28 lb/min on the Evo16g? Did I reinstall my T25 by accident?

[joemathews]

My highest logged airflow was 24.47 lb/min. I figure that puts me right around 28 lb/min actual airflow, which is much lower than I should be logging with this turbo...

If anyone has advice or insight into this problem, please feel free to offer it. TIA, as always

[joemathews]

I should also add that I will be getting my windows tinted today, and am hoping that will solve the problem.

[spyderturbo007]

Originally Posted by joemathews

I should also add that I will be getting my windows tinted today, and am hoping that will solve the problem.

That was going to be my suggestion! Either that or slapping on some nice spinners.

Anyway, it seems really weird that I am getting more airflow than you are (~30lbs/min @ 17.5psi). I thought mine sucked

I am still on the stock cat back and UICP so those are two huge restrictions that you don't have. Weird.

Maybe Kevin will chime in here and save the day.

[kjewer1]

Without a DSMlink log there is very little I can do to verify whats going on. Actually with WBO2 data I could do it. But I don't think I'll be doing any day saving here But lets run through a few items...

Any boost leaks would raise your airflow level. So hopefully we can say that you certainly don't have any appreciable leaks with that weak airflow

The change from 25 lbs to 28 lbs with a 2 psi increase in boost below 25 psi is exactly what I would expect, so at least something airflow related makes sense.

I don't think your low vacuum is low enough to worry about. I think the stock spec is 17 or 18 to 21.

The stock DP could be a factor, I was unable to break 30 lbs/min on the stock exhaust on my 2g before I sold it. I Think. I was at much lower boost though, like 16 psi. Hard to remember the details, maybe I posted the info somewhere else.

Clipping should raise airflow, not lower it (increased VE)

In my experience, the lower gears tend to move a couple more pounds than 3rd, and 4th usually isn't complete (until you are trapping high 120s (assuming AWD)). I always use 3rd gear.

At 7000 rpm the max timing is 17 degrees, over 2.1 g/rev. At that rpm you were certainly below that. Somewhere around 1.8. I don't know what timing is supposed to be at other g/rev values, but I think we can safely say that timing was being pulled for one reason or another. THe correction for 560s is -20%, so you were 4% richer, or about 9.1 target AFR. Its possible that with the low airflow at peak torque some knock was detected that porked timing for the whole run. The ECU will reset the knock when it detects a drop in throttle position, so it can often ruin the whole gear if it knocks in the midrange. What is your airflow, timing, and rpm, 500 rpm after you reach full boost?

[scottsee]

is it possable you could have an exhaust leak before your downpipe?

[spyderturbo007]

You seem to be hitting full boost late.

Also, for 9 - 10 s/sec you don't seem to have very many data points. Was timing, rpm and airflow the only parameters you were logging?

[kjewer1]

Airflow per rev will peak at the moment you reach full boost. Volumetric Efficiency generally goes down as RPM increases, so it makes sense. Airflow over time will peak at 6500 rpm or so on stock manifold/normal cam cars. With SMIMs and larger cams it moves up, or at least doesnt drop off as sharply if at all to ~8k. Torque will tend to follow airflow per rev (cylinder pressure) and HP will tend to follow airflow over time (HP is work done over time).

9 samples a second with 3 items becomes 3 samples per second, or a capture every .33 seconds or so. I'm not sure that correlates with the gaps in the RPM between datapoints. But this is one of the many drawbacks of using an OBD2 based system. Oddly enough, with DSMlink the 2g ECU is twice as fast as the 1G, so for once the 2G guys get the sample rate advantage. I get over 200 samples/sec, just for reference.

[JoeWagon]

We know that 560 evo injectors are supposed to flow nearer to 600cc (jeff @ dsmchips). You're looking at 25% if that's accurate, just to get to stock ratio. If they do flow only 560, you're still richer than a street tune. Kevin, can you explain how you got 4% richer from -20% fuel? How much fuel should he take out to get to 10.5:1 flowing 560cc?

[kjewer1]

I was using the 560 value, which is a -20% correction factor (450/560-1). 600s would be 25%. I'm surprised people dont know what these things flow. With DSMlink in the EVO now and my spreadsheet I can probably figure it out once I get the WBO2 installed

Since I was assuming 20% would get back to stock, 16% would be 4% richer than stock. Stock target AFR is 9.5:1 over 2.1 g/rev and at the higher rpms. I dont think there is enough airflow to be over that at those RPMs in this case, but we'll assume so for the sake of explanation. 9.5*.96 is where I got the 9.1 from. To reverse it, 9.5/10.5 - 1, or about a minus 9-10 percent change. If you think -25% is more appropriate for the EVO injectors, something around -35% would get you to 10.5. In reality at 24 lbs/min (at the ECU) at 7k rpm, or about 1.55 g/rev, target AFR might well be up in the low 10s. Without DSMlink there is no way to know. The calculation could be done from IPW, but I have never taken the time to figure out how to do that. But you see my point. This is all shooting in the dark, until you get over 2.1 g/rev at the ECU (tough to do on an AFC, getting worse as injector size goes up and/or correction value goes down), which allows us to know for sure what target AFR is.

[joemathews]

Originally Posted by 95GSXracer

I get over 200 samples/sec, just for reference.

Yeah, yeah, stop showing off

Anyone have links to directions for doing the byte timing test correctly?

I appreciate all of the input, guys. I'm thinking that my lack of flow/power is just because I am still running rich, but the only way to tell is to lean things out and find out. 2 percentage points at a time on the AFC should be subtle enough not to blow something up, should it not?

Kevin, if you're flowing 39 lb/min on your Evo8 turbo and stock Evo injectors (and stock evo pump, which is comparable to my 190), then my tune is even more likely to be rich.

I would be very interested to see DSMLink injector flow rate calculations and respective duty cycles, if you get around to it on your next tune, Kevin. While Jeff@dsmchips has given us a promising opinion, who knows if it's accurate or not?

Will report back when I have leaner logs...and hopefully not a broken car

[JoeWagon]

Where did you get your -15% global from anyway? You're running 8.55:1 AFR assuming 600cc.

I'd bump it to -25% then go slowly from there. This will be from 9.5:1 - 9.8:1 with the estimating due to injector flow. -30% is 10:1 - 10.5:1. That wouldn't be so bad either to just stay at. Turn the boost down to maybe 18psi, set fuel to -25 or -30 and start re tuning with the boost watching for knock. Hope your airflow skyrockets.

[spyderturbo007]

I max out at about 9 or 10 samples/sec on mine too.

If you want to do the byte timing test you need to be on the "main page" where it shows the boost gauge and gives the options for diagnostics, view and datalog. Then click on the bottom left side of the palm screen (right below the house). This will drop down a menu from the upper left of the screen and will have the option for the byte timing test.

Also, if you click on the same icon when you are on the datalog screen, you can change setting for alarms, click anywhere to start log, etc. Provided you are using V2.0 (I don't know about V1)

I normally end up with about 30 - 40 data points after a 3rg gear pull.

[kjewer1]

Turning down the boost will lower cylinder pressures due to airflow, but being below 2.1 g/rev, lowering airflow will also raise timing at the same time. Just don't assume that any timing change is knock related. Any change to the airflow signal changes timing, until you are over 2.1 at the ECU. Situations like this really help illustrate how lousy tuning is with an AFC, at low airflow at least.

I did some logs last night. But first let me make some qualifiers. The EVO8 MAF and 2G ecu don't exactly line up. Guage boost and DSMlink BoostEst match just about perfectly, which tells me the testing someone else did to come up with the MAF Comp values was valid. So airflow should be very close to being accurate. I need to get the WBO2 on there to verify.

Airflow topped out at 41 lbs/min at high rpm with 21 or so pounds of boost. This exactly the airflow I would expect at 21 psi from a cammed DSM. Which leads me to believe the EVO setup flows very well already (not sure how stock evo cams compare to stock dsm cams), or the number is ~3 lbs/min too high. The latter seems unlikely with boost and boostest matching, but I'm not ruling out the possibility. With the fuel all zeroed out and a global of ~20% (assuming 560s), target AFR was right around 10:1, and IDCs peaked at 114% or so I took out 3% fuel going for something in the mid 10s and let off at 7000 rpm, and IDCs hit 103%. So this is all pretty much expected, since my calculations say that 560s will only support 41 lbs per minute at 11:1, and with a target of 10.5 but IDCs of 103%, actual would be ~10.8-10.9:1. The Wideband is really the key component in validating these numbers though, and I can't get it back from the guy that bough the 2g yet...

I'm going to the track anyway tonight, but I'm going to have to run race gas since I'm flying blind without WBO2, and this car is just too damn expensive to take unnecessary risks

There are byte timing instructions on the pocketlogger site, http://www.pocketlogger.com/instruct...startobd2.html

[JoeWagon]

Bill has an EVO, dsmlink, and a wideband. Get some numbers from him.


EDIT>> Joe, you really need to set your S-AFC for your injectors before you start tuning. -20% to -25% is still rich, don't be a girl.

[scottsee]

4-4,5k is when max cylinder pressure occurs. isn't that also where your getting full boost? timings suppose to drop.

glad you've made improvments in your qest for the elusive evo16g safc2 tune. some day i'll stop breaking shit and will be there with ya.