run was not over my normal road, so no comparison should be made to previous runs. Around here, this stretch of road is considered "flat" but not to the degree of my test road.
I am going to have the 3.80's installed next week and will do another run to see the difference in performance.
Speaking of performance, there currently is a lot left on the table. On this run, AFR's varied from 18.8 to 11.6.
Bill
Anyway, we took it to Kansas City. Mileage varied from 27 @ 75 mph with air on about 50% of the time, to 30 @ 65 with 100% air. This is with the big tires, which cut rpm's about 7%. To me, the taller overall gearing does not appear to have helped mileage. The 3.55's are noisy and need replacement, so I've purchased a set of 3.80's.
I've decided to collect some data to evaluate the difference in performance between the 3.55 and 3.80 gears. So I did a little third gear 20(1518 rpm) -80(6074 rpm) mph run. In summary, the e.t. was 14.5 seconds, but some caveats are in order.
The run was not over my normal road, so no comparison should be made to previous runs. Around here, this stretch of road is considered "flat" but not to the degree of my test road.
I am going to have the 3.80's installed next week and will do another run to see the difference in performance.
Speaking of performance, there currently is a lot left on the table. On this run, AFR's varied from 18.8 to 11.6.
Bill
RootesRacer
Donation Time
I'll let Bill splain the why and the how, but my nomenclature for the pressure signal problem would be "integration", not attenuation.
I'll leave the nomenclature to the experts. What happened:
Way back when, on about my second try at tuning the thing, it was decided that due to the long runners, the vacuum signal to the ECU was fluctuating too much. So the line from the intake manifold log to the ECU was modified to include an orifice, (about a sixteenth in diameter) six feet of spaghetti tubing and 6 feet of 1/4" vacuum hose.
Three years later we re-examined that set up, by that time the 1/4" hose had been replaced with a vacuum chamber. I bypassed the vacuum chamber and got immediate positive results. Good off idle response and for the most part, reproduceable Lambdas. A few days ago I decided to do some fine tuning in the cruise range and discovered I still had problems at about 1500. So I removed the orifice. The vacuum line now consists of the spaghetti tubing. It seems to have made another improvement. But not so big this time.
In the meantime, with all the chasing of my tail, the max power tune is really bad. I will look at it again after the new gears are installed.
Bill
Way back when, on about my second try at tuning the thing, it was decided that due to the long runners, the vacuum signal to the ECU was fluctuating too much. So the line from the intake manifold log to the ECU was modified to include an orifice, (about a sixteenth in diameter) six feet of spaghetti tubing and 6 feet of 1/4" vacuum hose.
Three years later we re-examined that set up, by that time the 1/4" hose had been replaced with a vacuum chamber. I bypassed the vacuum chamber and got immediate positive results. Good off idle response and for the most part, reproduceable Lambdas. A few days ago I decided to do some fine tuning in the cruise range and discovered I still had problems at about 1500. So I removed the orifice. The vacuum line now consists of the spaghetti tubing. It seems to have made another improvement. But not so big this time.
In the meantime, with all the chasing of my tail, the max power tune is really bad. I will look at it again after the new gears are installed.
Bill
The 3.80 is installed. Good news and bad. I suppose the bad news is pretty simple, there still is not much power on top end.
Here are the results. 20 (1624 rpm) to 80 (6499 rpm) 15.1 seconds compared to the 14.6 seconds with the 3.55. This has to be balanced with some sort of temperature correction, the air intake temperature was 34 today, 22 with the 3.55. Also, there was a fairly strong (10 mph?) side wind. The 3.55 data was calm. Don't know how that shakes out.
To demonstrate the low end oomph, let's look at another 60 mph acceleration, 15 (1200 rpm) mph to 75 (6090 rpm). 13.1 seconds. I like that. I hope it is reproducible. The old attitude is back.
Tuning oddity, Lambdas vary from line to line. An example: .86, .89, .98, .91, 1, .85, .98, .88. kpa's were 98.8, 98.8, 97.6, 97.6, 99.3, 97, 98.5, 98.7. This is data taken every tenth of a second. The data is from 5372 to 5602 rpm.
Bill
Here are the results. 20 (1624 rpm) to 80 (6499 rpm) 15.1 seconds compared to the 14.6 seconds with the 3.55. This has to be balanced with some sort of temperature correction, the air intake temperature was 34 today, 22 with the 3.55. Also, there was a fairly strong (10 mph?) side wind. The 3.55 data was calm. Don't know how that shakes out.
To demonstrate the low end oomph, let's look at another 60 mph acceleration, 15 (1200 rpm) mph to 75 (6090 rpm). 13.1 seconds. I like that. I hope it is reproducible. The old attitude is back.
Tuning oddity, Lambdas vary from line to line. An example: .86, .89, .98, .91, 1, .85, .98, .88. kpa's were 98.8, 98.8, 97.6, 97.6, 99.3, 97, 98.5, 98.7. This is data taken every tenth of a second. The data is from 5372 to 5602 rpm.
Bill
RootesRacer
Donation Time
Bill,
You have about a 1kpa noise showing up in your WOT manifold pressure.
1% is not huge, but it indicates you have a fair amount of reversion or something going on in the intake.
This is probably how we ended up having you put on that pneumatic low pass filter that ended up turning into a phase shift oscillator.
So the wideband is telling you that you have about a .1 lambda (1.5AFR) variation in fuel ratio (this would be like 10% fuel difference per .1 lambda). The 1kpa noise can contribute to 1% so unless the ECU is showing it is changing 10% on the pulse width, something else is going on or the wideband needs more filtration (LC1s are not well known for being very accurate though Innocate says they have the fastest response time of their competitors).
If you can set the LC1s filter to a harder low pass, I would do so.
You have about a 1kpa noise showing up in your WOT manifold pressure.
1% is not huge, but it indicates you have a fair amount of reversion or something going on in the intake.
This is probably how we ended up having you put on that pneumatic low pass filter that ended up turning into a phase shift oscillator.
So the wideband is telling you that you have about a .1 lambda (1.5AFR) variation in fuel ratio (this would be like 10% fuel difference per .1 lambda). The 1kpa noise can contribute to 1% so unless the ECU is showing it is changing 10% on the pulse width, something else is going on or the wideband needs more filtration (LC1s are not well known for being very accurate though Innocate says they have the fastest response time of their competitors).
If you can set the LC1s filter to a harder low pass, I would do so.
Jarrid, the kpa variation is consistent with what it was with the orifice and reservoir. Maybe I've removed too much restriction and I should should use an orifice, but larger than previous?
Here is snipit of data for your enjoyment. I don't know what to make of it.
rpm Lam kpa PW
5372 .86 98.8 10.338
5404 .89 98.8 10.334
5435 .98 98.8 10.23
5460 .91 97.6 10.038
5490 1.0 97.6 10.037
5527 .85 99.3 10.152
5547 .98 97.0 9.821
5583 .88 98.5 9.989
5602 .99 98.7 10.011
5637 .85 97.3 9.784
Bill
Here is snipit of data for your enjoyment. I don't know what to make of it.
rpm Lam kpa PW
5372 .86 98.8 10.338
5404 .89 98.8 10.334
5435 .98 98.8 10.23
5460 .91 97.6 10.038
5490 1.0 97.6 10.037
5527 .85 99.3 10.152
5547 .98 97.0 9.821
5583 .88 98.5 9.989
5602 .99 98.7 10.011
5637 .85 97.3 9.784
Bill
MGBSS
Donation Time
Maybe I just "think"
I want programmable EFIMaybe I just "think"
Some more Lamdba numbers to dream by:
rpm 3.55 3.80
1.5k 1.23 0.83
2.0k 0.82 0.83
2.5k 0.82 1.0
3.0k 0.82 0.99
3.5k 0.86 0.98
4.0k 0.85 1.05
4.5k 0.85 0.99
5.0k 0.83 1.0
5.5k 1.00 0.86
6.0k 0.86 0.85
6.5k 0.86 0.86
These are runs taken a few days apart over the same stretch of road. Temperatures were very different, but the tune was unchanged. In general, the engine is running very good. Good power up to about 5k and very good drivability - best its ever been, so its not like I'm facing an impossible situation. Just can't optimise performance. To me, this is a continuation of the old problem, lack of repeatability. Only this time it is not impacting drivability.
I think I'll remove the air filter assembly and make a couple of runs, see how it does with a bare throttle body.
Jarrid, I have not had any luck downloading the LC1 software, but if I do, what would constitute a "harder" low pass?
Bill
Put a .045" orifice in the MAP sensor line and did two back to back runs. Here are Lambdas.
rpm - R1 - R2
1.5k -.89 -.83
2.0k -.98 -.83
2.5k -1.01 .98
3.0k -1.31 -1.02
3.5k -1.31 -1.05
4.0k -1.35 -1.25
4.5k - .95 - .94
5.0k -1.02 -.83
5.5k - .99 - .86
6.0k - .88 - .85
6.5k - .85 - .86
To me, that looks like an improvement. Certainly differrant set of numbers that those obtained with no orifice. The kpas are a little higher and more compact (but I have not run any statical analysis). However, off idle performance deteriorated. Hesitation and an occasional cough. So I'm going to try .062". Even though it is an improvement, .045" is not satisfactory. Maybe .062" will give even better Lambdas AND good throttle response.
Bill
rpm - R1 - R2
1.5k -.89 -.83
2.0k -.98 -.83
2.5k -1.01 .98
3.0k -1.31 -1.02
3.5k -1.31 -1.05
4.0k -1.35 -1.25
4.5k - .95 - .94
5.0k -1.02 -.83
5.5k - .99 - .86
6.0k - .88 - .85
6.5k - .85 - .86
To me, that looks like an improvement. Certainly differrant set of numbers that those obtained with no orifice. The kpas are a little higher and more compact (but I have not run any statical analysis). However, off idle performance deteriorated. Hesitation and an occasional cough. So I'm going to try .062". Even though it is an improvement, .045" is not satisfactory. Maybe .062" will give even better Lambdas AND good throttle response.
Bill
RootesRacer
Donation Time
Pulse width tracks nicely with MAP.
3% variation in PW cant cause a 10% shift in lambda.
You need to try to get steady state AFRs from your wideband or slow widebands output filter WAY THE HELL DOWN.
The LC1 doe not track dynamic AFRs very well.
3% variation in PW cant cause a 10% shift in lambda.
You need to try to get steady state AFRs from your wideband or slow widebands output filter WAY THE HELL DOWN.
The LC1 doe not track dynamic AFRs very well.