rkcarguy

Ok I have a 12A I've converted to EFI with a MS. I have mine up and running, however I've been having some problems tuning. In the req_fuel setting I initially setup for 14.7:1, resulting in a req_fuel of 9.3 for my setup. I found that my engine likes a 13.2:1 AFR best even at idle.

My question is, if I reset the req_fuel to the 13.2 AFR, and then lean out my VE table by multiplying the VE entries by(13.2/14.7) will this fix my problem and keep 13.2:1 the "target" AFR for the MS when it reads and adjusts via the O2 reading?

Right now it fights me and pulls my settings back too lean.

muythaibxr

If your settings are too lean, all you need to do is richen up the VE table. What's fighting you? What's pulling the settings back too lean?

KrisGXL

Are you using a wide band or narrow band O2 sensor? If your using a wide band you can use the "AFR Targets for VE Table 1", and set it to your desired mixture(depending on what firmware your using).

You can also scale this whole table by a factor if you want. In the AFR Targets for VE Table 1, click "tools">"table transform", and then enter the scaling factor.

rkcarguy

I am using a narrow band heated sensor from a Nissan SR20DET, and also a separate PLX wideband. So far the NB sensor seems to be reading right there with the wide band when I compare the AFR on the laptop and the wideband. I am using the newest megatune release. Thanks for the scaling advice that would have saved me alot of time

Muythai-my req_fuel was set at 14.7:1, where I found my engine wanted 13.2:1, and wouldn't even run at 14.7:1. The wideband showed around 13:1 under accel, but as soon as I try to hold an RPM above the EGO active RPM it leans out and stalls. I guess I can move the EGO point way above my engines RPM effectively disabling it but it should hold 13.2:1 as set in req_fuel.

renns

You cannot use NB with target AFR table. Consider the NB signal as binary, either rich (> 0.5V,), or lean (<0.5V), where the switchpoint is 14.7:1. If the NB is connected to MegaTune, then use it to tune your cruise area, which will be fine at 14.7. The idle and off-idle area is normally run open loop, via the "disable below XXXX rpm".

As for req_fuel, don't worry much about it. Just pick a reasonable value using the calculator, and tune your VE table accordingly. The req_fuel number gets multiplied by VE anyways, so it's the product of the two that matters.

muythaibxr

the leaning out is probably because you're running closed-loop with the narrow band. If you want to tune to anything but stoich in closed-loop, you'll need to do one of the following:

1) turn off closed loop and tune by hand while watching the AFR's on the wideband.
2) hook the wideband to the MS (if possible) and use AFR tables to keep the AFR reasonable, and use either autotune or VexME to make the table correct.

Or you can use a combination of the two approaches which is what I do.

rkcarguy

My engine will not run at 14.7:1 even just the car sitting holding it at 2500rpm, it leans out and stalls when the wideband reads leaner than 14:1. I thought it was a fuel pressure problem but fuel pressure is steady and as long as I'm accelerating it stays around 13:1 where I have the VE's set. Soon as I try to hold a constant RPM it goes lean and dies unless I pump the gas a bunch of times. I'm getting to the point where I'm going to unplug the damn O2 and see if it still does it.

renns

Seems odd that your engine dies at 2500 rpm leaner than 14:1. I've been running in the 15.5:1 to 16.5:1 range in light cruise (40-50 kPa, 2500-3000 rpm), and it runs just fine. What's your intake/injector setup look like? Do you have your O2 sensor after the collector? Is the engine warm, or are you trying to run closed loop during warmup? Like muy says, if you have a wideband that's operating properly, wire it's output to MS instead of the NB, and do your initial tuning that way.

rkcarguy

My intake is a racing beat Holley 4bl intake, on top of that I have a 1" carb spacer machined for the injectors, and then a custom upper plenum fed by a 240SX throttle body. The engine is fully warm when I am playing with it. I discovered that I was getting an intermitent signal from the o2 sensor possibly it is bad, when I unplugged it, it ran alot better. It's a little lean in the midrange and rich up top just need time to play with it and get it dialed in now.

KrisGXL

If you find any problems with your setup, could you please let us know? I ask because I am having a little bit of a stumble under light cruise also(14.7:1 NB-closed loop), and I've done the obvious plugs wires, fuel pressure, egr....

Thanks,
Kris

muythaibxr

that slight stumble is probably because of how closed loop is being done.

It doesn't keep the AFR right at 14.7:1, but instead oscillates slightly above and slightly below 14.7:1. If it's stumbling, that's probably the point where you're reaching 16:1 or higher. I would suggest changing the "Ignition Events Per Step" setting in the EGO settings. Post an MSQ while you're at it, so I or someone else can have an idea of your settings so we can tell you where to go from here.

KrisGXL

Here is my MSQ. I didn't want to jack this thread or anything , but it sounds like we're having similar problems.(I hope)

Kris

rkcarguy

I noticed this too when I had my O2 hooked up, the AFR would go up and down. I'm not that sharp on the computer end of things but if changing the "ign events" will make it respond faster and hold it closer to the right AFR it may help. I got put on 5 10hr days this week so I haven't played with my car this week, tomorrow and sunday I'm going to torture it some more. Amazingly I've gobbled up almost 5 gallons of gas in my tuning efforts and I haven't even driven the car around yet! I also noticed my RPM signal may be spiking even though I used shielded wire and everything, I'm going to add the recommended 50K ohm resistor to the tach lead and see if that helps. Next sunday is the next autocross event I'm hoping to get this thing tuned and get out there and whoop some a$$.

renns

Stick a 100k pot in the tach line instead of the 50K resistor. Adjust it about mid-span and try it out. If you have staring problems, or dropouts, lower the resistance of the pot. If you still have noise at high rpm, increase the resistance. Once you find an optimum settings, replace the pot with a fixed resistor of the same value.

As for AFR going up and down around 14.7:1, that's completely normal in closed loop operation. That's exactly how most oems run their feedback systems, even today. You can turn ego correction off by setting the 'disable until' rpm to above your redline, but it's really not necessary. Get the car running decent in the driveway, and then go out on some quiet roads and drive it.

rkcarguy

I made some progress and got rid of the off idle hesitation. I played with the accel enrichment and idle VE's until it ran better.

Can someone explain to me in more detail about the accel enrichements and TPSdot do? And what results from moving their values up and down? If I slowly press the gas about 1/2 way it revs pretty well, if I just stomp the gas I have a huge lean bog. I have the first accel BIN at about 8.0 and then I've just estimated the other values up in +1.0 increments.

Generally my setup has been very fuel hungry compared to the starting rotary map posted on the MS site, my engine initially ran better on the first startup on the 350 chevy VE table that the MS is preset with to give you an idea.

This engine is in a 1963 MG MIdget I auto-x, it's not street legal and I have no where to even sit a laptop while I drive around the parking lot. I've done alot of tuning on the carb I got rid of and I know what the "sweet spot" sounds like, so I will get it close just revving it and watching the wideband. Here's a pic of the car.
Any help would be appreciated, next auto-x is coming next weekend and I'd like to get out there and take the FTD away from that Turbo Miata