Hi there, new member here. I recently purchased a tuner version of Osiris for my 2006 Nissan Sentra SE-R Spec V, with the determination to street tune the vehicle before taking it to be dyno tuned in the spring. I have recently installed a new turbo, and switched from using an Apexi AFC NEO to the Osiris software, and have encountered a few problems with learning the tuning ropes.

First, a little information on my vehicle, with the hopes that this will help some of you answer my questions. I'm new to tuning, but understand the basics, and am just wondering if I'm doing this out of sequence which is resulting in failed attempts, or if I don't fully have a grasp on the software yet.

The car - '06 Sentra Spec V (mods you should know)
Precision PTE5857 turbo (wastegate spring at 7.5lbs)
Stock MAF
Subaru WRX 440cc blue injectors (410cc @ 4x psi, and 440cc at 51psi (stock SpecV fuel pressure)
Modded in-tank Walbro 255lph fuel pump
3inch intercooler piping on the cold side
3inch exhaust

Car was tuned on a NEO before switching to Up-Rev, and the PTE5857, so I have no even had a chance to make a part throttle pull, let alone a full throttle pull. I have been driving the car on a stock-ish base map, with increased rev-limiter, and no corrections to the fuel or maf table, currently.

I have read countless hours of threads from this site, and the Up-Rev user manual and tuning guide dozens of times, and all of the suggestions or corrections that I have read that I should make have negative side effects. Please correct me if I'm wrong, but I've been told and have read countless places that I should create a safe basemap (with target AFR's more rich than I would like for safety), and to set the fuel compensation table to 100 (ECU no longer adding or pulling fuel). Then start by tuning the K value, and then moving on to the MAF table after the K value has been scaled correctly.

The problem that I'm having is that with any decrease in the K value (decreasing due to using larger injectors (stock SpecV injectors are 310cc)), I get either an extremely lean, or an extremely sporadic idle, or both.

On my first attempt, I set the K value to 19200 (did the calculation of 310/440 = .7045, .7045 * 26800 (stock K value) = 18880), and the car idled in the mid/high 15 to mid/low 16 AFR. Put the K value stock, car idles stoich. Try a smaller (larger) K value of 22000 and get a more stable idle, but still very lean (almost solely 15.x AFR, reading as high as 16.2 and as low as stoich on 'rich' dips). Try a K value of 24000, and get a closer to stoich idle, but it is very unstable, loping from 13.2 to mid to high 15's on lean pops.

At this time, I decided to test with the MAF table to see if I can use the 22000 K value, and tune the MAF table to 'richen' the AFR's up. I used the ROM editor to log the MAF voltage values x RPM, and see that at idle, I'm showing ~.80v to ~1.12 volts (lowest and highest voltages during idle) and then I open my ROM and select the MAF table values from ~.74v ~1.25v, and then use the 'large increase' button (which should 'add' fuel at those voltages, correct?). I enable RTT, and select MAP 1 (map I'm making changes with), and the MAF values have made no apparent effect. Thinking maybe MAF values can't be changed via RTT, so I save and flash the ecu with the new ROM, and still, no apparent change in idle. I'm logging air/fuels with the stock wideband o2 (06 SpecV's use the primary o2 as a 'wideband') and double checking them against my actual wideband, and sure enough, they're correct in showing lean. The 'large increase' button for the MAF table adds somewhere in the ballpark of 150-200 to the total value at the given MAF voltages, so I went from reading from 314-769 at .8-1.25 volts to 500-900 at the same voltage range, so I would assume that there would have been a (noticeable) change made. I even went so far as to try the opposite and go lower than the stock 314-769 range (changed the values to 22x-5xx) and still no change. Stumped.

If I change the K value back to stock and change my target fuel map so that with any minimal change in load will trip the ecu into reading from the map, my air-fuels go pig rich (in my map I have the entire top row with stock settings, and everything from 1400 RPMs to redline set to 11.47 AFR, and the top two cells in the first column are set to stoich 14.7 AFR. With the map set like this, I can't make any pulls, or even pull out as the AFR's plummet to 10.0 and lower, and it actually threw my car into limp mode (no pedal response, car would not rev). I managed to get the car onto the road, and into fourth gear so I could play around with the K value again, to see if I could get the car to cruise stoich, but I couldn't get it exactly right. I turned the K value down to 24000, and the cruising AFR's were better, but not quite stoich (stayed pretty constant in the low to mid 13's, occasionally going to low 14's), and then tried 22000, and there was no super significant change (maybe .1 or .2 change if that much) at cruising rpms of around 2200 (right where the MAF volts tip between 1.9x and 2.xx as per the UpRev Nissan Tuning Guide), but just as the manual admits, the idle values sucked, and as soon as I pulled back over, the car idled extremely lean (high 16's, low 17's) and eventually died from lack of fuel.

At this point, I just gave up, flashed my stock-ish rom back, and drove home.

Can anyone point me where I'm going wrong, or if I need a better starting point? I realize ECU tuning is not as forgiving for a turbo car, but it's much better and safer than the SAFC/AFC even with my dumb-*** behind the keyboard, and I'm not entirely retarded in how tuning works. I would just like to be able to safely make some pulls to see how this turbo reacts, where boost onset is, what RPM full boost is achieved, etc, etc. I'm not looking for perfect AFR's, I'd be happy with a nice smooth curve down to a safe 11.0 just so I could enjoy this new turbo and being behind the wheel of my car again. Just some pointing in the right direction, and some guidance from some of the tuning gurus on here.

And also, I realize it's a QR25DE-T, and not a VQ35-T, but the mechanics behind UpRev are obviously similar, and I'm sure some of the tuners that post on here have tuned a QR on UpRev before.

I really appreciate any help anyone can give me. I'll be the first to admit that I'm a newbie to ECU tuning, but I'm competent and willing enough to learn, and can provide any information that is needed to help me get in the ballpark and get a better understanding of what I'm doing, namely what I'm doing wrong.

Thanks guys!

 

Did you properly scale your injectors before starting both latency and minimum time? All kinds of nonsense occurs if you don't.

That said (and once you are sure the valeus are correct), you can stabilize your idle by flattening out the MAF values in the idle areas. Mine is actually flat from about 1.0V and down. Sometimes what happens is due to the larger injector size the slope of the MAF table has to be reduced to prevent overcorrection (a/f bounce).

 

I was told by another UpRev user and Sentra tuner that K value corrections were the only changes I needed to make, and to leave all if not most of the latency values stock?

Currently min effective injector pulsewidth is set at 2.46ms (stock).

Latency values are stock. These injectors are BLUE Subaru WRX 410cc injectors, and I had found the latency values on NASIOC, but the values seemed substantially low, and were not labelled in what voltage order they were in, and comparing the values from NASIOC with other values found online, they are inconsistent, so I was unsure of what to enter.

Here are two sets of latency values I have found for the WRX 420cc injector (440cc at my fuel pressure):

WRX 440cc Latencies
10v-1.25ms
11v-1.01ms
12v-0.87ms
13v-0.74ms
14v-0.62ms
15v-0.53ms

and

420 cc 6442 2002-2003 WRX 2.79, 1.49, 0.98, 0.68, 0.38

The last set are straight from NASIOC, with no voltage values, and you can see the difference between the two sets. What would you suggest, djamps?

 

Also, I'm not even sure I can trust those values, as the WRX's run a lower fuel pressure than my Sentra does (43.5 vs 51psi), so if I'm not mistaken, fuel pressure directly affects latency value, correct? Latency is something I'd rather not try and tune on the fly, but moreso in my driveway.

Anyone have any suggestions? Sorry if these may be really dumb questions, but I feel like I understand most everything about the concepts behind tuning, it's just the practice that is failing me now.

I should have just got DW660's and not had to worry, because I'd have a latency sheet, hah.

 

Pressure has no affect on latency only flow rate and maybe dead time -- and you don't use dead time in osiris just IDC minimum. if you wanna be 100% sure what your values are you can use an oscilloscope. Uprev used to have a guide from what I remember but I can't find it... maybe this will help:

http://blog.nsfabrication.com/category/tech/

Best to ask Jared where his injector latency calculation guide went...

The stock nissan injectors (at least on a VQ) have way higher values both minimum IDC and latency than my DW's, at least double if I remember. I've seen similar for others as well.

Minimum injector time you can figure out on your own. Typically the 14V latency value is a good starting point then go lower if you have a hard time getting stoich idle when warm. But the latency you want to get dialed in right for sure since it indirectly implies dead time.

 

Okay, so let me make sure I'm following you correctly, djamps.

I should start the car, let it warm up, and then open the ROM Editor and use the latency wizard to adjust the 14v value in order to get the car to idle stoich? I'm assuming that I'll need to adjust down, but not positive, since it's technically not an 'aftermarket' injector, it's a stock Subaru injector.

What I don't understand, and my apologies if this is a stupid question, but how will I know if my values below 14v are correct? Obviously I can change the 'msec change per volt' value, but how will I know if I'm using the correct slope (msec)? Or is this not as important, since these are usually only values the ECU sees when cranking? But then that opens another question, wouldn't an incorrect value at lower than 14v cause hard starting, or hard cold starts?

I'm just trying to make sure I understand what the changes I'm making will have an effect on, and how to correctly change them.

 

Anyone care to chine in and help? I still have yet to make any positive progress in self-tuning. I have a dyno session coming up soon, and would like to have a reliable tune on the car for the drive down (4+ hours). Less work for the tuner, also, so more fun with high boost pulls!

 

A little update:

Still haven't got the car tuned yet, so it looks like I'll be trailering it the dyno session when we go. I'd still like to get a mild street tune on it, but after speaking with my tuner, I have decided to purchase larger injectors. Just waiting on my adapter harness to come in so I can install them.

The injectors I picked up are these, Siemens Deka IV 630cc (60lb) EV1 injector: http://www.siemensdeka.com/index.php...products_id=10

A few questions I have now:

It seems as if Siemen's website has a bit of information on the injectors, but I'm not 100% sure if everything I'm looking for is there? I see that the 14v value is listed as 1.14, but comparing the value to Injector-Rehab's website, the values are much different -

10v - .83
11v - .64
12v - .50
13v - .38
14v - .28
15v - .17

But then, as I mentioned before, the Spec V runs a static 51psi fuel pressure, and most injectors are tested at 3bar, or 43.5psi, so most of the values I find are for 3bar instead of closer to 4 bar. So I started looking some more, and came across this: http://siemensdeka.com/specsheets/FI114961cs.jpg

That page shows the latency vs voltage values. There are also compensation boxes for various functions to determine how to calculate the latency scale. Here's where I'm a little confused; there's a compensation multiplier for the high and low slope, and an offset compensation table, so which table should I use, if any? Is my assumption that I take the offset table (volts vs ms), and multiply each offset value with the compensation value that is closest to my fuel pressure (1.0791 @ 50psi) incorrect?

I'd really like to drive my car to the tuner, but I can't merit driving 6 hours on an untuned car, when I can just get a trailer and tow it and keep it safe. However, I would like to at least be able to drive the car with the 630's in it, and I know the ECU can't handle the huge jump in injector size without tweaking.

Please, anyone help?

 

Why not use a 'known' set of injectors with easily obtainable specs and known to work well not only with your engine but also with Uprev?

http://www.deatschwerks.com/products...spec-v-2002-10

DW has well known latencies (also shown on the flow sheet right in the box) and they behaved with Uprev flashed ECU's, at least the 440-800cc. There's a reason so many Uprev tuners always use DW.

For the injectors you already bought you need to bench them with a scope and calculate the values yourself. The widely varying data you found on them is the exact reason why...and hopefully they behave well with Uprev ECU and you know the correct values..

By the way there's a virus on the page you linked, I wouldn't enter your CC info on that site.

 

I didn't have the change to drop on DW's right now, and came across these on a really good deal (75$ shipped for all 4 with less than 5k miles on them). These have been on a Sentra that made 500+whp, so there's no problem with the injectors, they just don't have a fancy label on them. They're the same injectors that came in the Turbonetics turbo kits, and Ford uses them for direct drop in upgrades on everything from PI cars to Cobra's, as well as other manufactures that use EV1 style plugs, and even on cars that don't (they're popular in the Audi/VW world, too).

I doubt the latency values on Siemen's website are incorrect, but like you said, there are many different values out there, so it's hard to know. I could have them bench'd and tested, but that's more $$. If I'm going to go that far, might as well have bought DW's, which I still don't have enough money for.

Didn't know about the virus? I've been on both links and have no virus warning, unless there's something you can see that I can't?

I'd just like to get this figured out so I can get the car running on these injectors.. I plan to switch to DW's in the future, but if I know these injectors are fine in my application, can support the power goals I'm desiring, and have worked in my application previously, what's wrong in using them?

Not trying to be argumentative in any way. I just want help tuning my car, and am willing to learn, but I have to make by with what I have.

 

Without knowing your exact latency values you'll never be able to properly tune it (if at all!) That's why it's so important that you test the latency. The values posted on siemens are very low... not saying they are wrong, but they are so different than stock (by orders of magnitude) that your stock ECU might not even be able to properly control them especially at lower (idle) pulsewidth. What ECU did the sentra they came off run?

 

His car was an 04 (which uses an ECU very similar to the 06 like mine, sans a few features such as the onboard wideband on the 06+ ecu's, and some 04 auto SE-R ecu's), but it was tuned via a Split Second Box (FTC) by Vinny Ten Racing in New York.

I understand what you're saying about the values being so low. They are substantially lower than DW or Precision injectors, and that has me worried, but, Jim Wolf Technology offers a reflash for the 02-03 Spec V (different ecu, can't be flashed with Osiris) and the reflash handles injectors up to 630cc, including the Siemens Deka 60lbs that I have. So, obviously the ECU is capable of driving the injectors if JWT is doing them, and VTR was able to tune a Spec to 500whp with them.

I have contacted the previous owner of the injectors; he's a cool guy, so I'm sure he'll help me get this figured out. If not, I guess it's off to Justin Prado @ 2JRacing to get an Uprev tune on the dyno-monster, and see if he can figure out these silly injectors.

 

Yep you might be better off leaving it in the hands of someone with alot of Uprev experience. When I started out I would have been lost without knowing the exact values I was inputting, luckily my 440's flow close enough to stock I didn't have to change the latency at all, and later on when I upgraded the 600's were published widely so I knew exactly what to punch in. Small changes in injector latency change the entire behavior of the tune, and unstable idle (rich) is one of the big ones.