Where do I start When learning how to tune?
05-27-2014, 07:57 PM
#127
This is a good place to start.
http://foxed.ca/rx7manual/manuals/RE...amoto-1981.pdf
Look at pg 43 (48 of the PDF) as well as the timings that cause knock on pg 45(50)
Flame speed is very slow in the direction of the rotor trailing edge because of squish flow and quench. This is also (for that reason) the area where knock occurs. I call it the fuel/air heel.
At low rpm this isn't a problem because at 1000rpm there is 60ms available for a complete rotation. At 10krpm there is only 6ms available for a complete rotation in each case only a fraction of that time is available for each combustion. Since flame speed is related to temperature and pressure, it doesn't get any quicker as rpm goes up, which is why timing normally has to be advanced as RPM goes up. Unfortunately even with timing advance it gets harder and harder to burn the heel.
If pressure gets high enough (high load and advanced timing) and the heel hasn't ignited yet (high rpm) it causes knock. Add a third plug and you can eliminate some knock at high rpm and reclaim some power too. Alternatively you can advance the trailing timing, if you do then the trailing is exposed to combustion temperatures longer and will eventually get hot enough to cause preignition.
The problem is the linear size of the combustion chamber. You can have big combustion chambers, you can have high rpm engines, but you can't easily have high rpm engines with big combustion chambers.
Hope that helps you wrap your head around it.
http://foxed.ca/rx7manual/manuals/RE...amoto-1981.pdf
Look at pg 43 (48 of the PDF) as well as the timings that cause knock on pg 45(50)
Flame speed is very slow in the direction of the rotor trailing edge because of squish flow and quench. This is also (for that reason) the area where knock occurs. I call it the fuel/air heel.
At low rpm this isn't a problem because at 1000rpm there is 60ms available for a complete rotation. At 10krpm there is only 6ms available for a complete rotation in each case only a fraction of that time is available for each combustion. Since flame speed is related to temperature and pressure, it doesn't get any quicker as rpm goes up, which is why timing normally has to be advanced as RPM goes up. Unfortunately even with timing advance it gets harder and harder to burn the heel.
If pressure gets high enough (high load and advanced timing) and the heel hasn't ignited yet (high rpm) it causes knock. Add a third plug and you can eliminate some knock at high rpm and reclaim some power too. Alternatively you can advance the trailing timing, if you do then the trailing is exposed to combustion temperatures longer and will eventually get hot enough to cause preignition.
The problem is the linear size of the combustion chamber. You can have big combustion chambers, you can have high rpm engines, but you can't easily have high rpm engines with big combustion chambers.
Hope that helps you wrap your head around it.
05-28-2014, 09:21 PM
#128
Found this one, a lot of good and accurate information here:
Understanding Tuning! - Rx7 Forum and Owners Club
Or if you really feel like going in depth here:
Not2Fast: NACA Paper List
There is a lot you can do NA without blowing up your engine. Turbo is a whole other animal.
Understanding Tuning! - Rx7 Forum and Owners Club
Or if you really feel like going in depth here:
Not2Fast: NACA Paper List
There is a lot you can do NA without blowing up your engine. Turbo is a whole other animal.
05-30-2014, 12:53 AM
#130
Hybrid Greddy Boosted
So I flattened my maps to do some constant load logging to try and get some accurate results for MAF scaling. Pretty much a base map with the following changes:
- flat VE table on 1.
- Flat throttle enrichment tables of 14.7 (tip in, so shouldn't apply to constant load/speed)
- flat open loop fuelling tables at 13.02
After warming the engine up I loaded the above map then went for a drive. I went for a 10 minute logging session in 4th gear, 4200rpm. It was a flat road and my speed and load were constant. Strange thing is that the commanded AFR was logged (MazdaEdit) at around 14.81 with a bit of variation. I was expecting the exact figure of 13.02 with no fluctuations. WTF?
I can only assume that there are other tables that significantly contribute to the commanded AFR, even when at constant load/speed.
- flat VE table on 1.
- Flat throttle enrichment tables of 14.7 (tip in, so shouldn't apply to constant load/speed)
- flat open loop fuelling tables at 13.02
After warming the engine up I loaded the above map then went for a drive. I went for a 10 minute logging session in 4th gear, 4200rpm. It was a flat road and my speed and load were constant. Strange thing is that the commanded AFR was logged (MazdaEdit) at around 14.81 with a bit of variation. I was expecting the exact figure of 13.02 with no fluctuations. WTF?
I can only assume that there are other tables that significantly contribute to the commanded AFR, even when at constant load/speed.
05-30-2014, 10:33 AM
#133
There is a hidden closed loop that is driven by the rear o2 sensor. When in cruise conditions it puts a trim on commanded AFR to make the rear o2 sensor oscillate around 14.7:1. It's there to protect the cat and only applies below a certain load/rpm and above a certain temperature. I don't recall the numbers offhand. This even happens in regular open loop!
If you want to scale your maf you need have everything setup stock and then check commanded AFR vs actual AFR. Then just adjust the maf. The engine VE does change with RPM, the injectors become less effective with higher deamand and higher pressures (lower vac) and all these things can mask what you are trying to find if you null the tables. Above all realize that placing 12.5 (or any other value) AFR on a table is the equivilent of asking Santa, the numbers do not directly correspond to the AFR the car is going to give you, but the closer you get all the other values (MAF, VE, Injector scaling, etc) the more acurate they will be. The goal should be to have the AFR you want indicated out the tailpipe and not on a table.
If you want to scale your maf you need have everything setup stock and then check commanded AFR vs actual AFR. Then just adjust the maf. The engine VE does change with RPM, the injectors become less effective with higher deamand and higher pressures (lower vac) and all these things can mask what you are trying to find if you null the tables. Above all realize that placing 12.5 (or any other value) AFR on a table is the equivilent of asking Santa, the numbers do not directly correspond to the AFR the car is going to give you, but the closer you get all the other values (MAF, VE, Injector scaling, etc) the more acurate they will be. The goal should be to have the AFR you want indicated out the tailpipe and not on a table.
05-30-2014, 03:36 PM
#134
Hybrid Greddy Boosted
I think it may be a lot simpler than this. I'm now thinking the commanded AFR as seen and logged in MazdaEdit is the AFR after trim corrections have been applied. Big difference when the scaling is out by a bit.
I get the deal with arbitrary AFRs in the fuelling tables. While I'm learning, I'm going to shoot for the "Fazda" method using scaling and VE to get the table values as close to real world as possible. When I run out of patience, I'll do the quick and dirty fuel table adjustments
I get the deal with arbitrary AFRs in the fuelling tables. While I'm learning, I'm going to shoot for the "Fazda" method using scaling and VE to get the table values as close to real world as possible. When I run out of patience, I'll do the quick and dirty fuel table adjustments
05-31-2014, 05:11 AM
#138
Hybrid Greddy Boosted
01-29-2015, 04:17 PM
#139
Anyway, I'm not very interested in increasing power (for the moment at least), just want to up the omp rates, lower the fan temps (and sell the kit), and maybe up the idle. I'll try messing with it tonight a little.
Thanks again.
-EDIT-
Here are some things I have changed, am I on the right track?
-Edit- One says 80, but should say 60 (7k rpms / max load) just caught it.
Well that is worrisome indeed, as it is very simple.
However from what Jeff showed me there is a right way and a wrong way, but both should raise the idle.
1) You should change each of the tables equally
There are 5 idle tables.
IIRC (and the memory is not soooo good)
Table A is for vehicle stopped clutch out
Table B is for vehicle stopped clutch in
Table C is for vehicle stopped clutch out A/C on
Table D is for vehicle stopped clutch in A/C on
Table E is for vehicle moving clutch in
Attachment 183727
2) When you change a table change the entire table equally, for example 800 rpm to 900 rpm idle
Before you change
Attachment 183728
After change
Attachment 183729
Of course I'm assuming you know how to save the map and load it in to the ECU
However from what Jeff showed me there is a right way and a wrong way, but both should raise the idle.
1) You should change each of the tables equally
There are 5 idle tables.
IIRC (and the memory is not soooo good)
Table A is for vehicle stopped clutch out
Table B is for vehicle stopped clutch in
Table C is for vehicle stopped clutch out A/C on
Table D is for vehicle stopped clutch in A/C on
Table E is for vehicle moving clutch in
Attachment 183727
2) When you change a table change the entire table equally, for example 800 rpm to 900 rpm idle
Before you change
Attachment 183728
After change
Attachment 183729
Of course I'm assuming you know how to save the map and load it in to the ECU
01-30-2015, 03:00 AM
#143
I would suggest that you change them equivallenty. The difference is that there are some variations in idle maps A-E and IMO those variations should be maintained rather than making them all identical.
Also don't agree with changing all the speeds across each map line. You only care about the end value when the engine warms up. So why would you want to change the idle speeds when the engine is going through it's warm-up cycle? That was the MM method and after thinking it through it I concluded it was 't how I wanted to do it. The values you change ultimately depend on what the new idle speed is. In the case above changing the end value is fine for me. If the end value is higher than the cell before it then obviously that one needs to change too.
.
Also don't agree with changing all the speeds across each map line. You only care about the end value when the engine warms up. So why would you want to change the idle speeds when the engine is going through it's warm-up cycle? That was the MM method and after thinking it through it I concluded it was 't how I wanted to do it. The values you change ultimately depend on what the new idle speed is. In the case above changing the end value is fine for me. If the end value is higher than the cell before it then obviously that one needs to change too.
.
Last edited by TeamRX8; 01-31-2015 at 01:59 PM.
01-31-2015, 02:19 PM
#145
You're engine won't blow up if that's what you mean.
However, what I'm presenting to you is a philosophy of not just blindly following what someone else tells you. It's best to get your mind wrapped around what the software strategy is and what makes the sense to achieve your goal.
The factory PCM has a lot of complication and we only have access to a small part of it. We can't know all the dependency entanglements of how one setting may affect something else. I'm personally not a fan of making any more changes than are necessary to achieve 'close enough'
However, what I'm presenting to you is a philosophy of not just blindly following what someone else tells you. It's best to get your mind wrapped around what the software strategy is and what makes the sense to achieve your goal.
The factory PCM has a lot of complication and we only have access to a small part of it. We can't know all the dependency entanglements of how one setting may affect something else. I'm personally not a fan of making any more changes than are necessary to achieve 'close enough'
Its alot more complicated than that.
look how I put "A-E" I was just too lazy to paste all the idle tables.