Claudio Ludwig from the RX7 Club forum is very experienced with RX7 tuning. He frequents Guild of EFI Tuning and some other ECU manufacture forums.He posted a very interesting article on how getting DT exactly correct is often not as important as people may think. Andrew Simon from HP Academy also has commented on this. I'll try and find the link to Claudio's article a bit later...
Found it. Different ECU but same principles apply. Read Claudio's response #8
By applying Battery compensation and Dyno Tuning you can come up with a very good Tune without ever knowing the exact Dead Time of an Injectors. Many tuners use default the ECU's manufacturers default Dead Times and rarely change it. Andre Simon ( HP Academy ) was commenting on that exact method. As long as you don't swap injectors sizes. Yes, having the correct DT is desirable... but not always necessary.
Edited for Clarity:
2) The compensation map is absolutely needed. Some platforms are better than others at maintaining a consistent voltage, but you will always have variables. Unless you want to chase fluctuating AFRs, use the compensation map.
5) You do not need injector dead time values. You can come up with an effective compensation table without any injector information at all. As part of a scratch tune you'll get the car running and idling well and able to hold a given AFR while holding a consistent idle speed. We'll say 14:1 at 1000 rpm. At this point note the battery voltage. We'll say 13.5V. Now you can start manipulating battery voltage by turning on accessories. As the battery voltage falls, you will need to manipulate the values in the compensation table to maintain our 14:1 target AFR. It's possible to do all manner of things to alter the system voltage to hit the relevant spots on the table (usually 12-14 volts). And that's it. No exotic test equipment or calculations needed and it only takes a few minutes. You just need to make sure the base fuel map is accurately calibrated and the engine speed is consistent in the area you're doing the voltage test so that you're not fighting yourself as you go.
Last edited by BlackBird_SR71 on Sun Feb 18, 2018 2:13 pm, edited 1 time in total.
Except that the MS2 battery correction is not linear. The inputs required make it look linear and the documentation doesn't hint at anything different but it actually matches the real behaviour pretty well.
I read his response, his last point (number 5) is exactly what I tried to do, get a constant afr then add battery load and try conpensate till the afr stays consistent, not sure what rpm I did this at thou, it may have been 850-900, maybe a 1000 would of been better. I turned EGO off while I was trying to find the issue, I just wanted it out of the equation. the rest is how the dyno guy set it up, i've just been tweaking the VE table around the cruise area as he did a real lazy tune in those areas and just got the EGO to compensate. I also wonder if I need a new wideband sensor its done a few 1000 kms now.
BlackBird_SR71 wrote:Claudio Ludwig from the RX7 Club forum is very experienced with RX7 tuning. He frequents Guild of EFI Tuning and some other ECU manufacture forums.He posted a very interesting article on how getting DT exactly correct is often not as important as people may think. Andrew Simon from HP Academy also has commented on this. I'll try and find the link to Claudio's article a bit later...
Found it. Different ECU but same principles apply. Read Claudio's response #8
By applying Battery compensation and Dyno Tuning you can come up with a very good Tune without ever knowing the exact Dead Time of an Injectors. Many tuners use default the ECU's manufacturers default Dead Times and rarely change it. Andre Simon ( HP Academy ) was commenting on that exact method. As long as you don't swap injectors sizes. Yes, having the correct DT is desirable... but not always necessary.
Edited for Clarity:
2) The compensation map is absolutely needed. Some platforms are better than others at maintaining a consistent voltage, but you will always have variables. Unless you want to chase fluctuating AFRs, use the compensation map.
5) You do not need injector dead time values. You can come up with an effective compensation table without any injector information at all. As part of a scratch tune you'll get the car running and idling well and able to hold a given AFR while holding a consistent idle speed. We'll say 14:1 at 1000 rpm. At this point note the battery voltage. We'll say 13.5V. Now you can start manipulating battery voltage by turning on accessories. As the battery voltage falls, you will need to manipulate the values in the compensation table to maintain our 14:1 target AFR. It's possible to do all manner of things to alter the system voltage to hit the relevant spots on the table (usually 12-14 volts). And that's it. No exotic test equipment or calculations needed and it only takes a few minutes. You just need to make sure the base fuel map is accurately calibrated and the engine speed is consistent in the area you're doing the voltage test so that you're not fighting yourself as you go.
1992 Mazda Familia
BP26 Forged Motor with TD05 Turbo
Megasquirt 2 v3
You don't have to Turn EGO Control off. You just filter it so that it that it starts about 200 to 300 RPM above your idle RPM. The default of 1,300 is fine. Same with Auto-Tune. Either filter it so it doesn't touch your idle values, or lock out the Idle Cell area's so that Auto-Tune doesn't touch them.
You can get a much steadier idle by manually setting the idle Cell area's Tune in a 3x3 block ( Like Tic Tac Toe ) and shoot for around a 13.8 to 14.0 AFR at Idle. Tune for highest KPA value ( or lowest Vacuum ) and RPM values. Old school method that works very, very well.
You also have to make sure that there are not big value changes outside of your idle areas. Aut-Tune weighs the Cell around the running area, and adjacent Cell areas are averaged to come up with a Value. So Cell ranges that are never touched ( out side of running range ) can still affect the driveability. I see some of those areas in your Tune.
OK, I think now I am starting to make sense of what has been going here, for some reason in the tune I got back from the dyno guy I found the car would not start, the only way to keep it going after start was to put the ASE to 100%, when switching to semi-seq it is causing the engine to almost flood. Also the dyno had a theory that it is better to run at a lower than recommended req_fuel so that you can get a better resolution from the VE table. Now when I switch over the fueling is all over the place and I dont really have a starting point so the whole tune is completely a lost cause, and could end up doing some real damage, (bore wash comes to mind). But on one good note, after ASE ended, I managed to get the car to idle relatively stable at 14.3 even with changing electrical loads.
1992 Mazda Familia
BP26 Forged Motor with TD05 Turbo
Megasquirt 2 v3
