RomRaider

I already explained how ̶i̶t̶ ̶i̶s̶ ̶s̶u̶p̶p̶o̶s̶e̶d̶ ̶t̶o̶ ̶w̶o̶r̶k̶ it works.



That doesn't eliminate the possibility of variances between the individual tests, and does not remove the ambiguity I described between test conditions and in situ conditions.

The figures you are getting from DW are only applicable to a certain set of conditions, which obviously, are different than what is going on in situ, plain and simple.





A similar unit structure that you can plug ballpark figures into is much more preferable to some number that is not meaningful or easy to comprehend. Furthermore, using a u̶s̶e̶c̶ usec/g-air/stoich-rev value requires you do take an extra step to invert the difference in flow ratings. More complexity for no good reason.

Nobody can easily grasp what usec/g-air/stoich-rev means. I would much rather have timing table with degree units, even if its degrees retarded from 2.5* ATDC, rather than ATU clock cycle counts, wouldn't you?

Can you explain were the value 2707090 comes from in the formula?

True, but what percentage of variance would you expect to see between repeated tests on the same test system where most everything is controlled and/or known?

Yes they are. But they are more consistent and comparable to each other then trying to compare a derived estimated flow rate for gasoline from the ROM value to the measured flow rate of the injector with a different test liquid.
So you are agreeing with me that you can't compare the ROM flow rate with the test sheet flow rate from DW.

And nobody needs to worry or care about the units for the flow rate value in the ROM. We've already both agreed it is not a flow rate at all. So why not present it in the defs that way. It is a value that is adjust like any other value that needs adjustment when changing injectors.
Everyone knows their stock flow rate. There's enough test data on the stock injectors on various forums that we can reliably calculate the adjustment ratio.

For my case my stock injectors have been tested at 551cc/min. The injectors I installed where tested at 741cc/min. 551/741 is the ratio that all injector related values is adjusted by.


Here's an example (based on my ROM and data sheets) of what happens when you enter in the flow rate from the data sheet (which is what I suspect most people do when they change injectors, I did Did you?).
Stock ROM raw injector size value is: 5200
If I enter the data sheet mean flow rate (741cc/min) of the new injectors to be installed into the Editor def's formula, it converts that to a raw size value of:
2707090/741 = 3653

If instead I use the old/new injector flow ratio method to adjust the raw size value, what I get is:
5200*(551/741) = 3866

Now we can determine the difference in the calculation methods:
3866-3653 = 213
That's about a (213/3866) 5% error in the raw size value that is going to written back to the new ROM for flashing. You have to go make up that 5% difference somewhere. Most people will re-scale the MAF.

BTW: converting the ratio adjusted value back to a flow rate using the formula results in:
2707090/3866 = 700cc/min (again about 5% error in size)

So what I'm saying is the raw injector size adjustment using the ratio method might result in less injector size error to tune out. And therefore less of a change is required to the MAF to get back to 0% AFR correct.

From the unit analysis I posted:

60000000 (usec/minute) / [ 0.74 g/cc * 14.7 (g-air/g-fuel) ] = 5515720.

Accounting for the fact that we have two injectors firing per rev (4/cycle), we must divide by 2 and the result becomes 2757859.

My guess is that someone fudged the number to make it match X injectors flow rating, or used a different specific gravity of gasoline.



It depends how tightly they control the variables and how often they have their equipment calibrated.



Yes, but it seems for different reasons.



Nonsense. The units should be what the user can relate most to, even if it isn't perfect.



No, the real world meaning of the value is the static flow rate of the injector. The fact that it is expressed differently in the rom does not change that.



Because it then becomes a useless, meaningless number.


I have used this process.. It is common knowledge and can be done irrespective of the units that are involved. At the end of the day, you still need to fine tune it.

I have also plugged in the number from the sheet (reduced about 5% and dependent on the shop) as a baseline and tuned it in.

It seems like you're trying to box people in to this method of injector scaling by ambiguating the value, just because it doesn't perfectly align with every shop's. I don't agree with that on principle, and the fact that it complicates both styles of injector scaling.

At what variance between repetitive tests do you call for re-calibration?

We are still agreeing for the same reason. The ROM value as calculated by the current formula does not represent the flow rate on a test data sheet.

Why, it's fairly meaningless as you've described due to differences in testing methods. So why make the user think that it's accurate? Why not give them the raw value to adjust and disassociate it from the data sheet flow rate. I thought tuning was about accuracy.

OK I agree it is a flow rate, but not in the units of measure that you will find on the test data sheet.

Correct, it is without all the math that backs it up. And that's not needed to make an adjustment based on a ratio like what will be used on all the other injector related tables. And why would you use two different conversion methods, one for the injector size based on a tested cc/min of a different liquid and the rest of the tables based on a comparative ratio? All of the adjustments should be based on the same conversion method for best accuracy.

Yes, fine tune is the point I'm trying to make. It's easier to fine tune from a close starting point than to course tune if the value is off by a high percentage.

It is already ambiguous as you've pointed out with the math. Something's been fudged. I'm not trying to box anyone in, we've already been boxed by this formula. My initial suggestion was only to change the way the injector size value is represented. So that users don't think that the formula currently used in the defs is an accurate translation of the flow value from a data sheet to what's needed in the real world of your engine.

Here's the history of the current formula: viewtopic.php?t=503 It is stated there that it is an estimation and may not match flow sheets due to testing methods. This does not help the user tune his car...

And with further searches you can find a number of people discussing the inaccuracy of the current formula. Here's just one: viewtopic.php?t=1690

I'm done discussing this. It will be entirely moot once we get to user-selectable scalings.

[OT]

My... that looks decidedly like the 'Office Space' fax machine beat down scene