Injector Flow Calculation [Advanced-ish Tech]

[engineermike]

I've run across a couple of cases where either a) we wanted to know if brand X 95lb injector really flowed the same as brand Y 95lb injector, or b) why I have 2 differing sets of injector data for one set of injectors. You can compare the Ford injector data but no one number can be used to compare injector flow rate. Naturally, it's not that simple, so I set out to figure out how the math works. It's kind of crazy, but this is it:

First, determine the fuel request. This would be the greater of how much is required and the min pw delivered fuel:

Next, determine if you're above or below the breakpoint (this is the inflection point between the low slope and high slope).

If you're below the breakpoint, the equation is:

If you're above the breakpoint, the equation is:

The final step is to add the offset:

That said, I believe if you log your PW in something like VCMScanner, you'll get the PW before the offset if added.

These are the above parameter locations in PCMTec:
LowSlope - auF0119
LowPMult - auF30942
HighSlope - auF0121
HighPMult - auF30941
Breakpoint - auF0123
BPMult - auF30940
minpw - auF0124
OffsetTMult - auF61902
TMult - auF61903
Offset - auF0126
OffsetPMult - auF30943

Hope someone finds this useful.

Sponsored

 

[Optimum Performance]

Excess Injectors give you all the data for your ECM. Without this level of actual data you are just guessing.

 

[engineermike]

Excess Injectors give you all the data for your ECM. Without this level of actual data you are just guessing.

I'm pretty sure all injector suppliers give you all the data for your ECM. Some data is more accurate than others. Fuel Injector Connection, for instance, has better data than most.

The intent of this thread was to explain how the ECM uses the data to calculate the pulsewidth. Ford does it different, and arguably better, than other OEMs. If someone were using, say, a set of Ford 52# injectors and were unhappy with their trims, they could use the above math to compare the pw output yielded by the Ford Performance data to the Whipple data to the GT500 data, as an example. Or if you had some DW95's, you could compare the output of DW data vs 2018 Cobra Jet, that uses the same injectors but different data. There are many examples of ways to use it.

 

[engineermike]

Here's an example. We have 3 sets of data for the FP 52# injectors: FP, Trinity, and Whipple. Also, the FP 55# injectors are OEM in the Predator and quite popular in the centrif setups, but are very close to the 52s in flow, and data is only available in the OEM Predator tune AFAIK. Here are all 4 compared:

So you can see here that the FP data yields a very similar result to OEM Trinity data, with the exception of the very low PW range. Whipple data is similar to OEM Trinity data at low PW, but off at high PWs. The Predator 55s are capable of both lower and higher flows, albeit a small difference on either end.

 

[deanm11]

I love this level of tech. So related to your subject, I can't find anyone talking about this... injector flow, apparently, is always flowed with a test fluid or the like, not a fuel. I see it said it is for safety purposes but if the injectors are safe to fire in your car, why not in a tester that has no material heat in it. I guess because there's more risk because it's right in front of you?

My research in this area was due to the difference in viscosity amount fuels. "AI" claims 6 to 13 % less actual flow for E85 vs. 93 pump fuel. E85 is the highest viscosity, 93 thinner and the test fluid even thinner, flowing 5% more than 93. (according to AI) IF this is all correct, injector data should be adjusted substantially for all of us on E85.

 

[engineermike]

@deanm11 when you do orifice flow calculations, and it’s been a while and I’m going by memory, viscosity makes almost no difference. Something about the reynolds number or perhaps the speed of sound, I can’t remember for sure.

 

[Grimreaper]

Thanks for sharing, solid info.
On the Cobrajet they force the min pw high enough that it's always using the hi slope with the dw95's. Gen2 this needs a bit of tweaking if using stock pump control/basket. Gen3 probably not as much. Dw data is horrible imho, at least for these injectors. Would like to see the fic data if you have it to share.

If self tuning, plot your fuel flow from current hi/ lo/BP and across the range of PW your car will see. Then compare to your actual fuel trims. Helps immensely to see the impact if you change the low slope or break point and the expected impact on your trims.

The transient tuning was biggest issue I ran into on the 1000cc. Large over fueling swings when DFCO ended and slipping the clutch at take off. Best I could figure is the stock transient tables for puddle/wall fuel was no longer accurate with an open orfice vs diffused tip like the smaller coyote injectors. With pcmtech there's a lot of these tables for cam movement, not just manifold/ rpm change to.

 

[engineermike]

On the Cobrajet they force the min pw high enough that it's always using the hi slope with the dw95's.

Interesting observation. Though on the CJ data the low and high slopes are so close it only really needed one slope to begin with.

Gen2 this needs a bit of tweaking if using stock pump control/basket. Gen3 probably not as much. Dw data is horrible imho, at least for these injectors. Would like to see the fic data if you have it to share.

I have FIC95 data and here's how it compares to CJ and DW:

That said, I can't say for sure the FIC 95's are physically the same as the ones in the CJ which I know are DW.

Sponsored