E38 Injector Hard Limit
 

Hi all,

I'm trying to install ID1000 injectors into my 2008 LS3 (E38 pcm), and from what I can tell there is a 63.5 Lb/hr flow limit in the stock tables. I have noticed in other GM PCMs this limit goes up to 235lb/hr or so. It looks like the common approach is to scale all of the rest of the tables by 50% so you can theoretically install up to a 127lb/hr injector (DSX Tuning has a great article on this).

My question is two fold:
1. Most of the threads pertaining to this issue are pretty old, are there any OS patches that correct this hardlimit?
2. Are there any major disadvantages to scaling all of the tables as outlined in the DSX document?

Thanks everyone.

I had the same issue with my 08... ended up scaling the tune.

Is it swapping in a 2009 PCM a relatively easy process? I don't think I would exceed the 1.36g/cyl flow limit, I need the bigger injectors due to E85, but I think my air flow would be below the 2009 axis limit.

just modify the STOICH to "scale" the tune. No need to swap to an 09 pcm. I've tuned vehicles with 210 lb injectors with the 63.5 lb/hr limit

Wouldn't modifying the stoich table have run on effects in close loop operation?

no , o2 sensors read oxygen not fuel. They just want to target lambda.

Yea I was more thinking about the proportional and integral control terms of the closed loop controller. I'm wondering if scaling the target AFR would cause the error term to react twice as much.

Anyway, so you basically scaled the target AFR for normal operation, Power Enrichment, and Idle by 50% and then wrote your injector flow rate in at 50%?

Seems to work well for you? Do you run a SD tune or are you using the MAF?

closed loop will work normally, because the o2s read oxygen not fuel. Scaling the target AFR will not cause any errors. I do this all the time, I tune for a living.

with this stoich trick, you dont have to scale anything. You input the normal PE values that you desire. Lets say your injectors flow 127 lb/hr.. but your limit is half that. So now you have to put 63.5 lb/hr in your IFR table. That means that you now have to scale your airflow model by 50% decrease. Now the airflow model is jacked up by half of the actual values. Well.. you can command twice your fuel's stoich value, and you will "correct" the airflow model. This is why you should tune using lambda, it doesn't care what stoich AFR you use. It makes the fueling error calculation super easy when you use funky stoich values. You will have to halve the IVT values under the fuel tab if you double the stoich.

Subeone, first of all, thank you very much for your input, I really appreciate the help.

I've thought about what you said, and I'm starting to wrap my head around it. Let's work with a value of 50%. So the IFR table get scaled by 50% and therefor my ID1000 injectors (95.2lb/hr) would get entered as 47.6lb/hr. Now, if I just left things alone at this point, the car would run so pig rich it probably wouldn't run at all, it would be injecting twice as much fuel as required. So at this point, you tell the PCM that stoich is actually twice as lean as it really is. As you correctly stated, the lambda sensors work in EQ ratio / lambda, so provided the PCM now sends half the fuel, they're going read lambda 1 and send an all good signal to the PCM. Everything is happy.

Originally I thought this would screw up anything in the PCM that works with EQ ratio, but while I was typing this I thought it through. Let's say you're PE table is targeting EQ ratio of 1.2, converted to AFR (without a scaled stoich table) you would be targeting 12.25 AFR. With the scaled stoich table an EQ ratio of 1.2 would be targeting 24.5 AFR, and half of this places you right back at 12.25 AFR, so it should all work flawlessly. But then this leads me to wonder why you would have to make any changes to the IVT Gain table (which just multiplies EQ ratio). Could you explain this?

I guess even for idle where you DEFINE the target airflow, everything should still sort it self, because the PCM will just "spray half the fuel" while its targeting the modified stoich (even though it will be spraying double because of the injector scaling).

This post was definitely longer than necessary, but I needed to type it out in order to fully understand what possible run on effects this could cause. Now that I've thought about it, this should work really well, and if so, MAN the DSX method WAYYY over complicates things. I'm also very happy not be messing up the airflow model as I'm sure there are parameters of the model that I cannot scale with HPT.

So in short,
IFR divide by 2
Stoich table multiply by 2
Why do I need to touch the IVT gain table?

[ame]https://youtu.be/16mrQV1GGDw[/ame] sorry for the audio.. gets jacked up in some spots. Essentially you are 100% on board with what I am saying.

WOW, that's amazing thank you so much for putting that video together. I admit that I was looking at the wrong IVT table, the one you pointed out references AFR so yes that makes perfect sense.

However, you threw me off with reference to the spark table. The main high octane spark table has two inputs, Engine Speed & Airflow (Load). No where does it reference fueling, now of course if you mess with the airflow model, then you will have to redo the spark table. If we are changing the stoich definition I don't see how that would effect any of the inputs to the spark calculation. There are some fueling modifications to the spark timing, but I believe they reference EQ ratio, so it shouldn't need any modification. Thoughts?

Additionally, as I was trying to work this through I drafted out a flow chart to wrap my head around it. Maybe it will be useful to someone (hopefully it's correct)
https://cimg4.ibsrv.net/gimg/ls1tech...cfac2f8bc0.jpg
, I have attached it here.

I'm pretty familiar with engine dynamics so to speak, I'm a tuner in another realm lol. I've never used HP Tuners or done any tuning on a general motors platform, so I'm still learning. As I slog through it all I'm compiling a big document of basics that I hope to publish to the LS1 Tech community as a getting started document.

Looking at the chart above, it's definitely missing a few things, and I should probably add the spark calculation into the chart. For anyone looking at this forum topic please feel free to point out any errors in my logic, when I post the final documents I would love it if they were accurate :p

If you have DIC readouts for MPG, the double stoich/half IVT/half IFR method will make them inaccurate. Depending on where you read about it online, you need to take the fuel density table (if you have it, it will be on the engine>fuel>general tab beneath the stoich) by .5 or .6. to get you into the ballpark. The correct method would be to calculate you mpg over time and make adjustments over time until it is accurate.

I didn't have the fuel density table for an E67. I requested it and they added it for me.

Smart man! I didn't think about that. I feel like there are a ton of run of effects with this type of thing, just trying to cover all the bases.

I feel like so much of this will become obvious once the weather is nice and I actually try this in the car. Yes, this if for a 2008 C6 so it has the read out. HP Tuners has the fuel density table by default for my car.

Thanks for the tip!

There are always more than one way to achieve the same end result. The best advice I can give is learn to tune in lambda and not in afr no matter what method your going to use. Theres always pro and cons to each method, scaling, changing stoich, etc just take your time and pay attention to the data in the logs and use whatever method you understand the most. Just remember with tuning there is always a global effect, very rarely are you going to change a table and it not have an effect on another one.

100% agree with both of your points. Tuning in lambda is a must especially given that I will be running flex-fuel in the car. It would be nearly impossible to accurately use AFR, although AFR is what I grew up tuning with.

Same thing with the global effects of calibration, I agree that there are always run on effects. I'm really trying to wrap my head around each method and decide which method has the most run on effects. It's always great to here from experienced members who have actually put each method into practice, its really hard to think through all the effects in your head.

Not much driving to be done in snowy Michigan right now.

stoich trick is the most consistent and easiest BY FAR. You have to remember that there are still some tables that rely on air/fuel mass calculations that are not defined by hptuners, and the DSTECK/scaling method requires to address all the tables that have mass units i.e. lbm, lb/hr, g/sec, etc. Some tables are not defined, and thus your scaling method will not be 100% correct. You can however find the missing tables with tunerpro, winOLS, etc. then use the .xdf and the address to define the tables, but that is a ton of work. You would need a .bin most of the time too, to explore the rom.

You forgot to post the tune