Matt Cramer

Thanks for posting the data log. I see that everything seems to be operating normally from the electronics' perspective, but it's like the fuel suddenly gets cut off. I think it may be that fuel flow drops off with pressure. A system that can deliver the right volume of fuel at typical carb pressures may not be up to that at injection pressure.

Could you post the whole datalog file, not just a screen shot?

It will need an RPM signal to get a spark. You can try turning the distributor slowly and seeing with a voltmeter if you get a 0-12 volt signal on the tach input line.

What sort of signal is it they're missing? Are they not getting 12 volt power, or are they not getting a ground?

One trick I've used to avoid resets has been to put a radio noise filter in the 12 volt power line. Definitely cut down on the number of resets on my Dodge Dart, which has a very noisy electrical system. The engineers didn't expect it to power any electronics more sophisticated than an AM radio...

The difference is the programming on the Megaview's CPU - it needs an external programmer to change the code, unlike Megasquirt. I'll see if we can get a few of the MS-I kits reprogrammed.

An intercooler isn't really a precision machine - as long as it doesn't leak, you can copy an established design and reasonably expect it to work. But for something that spins at a hundred thousand RPM and glows red hot? Turbos are a lot less forgiving when it comes to sloppy tolerances. There are some really over the top horror stories, such as the infamous SS Autochrome glued together turbo. Some of the others just have porous castings, factories that don't know what a compressor map is, and questionable bearings.

You can use several different pins depending on how you have wired the Megasquirt internally. I see you are using MSnS-E - you will need to add the spark output B mod, but you can send it to any available pin you wish. I'm using pin 29 on my Megasquirt. The B pin is an override signal. With no voltage to the B pin, the eight pin HEI module acts like a four pin, firing the coil whenever it gets an input from the trigger. Apply voltage to the B pin, and Megasquirt can control the spark.

The bias resistor is inside the Megasquirt. Unless you've deliberately swapped bias resistors, use 2490 ohms.

You can connect to a PC by just providing power and ground. However, you won't get the lights to blink that way, as it needs a tach input for those lights to come on.

Yes to both - the wideband can be used for automatic tuning and for displaying the AFR through MegaView. For using a gauge, you'll either use a programmed analog output or a digital output (for the Innovate XD-16) from the wideband controller instead of hooking it up to the Megasquirt.

While a wide band oxygen sensor will give you more data, and I wouldn't want to tune a turbo car without one, Megasquirt can certainly use a narrow band sensor. It was originally designed to work with one, before inexpensive wideband systems hit the market. If you go with a wideband sensor, you will also need a controller, like an LC-1 or an AEM controller.

The MegaManual has an explanation under Configuring Megasquirt-II. I'll give you a short run-down here. Trigger offset: How soon BTDC you get the signal from the ignition input. Depends on where the distributor timing is set. Skip pulses: How many pulses it will ignore before cranking up. Typically 3, and generally only affects cranking. Predictor Algorithm and Gain: How Megasquirt tries to predict when the ignition pulse will arrive. The explanation for this is very long. Next Pulse Tolerance: These values are designed to filter out false triggers. Higher values filter out fewer pulses, while setting this very low can screw up the ignition during rapid RPM changes. Ignition Input Capture: Should be Falling Edge. Cranking Trigger: Unless you are running around 10 degrees of trigger offset, set this one to Calculated. Coil charging scheme: Set it to Standard Coil Charge. The only time you'll change this is when you are using a Ford EDIS ignition, where the pulse duration controls the spark advance - normally, the pulse duration controls dwell time. Spark Output - Should be Going High (Inverted) any time you are using the VB921 to drive the coil directly. DO NOT CHANGE THIS. Maximum spark duration: Used for calculating coil dwell.

You can fit a Megasquirt-I with six coil drivers, but not a Megasquirt-II. The standard MS-II firmware only has one ignition output channel, and even the alpha codes I've seen only have four channels. The trouble is finding an appropriate ignition trigger - I know many of the Nissan CASs use the length of the slits for cylinder identification, which does not work with current Megasquirt firmware. MSnS-E can trigger off a missing tooth wheel or a combination of a one tooth cam trigger and an equally spaced tooth wheel. Or you could fit it with a Ford EDIS ignition - that's the most well documented and popular method for distributorless ignition.

The biggest difficulty with a VG30DETT is that the trigger uses the length of a slot in the CAS to determine the number one cylinder. Megasquirt works by sensing missing teeth or by using a second trigger input. I believe there's a trigger wheel from Autronic that will convert a standard VG30DETT trigger into one with an outer ring of six holes and a single inner trigger, perfect for use with Megasquirt. Alternatively, you can use a 36-1 trigger wheel and run a Ford EDIS system or have Megasquirt trigger a wasted spark setup. If you want to have Megasquirt drive the coils, you need one VB921 per coil. The Megasquirt already comes with one VB921 (it's installed in our assembled units, but not jumpered), so you will need two more for wasted spark or five more for a COP setup.

Ah, that should work pretty well.

What exactly is it triggering from?

I've got a picture of the hose installation in progress on my personal blog, including the part number for the specific type of hose I'd ordered. One thing I noticed is that this hose really kinks up when you bend it. It's pretty visible in the photo. In some other spots, like on the exit of the turbo, I'm using rubber radiator hose.

I'm using neoprene / fiberglass hose from an industrial supply house on my Turbo Dart buildup. It's rated for the appropriate pressure, and being flexible it resists vibration too. I spent something like $65 for seven feet or so of it. So far I haven't tested it yet, but I don't expect any problems other than possibly it having a little more restriction owing to the external spiral reinforcement.

You may very well have to adjust the base timing a bit with the new distributor.

It looks like the problem is that you're running a trigger angle of 70 degrees, but you've set the cranking trigger to Trigger Rise. When you have this setting, if the RPM signal is below the cranking RPM threshold, Megasquirt completely ignores the trigger offset and fires the coil as soon as you get an input from the ignition trigger. So your crank timing will be way off. Change that to Calculated.

So far, only MSnS-E firmware has any support for boost control. Some people have tried using the standard spare output code on MS-II for boost control, but usually this results in boost that bounces around a bit. From what I've heard, it's common for boost to bounce up and down by 1 psi if you're using the generic spare output code.

Ok, for a locked advance, here's something that should work. 1. Build the Megasquirt the same way as described in our regular article. 2. Disable all advance mechanisms in the distributor. 3. Run the L wire to pin 24 (Tach if you're on a relay board) with a pull-up resistor from a 12 volt source (the BW or BR wire should work). Disconnect the L, but not the BW wire, from the coil. 4. Remove the LY and LW wires from the distributor. 5. Connect the pin 36 (relay board S5) wire to the negative terminal of the coil. While I haven't tested this one, it ought to work.

The non-turbo distributor is a completely different animal from the turbo unit. If you are using Randy's method and leaving the distributor in control of the spark, you would wire the relay board's Tach terminal (pin 15) to the negative terminal of the coil / L wire, with no resisotrs, pull-ups, or anything. If you've removed your advance mechanism altogether and want Megasquirt to control the timing, you would wire this up differently. I can post one possible solition if you are interested in that. But you would have to lock the advance inside the distributor and disconnect the vacuum advance to do this.

Shielding never hurts - although in my Dart, I didn't run shielded wires and it does work.