ALFAGTV6.COM

Post by **Michael** » Thu Feb 17, 2005 10:53 pm

Well it worked ... but I'm not sure which alteration fixed it

I replaced U4 Opto Isolator and installed MegasquirtnSpark-extra - my guess is that the latter fixed it. I know its not good strategy to make more than one change at a time, but I had some free time and could not resist installing the new U4 (with a socket) before testing MegasquirtnSpark-extra. The log shows some initial rpm fluctuation - down spikes at low rpm. This was during the time when I was adjusting the potentiometer (because the car was jerking). The pot is installed inline in the rpm wire. Lowering the restance fixed it - so now I can just measure the resistance on the pot and buy a fixed resistor to replace it permanently.
The rpm log is smooth now.
By the way, one of the biggest changes I found is the quality of the idle!! It will idle smoothly at 600 rpm's whereas perviously it has always had a pretty rough idle even ay 900 (with LJet I mean).

Here is the log:
http://alfagtv6.com/MegaSquirt/datalog20050217.zip

The MSNS-e code has a setting "False Trigger Protect" and though I can't find any documentation, I wonder if this is helping to smooth out the curve... I tried the Target AFR option initially, and switched it off after about 3 minutes since I really need to tune .. and it really works - I could see it pulling the mixture to the target I'd set in the target VE table - very cool!

I'll upload my files if someone want s to take a look at them:

VEX file
http://alfagtv6.com/MegaSquirt/Best_mapped12x12_001.vex
MSQ File:
http://alfagtv6.com/MegaSquirt/msns001.msq

This still needs lots of tuning - I've not done much yet since I was focused on the rpm spike issue and the stumble at 1600 rpms. With those (apparently) resolved, I can go forward with the tuning.

By the way, I had to subjectively mod steve's VE table from 8x8 to 12x12 - anyone know of a tool for doing this?

Cheers,

enzo · Post by **enzo** » Thu Feb 17, 2005 11:33 pm

Michael wrote:
By the way, I had to subjectively mod steve's VE table from 8x8 to 12x12 - anyone know of a tool for doing this?

try this one
http://1800philes.com/megasquirt/

Post by **Michael** » Fri Feb 18, 2005 9:56 am

Cheers!

bmacf · Post by **bmacf** » Thu Feb 24, 2005 4:09 pm

Hey Michael, congrats on fixing your tach spikes. Can you let me know what inline tach resistance you have?

I took a detour from my MS project to fix my exhaust (subject of a future and separate thread). I was sitting in my milano idling so long while tuning that the leaky exhaust was giving me a headache. That should be fixed so now I'm back to tuning.

Congrats and enjoy,
Bill

Post by **Michael** » Thu Feb 24, 2005 7:10 pm

Hi Bill:

I installed a variable resistor - its a 100KOhm resistor and I find it works well at about 47KOhm (where its set currently). I got it from Radio Shack. I noticed my MS box always buzzed a bit and this problem has been solved with the resistance set to where it is now. Anyone else notice this?

My MS project is also on hold right now as I'm working on the braking system.

Are you going to install MegasquirtnSpark-extra?
I can send you an MSQ if you like. Pretty rough, but it has most of the setting set.

Cheers,

bmacf · Post by **bmacf** » Mon Mar 14, 2005 4:51 pm

As a much belated follow-up, I installed MSn-E and no longer have tach spikes. I don't have the inline resistor like Michael has installed either.

Bill in MD

Post by **Michael** » Mon Mar 14, 2005 6:16 pm

Nice - it really seems like MSn-E prevents the tach spikes

http://www.alfagtv6.com/MegaSquirt/Msns ... 3_2005.zip

I also emailed you my MSQ file. It’s not very well developed as I only ran the car once before stopping MS work to install the brakes. It’s set for wideband, but otherwise is pretty generic.

Cheers,

Greg Gordon · Post by **Greg Gordon** » Mon Mar 14, 2005 6:26 pm

Bill, are you using Megasquirt with the big injectors? If so I am curious how that's working out.
Bye for now,
Greg

Steve R · Post by **Steve R** » Tue Mar 15, 2005 2:53 am

I have bigger 28.6lb high res injectors running via megasquirt and it runs great, in fact it has a more even idle than the stock small injectors.

& my 2c worth, the MSnS-e code seems to completely cure all tach spikes.

Big injectors courtesy of Greg, cheers !

ALFA GTV6 GP · Post by **ALFA GTV6 GP** » Tue Mar 15, 2005 5:45 am

The Lazy way out ? pre-assembled MegaSquirt (2.2) EFI ECU Fuel Injection controller w/Harness
New, tested, fast immediate shipping with 15ft harness .Sounds good to me !
Features of MegaSquirt include:

On board MAP sensor for NA or boosted engines upto 2.5bar absolute pressure!!
Speed density, Alpha-N, or even MAF via custom flash load.
Acceleration enrichment via TPS or MAP
Use any injector, high or low impedance
Log data on a PC at high speed for later analysis
Schematics, software all open source so you can customize it as you like
Compatible with output from narrow and wideband O2 sensors
Tuning software fully customizable and is free for download

Added features of this particular unit are:

The controller comes with a 15 foot long wire harness soldered to the DB-37 connector. Wire harness is wrapped with an expandable sleeve rated for 125C.
Aluminium case has 90 degree flanged ends, which make mounting easy.
Barbed bulkhead fitting is mounted on the case for the vacuum/boost hose, instead of connecting the hose directly to the pressure sensor. This reduces the risk of damaging the pressure sensor if the hose gets yanked. Also does not require opening up the case to attach the vacuum/boost hose.
All nuts are glued in place to prevent them from coming loose over time.
The crystal oscillator is glued to the PCB to prevent problems with vibrations.
The controller comes with the required DB-37 connector and the top/bottom hoods required for the wiring harness.

This unit can control ignition !

By way of a number of spin off software loads which are free to download that will allow this MegaSquirt unit to control ignition via Ford EDIS, GM HEI module, etc. There is tons of info on the net about this.

So Guys what mods need to be made to fit to a GTV6 with L jet or a Milano 3.0 with Motronic ?

ALFA GTV6 GP · Post by **ALFA GTV6 GP** » Tue Mar 15, 2005 6:02 am

More Pics of the pre -assembled Mega squirt(2.2)

David · Post by **David** » Wed Mar 16, 2005 12:33 am

I've just about completed my 2nd MS install, this time on a stock 2.5L 85 GTV6. I took the easy way out this time and bought a ready built unit. The GTV6 mods to the MS board are finished. No big deal as I had some spares from my first install on my 3.0L 85 GTV6. Just follow the GTV6 MS install in the technical section and replace/add the bits required.
This install I've wired the fuel pump mod exactly as per the diagram. Steve's mod worked fine last time so I'll let you know if it matters with the emitter or collector hole in the mod. It is easier to install than Steves mod as the transistors sit the right way up.

This time I'm using the L-Jet socket of an old ECU and mounting on an aluminium backing plate shaped to fit where the L-Jet ECU is mounted and leave the MS in the case.

The DB37 plug to the MS is to be wired to the L-Jet socket.

I'm also going to try inline resistors with the LowZ 2.5 injectors rather than the flyback board, again, just to see if it works.

Here are a couple of pics so far.

David

Post by **Michael** » Wed Mar 16, 2005 1:07 am

Nice mounting plan - is there enough space to connect the MS output cable to the LJet terminating plug?

I'm waiting for MS-II (3.0) - looks like it will be released any day now! When it is, I'll buy the kit and assemble it myself this time since its really not too difficult. Apparently, when the MS II box is built, it can just plug into the same connector as a 'seamless' upgrade. I'm also looking at this from the perspective of having a spare MS in the car in case something ever happens to the circuitry.

Cheers,

David · Post by **David** » Wed Mar 16, 2005 2:00 am

Michael,

I'm going to leave the plastic housing off the cable plug for now. The mounting board is actually longer than the L-Jet case as there is room to locate higher than the L-Jet but no room below. There may be enough room to make the back longer still and then use the plug complete. I'll check that when it's all together.

David

David · Post by **David** » Tue Mar 22, 2005 11:24 pm

I’ve completed my latest GTV6 MS build. As I mentioned in a previous post, I decided to follow Peter Florence’s Alternative Fuel Pump Diagram, taking particular care with the orientation of emitter and collector legs as per the diagram.

IMPORTANT - I have discovered that the fuel pump now shuts off after about 5 seconds if the ignition is left on and the engine has not started. If the engine stalls, again the fuel pump shuts off. I think this is what is supposed to happen and this was a safety concern someone raised in a previous post.

Can anyone with MS tell me if their fuel pump continues to run with the ignition on and the engine not running? Perhaps my first MS build was a bit dodgy.

I’m going to rewire the AFP circuit on my 1st MS board and see if I get the same result.

See below for a photo and diagram of the fuel pump mod. The install is easy as the transistors sit the right way up on the board. This may help others understand how the components sit on the circuit board. The diagram is a bit rough but I think I have the emitter & collector points marked correctly.

Michael,
It might be an idea to include the diagram in Steve’s GTV6 MS instructions in the Technical Section. I think it needs to be changed to match the new pinout if Steve’s OK with that.

Also, as you can see in the photo, I didn’t bother with the DB37 plug plastic case. I soldered all wires to the DB37 plug and the L-Jet socket pins and covered the connections with heat shrink insulation. It’s only a matter of removing the 2 L-Jet socket mounting screws to unplug the DB37. The Stim can then be used for testing the MS board, etc.

********
What NOT to do.
As an aside, the MS I purchased was partly built, just needed the FETS mounted and fitting in the MS case. I mounted the FETS on the case end and then when it was all finished went to test the fuel pump circuit. I hadn’t tried to start the car as I just wanted to make sure the fuel pump was turning on. Turning the ignition on and off several times, waiting for the pump to stop, etc, I noticed a strong smell of petrol. Pulled the spark plugs and found #4 cylinder full of fuel and fuel in all the other cylinders. Cranked it over (removed the coil HT lead and the fuel pump relay plug first) and fuel sprayed out of the plug holes up the walls of the garage. Sh#$%t

Turned out the Injector FETS I mounted were the metal backed ones, not the plastic case ones I used in the first MS and needed to be insulated from the case. The effect was all the injectors were spraying continuously the whole time I was testing the fuel pump circuit. The front pipes and muffler also had plenty of fuel in them. Thank goodness I didn’t fire

it up straight away. I let it air out for a couple of days and stuck a garden blower up the exhaust pipe to air out the exhaust system. Lesson learnt, read the instructions carefully and don’t assume everything will be the same as last time.
*****
I’m running the 2nd MS on my 3.0L with high Z injectors at the moment and will now work on adapting it for the 2.5L with low Z injectors using resistors in the injector circuit. My plan is to mount these on the aluminium backing board and wire them in between the DB37 and the L-Jet socket.

David

Help
The component diagram is a photo of my computer screen!. I have this as a pdf file but can't seem to be able to load it. Maybe someone can convert either the .doc or .pdf file to a format suitable as an attachment.

ALFAGTV6.COM

Success!

Re: Success!

Alternative Fuel Pump Wiring Diagram