351 TT tuning.. need some ideas, AFR seems somewhat random

srvblues00 · Post by **srvblues00** » Fri Mar 03, 2006 4:55 pm

My setup:
MS-1, V3.0 PCB, MegaTune 2.25b, & MSNsE 024s13c, LM-1 WBO2

My engine is a twin turbo 351w in a 88 Mustang hatch. You can see pics at http://www.binaryweb.com/scrapbook/

I have it running now, but it is very very rich.. like 10.5:1 and smoking rich. When I try to lean it out, it then goes too lean and the engine will die. If I have any O2 correctly enabled, it generally seems to run worse. I have O2 correction disabled now, but now I can't use autotune.

If I run with no AE, the car can barely get moving on its own (C4 automatic, 3000-ish stall). If I sloooowly roll into the throttle, i can get the car moving.

Any time I put the car under a load, it about dies from going lean. IE, 1st to 2nd shift (shifts high, no vacuum to modulator right now) However, looking at the datalogs I can find very rich and very lean conditions at the same RPM and MAP readings, so that doesn't tell me much about what bin to change in the VE table.

I've been playing with this thing each night after work now for about a week. It has gotten better, but progress is dropping off. Attached are 2 datalogs and my msq file.

TT350chevelle · Post by **TT350chevelle** » Sat Mar 04, 2006 12:09 am

Where are you getting your map signal? The map signal should follow TPS movement, yours shows very little movement.I take my signal directly from the manifold, very stable and correctly represents engine load.

Here's a couple screen shots to show how the Map almost mirrors the Tps, this is what you should see.

Brad J.

srvblues00 · Post by **srvblues00** » Tue Mar 07, 2006 7:51 pm

TT350chevelle, thanks for the reply.. you were right, I had a crappy source for my MAP signal. I got a nice signal going to the computer, and now it drives much better. I should have realized the MAP wasn't following the TPS..

Now when I put the engine under some load, it backfires or something, I'm not totally sure what its doing. Pretty much any time I am at 80+ KPa on the MAP, and I give it more throttle, it will pop very loudly and lose power. The last few times it has done it, I have been watching my LM-1 wideband when I knew it was about to happen. Every time it is 13.0:1 or richer, so it shouldn't be too lean to burn. It's never richer than 11.5:1. Doesn't matter if I do or do not run any AE. It almost feels like a spark problem, so last night I changed all 8 plugs (3 hour job.. yay!), and tied up the wires so that they were not touching the exhaust manifolds. I did find a 2 plug wire boots that had a small hole in them, but that was b/c they were too close to the headers. I put a 90 degree boot on them and that got me plenty of clearance. So, plugs and wires should be good now. Cap and rotor are brand new. Plugs are gapped at 0.028". I am running an MSD 6A box and a stock Ford coil for the moment. Rotor phasing is lined up nicely.. drilled a hole in the cap and checked the rotor with my timing light.

Attached is my latest log file, msq, and a mslvv screen shot of the latest log file.

TT350chevelle · Post by **TT350chevelle** » Tue Mar 07, 2006 11:47 pm

Are you running low impedance injectors? Your PWM settings would suggest low impedance injectors WITHOUT the flyback, but the V3.0 pcb has the flyback integrated, are your using the flyback circuit? If you are, then the 55% indicates that it is not working.If your not using the flyback circuit, then 55% might be to low to reliable operate the injectors, increase by about 5% and see if it effects your problem, but be aware that you could have a flyback failure running 8 low impedance injectors without the flyback circuit,... I did.

Brad J.
PS: Did you verify rotor phasing at different engine timing? You do have your MS controlling igniton right? Is the distributor used for triggering? Mechanical and vacuum advance are locked out? Just thinking out loud.

srvblues00 · Post by **srvblues00** » Wed Mar 08, 2006 9:56 am

TT350chevelle wrote:Are you running low impedance injectors? Your PWM settings would suggest low impedance injectors WITHOUT the flyback, but the V3.0 pcb has the flyback integrated, are your using the flyback circuit? If you are, then the 55% indicates that it is not working.If your not using the flyback circuit, then 55% might be to low to reliable operate the injectors, increase by about 5% and see if it effects your problem, but be aware that you could have a flyback failure running 8 low impedance injectors without the flyback circuit,... I did.

Brad J.
PS: Did you verify rotor phasing at different engine timing? You do have your MS controlling igniton right? Is the distributor used for triggering? Mechanical and vacuum advance are locked out? Just thinking out loud.

Yes, they are low impedence 46lb injectors (for now). If I lower the PWM to 40%, the engine will barely idle at all. (have to feather the gas so that the "send to ecu" function won't kill the engine when making a change) Haven't had a chance to drive it at 60%, my lunch break is almost over

How does the flyback circuit work? If it is there on the board, it gets used? There are no jumpers for it, right? Maybe tonight I can do a visual verification on the board to make sure it is there. Low impdenece injectors should run fine in the 30 to 40% range, right? Time to get back to work, I'll get back to this project tonight I guess.

TT350chevelle · Post by **TT350chevelle** » Wed Mar 08, 2006 10:43 am

I can't give you any specifics about V3.0 board (I built V2.2, 3 years ago!) but if you included the flyback components, you should start tuning the PWM settings from 30% and 1ms (my setup is running 20% and .8ms, 8-50lb/hr Seimens Deka injectors) If the engine will not even run at 30% and 1ms, then it is very likely that your flyback circuit is not working.

The V3.0 assembly guide installs the flyback components at step 69 ( http://www.megasquirt.info/ms2/V3assemble.htm )
Did you included these components?

Brad J.

PS: If increasing your current settings to 60% and 1ms improves/eliminates the problem then you need to double check your flyback components.Be warned that running your car without the flyback circuit working increases your chance of a flyback failure, increase your PWM to 60 or 65% for a quick test only, do not go cruzin and tuning with these settings.

Personally I would not run the car more than idle, if it will not idle with 30% and 1ms but it will at 70% and 1ms then figure out what is wrong with the flyback circuit before high rpm/duty cycle testing.

dieselgeek · Post by **dieselgeek** » Wed Mar 08, 2006 11:42 am

How is the spark timing? If you set Fixed Angle at, say, 20 degrees, does a timing light show 20 degrees on the timing marks?

Is this a TFI distributor? Sounds like a fun project, let us know how it goes!

-scott

srvblues00 · Post by **srvblues00** » Wed Mar 08, 2006 6:00 pm

Well, I got it running fairly well at 30% PWM. I added 3 degrees of timing at idle and it runs just as good or better than it did before at 60%. So, that issue is taken care of.

It is still cutting out hard when I get close to 90+ KPa. However, tonight I just wanted to see what would happen, and I wanted a different datalog, so I just laid into it hard and didn't let up for a while. It accelerated pretty hard, cut out, accelerated more, cut out, etc. According to the logs, I should have been on the rev limiter, but it still cut out/missed like it always does when I get around 0" HG. Interesting note, 1 datalog entry showed 3200 rpms and 111 MAP. The very next entry shows 6700 and 124 MAP. Entries are 0.371 seconds apart. Odd thing is that it says my TPS is at 44 and 47.. that seems strange b/c I thought I had it on the floor, but I suppose it is possible that the jerking of the car could have pulled my foot up a bit.

Anyway, attached to this post is the datalog from the above mentioned run. Click here for a sound clip of the part where it is cutting out. I stuck a mic in the engine bay and recorded it on the laptop along with the datalog. The sound clip is of the first very spikey part of the datalog. Click here for a sound clip ...some normal accel and cruise

dieselgeek, yes, my timing light will show whatever degree of advance I put in when I have the timing map disabled. I have also checked the timing at idle against what MS gauges say the timing is, and it is right on. Yes, it is a TFI dizzy, but I am only using the 3 wires for the hall pickup.. I'm not using the actual TFI module. Hall input to MS, LED17 output to pin 36 triggering the points circuit on an MSD 6a box.

TT350chevelle · Post by **TT350chevelle** » Wed Mar 08, 2006 11:02 pm

Hmmm... interesting..
Change your communications settings, timer interval (ms) to "0" (instead of 50 ms) and record another datalog.This should allow us to see more detail in your datalogs.
There is something going on with your MS timer, if you watch the SECL part of your datalog you should see a nice smooth timer count (the purple line at the bottom) your datalog is showing an inconsistant timer count, sometimes there is 3 seconds between recordings! If your MS loses track of time, then your fuel and more importantly your igniton, will be all fugged up.The screen shot below is from one of my datalogs.

Brad J.

PS: Did you calibrate your TPS in Megatune? (under tools) That will probably fix the odd TPS readings in the datalogs.

firefox · Post by **firefox** » Wed Mar 08, 2006 11:59 pm

hmmmmm that funny behavior between MAP and TPS reminds me on my problem:

got this in my datalogs....

look here

http://www.msefi.com/viewtopic.php?t=12 ... sc&start=0

Maybe that helps also... the trick with a resistor doesn't work in my installation.... don't know why... so I drop the TPS and drive it without it... with the rpm based accel enrichment I don't need it anymore.....
You could do a test without the TPS! If it doesn't bog anymore you've found the problem....