Look what Calibrated Success has been up to...

[eficalibrator]

Guys, I've gotten a lot of questions and requests for more information about how one gets "good" injector data. Up until this point, the only solution seems to have been to hope it's available from either the OEM (Ford, GM) or injector manufacturer (Bosch, Continental/Siemens) and in the public domain. I went through the exercise to convert the Ford Racing public data from Ford units and pressure into GM units and pressure, and the results are included in the data disc that accompanies my "GM Tuning Beginner's Guide" DVD.

Going a step further, I have been investigating what it would take to generate this data outside of the OEMs. What should the raw data look like? How do you run the test? What equipment is necessary?

So far, I've learned a lot about injector companies in the aftermarket. First and foremost, it showed me that only a couple of them actually have clue 1 about what they're selling. Seriously, this is a SHORT list.

Next, I got to learn about the test equipment available. A lot of places have generic injector flow benches, but this just doesn't cut it as delivered. Even the best digital bench out there today requires some "adjustment" to get useful data. ...but I'm getting close. I'm working closely with one test equipment company to see if we can manipulate their hardware to generate useful CALIBRATION DATA.



I still have a LOT left to do, but this graph visually shows the high slope, low slope, breakpoint, offset, short pulse adjust, etc. necessary to derive calibration data. IF this all works out clean, you'll see some new training material from me that will help shop owners willing to make the investment in the hardware how to start doing this too.

[Alvin]

I've trying to do the same thing in my spare time. I built a small little micro controller project to fire a injector at specific intervals of time and compare it against theoretical fuel delivered and actual. The injector I tested was supposed to be a either a 28.4 or a 26.4 (been a while, can't remember) and I came up with a high slope best fit of 2x.3. It worked out to a ~.0075% error

The controller side of my project is done. I just haven't had time to build something to test more than 1 injector at a time.

The one thing I've already noticed is temperature of the fuel makes a tremendous difference in results. In fact if I ever go ahead with this I thought I might use a nitrogen "pushed" fuel or test fluid supply instead of an electric pump.

[Alvin]

BTW. I didn't mean for this to be a one upper type of post. If you'd like to collaborate on some things for this or need any hardware/math side help let me know.

[DSteck]

ID actually uses actual gasoline for the test and precisely regulates fuel pressure and temperature to help standardize everything. Temperature control is definitely necessary.

[edcmat-l1]

ID actually uses actual gasoline for the test and precisely regulates fuel pressure and temperature to help standardize everything. Temperature control is definitely necessary.

I imagine that could get a bit sketchy.

[DSteck]

I imagine that could get a bit sketchy.

I've heard stories.

[LSxpwrdZ]

I've been wanting to build a test bench myself to do such tests and actually have the plans and model numbers of controllers and electronic regulation devices to do it all... Time is the biggest issue for me.

[eficalibrator]

No offense taken, Alvin. The trick here is to take data from more than just a sample size of one. Of the four injectors I was using to generate the plot above, I had two average units, one slightly higher and one slightly lower. Hence, the "average mass flow" above.

It's interesting to see WHERE most of the differences occur. These locations also help explain why some people spend a lot of time chasing their own tail on driveability fixes without matched sets of injectors and the right data.

As mentioned, temperature plays a role but it's not just fluid temperature that matters...

In case anyone is wondering, that plat is based on the stock LS3 injectors at 4 bar. As you can see at a glance, they're damn close to linear across almost the entire spectrum. You can verify this by looking at the stock GM calibration values in the short pulse adder table. See how GM uses numbers in the neighborhood of ~0.04 for adjustment most of the time? That means that they're hardly adjusting away from the linear "ideal" model at all because this injector is physically more linear than some of the older models. (Yet another reason why GM/Bosch went with these new injectors when trying to hit tighter emissions targets.) Yes, any single injector may vary from the "ideal" injector model, but that's why we characterize them as averages and hope that we bought a matched set. You guys are buying dynamically flow matched injector sets for your high performance applications, right? In the production world, GM recognizes that there will inevitably be variables and basically hopes that closed loop control and catalyst loading will be enough to cover the differences as far as emissions are concerned.

The punchline: Analyzing my raw data yields a flow rate (linear fit) of 5.24g/s at 400bar and an offset within about 0.2ms of stock. I'll be taking a closer look at the offset errors moving forward, but even GM revised that data over the years between LS7 and LS3 with the same injector so nobody's really sure what's right just yet.

[Alvin]

No offense taken, Alvin. The trick here is to take data from more than just a sample size of one.

Roger that.. My initial goal was just to try something and see if I could get measurable results out of it. Now that I did I just need to build the rest of the bench. My goal was to be able to test 8 at a time.

[Gh0st]

Great to hear guys. Nice to see people enthusiastic about tuning enough to help the community.

[Cantalope Kid]

Have you considered just incorporating the help of one of your existing followers (Injector Dynamics)? They have the equipment and know how to provide you with the data you seek. I'm certain they would be willing to share in the profits of good injector data. Well maybe not...Might be better for their bottom line to keep people in the dark regarding competing manufacturers injector data. Lol

In my opinion: By time we pay for the injectors, then pay more for yet another "how to tune" for new released injectors, we could have just sent sent our customers to Injector Dynamics. These guys learned to tune injectors from Greg Banish's own "Advanced" coarse. Then they went a step further and designed, manufactured, and tested their own injectors in house. They also include all necessary tuning data for the injectors they sell. Win-Win!

[DSteck]

Injector Dynamics did what they did independent of Greg's classes, although Paul did attend the classes. I'm 110% sure Paul has no desire to share methods or characterize other people's stuff (because I've asked him about it before).

[eficalibrator]

I'm actually good friends with Paul @ ID and have been in touch with him constantly on this project. I'm actually going to run a set of his ID850's as part of the study. I will eventually compare the on vehicle behavior of the SD60's with my math based data (from the DVDs), SD60's with new data (if it comes out different), ID850's with Paul's data, and ID850's with my new data (again, if it comes out different).

What becomes of my raw data coming out of this is still up in the air. The only thing I can say for sure is that I'll have another training video that focuses on injectors for both the repair and calibration aspect when I'm done. Cost will be significantly lower than the formal calibration training DVD's, but the content is simpler and more limited.

[eficalibrator]

This is as good a thread as any to add this...

Since I'm in the middle of building my new shop/studio, dealing with a new baby, and processing a pending legal concern with someone who thought it would be a good idea to violate federal copyright laws, I haven't had much opportunity to work on the next DVD. But I did have some smart-assed T-shirts made. Paul Yaw from ID gave his personal stamp of approval, so I guess I must have got the equation right.




These will be available shortly through Summit Racing Equipment. I'll also have a limited number of them with me at the PRI trade show this weekend. Those of you who read the books, paid attention to the DVDs or have been to a live class should appreciate the humor. Thanks again, guys! I promise to have more top notch training material soon.

[Eric@HPTuners]

This is as good a thread as any to add this...

Since I'm in the middle of building my new shop/studio, dealing with a new baby, and processing a pending legal concern with someone who thought it would be a good idea to violate federal copyright laws, I haven't had much opportunity to work on the next DVD. But I did have some smart-assed T-shirts made. Paul Yaw from ID gave his personal stamp of approval, so I guess I must have got the equation right.




These will be available shortly through Summit Racing Equipment. I'll also have a limited number of them with me at the PRI trade show this weekend. Those of you who read the books, paid attention to the DVDs or have been to a live class should appreciate the humor. Thanks again, guys! I promise to have more top notch training material soon.

That is awesome. I gotta order one!

[DSteck]

Congrats on the baby. I'm calling it quits after the two I've got. Lol.

[MMGT1]

Shortly? Meaning available before Christmas Greg?

[HartRod]

+1! Also, Greg when are you going to post the 2014 Tuning class schedule?

[eficalibrator]

Digging this back up to let people know that Summit is running a promo where you get a free T-shirt with the DVD.

Summit Racing promo #977

Details: Buy a Summit Racing EFI tuning DVD and get a free t-shirt!

Dates: July 21-August 31, 2014

[John Overton]

The shirt just came in, now I am just waiting for the DVD lol.