[roybfr]

I installed Helix 860's, I set the original scaler to 3932 based on the formula. Then based on when Cristian and others had posted, which was adjusting the value till the sum of the trims (correct/learned) was = +/-5%. At 3622 I got it to be pretty damn close to 0%, the learned would end up being .39, I even did a few ecu resets and that is where it came back to.

So I then moved on to the latency values. When I did this I turned on live tracing and noticed it was staying at 6 volts. I upped this to 8.96 and it seemed to stablize the correction closer to zero, it still bounced around between say -3% to +3% (it's been a long night so this could be slightly off). So I decided to up 9, 12, 14 adn 16 volts as well. When I did this my A/F dropped from around 14.7-15 down to 10.5 so I reset those 4. I went for a drive and the car seemed to run ok, no stumbling, thought I did not get in it much so I was out of boost most of the time. Even after the run the learned was close to where it started, maybe up to 1.18. So when I got home I let it idle a little more and did a few resets to make sure the correction + learned was sticking close to 0% and it was. I decided to make a few adjustments again on the 9 & 12 and bumped them up. I eded up taking the 6 from 7.66 to 8.96, the 9 from 2.69 to 2.79 and the 12 from 1.37 to 1.47. I left the 14 and 16v settings a long because any change and stuff got wacky.

First I guess I am asking if this seems right. From what I have read latency is trial and error and even those with the same injectors don't always have the same values. Next when I was running live trace it stayed on 6v even during my drive, but when I changed any of the higher voltage settings things got goofy. So does this mean that live trace does not work in this map? I know values scale between 2 values but one time I left 9, 12 and 14 a lone and just changed 16 and the trims changed with it.

Any help would be appreciated. Also I am wondering if I should bother doing my intake calibration now or after I install my turbo/intake (perrin rotated gt30r). I had the understanding that I should not do that until I put the other intake on but read something about doing it before and after which seemed a little redundant if not wasteful the first time around if making the change.

[R4ND0M_AX3]

Make sure battery voltage is displayed on the dashboard. Sounds like your trace isn't working. I don't recall if mine did. But I just look at the actual battery voltage and adjust based on that.

[WolfPlayer]

Two things here.

1) I will be evaluating a set of Helix injectors very soon. Currently, they are off at WitchHunter getting the full workup (flow AND latency testing ). Woohoo.

2) That Scalar looks off pretty good.

520cc/850cc * 5372 = 3286

The current scalar you are using (3622) is actually closer to what you would be using if you were running JECS injectors. But, for PEs or Helix you should be using something a couple hundred lower. I am betting that your latency values are too low and that you need to lower the scalar and raise the latency values.

No matter what, I'll report back with my final Helix results. I plan to install and tune the Helix injectors during the 2nd full week of June (wife is taking the kids and going away for the week!). I am only installing the Helix injectors because I want to evaluate them. I don't plan to keep them installed. IMHO, the JECS/Nismo injectors are the best injector for the STI. Also, please note that the stock injectors are manufactured by JECS

t

[roybfr]

I actually started with 3286 but in order to get he trims with in +/- 5% I had to raise it 3622. I will try it again tonigth at 3286 and put the battery voltage up so I can see what cell it pulling from. I did look at you spread sheet with the stock and nismo injectors. It's interesting that the latency values for the Nismo are less then the stock. I only say this as I had read aftermarket ones usually had higer values then stocks, but this could just be the JECS since they make the stockers as well.

Do you think 720cc is enough to support a GT30R?

[WolfPlayer]

Originally Posted by roybfr

I actually started with 3286 but in order to get he trims with in +/- 5% I had to raise it 3622. I will try it again tonigth at 3286 and put the battery voltage up so I can see what cell it pulling from. I did look at you spread sheet with the stock and nismo injectors. It's interesting that the latency values for the Nismo are less then the stock. I only say this as I had read aftermarket ones usually had higer values then stocks, but this could just be the JECS since they make the stockers as well.

Do you think 720cc is enough to support a GT30R?

Yes, definitely try lowering your scalar and raising your latency. I think R4ND0M_AX3 was running a 31XX scalar for his PE 850s. Yours will be similar ... somewhere in the 3100-3200 region.

As far as the GT30R is concerned, I would ask what amount of horsepower you want to push with that setup? The JECS can support 400whp with an acceptable duty cycle. If you want to run higher than I would go with the PE or Helix injectors simply because there is no other choice.

t

[R4ND0M_AX3]

I really need to get off my ass and finish tweaking that.

BTW, I thought we were calculating with a stock injector size of 480 or 490 and not 520.

[roybfr]

Well I changed it back to 3286. Put the latencys back to stock. Ended up with 7.76, 2.77, 1.57, 1.08 and .75. Will be interesting to see what you get back from witch on the Helix's. I am curious how often the 6, 9 and 16v areas get accessed as my voltage pretty much stayed around 13.75 so changing the 12 and 14v values made the biggest difference. The car starts and idles pretty well. I did some logging on some wot runs and things seemed to be ok but I was run rich compared to when the stockers were in, which I guess should be obvious. I went from 10.8-11.2 on the stockers at wot to 10.4-10.6, I ad never messed with the fuel so I guess I was a little on the rich side to begin with. Well hopfully this is enough on the injectors, going to start taking things apart and installing my perrin roated kit. Thanks for the help.

[R4ND0M_AX3]

now is the time to get this right before you change the intake. Otherwise you will drive yourself nuts fine tuning it.

[roybfr]

I am really not sure what more to do to get it right. The learned usually settles at idle around .39 and the correction is bouncing between -1.19 and 1.19, if not a smaller range. Based on what I read the goal is to get as close to 0 as possible and that is as close as I can get. Am I missing anything.

[WolfPlayer]

I know what to do First of all, as long as your are -5 to +5 under all situations (environment, load, etc) then you are AOK. Yes, shoot for 0. However, even when you get to 0 fuel trims the environmental conditions will change and your fuel trims will change. Thus, -5 to +5 under ALL conditions is fine.

If you are running rich at WOT compared to before then you need to lower your scalar more. Lowering the scalar tells the ECU that you have bigger injectors. Thus, the ECU tightens (lowers) the pulse width. This will lean you out at WOT. Furthermore, when you lower the scalar you have to raise the injector latency in order to keep the same idle fuel trims. Long story short, LOWER your scalar and raise your latency (12-14V). I'm fairly sure that you want to be in the 3100-3200 region (as I said before ).

t

[roybfr]

Well just so happens I drove my car to work and have the LT so I will give it a try today. When I am driving around the correction trim may go above or below the 5+/- but the learned is usally at .39 but may go as high as 2.75, then falls back down. Looking back at my logs pre install the learned was in the same range. Unfortunatly I did not log the correction value before so I don't know how much it bounced around. I assume that the correction is always going to jump a little but the goal is get the learned stable under thorttle which compared to my pre install logs it looks to be.

[roybfr]

Well I tried 3120 and other values bewteen 3100 and 3200 and could not get it idle with the correction and learned to be at +/-5%. The learned started out at .39 with the correction at 25, then the two met in the middle still totaling around 20-30, this was with the stock latency values back in place. So I went back to what I had last night, the car just seems to run better that way. I may recet the latency and go a little lower so that when I am WOT my AFR is closer to what it was with the old injectors.

It will be interesting to see what the Helix's flow when you get them back, especially since the PE guy was using lower scaler.

[WolfPlayer]

I don't understand what you are doing. If you lower the scalar as we have indicated then you need to raise your latency values. If you just lower your scalar then OF COURSE the car won't idle. You need to work the 2 values together. A decrease in scalar requires an increase in latency. An increase in the scalar requires a decrease in latency. Simple as that.

t

[roybfr]

At 3120 the car idles fine.

The instructions state to start with the calculated scaler, so in my case 3120, but then to adjust the scaler till the sum of the learned and correction trims is +/-5%. So when I set the scaler to 3120 the sum of my trims is 25 to 30 thus accorind to how I read the instructions I need to adjust the scaler. In order for me to get the sum of the trims to be +/- 5% I have to move the scaler up to 3268. I see what your saying but based on the following from Cobb

"For another example, let’s say you are replacing your factory 2004+ USDM STi injectors (525cc) for aftermarket 850cc injectors. The formula would look like:
New Scale Value = (500cc / 850cc) * 5372;
New Scale Value = 3160
Input this value as a starting point. You will likely need to tune from there. To fine tune the scale value if you are running a stock intake system, you can display the A/F Learning and A/F Correction values with the Dashboard.
With the engine idling at full temperature, make adjustments to the scale value until the A/F Correction + A/F Learning are as close to zero as possible, +/- 5% is generally acceptable. The closer you can get to 0% is ideal, but
do not begin tuning any other fuel tables until you at least have it within the +/- 5% window of total correction."

I would think what I am doing is right according to that, but there may just be two ways to get at it or I am just reading it wrong.

[roybfr]

Ok so I go to cobbs forumns and read this

"To put it another way, if almost all of your AFRs are consistantly 0.5:1 AFR leaner than your fuel targets, you need to decrease the size of your injectors. If you match all your fuel targets except one or two where you are 0.5:1 AFR leaner, then you need to increase the Airflow for the Mass Airflow Calibration for the MAF Voltages corresponding to those load points."

Which goes a long some what with what you are saying as my guess is that my afrs are off the same acrosst the board. This is a little frustrating as the information is kind of contraditory in what to do. Once says to adjust the scaler till you get to +/-5, the other seems to indicate setting at the calculated scaler, then set the latency to get the trims down, then fine tune with the intake calibration. I may call cobb tomorrow, I know they don't support tuing help but I would think they would at least answer the question as to what is at least the right method.