Nizpro Drilled injector data

[DarrylC]

Darryl you are 100% right regarding tuning the airflow model.

As a general observation for the community, if you put a set of IDs in a 100% stock car I very much doubt the fuelling will be 100% correct. In fact if you did this for the Deka 60 data you will be riding the minimum pulsewidth and have a rich condition as low engine speed. I did a full speed density re-calibration on my BA when I had standard cams and it took forever. There is so many 'load' scalers etc that need changing to reflect the recalculated load as a result of different airflow that it becomes very tedious exercise.

On a car that doesn't have a unique slope and offset for each cam angle it is much much easier....however for the masses I would just fudge the scalers and tidy up the airflow where needed and be done with it.

IH8TOADs, turbotrana, hiddeous, thinking about it I am doing what Toads says at the end so am just fudging as you say.

The way I get the injector dialled in was to find an injector that was close (I had Bosch 968s and used the data from the Ford M39 injector with the maximum value for Offset from the BA and M39 calibration). Increase Offset until the idle smoothed out. I would then adjust the Low Slope and breakpoint until I go the idle fuel trim under 2% (bit of a tedious process, but calculated by maths).

To Dial in the tune:
Drop boost to 10psi and lambda back enough so that the injectors were below 90% duty cycle (constant 0.75 from 2500rpm up was OK for them) and do a full throttle run to get the High Slope dialled in close. Another full throttle run and I would do a Lambda difference on the desired versus actual and multiply the "MAP per Airmass" table in those RPM ranges by the ratio of desired/actual to clean up the range (outlined in the Speed Density write up I did).

The car would be running well by this stage. Because I don't have access to a dyno then I would drive (200+kms) and get the LTFT values around all the breakpoints that covered the cruise range (1250-2000 covers the 60-110kph range on my car, so breakpoints of 1250, 1500, 1750 and 2000rpm). I log the LTFT/Cam Angle/RPM was for each point and adjust the "MAP per Airmass" at each of these points ("MAP per Airmass"/(1.0+LTFT) so if fuel trim is -7% I would divide by 0.93).

This last step was the key for me getting the fuel economy as I would get my cruise fuel trims under 2%. If I had a dyno I would have done what Tony/LaSota racing suggested and set the LTFT off and measure the STFT to get the amount out at cruise. Note you can do a quick and dirty here and just work out the fuel trim adjustment that will get all the load points close to 0% if you wish.

So in all a fudge that will get you really close.

[JETURBO]

Darryl u can get the LTFT to cut in at idle to replicate what trimming on low slope it will do ( cruise lo load ) dont know if u knew but it does take a while, car has to be at full running temp and let it sit idling ( and keep logging ) then after ( this time can obviously differ each time ) it will cut into LTFT and logging will show u where to go from there.

This was the easiest way for me, even sitting on the dyno just low loading would take to long and to many variables imho

[DarrylC]

Jet, I think I am doing what you are saying just did not put in all the details. I should have said get the idle LTFT to under 2% and yes it does take a while for them to settle. Once I get the car warmed up I use a couple of load points for my calculations (neutral/in gear on the auto).

I think what Tony does might be quicker as you don't have to wait for the LTFTs to settle, just get it on closed loop for a short period.

Is there a better way for dialling in injectors that are already off the low slope at idle?

And then I have to tune for the LPG injector and get a balance between petrol and LPG tunes (compromises the petrol tune).

Great thread, learning heaps

[JETURBO]

Haha yeh ok i see i thought u ment u went for a 200km drive at cruise, was thinking how the fook do u do that !

To answer your question .... Maybe i personally dont know but half of the cars that have "issues" out there can 8 times out of 10 come back to the scalers and doing that vital initial 1/2 hour at the start gives me at least a great base to start tuning off
Seen all sorts of crazy scales out there for typical injectors and most common is leaving stock everything but upping slopes

Dont know how some tune shops think this is ok but it takes all types but room for us to improve on !

Also some tuners have a different thought process to why and how they come up with there figures, at the end of the day as long as iam trimming as close to 0% with a swing each way by 2% and my lambda is reading out the tailpipe exactly what each throttle % is in my fuel map iam happy and the car responds brilliantly

[Huddo]

Micks wifes car had the ID725's replaced with a 1000 cc injectors of different brand to what everyone normally uses and they had some issues, they were replaced with a set of KPM1250's.
He would not say what the brand was of the 1000cc as he does not want to run down a brand at all

[hiddeous]

I'm not saying the other 1000's where xspurts but im also not saying they weren't.

[DarrylC]

Micks wifes car had the ID725's replaced with a 1000 cc injectors of different brand to what everyone normally uses and they had some issues, they were replaced with a set of ID 1250's.
He would not say what the brand was of the 1000cc as he does not want to run down a brand at all

Huddo, what did Mick actually say?
I would be surprised if he said ID 1250s as they are ID 1300s or perhaps they were the KPM 1250CC injector which were later changed to 1500CC for 4 bar systems (from memory). I thought he was a KPM reseller but might have this wrong.

[JETURBO]

You have those figure and stats correct

I think he may of been playing around with the "BLACK OPS" injectors being 1000cc and typically come with zero data

[Huddo]

sorry, he said KPM 1250's, for some reason I typed ID's, sorry if I caused any confusion

[turbotrana]

This is quite a good piece of info http://www.myo-p.com/Ford-EEC/EEC%20...reakpoint.html

The problem with our injector type is that the start flowing differently under smaller pulsewidths which throws out the actual amount of fuel delivered versus the calculated. One could have a function or table of injector flow versus pulsewidth but instead there is a second injector value available for the EEC to use in its calculation. When the amount of fuel to be injected is less than the injector breakpoint, the EEC starts using the "other" variable, the low slope, to calculate the pulsewidth. I you graph the fuel delivery versus the pulse width, you will find it pretty linear with the exception of the very low end of the pulse width where the slope of the curve (line) is changing. The linearity is not the same and the fuel error gets bigger and bigger with smaller pulse width. The bigger the injector, the worse it is.

The key is to find the point where your injectors (average of all 8) curve linearity is changing. Bench testing is probably the best way to do this.
For example:
To get my breakpoint set to 2.5ms I had to work out how much fuel (lb mass) is delivered in 2.5ms
First I worked out how much fuel is delivered in 1ms:
1000ms in a second
60 seconds in a minute..
60 minutes in a hour
42.29lb/hr injectors
lb mass per ms=42.29/(1000*60*60)
which works out to be 1.174722222e-5 AKA 0.0000117472 lbs of fuel per ms.
so to set it to 2.5ms I multiplied 0.0000117472 by 2.5 which gave me 0.000029368
Set the breakpoint to that and tada! A useful injector low slope!



The breakpoint tells the EEC when to change between the high and low slopes. Both values are used at all times, but the low slope value dominates the calculation below the breakpoint and the high slope dominates above the breakpoint. If you're rich at pw below the breakpoint increase your low slope and vice versa. If you're rich at pw above the breakpoint increase your high slope, and vice versa. If you're rich across the board, increase both, and vice versa. Have you tried plotting your KAMRF vs. PW using Walt Barnes' spreadsheet (in the files section of this list on Yahoo Groups under "Other". The graph makes it very clear what is happening. It's also important to be aware that the breakpoint value is in lbs. of fuel, not ms.
See also:

[turbotrana]

Some strategies allow you to datalog pulsewidth. Mine doesn't . So I use a Multimeter that measures pulsewidth in MS and run wires from the injector to inside the car to measure pulsewidth.
If you look at Injector data that Injector Dynamics puts out, it seems that the breakpoint in milliseconds is usually around the 2ms mark. From that you can use the above calculation to work out a breakpoint in lb per fuel injection cycle .
From the above calculation you can determine when you should be in and out of the lo slope area.From reading the your multimeter pulsewidth you can determine what slope you are in and need to adjust.

Some more good info http://info.efidynotuning.com/fuel101.htm

[DarrylC]

This is quite a good piece of info http://www.myo-p.com/Ford-EEC/EEC%20...reakpoint.html

....

Tony, IMHO that article is from someone who does not understand what they are talking about.

Injectors flow nothing from 0 to Offset,
From Offset to Offset+minpw they flow somewhat erratically but generally around the low slope
From Offset+minpw to the Breakpoint time they flow close to the Low Slope (model the Low Slope flow well)
From Breakpoint time to rated duty cycle (usually 90+%) they flow at High Slope

This info is available from the Injector Dynamics web site.

You can use this information to accurately calculate injector pulse widths/duty cycles from the speed density tables to see what is the capability of your injectors are.

What can be seen from this is that if you assume, like the article doses, that injectors flow linearly then you are way off.

I once modelled injector pulse width using a linear model and it was not what was logged by SCT. So I modelled the injector flow as per above and my predicted pulse width was within a couple of % of what SCT logged.

[turbotrana]

Thanks Darryl, have you got a link to that

[DarrylC]

Tony,
Ford characteristics:
http://injectordynamics.com/articles...racterization/

The GM one:
http://injectordynamics.com/articles...racterization/

which talks about what I was saying about calculating the injector flow based on a start point of zero.

IMHO the Injector Dynamics web site has the best information I have ever seen, it is especially helpful about how to calculate VE on a Ford (which is what I based a lot of my SD write up on).

[turbotrana]

Have skim read that a few times in the past. I guess got to do a slow read.

[turbotrana]

I rang Nizpro to get the scallers for the FG XR6T injector.
He gave me the following
lo Slope 70 lb/hr or 0.0195 lb/s
hi slope 79 lb/hr or 0.022 lb/s
breakpoint Stock 0.0000148lb
offset 13v Stock 0.86ms
offset 14v stock 0.76ms
Min Pulsewidth Stock 0.95 ms

The guys reason for keeping the last 3 stock is that the " electrical characteristics of the injector have not changed"

The following are the figures I have come up with thus far. The Hi slope has not been dialed in 100% tho just approximate. The other figures trim not too bad +/- 5%
lo slope 93lb/hr 0.025833 lb/s
hi slope 76 lb/hr 0.02116 lb/s
breakpoint 0.00005lb
offset 13v 1.694
offset 14v 1.06270
min pulsewidth 0.3ms

I am only mainly playing around with these Nizpro Drilled injectors as an exercise in learning better how to scale injectors but geez the Nizpro guy made me feel bad when he said it should only take 10min for a good tuner to get them fully dialed in!!!!!!

[DarrylC]

I rang Nizpro to get the scallers for the FG XR6T injector.

...
The guys reason for keeping the last 3 stock is that the " electrical characteristics of the injector have not changed"

Might need a B@##^&*% filter on this one. I have heard this before and will trust what both KPM Motorsport and Mr Paul Yaw (Injector Dynamics) says:

http://www.fordforums.com.au/showpos...2&postcount=35

and in particular this excellent article titled "Drillbits and Dipshits"

http://wpdev.injectordynamics.com/ar...-and-dipshits/

Lets face it, 1000cc injectors are usually machined 550cc injectors and suddenly the minpw has gone from 0.7ms to 0.272, I think things are changing. I have asked KPM to explain but as of yet there has not been a reply. Might just email David.

I am only mainly playing around with these Nizpro Drilled injectors as an exercise in learning better how to scale injectors but geez the Nizpro guy made me feel bad when he said it should only take 10min for a good tuner to get them fully dialed in!!!!!!

10mins, you feel bad! I must be terrible it has taken me forever... How do they do it

What a good idea, ring the vendor. Impressed by Nizpro giving out the data.

Well done Tony. You are certainly helping me out.

[JETURBO]

It would be because nizzy for 1 would have years worth of data to "knowingly" plug in and then the fact is most "tuners" will accept +15% trim hence the 10min slut job u see go out and then cars stalling, hard starting, stalling at pull up,using excess fuel and so on

Yes theres heaps of 10min scaler jobs out there as tony first found out with his car, no one has time to do things more than once in this industry it seems

We heard of a well known "tuner" say to a customer that "genuine id1000's leak and dont work well and does not recommend them to anyone"

Ummmm yeh ok then !!

[turbotrana]

Toads saw the Nizpro scallers used in my tune a couple of years ago and immediately pointed out the unusual slopes and suggested to increase offset and raise the lo slope. I was green at the time and whilst I took note, did not understand much about these scalers at the time.

Now that I have some idea about the scalers I do scratch my head as to why Nizpro have done them this way as it is clear from when you look at every injector flow pattern they follow the same trend. All my "top tuner" did was use the Nizpro injector values but still scratch my head as to why he used them also if he is a "top tuner".

This guy said it best http://info.efidynotuning.com/fuel101.htm

Injector Low Slope
The Injector Low Slope accounts for non-linearities at low pulsewidths, at low pulsewidths an injector acts as a larger injector spraying more fuel, so to compensate you will have a low slope with a value HIGHER than the high slope. Typically the low slope is approximately 15% larger than the high slope. So a 30lb injector would typically have a low slope of (30*1.15=) 34.5, a 60lb injector would have a low slope of (60*1.15=) 69 Now the further apart the low slope is from the high slope the more influence the breakpoint will have, thus IT IS IDEAL AND HIGHLY RECOMMENDED to have the low slope AS CLOSE AS POSSIBLE to the high slope, this will insure your fueling is not doing anything odd at the breakpoint. However, the low slope MUST NEVER be less than the high slope but can be equal to.

[IH8TOADS]

ID1300 and KPM1500's have a quoted low slope lower than the high slope. These values represent the characteristics of the injector so it is what it is. I've had to adjust my thinking!

In regards to the Nizpro Injectors I have always added to the battery offset table as this table is an adder. With the holes in the director plate being enlarged, my theory is to adjust the table values as more fuel would theoretically flow in a given dead time. But as Nizpro have said, the electrical/mechanics of the injector have not changed and the opening/closing time shouldn't have either. Its just how you interpret the table. I have learnt over the years that there are many ways to skin a cat....and sometimes there is no right or wrong way, especially when you don't have the proper equipment to obtain the necessary data. The Nizpro director plates are actually pretty good and shouldn't be put in the same category as those Injectors in Paul's article....they're in their own league!!!