Fuel Injector Sizing

[4pointslow]

Fuel Injector Sizing
x is multiplication symbol
/ is division symbol


For naturally aspirated engines there is a fuel injector sizing equation all over the internet and also found in books written by respected experts in the field. The equation is this:
horsepower x brake specific fuel consumption / (.8 x number of injectors)
Note that horse power would be crank horsepower, not wheel horsepower

For boosted engines there is a different equation I have come across. It goes like this:
airflow cfm x .44298 / number of injectors (possibly x .8 for 80% duty cycle of injectors?)

To get CFM for a boosted engine:
cubic inch displacement x max power rpm x naturally aspirated volumetric efficiency x pressure ratio / 3456

My max horse power is at approximately 5200 rpm's (see first picture below)
Naturally aspirated volumetric efficiency of a two valve engine is approximately .85 (85%)

To get pressure ratio:
Atmospheric pressure + boost / atmospheric pressure
My pressure ratio:
14.7 + 16lbs boost / 14.7 = 2.09 pressure ratio

So on my engine....
244 x 5200 x .85 x 2.09 / 3456 = 652 cfm

652 x .44298 / 6 = 48lb injectors
if I use the .8 x 6 injectors to be at 80% duty cycle then....
652 x .44298 / 4.8 = 60lb injectors

So how does this math check out in the real world?
I have 60lb injectors and have just checked to see what the duty cycle was on them at WOT.
If this formula is correct (I think it is)....
Fuel PW in clock tics / (number of cylinders x time between pip edges in clock tics)
Then according to the datalog at around 6,048 rpm's.....(see 2nd picture below)
2290 fuel PW / (6 cyl x 605 Time between pip edges) or
2290 / 3630 = .63 or 63% duty cycle
I also checked at 5200 rpm's to see if injector duty cycle was higher, according to the numbers it was around 58% duty cycle.

So my 60lb injectors are running at most 63% duty cycle. Preferred would be 80%, you do not want 100% because it would be bad for the injectors and PCM.
So why am I going through the trouble of all this?
I want smaller injectors.
One of the oxygen sensor test called the rapid response test wont run/pass with the 60lb injectors installed. (see 3rd picture below)
I have to go through inspection once every two years and the last time I had to put some older 30lb injectors in to get the monitors to complete/pass. It sucks changing the injectors because I have to remove the supercharger.

This also has shown me that fuel control is not optimum at lower rpm's with the 60lb injectors, even though no codes ever set. I plan to try smaller injectors to see if the PCM will be able to run and pass the O2 rapid response test with them. I am wondering what is the largest size that the PCM will be able to complete the O2 rapid response test with. So far I know it will with 30lb injectors, and not with 60lb ones.

 

[vroomzoomboom]

the more math i see done to figure stuff like that out, the more stupider i become then what i was before......:scratch:

 

[4pointslow]

Math

I swear it is easier to type it than to read it. lol.

 

[2000StreetRod]

oversize injectors

I agree that
DC = injector pulse width in tics/(6 * pip to pip interval in tics)
And that your duty cycle at 6048 rpm is 63%.

I would also like to try lower flow injectors since I was unable to slope tune the AFR with my 60s and had to revert to MTF tuning. I thought about using them in my distant future supercharged 4.6L DOHC but even with high boost the demand would probably be about the same as for my boosted 4.0L SOHC. The 4.6L only has 0.575 liters/cylinder while the 4.0L has 0.667 liters/cylinder.

I'll be interested in what you select.

 

[4pointslow]

Stock injector duty cycle

Thanks for the confirmation on that formula 2000streetrod, it is funny how I searched for that formula multiple times, on the internet and even your thread and didn't put 2 and 2 together until last night. I am thick headed sometimes but when I finally get it, I GET IT. lol.

I datalogged my 4 door explorer to see what a stock 4.0 SOHC injector duty cycle looked like. It shifts at 5700-5800 rpms. The highest duty cycle I see was around 5769 rpm's.
2790 / (6 x 644) = around 72% duty cycle (see picture below)
Keep in mind when it is colder outside, and the air is denser, you would probably see higher duty cycle.

 

[4pointslow]

More math

So....
stock injector duty cycle aprox 72%
60lb injector duty cycle aprox 62%

62 / 72 = .86

.86 x 60lb = aprox 52 lbs injectors

I think anywhere from 48-52 might be the sweet spot, of course I would still have to retest to see if the O2 sensor rapid response test runs/passes. I am more inclined to buy a size that SCT might have a good value file on. I will check their files.

 

[2000StreetRod]

optimum injector flow

For your current injectors and max duty cycle:
60 lb/hr * .63 = 37.8 lb/hr @ 40psi

If you want the max duty cycle to be 80%:
(37.8 lb/hr)/.8 = 47.25 lb/hr @ 40psi

Siemens makes a Deka 44 (44.5 lb/hr @ 43.5 psi) which would have a max DC of about (37.8/44.5) * (43.5/40) = 92%

They also make a Deka 53 (53.5 lb/hr @ 43.5 psi) which would have a max DC of about (37.8/53.5) * (43.5/40) = 77%

Both of the above are high impedance but EV6/USCAR connector. I haven't checked the lengths. The reason I let James convince me to go with the Deka 60 is because of their linearity and low duty cycle controllability. But that's true for all of the Deka injectors. I wish I had gone smaller.

 

[4pointslow]

2000StreetRod
What do you think of these?

Injectors

 

[4pointslow]

minimum pulsewidth

I was looking at the minimum pulsewidth between the above and the 60lb I have now. Check out the difference, 60lb is 1.391 and the 52lb is .687
I could not find the calibration summary for the siemensdeka 52lb

 

[4pointslow]

min pulsewidth

Whats your take on the minimum pulse width in the SCT advantage3 software.
It cant be milliseconds, doesn't make sense when looking at the stock numbers.
I have .0000915
Edit:
I read it is ( / 1000) in advantage3

 

[2000StreetRod]

Bosch 52 lb/hr

Those specifications are for the Bosch 52 lb/hr @ 39.5 psi injectors with EV1 configuration.
37.8/52 = 72.7% duty cycle
Their minimum pulse width looks good. The Advantage 3 minimum pulsewidth is in seconds with the default maximum of 1 ms (.001). I had to use Tools, Options and uncheck Check Min/Max value limits to initially enter the specified 1.391 ms (.001391). Remember that the Deka 60 min pulse width of 1.391 ms is the linear range but I had to set mine at .2 ms to avoid clipping at idle.

I don't know if 52 lb/hr injectors will allow you to pass the O2 rapid response test but they have flow capacity for power increase (more boost, etc.). How do the 52 lb/hr specs compare to the 30 lb/hr specs that pass? What part of the test is causing the failure (idle, cruise, etc.)?

 

[4pointslow]

injectors

The test is run at cruise, approximately 20-30 mph. You can see in the 3rd picture above in post 1 it was around 30 mph when it was trying.

Comparing the 60 to the 30 I have now didn't even cross my mind, I am just thick headed sometimes. lol.
I am going to look for a part number off one to see if I can get my hands on the calibration summary for them. It may not be available though. I guess I could look at the value files for some 30lb ones too.

The factory minimum pw in seconds for my truck is .0000915 or .0915ms
That's a very low number compared to the 60lb minimum.
(1.391ms or .001391 seconds)

The 52lb Bosch would be .687ms or .000687 seconds (much closer to stock than the 60lb ones but still doesn't guaranty the O2 monitor will work)

I entered the 52lb minimum PW in a stock tune and there was no error about it being too big. Makes me wonder if that is why they put that limit in there in the first place.

 

[4pointslow]

30 lb injector value files

Here is the minimum PW info from the value files

2007 value file 30# Injectors = 0.000400543212890625
Ford racing 2009 valuefile 30# [M-9593-BB302] = 0.000774383544921875
Ford Racing 2011 valuefile 30# [M-9593-B302] = 0.00107955932617188

I don't know what brand the first ones are.
Here are pictures of the other two. Mine look like the second picture below on the right (M-9593-B302). I will still try to check for a part number.

 

[4pointslow]

30lb Injector part number

The numbers on the old 30lb injectors are
280150756
E9SE-B1B

I could not find minimum PW or a calibration sheet for these numbers.

 

[2000StreetRod]

unreasonable min pulsewidth

. . . The factory minimum pw in seconds for my truck is .0000915 or .0915ms
That's a very low number compared to the 60lb minimum.
(1.391ms or .001391 seconds)

That's the same min pulsewidth in my stock CDE4 strategy. I doubt it is representative of the injector capability. It was probably arbitrarily set that low by Ford to eliminate it as a limitation in the strategy. I just can't believe the solenoid response time is that fast.

. . . I entered the 52lb minimum PW in a stock tune and there was no error about it being too big. Makes me wonder if that is why they put that limit in there in the first place.

The Advantage III default minimum pulsewidth minimum is 0 and the maximum is 1 ms so there's no problem with .687ms. There was only a problem with the Deka 60 min pulsewidth of 1.391 ms which exceeded the maximum.

 

[4pointslow]

Fuel

It would be interesting to see how much fuel was coming out with the 30lb injectors during the O2 monitor test and compare it to the amount of fuel that would come out in the minimum repeatable PW of the other injectors. Probably too much math anyway. If I didn't have to remove the supercharger to replace the injectors I would have changed them by now just to see if it would work.

 

[2000StreetRod]

M-9593-b302

I found a post on the internet stating that M-9593-B302 is Ford's injector equivalent to Bosch's 280150756. Here's the calibration summary:

Although I don't see how that helps right now.

 

[4pointslow]

where

Thanks,
Where did you find that? I was looking for that.
Have a link to the thread?

 

[2000StreetRod]

link

Thanks,
Where did you find that? I was looking for that.
Have a link to the thread?

https://fordperformanceracingparts.com/parts/ics/m-9593-b302.pdf

 

[2000StreetRod]

rapid response test

Have you tried adjusting the Closed Loop Bias, the Min Allowed Closed Loop Bias and the Max Allowed Closed Loop Bias to see if you can pass the rapid response test?