The definition in the FAST XFI manual for MAP (psia) is as follows.

MAP (PSIA) - The calibrated reading from the manifold absolute pressure sensor expressed in pounds per square inch absolute. With the engine stopped, it should read around 14.7 psia.

I have a two bar MAP in my car. With the engine off it reads 0 not 14.7. I double checked and the software is set for a 2 bar MAP. I tried using a MityVac that I confirmed as accurate on a known good mechanical gauge and a vac/press test gauge. It gave the same readings on both, dead on the same as the gauge of the Mityvac. I have two 2 bar MAPs, one for the Autometer boost gauge and one for the XFI. In testing the Autometer gauge (I did let it reference atmospheric at power on) I got weird results. 5 psi on the MV was 2 on the Autometer, 10psi MV was like 4 on the Autometer. The more vacuum I applied the farther off it got. The gauge reads to 30in Hg of vacuum but with 30 applied with the MV I was showing only about 12 or so on the Autometer, maybe a bit more.

I haven't checked on the other MAP sensor for the XFI since I just noticed it's 3am and I shouldn't start a trouble shooting session now but I can't help but think that if the XFI is expecting 14.7 at engine off and is seeing 0 something is going to be out of whack. Or is this just some weird GM MAP sensor thing. Thanks for any ideas.

 

I previously used the two bar MAP sensor from FAST (which is a std GM unit). My understanding is that the 2 BAR MAPs are designed to read 1BAR vacuum (1 BAR = 1 ATM = 14.7 psi) and 1 BAR pressure. With this being said, when the engine stops, the indicated value should be directly in the middle (or 1 BAR), equivalent to 1ATM (14.7 psia). REad this as no indicated vacuum and no indicated psi.

As we walk on this earth, each and everyone of us is exposed to 1ATM or 14.7 psia (yes I know, some are slightly higher and lower depending on where one lives). PSIA (Pounds per square inch absolute) means as refereneced from 0ATM or 0 psia. It is all a matter of reference point and in this case they are referenced from 0 ATM.

I currently run a 3 BAR MAP, which allows 2 BAR (2 ATM = 29.4 psia) pressure on the boosted side. These are really a pain, as the gradiations within FAST or DFI decrease when you go to the larger scale of 3 BAR.

I hope this helps.
Aaron

 

Yeah, I have figured that was how it worked since with the motor off if you look at the Base VE table that little floating bubble is sitting at around 101 kPa. But it makes me wonder then 1. If the motor makes more than 14.7 pounds of vacuum, which most seem to at idle then how does that show on the table? I can't check visually since my car is not making more than 8 in Hg of vacuum no matter what I do with the car undriveable at this point. Not to mention coast down where it can jump to nearly 30 pounds of vacuum. Does the XFI just ignore it? Does it just use those last values at like 14 pounds of vacuum no matter what it goes to? I looked at the Dakota datalog sample they provide and it actually shows yellow at low double digits and it turns red at single in kPa. Is high vacuum just assumed to be no load and so any default value will be fine for fueling?.

Number 2 would be if the reading is supposed to be 0 with engine off why do they say in their online manual which is updated constantly that it should be 14.7? I am seeing vacuum numbers of 2 to 4 in Hg negative at idle, should I be adding these to 14.7 to get my actual vacuum at idle or is their manual just screwed up? And since it's a year old online manual that can be updated in less than a minute that's pretty sad. Thanks for helping to clarify Aaron.

 

Oh wow, where to start.

Lets get back to basics. When you are idling the car you have this big engine pumping 5.7L of air every other rpm at 800 rpm or so. Thats alot of air. When the throttle blades are closed the IAC port will not be able to supply that much air so you get a lower pressure inside the intake manifold. My car runs about 12 inHg on my vac gauge at idle which equates out to 60kpa(101 - 40kpa). The gauges may be a little off, but they should be close. Remember this is negative vaccume so if you just convert the numbers 12 inHg is 40kpa, but you have to take that from the reference point which is 101 kpa.

Why is the reference 101kpa? Well, what happens when you go to wide open throttle. Your tb blades open all the way and then you can flow as much air into the manifold as you can into the cylenders. Since this happens you will equalize pressure with the atmosphere. And since the air going into the cylenders is the same as going into the manifold when the engine is stopped you will also see a similar WOT reading of 101kpa when the engine is stopped.


AKS is right about the map sensors. 1 bar is negative, 2 bar is 1 negative 1 positive, and 3 bar is 1 negative and two positive. If you are using a psi map with a 2 bar map sensor the axis should read -14.7 to 14.7. And at engine off you should read 0, because is equalizes with the atmospheric pressure.

If you are really reading 30 psi vacuum there is somthing wrong with your car. even if its 30inHg that seems like alot of vacuum. The XFI will only do what the it can with the sensor data and the map sensor can only read to -14.7 psi. So the sensor will give this value to the XFI and it will go with that value.

My advice to you is to only look at the XFI outputs. The boost gauge is good to see what is really going on, but for tuning purposes only what the XFI sees makes a difference.

Good luck man.

Edit: By the way, lets hear what you have for an engine setup, especially the cam, and what kind of FI setup. That may help answer alot of questions.

 

My engine combo is almost exactly the same as one of Aaron's old ones. It's a 406 with ported AFR 195's and a P600b running the 12lb pulley. The cam is a EM380R, which is a 224/230 at .050. Same Hooker SuperComps everyone runs on the L98 motors. 72lb Siemens Deka low impedence injectors. Intake is a SuperRam with a matching 58mm Accel throttle body. Using a FAST XFI to tune it.



I started trying tune it myself about 5 months ago but with the car was immobile then and the fumes got out of hand fast. I spent the time up until a month ago putting the suspension back in so I could roll it outside and avoid asphyxiating myself. I decided to remove my lame tuning abilities from the picture of trying to find the problem and bought a baseline tune from TJ Tracy of Advanced Performance. At least I know it would be a good base tune. I got that about a week ago. I have been chasing gremlins since then.

The first one was that on the inductive pickups in the distributor, MSD 2345, both crank and cam position sensors, it turns out that FAST and MSD have a different idea of what pos and neg are wiring wise. What MSD says is positive, FAST says is negative. Swapped that around and the car started idling a lot better. But still it's rough and as it warms up it's blowing tons of overrich smoke out the tailpipes.

Now that I am not using my own lame tune I am trying to find out why it's running this bad, this tune is based off a customer car of his that was near identical to mine, with changes made for the small differences. First I notice the vacuum is very low 2 to 4 PSIA on the XFI readout. I figure I must have a vacuum leak, but before I tear into a incredibly annoying to find problem I figure I will make sure the 2 bar MAP I got from Exotic Muscle is working correctly. That is when I notice the the whole XFI manual says 14.7 PSIA motor off, actual reading is 0 with motor off. Ok, maybe their tech writer is asleep. So I get a MityVac and gauge to test it.



I splice in another gauge just because I don't trust the MityVac one even though they read the same, I trial test it on a known good Stewart Warner mechanical boost/vac gauge I have lieing around. Tests dead on. I then try it on the 2 bar MAP sensor for the XFI. I get.

MityVac 5in Hg...XFI PSIA(-3.5) kPa(77)
MityVac 10in Hg...XFI PSIA(-6.1) kPa(59)
MityVac 15in Hg...XFI PSIA(-8.7) kPa(41)
MityVac 20in Hg...XFI PSIA(-11.2) kPa(24) yellow text
MityVac 25in Hg...XFI PSIA(-14.3) kPa(2) red text

The Autometer gauge which functions with an identical looking MAP sensor gives similar type readings. They are wired and plumbed completely separately including manifold hookup. So if a speed density system is basing its tuning first and foremost on this MAP sensor and it's not actually giving a reading that seems to correspond to reality I am not surprised it is blowing tons of smoke on a known good tune.

Now I am just trying to figure out what to make of these figures before I waste time looking for vacuum leaks or swapping MAP sensors or other things. I have a 27k datalog of my last run. Basically just the car idling for 5 minutes with me revving it a few times in the middle if anyone wants to look at it.

P.S. another can of worms, with the datalogger set to 20 frames a second it takes 2.1 to 2.5 frame per second. No clue why, might just be broken in the software version I have, I will have to talk to FAST about it. Thanks for any ideas guys.

 

That MAP Kpa reading sounds right to the Mityvac readings. Have you verified the crank reference angle ?

How many sensors are you datalogging at a time ? Sounds like you're datalogging one of the high resolution sensors.

Also, one of the things with the FAST distributor network is that the selling vendor is supposed to supply a start-up program and provide support.

The XFI will support up to a 5 bar MAP sensor. There's a 3.5 bar sensor out there which is great for turbo guys - gives resolution to 40 psi. You're able to define the areas of the tables. Basicly it gives you the ablity to focus the area. If you want to devote 10 lines of the VE table to the areas between 30 and 70 KPA you can. Same for the rows.

Some real nice features with the new system. The processor speed has been increased and is noticeable in how correction is applied.

Damn, I'm getting into a sales pitch If you can't tell I'm impressed and want to upgrade my SEFI box.

 

I log mine at 10 frames/sec, and I get a reading every 1/10th of a sec. I do find that the data logs can get very big quite quickly. I log for 20 secs tops and that covers a few secs in staging, through a full run, then a few secs in the shut down area. I also have a std variable set that I log which has 11 variables (I think).

If I remember correctly, if you are recording too much data into a file, it reduces the capture frequency. Reduce the length and the number of variables, and you should see the capture rate return to normal.

Regarding the psia vs psig debacle, 14.7 psia is neither vacuum or boost when the engine is not running.

I will agree, your engine is very similar to my old combo when I first tried the ProCharger line. Give me the pulley size in diameter vs ATIs BS boost values. BTW, your engine looks good.

Aaron

 

Aaron I have the 2.55 pulley. They list the default as 2.8 for the non intercooled kit at 8psi and the 2.65 as 10psi for the base intercooled and the 2.55 as 12psi. But as you said it so dependent on motor that their psi number is basically worthless.

All that being said then, does the reading of 2 to 4 MAP (psia) sound right for my cam or is that too low? I am trying to figure if I have a vacuum leak or if my timing is just messed up and making the car have low vacuum. The ignition tables are weird, the first vertical row is like 15 for like 400 rpm then the next one for 900, around idle speed, is 30. I looked at a few demo tunes and the ones they show in the manual and they all look like that. Isn't 30 degrees of advance at idle a bit much? Or am I missing something. I really find it hard to believe that TJ Tracys tune is off so far that it would be blowing the amount of smoke out that it does as it comes closer to operating temperature. So I must have something out of whack elsewhere. I do not want to start messing around with his known good tune unless I know there are no mechanical or MAP or ignition issues first. I need to get some clarification from FAST as to the "It should read 14.7 psia with the motor" off thing from the manual when it actually says 0. I could have the wrong MAP sensor or something.

As to the logger I am logging 14 channels which is the max, but even my five minute long idling one is only 27k. With like a 160GB hdd on my laptop I could have hundreds of those and not even dent it. I did set the max length of the sample to 20 minutes because it take nearly ten for my car just to reach operating temperature sitting in the garage before I start trying to actually play with things, so it was getting annoying having it cut out right before I start testing things. If the length does automatically reduce the sampling rate I will just shorten the sampling time and trigger the sample later after I get it to operating temp. I will call FAST tomorrow and see what they have to say about it. And suggest that they put that in the online manual.

Thanks for all the helps guys.

 

For what it is worth, I was running the 2.4" pulley and also hadd the 2.25" pulley. Unfortunately, the 2.25" just resulted in more belt slip, no additional boost. The radii just gets to small.

I am running 36° at idle of 870 RPM, so no, this is not too much advance.

Save his file as a specific name, then save it again as something else. I save all my files in date format (i.e. 87 Vette 060604a.gct)

Let this go. 14.7psia is the same as 0psi on a normal vacuum guage.

It is not the computer limiting you, it is the total number of data points allowed to be collected. Try resetting your total time to something like 30secs and see what it does to your capture frequency.

I hope this has helped.
Aaron

 

Here's a couple of quick questions for you.

1.)What color is the plug end that sticks into the map sensor, green or yellow?

2.)If you are idling at 2-4 psi of vacuum, what rpm is it idling at?


by the way these readings look correct.

 

Here is a pic of the MAP sensor connections. The Autometer gauge one is on the left and the XFI one is on the right. The Autometer one looks black and the XFI one looks to be orange. I got the integration wiring harness from Mac at Fast Track so I think it is a generic connector and that the color isn't too accurate a test of what they are. He also included a green colored connector that was for a different MAP, either 1 or 3 bar, I can't for the life of me recall anymore. Figured a pic would be worth a thousand words on that color. I did do the sensor calibration from the FAST manual on it and it is a 2 bar.



As to the idle its set at 800 now, since that is where it will have to be to pass smog, though TJ had the original setting at 910. It flucuates between 800 and 950. I have the vacuum down to an average of 61 kPa now. I was told the cam should pull 12 to 16 in HG which is around 52 (12) to 38(16) but even with the flucuations I never saw it drop below 51. I figure it should be lower but whatever.

Right now the distributor keeps loosening up after like 15 minutes of idling and leaking oil and I have to tighten it which is annoying. In playing with the spark table a bit, as I said the first vertical row is 15 and the next is like 32 it seemed to settle down a bit when I brought the next two rows down in the idle area to like 20. But then the distributor thing started happening so I had to stop.

It is blowing no smoke at all anymore. Matter of fact kind of smells a bit lean if anything. I find that odd since the O2 sensor is reading like 12.5 average at idle with the A/F target table set at 14. You would figure that would be way rich. I started with TJ's default VE of around 50 at that idle spot but dropped it down to 27 before it started the lean hunt action below that. But even at 27 VE at idle spot the A/F actual wouldnt go leaner than 12.5. The XFI guide says 13 to 14 but if I get leaner than 27 it starts to hunt and if left below 25 long enough it will start to backfire a bit.

So I guess all things considered it is running vastly better than it was a week ago. Still needs to be smoother and according to the O2 sensor leaner if I ever hope to pass smog but at least some slight progress is being made. Sad part is the timing must be set to 8 at idle or it's instant fail on smog so I will have to play with that as soon as I get it under control. Been making backups of the GCT when I make changes and datalogging every run so I can always go back if need be. Thanks for the insight in to this stuff by the way guys, as a newb at this all the information helps. I have the last datalog if anyone wants to see it. It's 150k for like ten minutes of idling, like I said they dont get too big at this point.

Oh, and btw, the length of the sample didn't seem to affect it as much as the sensor type. As soon as I cut out like MAP (psia) and just used MAP kPa it went to .11 instead of .42 a frame. Just put the MAT sensor in there but its still at .11 so I guess that one doesn't bug it much. Didn't get a call back from the FAST guy for a confirm on any of this, just anecdotal atm. Later guys.

 

Sounds like things are running alot better now. All that from turning the timing back? Interesting. I have mine set around 20 too. You should double check everything with a timing light just to make sure.

 

Here is a quick screenshot of the datalog. I know you can't get much from just looking at it as opposed to being able to page through it with the software. The Cam Sync freaking out there when I start to bring the timing from 32 down to 20 is my fault, I have the Cam Sync set too close to the timing event and it got triggered, I just need to back it off some so its not in the way of the reduced timing. Just its a huge pita to get to the screws to loosen the Cam Position Sensor when the distributor is installed, I can't even see the one because of the EGR riser from the headers. But as you can see from the spikes it has to be moved back about 20 degrees so I can play with the timing without running into it.

 

Jaxian,
Why is your TPS reading 0? This may be a good place to start. I set mine at 19% and it scales from 19-98. Make sure you are able to get full scale without going to the limits on either end. Set your minimum idle at 20%, that way you are not chasing your tail with idle enrichment.

As for logs, the following 14 variables are what I log always (unless I have a specific issue that I am hunting):
- RPM - TPS (%) - Coolant (°F) - Air Temp (°F) - Spark (°BTDC) - ESC Rtd (°) - MAP (kPa) - MAP (psia) - Target A/F - Actual A/F - O2 Corr (%) - VE (%) - Inj DC (%) - MAP AE

If you detail how you loaded the file to show on the forum, I will gladly upload a file.

Aaron

 

Email me your program and datalog. As it stands you're not going to be able to get your idle to be any leaner than you have. 1.3 ms pulse width is pretty much as low of a pulse width thats possible. There are some other things in that datalog that aren't looking kosher to me.

As for idle timing and emissions. On a L98 based motor the base timing in the distributor is set at 8 degree BTDC. When the EST takes over typically timing flucuates from 19 to 25 degrees BTDC. When they do the emissions testing in California do they disconnect the timing connector to check base timing ? If they do I'm pretty sure I've got an idea on how to get the XFI to do what you need.

 

Hey Aaron. The reason the TPS is at 0 is because it was just sitting idling and I didn't have my foot on the throttle. In the FAST book they say to calibrate the TPS so it is 0 with the throttle on the stops and 100 with it flat on floor. Then you set the points between it on the TPS calibration graph so it's a straight line from lowest voltage (TPS 0%) to highest voltage (TPS 100%).

On my car the range was .530 to 4.63 volts so they said subtract .53 from 4.63 to get 4.1 then divide that by 7 since that's how many set points there are on the graph. So I had a plot point every .58 volts. Since it was a straight line that is supposed to give an even throttle response. If they have this set up wrong I would love to know. At the moment even breathing on my throttle will show TPS reaction and floored is 100% so it seemed to make sense.

As to how to post the datalog what I did was open the CCom program pull up the datalog then take a screenshot. (Shift <print scrn>). That puts it on the clipboard, then I opened up Paint and hit <paste>. I took the little cut tool and cropped just the datalog instead of the whole screen. I opened it in Photoshop at that point just to convert it to a jpg since they are usually smaller, I think you can just save it as a .jpg in Paint but Photoshops filters are much better and the image quality is usually better. After that I just threw it up on Photobucket like any other picture.

SloR, I can email the last datalog there and the gct. It is a XFI version 1.205 of CCom with the reversed tables. If yours doesn't have the reversed tables update basically anything with Coolant Temperature will be backwards. Very noticable, just remember its backward when you examine it.

I must admit, every time I tried to lean it out more than it was it started to pop and do the lean hunt thing so I just didnt see how I could get it leaner and it was still at like 12.5 average for the 6 minute run. I found one from like 4 months ago before I stopped to finish the suspension so I could roll it outside to tune it and it was pulling like 14lbs of vacuum (42 kPa) and the BPW and Inj DC where both near 0. I didn't even see how that was possible but it was pulling nearly double my present vacuum. Since they are small maybe I will throw that one in, the datalog files are named by date so there will be one from today and one from like 4 or 5 months ago that I was trying to do myself. 95% of my present tune is the one I got from TJ Tracy at Advanced Performance so I can't take credit for it, mine was kind of hacked together from the default one that is on the CD when you get the CCom XFI software.

I don't recall whether they pull the EST wire to smog it, it has been over two years now since mine was smogged and I wasn't watching last time. That would make a big difference. Guess I better find out. Thanks for all the help guys, it would be nice after all these years to be able to actually drive the car around with this motor. Just have to get the distributor to stop loosening up too.