Related to the Bowling & Grippo Fuel Injection system.
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

MSn-E

Post by P.Webb »

I'm taking the plung down the MSnE road. I got the S/W and the MT2.25 rev with the MSnE enhancements. I took my 8x8 table and adapted to a 12x12.

I'm going to run fuel only for now but I'd like to adopt the 12x12 tables now on standard hardware, MSn-E firmware and MT2.25 to get my feet wet with it before I implement ignition.

This is in anticipation of installing the EDIS in my car. Things are on their way...

Links to what you need...

http://not2fast.wryday.com/megasquirt/m ... 25b296.zip
http://megasquirt.sourceforge.net/extra ... ra020i.zip
http://megasquirt.sourceforge.net/extra/setup.html
http://megasquirt.sourceforge.net/extra ... lonly.html
http://megasquirt.sourceforge.net/extra/setup-edis.html
http://megasquirt.sourceforge.net/extra ... dware.html

You can upgrade MT to 2.25.326 at:

http://not2fast.wryday.com/megasquirt/m ... 25b326.zip

Also check out the latest MSLVV with some great new features:

http://not2fast.wryday.com/megasquirt/m ... v225b9.zip

I'm going to try this out tomorrow. I'll post how it went after a couple of test runs.

-Peter
Last edited by P.Webb on Sat Apr 30, 2005 3:28 pm, edited 1 time in total.
The true joy of an Alfa Romeo is behind the wheel.
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

Post by P.Webb »

After a little trouble, I got the MSn-E working with the 12x12 tables. It's a lot more complicated than the 2.98 and MT2.20 combination because of all the extra spark related parameters and figuring the interpolation to fill the 12x12 table retaining the key points of the 8x8 table.

The latest msq is posted on my site per the link above.

What I learned...

1) When you download the MT2.25, download the 296 then the 326 over it but retain the msns-extra.ini from the extra020i download because the 326 release is looking for extra021 signature. This results in the tables getting corrupted because of the signature mismatch.

2) Update the code in the MS before connecting MT2.25 to it. MT has some garbage in the variables that only get initialized from the MSn-E firmware on connection.

* Connect to the MS and save the current config as msq
* Under tools, select 'dump' to save the settings
* Insert the boot jumper and power cycle
* Use download.exe to flash the MS with the MSn-E 020i code
* Copy msns-extra.ini from the code folder to the MtCfg folder
* Remove the jumper and power cycle the MS
* Start MT and check all the default parameters
* Open the dump file from the 2.98 config and set them in:
Constants 1
Enrichments 1
VE Table 1
* Don't worry about the other tables for DT or MSnS
* Set the code config to fuel only, turn off all spark options

Start the car and see how it runs.

Hope someone gives it a try.

-Peter
The true joy of an Alfa Romeo is behind the wheel.
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

Post by P.Webb »

Replying to myself again...

Spent the last 2 days tuning the 12x12 tables on the MSn-E. It was difficult to get the resolution right. Finally got it running well this afternoon. I had to advance the timing quite a bit to get good vacuum @ idle due to the cams. It idles between 1000-1100 RPM @ 53kpa give or take (fluctuation because of the reversion on overlap).

On top end rebuild, I took another .040 off the heads to gain a point in compression. I had to leave the keys out of the cams to get them timed. I think the cams were a little off before because of the head thickness and using the keys. I guess they were close enough but not right on.

If anyone is interested in the MSn-E config and my MSQ file, drop me a line.

-Peter
The true joy of an Alfa Romeo is behind the wheel.
User avatar
Steve R
Platinum
Platinum
Posts: 237
Joined: Wed Nov 24, 2004 1:06 pm
Location: Dorset UK

Post by Steve R »

Peter, I've been working on mine recently too, but last night I backed off to a prevous working version fuel only set up whilst I think about the installation further.

My version will keep the existing distributor set up and use MS to trigger the ignitor to fire the coil. Purely because I wanted to keep changes minimal and looking stock from an aethetic point of view. (although I agree EDIS is a better system) The GTV6 ignition unit isn't directly compatable so I have another ignitor, coil & wiring from a scrap VW, plus I have a spare dissasembled GTV6 distributor to weld up and lock down... none of this is fitted yet whilst I sort out the ECU side first.

ANyway, I had similar problems with the MT versions to those described above, but got around it by copying & pasting config files too - can't remember exactly what I did but Peter's description sounds very very similar. Got the MT working a couple of weeks ago and built a 12x12 table & also constructed an initial 12x12 spark table too (based around the Alfa's stock ignition curves a s a starting point).

However, the real problems came when I re flashed the new MS code to the ECU. All seemed to go OK, and after some trial and error re megatune versions and signatures I had a fully configured version. Tried it in the car & no start. :(

So I reflash the ECU again, then load my config & all seems OK on the PC. I power cycle the MS and view again with MT to ensure all is OK, but it's not, the values displayed indicate something has corrupted, plus my centre LED is flashing on/off once per second (never done that before!), obviously won't start.

Tried this a few times and each time it got corrupted, hence I went back to a previous .s19 file and my most recent working config. With this in place the car starts and runs perfectly as before.

I have to think this through a little more !! I'm sure it' something I'm doing wrong.

Oh by the way, I've never been able to get "download.exe" to work, I use hyperterminal - perhaps that's the problem?
83' GTV6, 3.0 24v supercharged
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

Post by P.Webb »

Yep, I had a problem using hyperterm flashing the MSn-E code. The center LED flashing indicates a config error for MSn-E so the code is loaded.

I had the EXACT same scenario. Here's how a solved it and you need to do these steps exactly and in order. The problem is with the 020i and 021 MSn-E signatures. MT2.25 starts in a default 021 signature with the msns_extra.ini. And BTW, the dump feature in MT2.25 with MSn-E only dumps 8x8 of the 12x12 tables. I need to email Eric about that.

Download the MT2.25B9 and unzip it. Don't use the setup download.

Change settings.ini to unset MS-I and set MSEXTRA

Download the complete zip of msns_extra_020i.zip

Flash the ECU using the download.exe supplied in MSn-E zip

Copy the msns_extra.ini from the 020i zip to the MT MtCfg folder

Start MT2.25

Connect to the ECU. There's garbage in the 115 byte tables in the initial start of MT until the ECU fills them out with defaut values and turns MSnS on (spark) by default.

Input your constants from the dump from MT2.20

Input your enrichments from the dump from MT2.20

Input your 12x12 fuel table. I did manual interpolation to figure out some starting points. I set the RPM every 500 but my MAP points are a little strange because of the lack of vacuum from the cams.

Input your 12x12 spark tables.

Check all the fields in general and code config tabs for good (default) values. Make sure it's set for table 1 only (not DT) and for fuel only (all MSnS field set to off).

Save the msq file with the ECU connected.

That should get that problem fixed. It took me 2 days to figure all that out by trial and error but it runs now. Still working on the 12x12 fuel points in the table before I tackle spark. The wheel is turning out to be the hardest part. I have all the EDIS parts and once I figure out how to mount the wheel on the pulley, or drill the holes to simulate the 35 VR pulses it should be on the way. It's already snowed twice here so it looks like I have a few months to get this figured out.

Did you use the ML4.1 ignitor from the MSnS pages? The pinouts look similar. I was looking at the wiring diagram and noticed the ignitor is almost the same as the Motronic one mounted under the coil. I could literally install the Motronic ignitor and swing the wires from the advance box, which is located next to the ECU on the Milano, to the MS DB37.

That might be a simple way to get spark working immediately before I jump into distributorless. Though I really like the idea of dumping the distributor. It's a lot of work on the 3.0 because I have to install a 164 oil pump drive shaft to lock down the distributor drive so it doesn't ride out of its housing. If it rides up and loses contact with the gear, I lose the oil pump drive, and I hear that's bad :0

HTH

-Peter
The true joy of an Alfa Romeo is behind the wheel.
User avatar
Steve R
Platinum
Platinum
Posts: 237
Joined: Wed Nov 24, 2004 1:06 pm
Location: Dorset UK

Post by Steve R »

Peter, imbecile question time from me... :oops:

How does download.exe work? As whenever I try to open the program is momentarily flashes up a window which then vanishes before I can even see what it looks like, let alone use to reflash the code ! This happens on any of 3 machines I've tried.
83' GTV6, 3.0 24v supercharged
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

Post by P.Webb »

download.exe is a command line program. open a command line window cd to where the download.exe and s19 file is. type

download <filename>.s19 and specifiy a com port (just the number) if you don't default to com1.

I don't see why hyperterminal wouldn't work though. Make sure you do a 'W'ipe and then an 'U'pgrade in the bootloader.

Let me know how it turns out.

HTH

-Peter

P.S. I've attached the MSQ to this message. Not sure how the forum will deal with it but trying anyway...
Last edited by P.Webb on Sat Apr 30, 2005 3:29 pm, edited 1 time in total.
The true joy of an Alfa Romeo is behind the wheel.
User avatar
Steve R
Platinum
Platinum
Posts: 237
Joined: Wed Nov 24, 2004 1:06 pm
Location: Dorset UK

Post by Steve R »

I'll give the code another try over the weekend - presuming I have any freetime ! & feedback.

There's a range of coil ignitors that will work, basically any that have active dwell control and are "6 wire" units will be fine. Note these units have a 7pin connector, but only 6 wires are ever used. There's a list on the MS forum, but my experience from looking round a local VW only scrapyard is that many petrol engined 80's & early 90's VW's have suitable parts to salvage, I read that many Saabs of similar era do too. If the motronic unit is 6 wire it's highly likely to be fine, however the original bosch unit on my 83 GTV6 (p/n 0227 100 117) definately isn't suitable. Here's the link to the list :

http://www.msefi.com/viewtopic.php?t=4745

regards
83' GTV6, 3.0 24v supercharged
User avatar
Steve R
Platinum
Platinum
Posts: 237
Joined: Wed Nov 24, 2004 1:06 pm
Location: Dorset UK

Post by Steve R »

I also meant to say using distributor based ignition is in essense simple: Interupt the signal output wire from the sensor on the distributor & send to the MS, and the output from the MS is send to the input on the ignitor instead.

The devil appears to be in getting the signal recognition and output correct for a specific application as there's numerous permutations!

If anyone's interested, when (if!) I get it sorted I'll write up another "how to..." to ease the learning curve for anyone else who wants to try ?
83' GTV6, 3.0 24v supercharged
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

Post by P.Webb »

The 137 and 139 modules are from a 164. Looks like the Saab used them as well. Dwell and advance are ok but I really like the multi-spark of the EDIS.

Since the ignitor on the Milano is the same pin and plug configuration I was thinking it would be easy to just take the 137 module and put it in place under the coil and move the wires from the digital ignition module to the MS running MSn-E for a quick fix to do spark while I work out the wheel issues.

Right now it looks like modifying the pulley by machining in the teeth is the best way to go. I still need a quote from the machine shop to do the work and a good 3.0 pulley.

Anyone have a 3.0 pulley to sell? I suppose I could have them make one from scratch.

-Peter
The true joy of an Alfa Romeo is behind the wheel.
User avatar
Steve R
Platinum
Platinum
Posts: 237
Joined: Wed Nov 24, 2004 1:06 pm
Location: Dorset UK

Post by Steve R »

Got mine running on the latest MSnsextra code & megatune version and went for test run last night on 12x12 fuel tables. All is well, I need to fine tune some of the extra table areas - but only very slightly by the feel of things.

On a purely enjoyment note I'd like to add that yesterday evening was cold (1C), humid enough to be slightly misty, high air pressure (4-5 kpa more than usual), minimal traffic and the car absolutely FLEW, like wow! At least I bet the guy in the Porsche Boxster :oops: thought so hahahahahahaha :wink: !

Now that the firmware and software are running OK, next job is to interupt the ignition circuitry and get it running fully mapped, roll on the Christmas break.... :D
83' GTV6, 3.0 24v supercharged
User avatar
Steve R
Platinum
Platinum
Posts: 237
Joined: Wed Nov 24, 2004 1:06 pm
Location: Dorset UK

Post by Steve R »

Yesterday I got mine running with spark control from MS ! :D

However, there was a LOT of trial and error involved in getting it to actually run & it then ran like an absolute pig and nothing I tried made it come good. So for now I've swapped back to my normal dizzy whilst I figure out what to do next. :(

The MS forum suggest I upgrade the CPU code to a version that does "next cylinder" timing calculations, something for next weekend...

Anyway thought I'd just keep you advised of work in progress, nearly there.
83' GTV6, 3.0 24v supercharged
joey
Gold
Gold
Posts: 192
Joined: Fri Dec 03, 2004 5:36 pm
Location: in a state of limbo

Post by joey »

yep, it'll leave it to you two gurus to work everything out and provide us with a simple step by step guide on how to do it ! :oops:

hey i'm no pioneer, but i'll happily mooch off your hard-work, if that helps at all ! :lol:

now get cracking ! :wink:
bmacf
Gold
Gold
Posts: 114
Joined: Sat Dec 18, 2004 4:11 pm

Post by bmacf »

I'm a little confused about whether MSn-E is the same as MegaSquirt 'n Spark -E. Can anybody provide clarification? Right now I'm reading through msefi.com about MegaSquirt 'n Spark -E - do I only need to change the processor to an HC908 and make the software changes to use MegaSquirt 'n Spark -E?

Thanks,
Bill in MD
Bill in MD
'88 Alfa Milano Verde x 2
'77 Fiat Spider
User avatar
P.Webb
Gold
Gold
Posts: 182
Joined: Mon Nov 29, 2004 9:59 am
Location: Chicago, IL
Contact:

Post by P.Webb »

MSnS-E is different to the MegaSpark or MSnS code. It's a conglomeration of all the variants rolled together into one firmware. MT is configured to work with it through the .ini files for spark tables, dual table, MAPdotAE and other such enhancements.

MSn-E is the most complete of all the packages.

-Peter
The true joy of an Alfa Romeo is behind the wheel.
Post Reply