where to get LS9 injectors?

[Iam Broke]

OK, I'll try again Sorry, no excel here to play with.

[RWTD]

I've never noticed a transient problem with LS9 injectors.

How many non-LS9/LSA vehicles with LS9 injectors have you actually tested for variances on transient? I realize you mentioned your friends vehicle earlier, but you stated you only tested WOT before/after, and not anything non-WOT related. If you have any data I would love to see it (I'm a data whore). I'm not trying to be argumentative, but I'm just perplexed how an injector with a different spray angle vs factory can have zero transient tuning changes and there not be any logged differences. If say a Siemens 60 that works perfectly in an LS7, but same injector in an LS9 requires transient tweaks, why wouldn't an LS9 injector require transient tuning treatment in an LS7, for example? My main concern is emissions, and of course potential drivability concerns brought forth by the customer afterwards. I have some *picky* customers. Then again, maybe I'm just that obsessive compulsive when it comes to tuning? LOL!

[RWTD]

OK, I'll try again Sorry, no excel here to play with.

Much better!

P.s. I prefer the IFR values to be in #/hr, and with a 4 place decimal. LOL!

[Iam Broke]

RRRRRRIGHT! I'll send you the stock file then.

[RWTD]

RRRRRRIGHT! I'll send you the stock file then.

Hehehe!

Everyone, there should be several LS9/LSA files in the repository, which would make copy/pasting a lot easier.

[DSteck]

How many non-LS9/LSA vehicles with LS9 injectors have you actually tested for variances on transient? I realize you mentioned your friends vehicle earlier, but you stated you only tested WOT before/after, and not anything non-WOT related. If you have any data I would love to see it (I'm a data whore). I'm not trying to be argumentative, but I'm just perplexed how an injector with a different spray angle vs factory can have zero transient tuning changes and there not be any logged differences. If say a Siemens 60 that works perfectly in an LS7, but same injector in an LS9 requires transient tweaks, why wouldn't an LS9 injector require transient tuning treatment in an LS7, for example? My main concern is emissions, and of course potential drivability concerns brought forth by the customer afterwards. I have some *picky* customers. Then again, maybe I'm just that obsessive compulsive when it comes to tuning? LOL!

Probably 20 cars with LS9s that don't use LSA/LS9 heads. That includes Gen III and Gen IV stuff.

An OCD person would appreciate the fact that real data doesn't exist for SD60s.

[RWTD]

An OCD person would appreciate the fact that real data doesn't exist for SD60s.

I can take that shot, but I can assure you his data is >95%, and the rest can be baked. Btw, shouldn't we allow Greg to discuss and defend his injector data himself?

I recall 2 years ago you stating you've never touched/tuned with transient fueling before, and you asking Greg for help on approaching where to start for such (). One can have injector data for days, but that still doesn't mean the rest of the fueling curve is proper. I have to ensure all of my customers pass emissions. One can't get a tag in Denver if that doesn't happen (I believe that's the case in St. Louis, as well?). I've seen too many do wonderful on steady state testing, but easily fail overall due to transitional issues. Yes, the boogie man does live, and he causes a ton of headaches. A goal of mine is to see the vehicle pass in all categories, and do well in all aspects, not "just get by".

[DSteck]

My stuff passes. *shrug* I'd never pass a sniffer because of the cam though. Lol. I'm not a "just get by" person either. If that was what your insinuating, I'd say that's an unfair claim.

I agree, his data is better than what's out there, but even he has admitted it isn't 100%. That's the inherent error in converting Ford data to GM. Short Pulse is where it suffers most. Further, Paul Yaw has tested those SDs (the short and long 60s and the 80s). The 80s are horrible. The 60s aren't ideal either (better though). I like injectors that have passed CARB testing (the new Noble M600 uses ID725s actually for that reason).

Yes, I asked Greg about how GM handles transient issues. However, with doing both VE and MAF together as opposed to MAF only, I've never had to tweak transient fueling to get a good feeling response. I was lazy for a while doing MAF only until EQ_VE became prominent for use. Even PD blower cars aren't too bad. I still make adjustments, but its not an area I scrutinize over, especially when the butt dyno can't pick it up. Transients get screwed up for sure when doing the stoich trick for IFR, so that instance definitely requires adjustment until Chris adds the coefficients into the calibrations to edit the IFR calculation.




Having said all that, the LS9s work fine. If I get some time (time? What's that?), I'll post a couple files with the data in different formats. I went through and did sixth order polynomial refitting of the data to other axis labels since I use them on a wide variety of platforms.

[HNK]

Any chance a more accurate data for the Long SD 60s?

[Jabbott]

Actually it is on my list of data to do and turn over to Greg. The one thing that was not right about the data on the SD60 was not he actual injector data itself, I don't think the new data is going to be changing much. The issue is the model around the injector in the ECU is made for a Bosch with a pencil stream and the SD 60 has a larger pattern to it. There are models for IVT INJT that also have an effect and they are completely wrong. I am working on having the correct model for those also.

[eficalibrator]

Dave, did I piss in your Cheerios or something?

As far as data for SD60s is concerned... Translating Ford style data in GM might be close, but its not right (especially the short pulse adder).

An OCD person would appreciate the fact that real data doesn't exist for SD60s.

What leads you to believe that data generated by Ford's SciLab, based on a large sample size across all applicable conditions, is not accurate? The same characterization methodology has been in use at Ford since at least 1987 and currently works to pass TierIIBin5 emissions on many applications. I would venture a guess that their two-slope characterization is "accurate enough" to satisfy potential driveability concerns with your average aftermarket enthusiast.

All I did was take their data (derived from a strong numerical sample and average) and apply it to new breakpoints that align with GM vehicles and ECU parameters. My resultant relationship of fuel mass versus activation time is IDENTICAL to Ford's data. Don't shoot the messenger.

Regarding your concern for short pulse anomalies, do you have error data to show discrepancies between my provided data and actual fuel mass delivery at short pulses? Is your data based on a statistically significant amount of samples? I will concede that the Ford method "cuts the corner" so to speak near the breakpoint. The real question are, "Is that enough error to make a noticeable difference?" and "Am I even operating in this region?" You either have data or an opinion here. If you can demonstrate that the Ford measurements are improper, I'm all ears.

I will be among the first to acknowledge that the SD80's are not exactly consistent. I tested them first hand and found them to be all over the place. I'm honestly not sure how many Ford would have had to have tested in order to get useful data. Even then, the part to part variation that I saw was pretty bad. This is NOT the case with the 60's.

If anyone wants a good demonstration of the accuracy of the SD60 data that I provided, look no further than my GM Tuning Advanced Guide DVD. As part of the instructional scaling exercise in that video, I calibrated the engine first with stock injectors and then followed that by observing what happened when I ONLY changed the injectors to SD60's (using my data from DVD#1). The result was lambda errors of less than 3%, indicating a strong correlation between my injector data and reality provided you believe that the OEM 41# injector data was right to begin with.

This is not a Calibrated Success vs Injector Dynamics argument. I too am on speed dial with Paul Yaw. Ask him who helped him understand GM/Ford OEM injector models... ID doesn't make an injector smaller than ~70+lb/hr, so that leaves a big gap for the average boosted LSx street car guys before they hit "big power" (1000+hp) territory. Getting good driveability sometimes requires that one doesn't blindly put gigantic injectors in the engine if they're not needed, even if you have 100% perfect data for those gargantuan injectors.

[DSteck]

Nothing ill intended. I've seen you mention that the data isn't 100% accurate. You know the short pulse is off. It's just the nature of converting from low/high slope format to GM's short pulse adder setup. I'm not shooting the messenger by any means. I understand exactly how you did it, and consequently understand where error comes from.

As far as the short pulse, I don't have anything saved, but I can vouch that the 80s are not fun to use (because the injector itself just isnt that great). The 60s are close, but the idle area never seems proper given trims everywhere else when using the MAF. The VE helps to wash that out though. The saving grace is that they don't dip far into the short pulse table. They just don't work as well as factory from my experience. Do I still suggest them to bridge the gap between factory and ID850/1000s? You bet. Unless a person doesn't care about saving money.

I dedicate the vast majority of my time to drivability. I can appreciate what's involved with making it work. That said, a gargantuan injector can offer pretty stellar drivability anyway. Plus there's the fun of playing with EOIT since the entire shot can be delivered in an extremely short time period (my buddy's 441 with 850s was only 37% IDC on gasoline so we pushed EOIT around I squeeze every ounce out of it at WOT).





The point remains that a lot of people are flat out afraid to use the LS9 injector, and likely will never try it. I was skeptical because of the design, but tried them. After they worked fine on a couple cars, I stopped even thinking about the issue and moved on.

[pontisteve]

LOL, I knew Greg would chime in on this conversation. And seeing as how I sat thru an injector seminar that Paul and Greg co-hosted at PRI this year, I know them two are definitely on speed dial. By the way Greg, I was hoping that seminar would find it's way onto social media but haven't seen it yet.

As for the inherent flaw in converting from Ford to GM, we're really talking about short pulse adder vs low slope, right? So the issue at hand is a curve with a whole bunch of data points, versus a straight line average of that curve, right? I would have to think that this shouldn't be any more of any issue than every Ford would have on a factory car, right?

Greg, I did notice (if memory serves correct) that you had the Min PW on the 60's down to .8 ms, and that the Continental data says "Minimum Linear PW 1.5 ms". If I've got that right, I figured you were fudging the Min PW down a bit, trading dead nuts consistency in for the capability of 1/2 the fueling. The Ford data says 1.391.

While I'm at it, I might mention to Greg that we have a real shortage of fuel injectors in the Ford camp now. Ford Racing discontinued the EV1's and EV6's, leaving us only with their newer lineup that was designed for dual intake valve engines, and therefore have dual cone spray patterns. Maybe the world is changing, and we need to find a way to stick some of these GM injectors into a Ford! Is it time to add a new sheet to the Injector Rosetta Stone?

[DSteck]

As for the inherent flaw in converting from Ford to GM, we're really talking about short pulse adder vs low slope, right? So the issue at hand is a curve with a whole bunch of data points, versus a straight line average of that curve, right? I would have to think that this shouldn't be any more of any issue than every Ford would have on a factory car, right?

That, plus taking a set of data, generalizing it by coefficients and multipliers, then converting that generalization to a totally different format.

[Michael_D]

I've been surfing the glorious interweb, trying to find out exactly what this "different spray angle" is for LS9 injectors. This thread is about as 'iinformative' as my mad google skills have found.

I keep seeing this statement: "They have a different spray angle because they sit on the intake at a different angle"..... What the hell does that mean? Is it the spray cone angle that is different, or do the LS9 injectors actually spray the fuel at an ange that is not parellel to the injector body?? If they are actually 'angled', how does a person clock them securely in the intake?

I have to get a freind some bigger injectors for is LS7 that I've been dialing in. I'm sick of worrying about incorrect injector data, and he can't afford ID's....

This is the last WOT run.....4th gear, IAT is 50 F. Sea level.

Duty Cycle Screen Shot.png

[cavy.417]

They spray at like a 13* angle iirc, to hit the back of the valve stem on the lsa engine. This is because of the positioning of the injector in the intake due to the space constraints with the blower. So instead of a normal injector spraying directly down. These spray at an angle. I've been running these in my n/a 416 since it was new. Going on about 3500 miles now and running fine.

[pontisteve]

I've been surfing the glorious interweb, trying to find out exactly what this "different spray angle" is for LS9 injectors. This thread is about as 'iinformative' as my mad google skills have found.

I keep seeing this statement: "They have a different spray angle because they sit on the intake at a different angle"..... What the hell does that mean? Is it the spray cone angle that is different, or do the LS9 injectors actually spray the fuel at an ange that is not parellel to the injector body?? If they are actually 'angled', how does a person clock them securely in the intake?

I have to get a freind some bigger injectors for is LS7 that I've been dialing in. I'm sick of worrying about incorrect injector data, and he can't afford ID's....

This is the last WOT run.....4th gear, IAT is 50 F. Sea level.

Duty Cycle Screen Shot.png

I think they call that "bent angle". The injector sprays out at an angle to compensate for the injector itself being at an unusual angle. Google some Bosch white papers on their injectors, and you'll find more info about the various Bosch terms. So if you take this offset spray angle and put it in a traditional intake, it ends up spraying half on the valve and half on the port walls. This screws up transient fueling and atomization.

in addition to bent angle, there is also the cone angle, or how wide the injector spray cone is, in degrees. The bigger the intake valve, the bigger the spray cone angle.

Any time I start hearing about how an injector is not really made for what I'm trying to do, I just move on to a different injector that is made for it. My personal rule is to use the Siemens 60's, and anything bigger gets ID's.

[Michael_D]

Thanks. I'll look for the specifics from Bosch. I have not had my hands on an LS9. How are the injectors clocked into position? If this spray pattern is 'bent', then there must be some way to get the injector positioned correctly, or the spray could end up pointing the wrong direction, no??

[pontisteve]

The bent angle is to compensate for the injector not being able to be installed at the correct angle on that car. So when you put that same injector in a regular car with a straight intake port, the thing sprays sideways. Short of welding an injector bung onto an intake at the same wrong angle, I don't think there's a good way to fix that. It's just a poor choice for a normal engine. I could be wrong. I'm not there physically checking the injector specs, and looking at the LS9 intake on a work bench. I'm just pointing out the why behind the concern. Now you can go verify this info yourself.