"Digi-John" Megasquirt-2 Project

[Run8Stud]

My Digifant ECU was working mostly-ok on my '87 westy, however I could never seem to get the idle box to play nicely with the rest of the system, resulting in removing the idle controller completely. It would idle too high, or kick itself out of idle and never be able to settle back in, all of this despite having everything dialed in as per VW\Bentley. My Digifant also had a couple other quirky behaviors, so that combined with my insatiable desire to **** (mess) with stuff led me to my latest project: Megasquirting my van.

I bought a MegaSquirt II (MSII) from DIYAutoTune, pre-assembled. My initial goal is just a simple replacement; ie, if needed, I could just unplug the Megasquirt and replug my Digifant without having to reconfigure twenty other things. I'm happy to say I've achieved this. I've also been able to wire up the idle valve and get that working properly-ish.

I'm using the existing distributor and hall-sender signal to supply the MSII with the rpm signal it needs, and I've tee'd into the existing vacuum hose (the one that supplies the fuel PSI regulator) for the MAP sensor built in to the MSII.

For the other sensor and power connections, I made some quick-dirty jumpers that plug into the corresponding pins on the van's Digifant plug, each one soldered to its necessary pin on the MSII's DB37 plug. I had some 17\64 female crimp connections, and using a grinder, turned them into male connectors that slip nicely into the female Digifant plug pins.

It was a bit difficult at first getting the initial MSII settings correct, but eventually got it right. I'll post a photo of some of these settings in case the next guy is interested. After a couple weeks of driving around and tweaking settings\tables, I'm happy to say it runs pretty well, solving the problems I set out to solve, and creating some new ones.

I still need to figure out if my fuel mileage is better\worse; on my Digifant, I was getting 23-ish on the freeway at 55-60mph. I have the existing narrowband O2 sensor hooked up and running, so I would assume the mileage should be similar to what the Digifant was giving me.

I'll post some photos below with some more explanations shortly...
_________________
John G. - 1987 Westy 2.1 5-Speed

[Run8Stud]

The first step was to mount the MSII unit. I just screwed it to the sidewall. The unit is grounded via the wiring, so it's not a problem having it screwed to the wood here:

The next step was to secure the existing Digifant plug so that I could run my jumper wires to it. Please forgive the messy wiring, it's mainly for prototyping purposes; I ultimately want to get my hands on a dead Digifant ECU so I can scavenge the plug out of it, and then I can make a proper cable to take the place of this mess.

You can also see my MAP vacuum hose poking thru the body. The MSII unit has a built-in MAP sensor, so I just needed to run a vacuum line out to the intake manifold. I have installed a grommet since this photo was taken so as not to chafe the hose.

The MAP hose tees into the existing vacuum line to the fuel-PSI regulator, so all in all, this was a simple job.
_________________
John G. - 1987 Westy 2.1 5-Speed

[10degnorth]

I love seeing mega/microsquirt projects in the Vanagon!!! Props to you for making it reversible. When I did mine I totally gutted the stock harness which I kinda regret for the sake of longevity for the next owner.
_________________
1984 Wolfsburg Vanagon (Current Project): https://www.thesamba.com/vw/forum/viewtopic.php?t=773912&highlight=

1988 Microsquirted, Wolfsburg Vanagon (Sold): https://www.thesamba.com/vw/forum/viewtopic.php?t=708000&highlight=

Microsquirt Conversion Guide (Work now Discontinued): https://docs.google.com/document/d/1laXykNPbRYS8MT...2imqd2pnx7

[Run8Stud]

The black thingy I have screwed to the MSII case is an Ignitor, needed to fire the coil. The MSII can be configured to drive a coil, however, it requires some electronics work internally that I'm not comfortable doing, so another option is to use the existing ignition logic-output to drive an ignitor which fires the coil. This is a common ignitor unit and seems to work well with my van's coil:

The ignitor and the plug\pigtail for it were easily acquired at my local FLAPS. The ignitor normally mounts onto a heat-sink using heat-grease similar to how a CPU mounts. Not having a heat-sink, I simply mounted it to the MSII case using the heat-grease supplied with the ignitor, and it's running just fine, doesn't even seem to get noticeably warm at all.
_________________
John G. - 1987 Westy 2.1 5-Speed

[Run8Stud]

Here are the settings I'm currently using:

Below are the closed-loop idle settings (what controls the idle valve). These are still subject to change. It mostly works correctly, but needs to be fiddled with some more. The idle valve does more than just idle the motor, it helps during stick-shifting by allowing the engine rpm to drop more slowly, it also helps during startup, especially cold-startup. There are a number of different curves and settings necessary to get it perfect; I'm about 73% there I think:

I did need to run two wires from the valve to the MSII unit via my grommet I used for the MAP hose. These two wires are spliced to the existing idle-valve connector. I could not find any info on whether or not polarity mattered for these valves. Reluctantly, I tried it both ways and it operated the same both ways, so I guess in this application it didn't matter.

Not shown above is the Frequency: I'm using the 150hz setting and it seems quite happy, and feels\sounds like it's humming just as it used to with Digifant. The duty-cycle on these valves seems to be about 15% for closed, and around 45% or so for full-open.
_________________
John G. - 1987 Westy 2.1 5-Speed

[Run8Stud]

Below is my current Ignition table (showing total-advance). Digifant uses 35 degrees advance on top of 5 degrees initial advance at the distributor, so total is around 40ish. I previously found my engine responded much moBetta with another 5 degrees or so of advance, hence my total values below:

I derived this table from interpreting the 3D chart shown in the Digifant training manual. I really don't know if that chart is accurate, but it seemed to match what I observed with my timing light with Digifant hooked up. I also found some other timing tables from air-cooled distributors, and with some study on mechanical and vacuum advance and why its done, I came up with this. I'd be curious if and how I can make it even better. The engine seems to run fine, but I'm sure this table could use some work.

The 5-degree areas are so the engine will get down to closed-loop idle. If this part of the chart gets advanced too much, these engines will never get back down to idle-rpm, ESPECIALLY if you have the idle valve hooked up as Digifant seems to keep it partially open at all times. Even with MSII, I found it initially tricky to get the timing and idle valve settings correct so that I wouldn't get stuck with 1500+ rpm idle.

Below 1000 rpm, I have MSII's Idle Spark Advance set up. Digifant also did this. This is what makes your timing jump around seemingly erratically at idle, but it's in fact the ECU stabilizing idle rpm using timing. I have my engine set to idle at 935rpm (~52 MAP) and it seems quite happy there. Idle Spark Advance helps quite a bit when clutching into first, or when the AC compressor kicks on. You can cover a multitude of sins with this feature.
_________________
John G. - 1987 Westy 2.1 5-Speed

[Run8Stud]

Below is my current VE table; this table decides how much fuel gets squirted at any given RPM\MAP. This table was originally auto-generated using TunerStudioMS with values appropriate for a stock 2.1 WBX motor, and has been adjusted from there using the auto-tune feature. It is far from ideal at this point, but not too bad.

I'm only using the existing narrowband O2 sensor for EGO correction, and I have the MSII configured to use EGO correction below 70 MAP and above 1000rpm. Using EGO correction during idle causes my idle to get a bit touchy; these motors (or at least mine) don't seem to like to idle near stoich 14.7; it idles so much more smoothly a bit richer at what I figure is about 13.5.

I have this table intentionally richer than stoich so that above 70 MAP, EGO disables and the engine gets a richer mixture for higher-throttle applications.

My Butt Dyno indicates the engine runs as well as it did on Digifant, if not a bit better. I'll need to take a longer road-trip to get my fuel mileage values; at normal cruise, the O2 sensor is trimming the mixture towards stoich, so it should be just as good as the Digifant was.

The next step would be to install a Wideband O2 Sensor with its corresponding controller, and then I'd be able to more finely tune the VE table for different mixtures. But for now, Narrowband will do.
_________________
John G. - 1987 Westy 2.1 5-Speed

[DanHoug]

great job, documentation, and info!!! it'll be interesting to see this over the next month and as you head into colder weather, assuming you leave where it is always Sunny and 78!
_________________
-dan
60% of what you find on the internet is wrong, including this post.

'87 Westy & '89 Westy both 2.1 4spd

Past projects can be found at--
www.thefixitworkshop.com

[Run8Stud]

I've not been able to get the cold-start settings squared away just for that reason, the WX here is just too damn nice. We had 3 feet of sunshine this morning and it doesn't seem to wanna let up.

Temperature may also affect the idle-valve initial settings, so once it does cool off, I may end up having the high-idle-can't-recover problem again.
_________________
John G. - 1987 Westy 2.1 5-Speed

[Geato]

Very useful data. I have been kicking around my own design based on the Speeduino open source project. I want to get an engine on a test stand to get the course settings in place. Board is designed, enclosure prototype done, engine stand ... not so much.

If you want the connector used in the ECU, you can get them from Newark. Here is a link to the manufacturer showing Newark's price and link to purchase. https://www.te.com/usa-en/product-827050-1.html

If anyone is looking for the cable end, here is the part number. https://www.te.com/usa-en/product-827249-2.html

[DanHoug]

[ragnarhairybreeks]

Yeah Dan, Geato’s tips on this kinda of thing, golden.

Alistair
_________________
'86 7 passenger syncro, converted to westy pop top, project still in progress
'82 westy, diesel converted to gas in '94, now gone...
https://shufti.blog/
Old address still works...
http://shufti.wordpress.com

[Run8Stud]

Geato, thanks for the Newark link!
_________________
John G. - 1987 Westy 2.1 5-Speed

[MayoMan]

This is good information. I like the idea it is reversible as I reside in California and would need to pass SMOG.
_________________
~~~~~~~~~~~~~~~~
1965 Sundial Camper
1961 Bug
1986 Westy Syncro

[GoEverywhere]

Very nicely done! I've got a MS2 running in my Vanagon as well, but I opted to gut the harness and start over to do things like crank fired ignition and sequential injection. And my camshaft is too big to play nicely with the stock ECU anyway! Nice to see how you did it though!

How's the temp2 sensor working for ya? I had so many problems getting it to work with the MS that I moved to TheDubShop's temperature sensor, which worked much nicer.

[tjet]

That's a great idea not messing with the factory wires & connector. Ideally, you would need a connector that matches the ECU side for your new MS ECU plug

Bookmarking this

[valvecovergasket]

[Run8Stud]

Firmware is MS2/Extra 3.4.4

If I up my idle area too much, I sometimes get hung up around 1100 rpm or so and it won't fall back into idle unless something drags the motor back down. I had the same problem in Digifant if I had the dizzy advanced too much beyond the stock 5 degrees.

I could adjust the Idle Adaptive Advance feature to kick in at a higher rpm, say 1500ish, right now it's kicking in at 1100rpm (coincidence?? lol). I'll try that today and see if I can't up the timing values in the chart.

My only objection is during low-throttle crawling, I seemingly start getting too much of a timing jump between Idle-Advance mode and throttle mode, and it makes the van jumpy and starts a driveline oscillation.
_________________
John G. - 1987 Westy 2.1 5-Speed

[Run8Stud]

GoEverywhere, for both my Coolant (Temp2) and the air-temp sensor in the old AFM, I'm using the following values in the Thermistor Table calibration:

- 3-Point Therm Generator
- Bias Resistor Ohms = 2500
- 0c = 5500ohms, 50c = 800ohms, 100c = 200ohms

These values were interpolated off the charts in the Digifant training manual. The only one I had to guess was the bias resistor value. They seem to be pretty close to correct based on what I've observed during driving.
_________________
John G. - 1987 Westy 2.1 5-Speed

[Run8Stud]

MayoMan, I'm down here in San Diego. Thankfully my Digifant just recently passed smog fine, so I'm good for a couple years. I might just try passing on the MS2 next time. I'd probably have to enable the O2 sensor to trim the fuel at idle; right now I have it trimming above 1000rpm so as to keep my idle more stable, idling a bit richer than what the O2 sensor would allow.
_________________
John G. - 1987 Westy 2.1 5-Speed