Im looking to see if anyone has successfully assembled the $6E code from the hack listings that are available. I've been working with both the motorola assembler & c16 assemblers with no success yet. It seems there are a lot of edits required to get it to work with either of these assemblers. If you've been successful, what platform are you using?

 

I haven't done $6E, but have done others of this era. It did take a lot of editing to get the hac to assemble. Did this on a PC with the Dunfield cross assembler.

RBob.

 

Thanks for the reply and encouragement. Ikll keep plugging along. Only about 6000 lines out of 9442 left to edit. Its been quite a while since i’ve needed to write in assembler....

 

Here's an update if anyone is interested. I have it working. It took edits to almost all of the lines in the hac to clean up the comments and get it working with my Cross-16 assembler but it works well. I have revised the code to log both my right and left wide band O2's in the data stream and have improved the reliability of the switch from Double Fire to Single fire mode. Now my 427 idles & cruises in SF mode with the 42 lb injectors at reasonable pulse widths like it's supposed to.

Now I'm on to installing a 100mm ford slot Maf . I've worked out a simple signal converter to allow it to drive the existing MAF input on the 1227165 ECM if anyone is interested.....

RBob - thanks again for the encouragement.

 

I find this very interesting.

 

I'm installing the MAF this week - It probably should be in its own thread, so I'll start one for it. If there is anything more about $6E, post on.

 

A frequency converter? Did you use all of the available AD channels for the widebands ?

-- Joe

 

The (hitachi MAF 0012) ford slot maf is 0- 5 v analog just like the early gm maf. Problem is the output can’t drive the 1227165 maf input because it has a 1k pullup resister in the ecm tied to + 5volts. Two options to fix that: 1) open up the ecm and cut the pullup resistor or 2) build a converter. I made a converter using an LM324 op amp configured as a voltage follower, and a small 12v to +/- 12v power converter. This works great on the bench and should provide the option of switching between gm maf and ford maf without modfying the ecm.

I used the MAP and MAP2 inputs for the widebands. - analog channels 0 and 3 on the ecm. My code patch is where the egr code was, but it could go anywhere that 16 bytes will fit in...

 

I ran one of those briefly with my Megasquirt. I didn't realize the early MAF was 0-5 volts, I assumed when guys were running the Ford mafs using modified $32 and $6E they were using a spare ADC to read the MAF voltage.

I wasn't aware of the resistor on the factory MAF input.

What does blowerworks do? Build a converter, or use a spare ADC ?

The whole MAF thing is interesting. I couldn't really get it to work properly on my car, and I'm told it was due to my MAF location and tube sizing. But if you look at all the OEMs, they all use hybrid MAF/MAP. My LT1 car is a prime example, all LSx cars, etc. Surely GM wouldn't be doing it if it wasn't ideal.

Please share your results on the project. I know a lot of the Vette guys buy the blowerworks MAF and calibrations, and I'm sure it would be nice for a simple DIY method for those who want it.


-- Joe

 

Have you found an assembler that will work with 64 bit windows? The old stuff I used to use in the xp days doesn't want to work with windows 7,8,10...haven't found a good work around yet.

With an analog Ford MAF and 165 ecm, I like to use one of the available MAP inputs for the MAF signal.

I have removed the pull-up resistor for testing with an 870 ecm. It works, but data and knowledge of the $1F source is limited, so kind of a dead end. This could be a viable approach for the 165 ecm since the vmaf circuits are the same, but not convinced its worth the effort so I haven't continued down this path.

A signal converter adds complexity and may affect the signal quality depending upon the design.

Either method can work, just depends upon application needs and preference for working with software or hardware. Tuning will always be required. I would sacrifice one of the widebands and use the other MAP input for the MAF, but to each his own.

Curious about the single fire/double fire mode switch work...This transition can be pretty harsh with big injectors, so I'm interested if you've found a way to make this a bit more seamless, or if you can live with the occasional jolt.

 

Does a '165 have enough inputs for a MAF, MAP, and wideband?

Boost spark retard would be nice. If the spark advance table was MAP target, and maybe a few other things like like overboost protection.

-- Joe

 

Tequillaboy - thanks for your comments on this. Your posts and threads are what got me here. I probably wouldn’t have tried this otherwise.

I agree that the MAF adapter adds complexity and reduces overall reliability, especially if the adapter is in the engine compartment. My long term plan is to cut the pullup in the ecm once i’m happy with the new maf. The adapter is a very simple unity gain op amp to add drive capability - bench testing against a 1k load shows no distortion up to 4.8 volts which is beyond max airflow for my engine.

The problem i found with single fire mode was in getting it to switch back from double to single fire mode. Once the engine is started, PE mode puts it into single fire and only DFCO puts it back into single fire mode. So, while any robust accel maneuver puts it into DF, when you let off the throttle, decel enleanment doesn’t switch it back to SF. you have to coast for the minimum DFCO time to get it back into SF. This leads to accel into traffic (DF) then level off into light throttle cruise stuck in DF mode at rediculously low pulse widths & usually too rich. Sometimes, coming back to a stop it doesn’t switch back & tries to idle in DF @ minimum pulse width (very rich).
Coasting to DFCO does produce a noticeable jerk. More from fuel cut off (I think) than from the change in injection modes. My code patch is set up to switch down to SF when pulses reach ~ 3.5 and back to df if pulses reach 7 msec. This stays in the linear slope region for my injectors an should make the transition smooth. I also left the stock $6E transitions in place. In my limited test driving there was no noticeable jerk when it switches back to SF.

As for an assembler that works with windows 7+ I haven’t found one. My garage is running on XP & Vista.

I’m running NA for autocross & road course stuff. in order to run map with two widebands on the 165 you would need to give up some other analog input.

A couple of power train Calibration engineers stopped by my shop last year & looked over my issues. They recommended two things — 1. Know your injector flow characteristics & 2. Put widebands on both banks. For now im sticking with two wb.

 

Does the '165 have 2 injector drivers or one? I know they are 'tied together', but is that at a logic level or literally 1 driver to both pins?

I've experimented a lot with injector strategy with the aftermarket stuff, and found that double filter alternating bank to bank works best for large injectors.

In regards to removing that pullup, do you have pictures showing the process (from other folks that have done it) ?

-- Joe

 

http://home.earthlink.net/~cliff_har...7165sheet4.gif