[MalcolmV8]

So when I had my ID1000 injectors I tweaked injector data values to correct A/F and then made final minor touches to MAF curve. The car drove great. Here's how my MAF looked. Blue line showing stock curve for my car and green been my curve.



Only problem is I was running out of injector so I got some ID1700s. So initially I scaled the injector values by the same amount as I've had to scale my MAF as a baseline starting point. It was WAY off. Car was pig rich. So I proceeded to pull air from the MAF curve to get a baseline of how off it was. A lot.

Here's how my new MAF curve looked



I decided to plot it out and see the differences. As you can see in the pic below the bottom part of the curve was cut almost in half. 90% on average of what it was before. Presumably that's the Low slope. The top half I had to cut down by 50% on average, presumably the High slope.



So to make that setup work I adjusted my "Engine Size" to bring load settings normal which would be idling around 20 ~ 25% and that put timing tables where they should be on driving and WOT pulls. That part seemed fine. Of course since the MAF curve was cut down so much the LWFM table is completely off so hammering on the gas rapidly especially taking off aggressively from a stop caused the car to go pig rich and buck and miss.
Also deceleration was strange. It appeared to be shutting off the gas now and going completely lean and the exhaust sounds like crap as well as the car having an odd minor surging in those conditions. My tune is not set to shut off fuel on decel.

So I proceeded to put the MAF curve back to its prior scaled and calculated value. It's an OEM GT500 sensor in a 127 mm housing so it's easy to do the math of what it should be.

The problem is the ID1700 injector values exceed the maximum values the ECU will accept. So both of these are maxed and not where they should be



I figured I could get around it by adjusting the low and high slope multipliers since those values are maxed out but that does not appear to be true. It seems once those values are maxed that's it you're done. Regardless of what the multipliers get set to. Is that correct?

As you can see here I multiplied the multiplier table out many times over, 10% at a time, and could not change the fueling of the car. It was un responsive. It was just stuck barely idling and missing around 11 A/F. I attempted both high and low slope multipliers with no luck.



So I guess I'm trying to make a decision as to the best way to tackle this. What have others done? It looks like I may have to live with a skewed and reduced MAF curve and then adjust LWFM and possibly other things to compensate for it.

[decipha]

how far back did you scale it?

what did the low slope come up to?

[MalcolmV8]

how far back did you scale it?

what did the low slope come up to?

On the left is what the OEM GT500 slot sensor calculates out to for a 127 mm MAF housing. My ECU/strategy (FBGI0) has a 63.9 hard limit so I had to scale by 0.627 to make it fit.



I got the ID1700 injector data directly from ID's website. This is what it was



I also updated all the other multiplier tables with the data exactly as ID supplies it. I then scaled my High and Low slopes as well as breakpoint by the same 0.627 to get this



Fired up the car and it was pig rich. Knowing I had injector hard limits I first attempted to remove air from the MAF curve as mentioned in the first post but found I had to remove so much. That's when I went back to the correct MAF curve and adjusted injector values only to find I'd hit the hard limits and the multiplier tables would not have any affect at that point.

[MalcolmV8]

This btw is the raw data from ID for the 1700s. I had to convert a few values as you can see.

[decipha]

what did the low slope come up to?

[MalcolmV8]

what did the low slope come up to?

Perhaps I don't understand what you're asking for? Was the 76.596 not what you're asking for?

[LeadHead]

He's saying your high/low slopes haven't been properly calculated. Injector Dynamics ow slope information for the ID1700s is basically useless

[MalcolmV8]

He's saying your high/low slopes haven't been properly calculated. Injector Dynamics ow slope information for the ID1700s is basically useless

I see. I would agree based on the results I saw. However I can't adjust the low slope to a value that puts the car at stoich at idle. I reach the maximum value for low slope allowed by my ECU/strategy. So I'm looking for some "tricks" or work arounds.

[LeadHead]

Well you can scale your MAF and Injector size as you tried earlier, but you can't use the supplied data from Injector Dynamics to do it. They even state on their website that these injectors on pump-gas won't be linear at all down in the idle/cruise range. They say it *should* be fine on E-85, but a small engine like a Modular could still well in the non-linear range at idle, even on E-85.

You can do some extrapolation based on the ID1000. The ID1000's low slope is about 16% higher than the high slope, so applying that to the ID1700s, you're probably looking at a low slope in the 140-150 lb/hr range, maybe even higher.

[MalcolmV8]

Well you can scale your MAF and Injector size as you tried earlier, but you can't use the supplied data from Injector Dynamics to do it. They even state on their website that these injectors on pump-gas won't be linear at all down in the idle/cruise range. They say it *should* be fine on E-85, but a small engine like a Modular could still well in the non-linear range at idle, even on E-85.

You can do some extrapolation based on the ID1000. The ID1000's low slope is about 16% higher than the high slope, so applying that to the ID1700s, you're probably looking at a low slope in the 140-150 lb/hr range, maybe even higher.

They idle perfectly fine on E85 on my car. So that doesn't seem to be an issue. The issue is I had pull air from my MAF to make the ECU do what I wanted which is the wrong way to go about it.

As you mentioned 140-150 is a possible range. However I cannot enter anything bigger than 112.5. So instead I pull air from the MAF curve. Unless I can think of a better way that may end up been my solution. Then I'll have to fix a few things like LWFM table and everything else affected by skewing the MAF curve.

[decipha]

what i was getting at was that you hadnt scaled back far enough

Scale it back more so that the injector slopes are below 60

[MalcolmV8]

what i was getting at was that you hadnt scaled back far enough

Scale it back more so that the injector slopes are below 60

When I scale though the MAF and injectors scale in sync so the A/F stays the same. So it doesn't seem to matter if I scale 10% or 70% the results are the same. I end up having to pull air from the MAF curve to lean it out and end up with a skewed MAF curve that claims very little air going into the engine.

[cgrey8]

Is that only at idle or all across the curve?

If only at idle then you likely need to reduce injector offset.

[MalcolmV8]

Is that only at idle or all across the curve?

If only at idle then you likely need to reduce injector offset.

Entire curve.

[decipha]

what i was getting at was that you hadnt scaled back far enough

Scale it back more so that the injector slopes are below 60

When I scale though the MAF and injectors scale in sync so the A/F stays the same. So it doesn't seem to matter if I scale 10% or 70% the results are the same. I end up having to pull air from the MAF curve to lean it out and end up with a skewed MAF curve that claims very little air going into the engine.

that is incorrect, the 03/04 cobra ecu has a 27mhz clock, it only accepts a max slope value of 60, although you can put a higher value in there the ecu cannot calculate it

scale it back farther so the slopes are below 60 and post back with your results, no doubt you'll have to adjust it but it should be closer from the jump

[cgrey8]

...the 03/04 cobra ecu...only accepts a max slope value of 60, although you can put a higher value in there the ecu cannot calculate it...

Why is that?

I know the GUFx stuff has a limit of 112.5 that can be entered for AHISL and ALOSL. But I thought the newer strats that could accept a larger value were actually able to use that value. So is this 60lb/hr limit only an 03/04 strategy limitation or a range of strats that have this limitation?

Or is this a definition problem? My guess is it isn't a definition issue.

[MalcolmV8]

what i was getting at was that you hadnt scaled back far enough

Scale it back more so that the injector slopes are below 60

When I scale though the MAF and injectors scale in sync so the A/F stays the same. So it doesn't seem to matter if I scale 10% or 70% the results are the same. I end up having to pull air from the MAF curve to lean it out and end up with a skewed MAF curve that claims very little air going into the engine.

that is incorrect, the 03/04 cobra ecu has a 27mhz clock, it only accepts a max slope value of 60, although you can put a higher value in there the ecu cannot calculate it

scale it back farther so the slopes are below 60 and post back with your results, no doubt you'll have to adjust it but it should be closer from the jump

I've never heard of 60 been the limit. I mean I was able to put in ID1000 values (98 lb injectors) in without any scaling and it worked. As shown in my first post the MAF curve was almost perfect right off the bat. I've always heard a 108 lb injector was as large as the 03/04 Cobra ECUs would accept.

I'm willing to give it a shot. If I scale by 0.48 I'd have a high slope of 60.095 and low slope of 58.636 and my MAF curve would peak at 48.9 instead of 63.9. Right now my MAF curve peaks at 45 so almost identical (3.9 lb/min difference) which brings me to my prior statement. No matter how much you scale they stay in sync because you're scaling both MAF and injectors together. I've tried this back when I had my ID1000s. I scaled (for the MAF, not the injectors) by 0.7 or so all the way to 0.3 and A/F any many ratios in between and always got the exact same results and A/F and always had to adjust my MAF but the same amount after the scaling to correct A/F.

Anyways, I will try scaling to where my injector slopes are 60 or below tomorrow and report back.

[MalcolmV8]

OK I didn't have much time today but I got the car out quick to do a little playing around. With it warmed up and idling in driveway I started changing the low slope. Anything from 10 all the way to 100 made no difference. I then realized my car is idling on high slope and only using high slope.

My high slope was currently set to 78. So I tried changing it to all random numbers up to 100 and then down to 60 with no changes at all. I then went below 60 to 50 and suddenly my idle A/F dropped down to around 13. So Decipha is exactly right. A number greater than 60 has no affect. Thinking back this certainly explains some weird anomalies I've experienced in the past.

So to see if it was simply my break point making me idle on high slope I put my high slope at 60 so car was idling normal. I set low slope to 30 so I'd see a drastic difference if the car switched to low slope. I cut breakpoint number down in half and sure enough it changed to low slope and car went pig rich.

This leads me to a follow up question. At what point should the injectors switch from low slope to high slope? I don't fully understand the difference between the two and when the car should transition.

I'm going to attempt to scale down the injectors and MAF as previously suggested by Decipha to a point where injector values are below 60 and see what sort of results I can get.

[MalcolmV8]

Another thought is base fuel pressure. I'm running 50 PSI. Should I go back to 39 PSI base to make the 1700s a little smaller? I've always been on the 39 base but had recently switched to a 50 base when my ID1000s were running out but ended up maxing them even with a 50 base. I have plenty of head room to drop back to 39 base if the ID1700s will spray properly like that.

[LeadHead]

The breakpoint is the point where the injector stops responding linearly to pulsewidth. Below the breakpoint, the "mechanical-ness" of the injector starts becoming an issue.

It takes a certain amount of time for the solenoid to move, for it to open the valve, for it to release, for the valve to fully close, etc... This "mechanical-ness" in effect causes the injector to deliver more fuel than expected given the pulsewidth. Above the breakpoint, the time it takes for these events to happen is so short relative to the pulsewidth, it doesn't matter. Below the breakpoint, they start becoming a significant issue. Ford fixed this issue by creating the high/low slope. The "high slope" is the point where the injector is operating linearly, where twice the pulsewidth delivers twice the fuel. The low slope approximates the non-linear operation by assuming the injector flows more fuel than its specified to below a certain pulsewidth.

To find the high/low slope, you need to get your fuel mostly dialed in so your LAMBSE matches your AFR under a heavy load and lots of throttle. Then, you need to do a low throttle high RPM run, followed by a high throttle low RPM run. The goal is to achieve the same MAF flow (and thus fueling requirements). So say 5% throttle 6,000 RPM and 30% throttle 2000 RPM both yield 300 kg/hr MAF flow. If your low slope is too low, you should see the 5% TPS/6,000 RPM fuel error to be significantly higher than the 30% TPS 2000 RPM run.

[MalcolmV8]

OK that makes sense. I looked at the stock tune for my car and even the ID 1000 values and the low slope is larger which would make sense per your explanation. However interestingly enough the ID1700s are just the opposite. Their low slope acts as a smaller injector than the high slope. So they actually deliver more fuel as they get to opening and closing faster. Quite interesting.

[decipha]

its not uncommon for some larger injectors to actually spray less fuel == lean out when you give them MORE PW

from a logical stand point it doesn't make sense, why would having the injector open longer spray less fuel? well its cuz of the shitty drooling you get from them injectors down low

might as well drop it back to 39 psi base

[MalcolmV8]

OK set base pressure back to 39 PSI.

I scaled the tune down enough that my injector slopes were below 60. 59.469 high slope and 58.025 low slope. Break point was 0.00000950. My MAF curve only went up to 48.41.

First thing I noticed was the car was idling on the high slope. I could tell by adjusting the slopes while idling and high slope affected the idle, not low. So I started multiplying out the low slope by 10% at a time till it took affect at idle then 20% more. I had to multiply it all the way to 0.00002027 just for idle to kick over to low slope. 20% more and I left it at 0.00002432

As expected I had to tweak the MAF curve to bring A/F in check. Not as bad as I thought though. The car actually drives a lot better than before.

So "Scaled" is what my MAF scaled out to in theory. Actual is what it ended up been adjusted to. % difference is what percentage I had to drop the MAF to get AF in check. It's not perfect, just had a little bit of time to mess with it but its fairly close. I had to pull more than double from down low than up top. Presumably injector data been off.



Engine size or displacement came to a scaled value of 15.43 (32.49 stock value). I ended up with 14.3 to keep load in check when making WOT pulls. Otherwise it was only hitting 1.3, now it goes all the way up to 2.0 and correct timing tables are used. It also put the LWFM about in the right spot so tip in throttle was much smoother.

Manifold volume scaled from 11.875 to 5.64 but I'm thinking since I had to pull additional volume from the MAF curve perhaps pulling some more from it will help put transients back in place.

Still lots of work but it feels like progress since car drives way better than it was before.

[decipha]

you should never let LOAD hit 200, that gives you no room to grow or for colder nights, kick up the sarchg (engine size) so load stays below 200 and adjust accordingly

[MalcolmV8]

OK I can do that. Reviewing the log it would appear load was hitting 2.0 at 5000 RPM. That's under boost of course deep into the throttle. Just for my understanding what's wrong with that? What's different on a colder night with regards to load? I suppose air would be denser and I'd have a little more in the engine. Thanks.

[cgrey8]

...Just for my understanding what's wrong with that? What's different on a colder night with regards to load? I suppose air would be denser and I'd have a little more in the engine...

Without additional mods, that's the only difference.

As I understand it, you are maxing the EEC out at 200% thus there is additional engine load capacity above 200% that is not being represented and instead is being reported as 200%. My comments below assume this...

From a proper tuning standpoint, he's right. You want to leave headroom to never knowingly hit 200% thus leaving room to hopefully never hit it, but if it happens, it only unknowingly barely hits/exceeds 200%.

But from a practical standpoint, it's a question of how engine management should change as actual engine load moves into this unrepresentable area. So lets say you hit 200% but the engine was capable of, say, 230%. The question is do you expect, say spark advance, to change as actual engine load goes from 200% towards its actual WOT? Or would you be commanding the same spark value at what's registered as 200% and WOT?

If you DO NOT intend/expect the engine to be controlled differently (i.e. different spark advance, different fuel enrichment, different cam timings, etc) as the engine progresses through this unrepresentable load area above 200%, then I don't see that there's a practical reason to bother changing anything if the engine otherwise runs satisfactorily and as you expect.

But that's just me typing out loud without experience with an engine capable of what your is or experience with your strategy.

[decipha]

you cannot dial in pct_load / totldst at 200

[cgrey8]

you cannot dial in pct_load / totldst at 200

You made a statement as though the explanation was self-evident. But it wasn't, at least not to me.

[decipha]

sry

u no love me no more?

[sailorbob]

Why is that?

The injector size limit is a result of the conversion to alosl_tick and ahisl_tick. All eec-v ecu's are affected, the slower ecu's clock speed the lower the injector size is before the a**sl_tick values become maxed out.