Pages: 1 ... 5 6 [7] 8 9 ... 21
Author Topic: Bosch EV14 Injector Migration  (Read 361578 times)
nyet
Administrator
Hero Member
*****

Karma: +604/-166
Offline Offline

Posts: 12232


WWW
« Reply #90 on: November 28, 2012, 01:45:15 PM »

Awesome. Thank you for the summary.

lambda regulation has been also altered.

Can you elaborate?
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
rnagy86
Sr. Member
****

Karma: +6/-11
Offline Offline

Posts: 367


« Reply #91 on: November 28, 2012, 02:18:31 PM »

The relevant bits for an M-box:
Lambda regulation

ZKRKAT @ 0x1C2FE: Integration speed integrator - rate of RKAT adaption (long term)
(%/s), 16bit, float, precision 4

ZKRKAZ @ 0x1C300: Integration speed integrator - rate of RKAZ adaption (short term)
(%/s), 16bit, float, precision 4

MLO1 @ 0x1C2E4: upper level air flow range 1 - LTFT Active at IDLE
(kg/h), 16bit, float, precision 3

MLO2 @ 0x14EFC: upper level air flow range 2 -- FRAU small area regulation
(kg/h), 16bit, float, precision 3

MLO3 @ 0x1C2E6: upper air threshold quantity range 3 -- STFT Active Idle
(kg/h), 16bit, float, precision 3

MLU2 @ 0x1C2E8: lower air flow rate threshold range 2 -- FRAU Small Area Regulation
(kg/h), 16bit, float, precision 3

MLU4 @ 0x14EFE: lower air flow rate threshold range 4 -- FRAO Large Area Regulation
(kg/h), 16bit, float, precision 3

RLO2 @ 0x18CD7: upper load threshold range 2 -- FRAU Small Area Regulation
(%), 16bit, float, precision 3

RLU2 @ 0x18CD9: lower rl (load) - Threshold range 2 -- FRAU Small Area Regulation
(%), 16bit, float, precision 3

RLU3 @ 0x18CDA: lower rl (load) - Threshold range 3 -- STFT Active Idle
(%), 16bit, float, precision 3

RLU4 @ 0x18CDB: rl (load)  lower threshold multiplicative upper range -- FRAO Large Area Regulation
(%), 16bit, float, precision 3

---- the following is a list of modified ones, all the others are stock:

ZKRKAT: 0.2490
ZKRKAZ: 0.1025
MLO2: 285.000
MLO3: 115.000
MLU2: 40.000
MLU4: 340.000
RLO2: 114.750
RLO3: 54.750
RLU2: 30.000
RLU3: 28.500
RLU4: 87.750

Logged
rnagy86
Sr. Member
****

Karma: +6/-11
Offline Offline

Posts: 367


« Reply #92 on: November 28, 2012, 02:23:10 PM »

guess we should really start to have a central place for definition files ... netmofo github repo?  Undecided
Logged
nyet
Administrator
Hero Member
*****

Karma: +604/-166
Offline Offline

Posts: 12232


WWW
« Reply #93 on: November 28, 2012, 04:24:59 PM »

https://github.com/nyetwurk/ecuxplot/tree/master/data

I really need to get on ASAP2 conversions
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
nyet
Administrator
Hero Member
*****

Karma: +604/-166
Offline Offline

Posts: 12232


WWW
« Reply #94 on: November 28, 2012, 04:29:44 PM »

Can you elaborate on the reasoning for the lambda regulation and wall film changes so I can do a s4wiki writeup?
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
Bische
Sr. Member
****

Karma: +25/-4
Offline Offline

Posts: 396



WWW
« Reply #95 on: November 28, 2012, 08:34:25 PM »

Im also curious why one wants to zero acceleration/decceleration enrichment?

I have always been under the impression that if an aftermarket injector is calibrated correctly it should give you the same fueling charateristic as the stock injectors? Im talking about warmup, wallfilm, acc/dec enrichment, trims and all that jazz.

Take the wall film for example, if we were talking stand alone ECU's we would add injector PW based on RPM and TMOT -here we would need to calibrate this each time we change injectors or change fuel.

But in the ME7 the injectors are setup to inject x amount fuel based on a fuel mass request, rk_w. The wall film(and all of those other above mentioned functions) is calculated in the requested rk_w, so regardless of what injectors/injector calibration we have, the requested amount of fuel is still going to be the same. So when the new bigger injector is calibrated to give that x amount of fuel at the right voltage and pulsewidth, the injected fuel mass will be the same as if we were still running stock injectors?

This is just my own observations  Smiley
« Last Edit: November 28, 2012, 11:37:50 PM by Bische » Logged
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #96 on: November 29, 2012, 06:35:00 AM »

Can you elaborate on the reasoning for the lambda regulation and wall film changes so I can do a s4wiki writeup?

fuel droplet size, fuel spray spatial distribution and local crossflow velocity are influencing variables for wall film formation and breakup that are quite different on a stage 3 type application.  while it is true that the delivered fuel parcel (mass) ought to remain unchanged compared to oem (if the larger injector is properly calibrated), the formation and breakup of the wall film are quite different as this action is influenced by droplet size, mass and velocity (magnitude and direction).  the volumetric flow of droplets emerging from the wall film is directly influenced by the size and velocity of the particles impinging on the surface and, of course, by the number of fuel particles that are able to escape impact with the wall at a particular crossflow.  the ESUK function is meant to provide a model of wall film behavior and a means to preserve the wall film mass at close to steady state through load transitions (gross injector transitions) so as to prevent a momentary lean condition on throttle tip in (large air mass reduces wall film) as well as a momentary rich condition on deceleration (reduction in air mass/velocity allows an increase in wall film formation). if left at oem calibration the result is an under-compensation on cold starts at cold ambient temperature and an over-compensation at operating temperature. the cold scenario presents as bucking and uneven throttle response following a cold start - perhaps to the point of engine stall - as much of the fuel condenses as wall film leaving an insufficient amount of atomized fuel for combustion.  the 90*C+ scenario presents as less than ideal combustion at light throttle while at operating temperature (felt as engine/driveline vibration especially noticeable with upgraded mounts) as the wall film mass is erratic.  this inconsistent wall film mass will also result in an oscillation in lambda regulation (short and long term for small and large air masses) which will exasperate the uneven wall film compensation.  all of this reduces driver satisfaction.  the vehicle will be silky smooth one minute then driveline vibrations the next.  at steady-state cruise the system will eventually equalize, however, once the vehicle slows for traffic then accelerates again the driver will notice the subtle vibration until equalization is again realized.

rnagy posted my current KFBAKL.  KFVAKL is 0’d for now until I have time to properly calibrate it (though I may just leave the deceleration compensation off as the results are excellent).

as far as lambda regulation – i calibrated small and large areas for a k04/85mm MAF application and the associated increase in air masses (using RS4 values as a reference) and then logged many thousands of miles observing lambda regulation (lots of road trips) until i found settings that satisfactorily adapted for large and small air masses for both short and long term values in a timely fashion whilst avoiding the tendency for oscillation upon both slight and large load transitions.


ALSO - i've attached my .xdf which has all of the Lambda Regulation and Injection Transition maps/scalars defined.  thx to Nye for all of his work on the base xdf.

« Last Edit: November 29, 2012, 12:56:03 PM by berTTos » Logged
imolasb5
Jr. Member
**

Karma: +5/-1
Offline Offline

Posts: 38


« Reply #97 on: November 29, 2012, 03:42:12 PM »

fuel droplet size, fuel spray spatial distribution and local crossflow velocity are influencing variables for wall film formation and breakup that are quite different on a stage 3 type application.  while it is true that the delivered fuel parcel (mass) ought to remain unchanged compared to oem (if the larger injector is properly calibrated), the formation and breakup of the wall film are quite different as this action is influenced by droplet size, mass and velocity (magnitude and direction).  the volumetric flow of droplets emerging from the wall film is directly influenced by the size and velocity of the particles impinging on the surface and, of course, by the number of fuel particles that are able to escape impact with the wall at a particular crossflow.  the ESUK function is meant to provide a model of wall film behavior and a means to preserve the wall film mass at close to steady state through load transitions (gross injector transitions) so as to prevent a momentary lean condition on throttle tip in (large air mass reduces wall film) as well as a momentary rich condition on deceleration (reduction in air mass/velocity allows an increase in wall film formation). if left at oem calibration the result is an under-compensation on cold starts at cold ambient temperature and an over-compensation at operating temperature. the cold scenario presents as bucking and uneven throttle response following a cold start - perhaps to the point of engine stall - as much of the fuel condenses as wall film leaving an insufficient amount of atomized fuel for combustion.  the 90*C+ scenario presents as less than ideal combustion at light throttle while at operating temperature (felt as engine/driveline vibration especially noticeable with upgraded mounts) as the wall film mass is erratic.  this inconsistent wall film mass will also result in an oscillation in lambda regulation (short and long term for small and large air masses) which will exasperate the uneven wall film compensation.  all of this reduces driver satisfaction.  the vehicle will be silky smooth one minute then driveline vibrations the next.  at steady-state cruise the system will eventually equalize, however, once the vehicle slows for traffic then accelerates again the driver will notice the subtle vibration until equalization is again realized.

rnagy posted my current KFBAKL.  KFVAKL is 0’d for now until I have time to properly calibrate it (though I may just leave the deceleration compensation off as the results are excellent).

as far as lambda regulation – i calibrated small and large areas for a k04/85mm MAF application and the associated increase in air masses (using RS4 values as a reference) and then logged many thousands of miles observing lambda regulation (lots of road trips) until i found settings that satisfactorily adapted for large and small air masses for both short and long term values in a timely fashion whilst avoiding the tendency for oscillation upon both slight and large load transitions.


ALSO - i've attached my .xdf which has all of the Lambda Regulation and Injection Transition maps/scalars defined.  thx to Nye for all of his work on the base xdf.



Thanks man!
Logged
Bische
Sr. Member
****

Karma: +25/-4
Offline Offline

Posts: 396



WWW
« Reply #98 on: November 30, 2012, 11:55:52 AM »

text

Thanks for your explaination, very thorough Smiley

Dont take this the wrong way, I totally see were youre coming from, but please explain why this would crave for altering the request? Isnt the request formed around the head ports size/shape, thermal conductivity and airflow rather than injector spray pattern efficiency?

I have battled my 1000cc EV14's for months, I have been all over the place trying to get them right with tuning wall film, acc/decc enrichment, KRKTE, FRLFSDP - you name it. Now I have them running very close to perfect, but if there is something I am missing, I am eager to learn Smiley
Logged
berTTos
Full Member
***

Karma: +24/-0
Offline Offline

Posts: 91


« Reply #99 on: November 30, 2012, 04:09:39 PM »

Thanks for your explaination, very thorough Smiley

Dont take this the wrong way, I totally see were youre coming from, but please explain why this would crave for altering the request? Isnt the request formed around the head ports size/shape, thermal conductivity and airflow rather than injector spray pattern efficiency?

I have battled my 1000cc EV14's for months, I have been all over the place trying to get them right with tuning wall film, acc/decc enrichment, KRKTE, FRLFSDP - you name it. Now I have them running very close to perfect, but if there is something I am missing, I am eager to learn Smiley

the spray pattern and atomization characteristics significantly affect wall film development and breakdown AND the amount of wall film present at steady state.  it is true (or ought to be) that the effective fuel mass is the same, regardless of injector in use and i believe this is true in the macro sense.  however, when we examine combustion on a smaller time scale, the amount of fuel present in the film, atomizing out of the film and injected fuel particles escaping contact with the film we end up with an inconsistent fuel mass (especially at light throttle) for each combustion event - not a huge difference, but enough to be felt by the driver and enough to upset lambda regulation.

while on this topic - it is shocking how much driveline vibration comes from the tune.  i remember an AW forum member running a popular stage 3 tune trying to troubleshoot driveline vibration.  he went so far as to replace his entire driveline and the vibrations persisted.  i guarantee it came from the tune.

the proper way to correct this is to recalibrate the wall film model.  this is a difficult task. i suppose one could start with the recommended Bosch reference values and simply begin trial and error testing.  the next best thing (and considerably easier) is to leave the model alone and to alter the compensation.  it works consistently at all temps and altitudes and that's good enough for me.

and, as they say - 'the proof is in the pudding.'  this EV14 file is running smoothly at various altitudes and temperatures on many stage 3 vehicles without any cold start issues whatsoever, no strange rich or lean spots on throttle transitions and no strange driveline vibrations. these engines have various cylinder heads, supporting mods and run on different fuel formulations (all at least 93 octane).  they all run Ford Racing EV14 52lb injectors.

one last issue - the 1000cc injectors.  are these 'modified?'  i've had nothing but trouble calibrating drilled injectors.  their response deviates considerably from their original specs and each set is a little different.  if this is the case i don't doubt that you're having to compensate all over the place to get them dialed in.  also (and i'm sure you're already aware of this) be certain to stay in the linear pulsewidth range and out of the area below the 'knee' (nonlinear range), as best you can, in the injector pw curve. the problem is knowing what the curve is.  with drilled injectors the only way to know for sure is to send them in for examination.  and the bad part? - the data collected is only valid for that set and will likely vary from another set.

if you'd like a copy of my bin pm me - you're welcome to it.
Logged
nyet
Administrator
Hero Member
*****

Karma: +604/-166
Offline Offline

Posts: 12232


WWW
« Reply #100 on: December 09, 2012, 05:23:22 PM »

berTTos found out that the stock TEMIN, TEMINVA are too low for these injectors so we have been running with the following values:
TEMIN: 0.7788
TEMINVA: 0.7788

I don't get this either; why would you want a HIGHER TEMIN?
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
prj
Hero Member
*****

Karma: +903/-420
Offline Offline

Posts: 5787


« Reply #101 on: December 10, 2012, 07:22:53 AM »

I don't get this either; why would you want a HIGHER TEMIN?

Because they are mixing up TE and TI and doing it wrong.
Logged

PM's will not be answered, so don't even try.
Log your car properly.
rnagy86
Sr. Member
****

Karma: +6/-11
Offline Offline

Posts: 367


« Reply #102 on: December 10, 2012, 08:28:41 AM »

I don't get this either; why would you want a HIGHER TEMIN?

Well my theory is if you calculate TEMIN purely mathematically after switching to larger injectors
you should end up having a lower TEMIN, but the EV14s are way different than the stock injectors
or the Dekas so in real life this rule does not seem to be a pratical approach.
And by having a larger TEMIN I guess I was supposed to be rich at idle and at really light load areas, but I am not,
everything seems to be spot on -- btw i am running with nehalem's KRKTE at the moment which zeroed out my LTFTs completely.
« Last Edit: December 10, 2012, 08:30:51 AM by rnagy86 » Logged
prj
Hero Member
*****

Karma: +903/-420
Offline Offline

Posts: 5787


« Reply #103 on: December 10, 2012, 09:07:01 AM »

Well my theory is if you calculate TEMIN purely mathematically after switching to larger injectors
you should end up having a lower TEMIN, but the EV14s are way different than the stock injectors
or the Dekas so in real life this rule does not seem to be a pratical approach.
And by having a larger TEMIN I guess I was supposed to be rich at idle and at really light load areas, but I am not,
everything seems to be spot on -- btw i am running with nehalem's KRKTE at the moment which zeroed out my LTFTs completely.

Your theory is not correct. TEMIN is calculated for effective injection time, latency has little to do with it.
Unless you want to say that your big injector flows less at low pulsewidth than the stock injectors due to non-linearity, but this is not the case.

The reason you are still fine with TEMIN, is because it's quite low stock. Yours is tuned incorrectly, end of story.
Logged

PM's will not be answered, so don't even try.
Log your car properly.
rnagy86
Sr. Member
****

Karma: +6/-11
Offline Offline

Posts: 367


« Reply #104 on: December 10, 2012, 09:18:04 AM »

Your theory is not correct. TEMIN is calculated for effective injection time, latency has little to do with it.
Unless you want to say that your big injector flows less at low pulsewidth than the stock injectors due to non-linearity, but this is not the case.

The reason you are still fine with TEMIN, is because it's quite low stock. Yours is tuned incorrectly, end of story.

I thought EV14s are not stable below 0.7ms and Dekas aren't stable below 1.5ms. -- but hey I am just learning this Smiley
Logged
Pages: 1 ... 5 6 [7] 8 9 ... 21
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.365 seconds with 17 queries. (Pretty URLs adds 0.001s, 0q)