Winter Upgrade - Megasquirt

[bobcowan]

Time to be thinking about winter upgrades. I like to plan early to minimize down time, especially with a big project.

My EFI works pretty good. Easy to tune, pretty accurate. But, there's two features I would like to have, and my Haltech F10 won't do them.

First, I want to be able to run E85 at the track. More specifically, E70. It's cheaper, and runs cooler. Race fuel is up to about $8.50 a gallon. E85 is $3.19. The haltech F10 can't support E85 because of the accelerater pump. The F10 limits the pump to 100% of the injecter pulse width. With E85, I need about 120%.

Second, I want better control of the ignition advance curve. The old mechanical weights and springs are OK, but we can do better.

There are some really nice systems on the market right now, like the BigStuff3. Sequential EFI, distributerless, Wide band O2 sensor with self tuning, and full igntion control. But, that's about $2,500-3,000! Yikes! How much E85 would I have to burn to break even there?

So, my current plan is to build a Megasquirt II or III. What I read on the web sites is a little confusing and contradictory, so I'm not yet sure which one I'll build. Sequential injector firing would be nice, but certainly not that important; batch fire works just fine. And I don't need an idle control motor; might be nice, but I don't really need it.

What I do want is to use the Ford EDIS-8 distributerless ignition without an external exciter. And I want wide band O2 without an external control box. I think I can get both of those on II, but I'm not sure yet. To get both of those, I might need to step up to a III.

One feature I think could be real handy would be a dual map feature. To be able to switch between two fuel maps with the flip of a switch would be really nice. Like 91 pump gas and E85 street tunes.

Which makes me think "Why Not?". Factory dual fuel vehicles use a sensor in the fuel line to tell the computer what mix of fuel is being fed to the injecters. But I havn't seen any of the aftermarket computers with anything like that. I wonder why not?

To meet the budget, I'll probably build it myself. That's a pretty daunting task, so I'm not positive on that point yet. I havn't done anything like that in 40 years - and that effort didn't turn out so well. Remember Heath Kits? It's probably worth $150 to have a pro build it. But I'm a cheap bugger, so we'll see.

I read through various build threads, and that was pretty helpfull. And I'm spending time reading on the various megasquirt boards. When I understand it better, I'll start ordering parts. They need to write a "Meagsquirt for Dummies" book

Doing some preliminary shopping, I can do the whole job with all the features I want for about $600'ish. That's not chump change, but it's a whole lot cheaper than BigStuff3 or something similar. And that includes the new ignition system.

As I read through the on line manual, MS-II will manage EDIS-8 directly, with no exciter boxes needed. It only needs a crank signal. You need a cam signal for COP and independant coils. But EDIS is a wasted spark, and you can use the factory 32-1 trigger wheel system by itself. MS can even supply coils with a standard GM style cable connecter instead of the factory clip on style. That way I can make my own cables and route them the way I like.

I ran across a company that can accuratly mount the trigger wheel to the back of any balancer. Since I'm using a zero balance racing part, that might be the way to go. You can also use the factory crank sensor on a factory front cover; I like that idea. It will probably be more reliable than the Mallory Unilite I have now.

[Luce]

I'll be running megasquirtII soon. I haven't started my engine yet, but I have built the Megasquirt and runs on the simulator board. I managed to solder it together and upload the firmware and base maps with little drama, but I wouldn't recommend that challenge unless you have a nice solder station and experience. The board is fairly densely populated.

What I like most about the system is it's not at all "black box", and there are additional outputs that are easily configurable. I'll use one of the extras to override my fan speed controller if temps get too hot.

I'm keeping the mech advance and distributor with a MDS 6A box to give a base curve and multiple sparks, but will also let Megasquirt "adjust" the timing for the vac advance and fine tuning. I will also have a breakout box so if the MDS box fails, with the flip of a switch, I can be back on the road with Megasquirt's coil driver.

I highly recommend DIYAutoTune.com Megasquirt Kits / Assembled Engine Management Systems, Wideband o2 Sensor Systems and tuning products as a supplier.

But take this all with a grain of salt...it hasn't run yet.

I'll be watching to see how it comes out.

[bobcowan]

Why did you choose II over III? II is reasonably priced, and has pretty much all the features I want. It will drive a Ford EDIS module. But it appears the III will drive the coil packs directly, and won't need the module.

Unfortunatly, both the II and III need an additional module for a wide band O2 sensor.

[lovehamr]

Luce, I don't know if it will matter to you but you could also use the MSD6AL Programable and a locked dizzy to set your curves. They even come set up for a MAP sensor. I'm using one but in a carbed application.

[Luce]

Yea, I could also lock the dizzy and do all of the timing curve in megasquirt too. The problem I see with that is, if you advance or retard too much, the rotor may fire to the wrong plug.

I didn't want the MSD box handling the timing. Then I would have 2 computers to hook up to for tuning.

MS3 has a few more features than MS2, non of which sounded like things I would use. Sequential injection's only advantage is better idle and lower low speed emissions. If you're making any power, sequential is pouring half of it's gas on a closed valve too.

MS3 also has logging and a few other neat things. And it's surface mount components, so the board comes pre populated. But I don't think either of the MS units do multiple sparks. That's the main reason I went with a MSD box.

[bobcowan]

I don't think MS does multiple sparks, either.

I planon using the Ford EDIS-8 system, and getting rid of the distributer from the ignition system. That's supposed to good for up 8,000 rpm's, and a good aftermarket coil will produce a really hot spark over 7,500. Although CD boxes are available, I'm not sure they're a big advantage with EDIS - especially considering the cost.

Regarding the EDIS, I found this in the MSIII documentation: "It is important to be aware that while Ford EDIS and GM DIS both have special toothed wheels, the module handles all the decoding and presents a signal to the Megasquirt that looks like a distributor input. With these two modules, the Megasquirt does not know or care how many teeth are actually on the wheel, so do not use the "trigger wheel" setting. This also means that normally you cannot use sequential fuel with these systems as no engine position information is available to the Megasquirt. "


Looking at the comparison chart, I'm not finding any compelling reason to use III. And some good reasons to use the II. Oddly enough, the II does not have a fan control, and the III does not. How odd. But, you could probably configure one of the switch outputs (of which there are many) to control the fan relay.

Megasquirt Product Comparison

I'll probably save the dough and use the II, with a Ford EDIS module. Decent price, good performance, and good reliability.

[Luce]

Well, MS2 doesn't have a hardwired dedicated fan control, you can add a 2n222 transistor to (I forgot which one I used) one of the 3 LED's and set it to turn on at a temp point and off at a temp point and wire it to a relay. The setup is super easy in Megatune. It works with my sim card. I made a controller box to PMW my fan below meltdown, and incorporated a relay so if it fails, MS can step in and save the day. Lonestar Classics Forums • View topic - Using my other hobby.

For 40 bucks or so, you can get a simulator that has an oscillator that acts like a distributor signal. An it has 5 or so pots that pretend to be the air temp sensor, coolant sensor, O2 sensor, tac speed, and throttle position. With this, you can simulate all of the sensors to verify that the box is working.

If you wanted to, you could set a relay to turn on the blinker if the RPMs were above 5000, water temp is below 200, and atmospheric pressure is above .95 KPa. That's what I like about MS.

[Luce]

I think the ford EISD is super cool, but my 427W's HP peak is below 6000 RPM. My thought was the multiple spark and a top notch dizzy would give me a better light and not run out of steam before redline. And, I'll have the MS ignition control that can be subbed in for the MDS box if it crapps.

[bobcowan]

I'm currently using a Mallory ignition, with CD box. It has mechanical advance only. It does a pretty good job. But with such a radical engine, I'm pretty sure it will run better if I can get better control of the ignition events. Since I want to upgrade the computer anyway, I figured I could kill two birds with one stone.

My peak HP and torque is right around 6,000 rpm. It will still make power above that, just not as much. I currently have the redline set a 6,400 - that matchs the hydraulic roller cam. But I'm seriously considering a solid roller cam for the next overhaul. Although I can't use it today, I want an ignition system that will handle 7,500rpm's.

I always carry spare ignition parts for the Mallory. Failures are not uncommon. I'd like to use something more reliable.

Besides, upgrades like this are fun!

[Luce]

Unless you really want to get rid of the distributor, you may consider my option.

I plan to have the dizzy hooked to the megasquirt, and it feeds the CD box. I also don't have the vacuum advance in my dizzy, but megasquirt can add it. And with the flip of a switch, the megasquirt can be hooked directly to the coil instead of the CD box if it (the CD box) fails.

If I remember correctly, the megasuirt sends a signal to the ford DIS to control the timing. You're gaining 3 more coils with the DIS system, but I don't know how much real control you're gaining over a good distributor. I'm keeping the mechanical advance on the dizzy, and just fine tuning it with MS, but you can totally block off the dizzy advance and control it all with MS. You can also use the dizzy with a crank trigger wheel with just 4 ticks per rev, but you have to be sure to keep the dizzy and trigger synced so it doesn't throw the spark to the next or previous cylinder. If I did that option, I would cut a big hole in a cap so I could twist the dizzy around so it was firing right when the rotor was on a terminal.

Yep, that's the cool part of MS. You can do it however you want.

[bobcowan]

I'm going to remove the distributer from the ignition altogether. With 4 coils, they'll have a longer dwell time, and should be able to produce a hotter spark at high rpm's.

I have to keep the distributer (for now) to run the oil pump. At next rebuild, I may use a 5.0 oil pump and drive shaft, and then use an Explorer stub.

My initial intention was to use the MSII to manage the Ford EDIS8 module. But the MSII can be set up to fire the coils directly, and not use the Ford module; rather than have a computer that controls a computer that controls the coils - the computer will control the coils directly. I like that Idea. Simpler, easier, smaller, more accurate. Cost is about the same, too. Maybe even a little cheaper.

I'm going to have a custom 32-1 trigger wheel made, and build a bracket for the sensor. That should cost me <$20, and should be dead on accurate.

I'm also considering a knock sensor. Not sure about that, though. I have heard they are not that accurate in a "noisy" engine. I'm considering a solid roller cam, and that may produce enough noise to activate the sensor.

OOne of my main info sites is DIYAutoTune.com . I'm also interested in the MicroSquirt. But the $99,900.00 price tag seems a little steep I'm also looking at the Sequencer, but can't really find much info on that.

Options I'm considering:
- Seperate connecter for the ignition control
- 2 barro sensors - I believe one internal is standard
-- and another can be hard wired on one of the extra inputs
- Fan control
- shift light control
---- These last two I think I can do myself. It's a matter of programming the outputs.

I'v been doing a lot of reading and research - which I enjoy. It's a challange, and I'm looking foreward to it.

[Luce]

I haven't seen a schematic to study and see if it's true, but it's my understanding with the MSD box it can output up to 480V to the coil, so it only needs 1/20th as much dwell to get the same spark as a simple make/break switch at high RPMs when it's cranking the volts. I'm sure either set up is a 10 fold improvement over points and condenser.

MS checks the barro on the one sensor before you crank to learn the current barometric pressure. For mountain driving and aircraft use, the second barro is almost a must unless you run closed loop and give the O2 lots of leaway to fudge the map.

The fan control and a rev only based shift light are trivial in the set up. If you wanted the shift light to take into account MAP as well (so it became more an economy shift light instead of a race type rev limit pre warning) I think that's even possible.

Keep us posted. I'm interested to see how your turns out. I met a guy at a car show a few months back who had MS on a pro street 60's Mustang with a Cleavor. I wish I would have got his email. There's not to many of us brave enough to step outside of the box of what Summit offers in EFI.

I think the knock sensor will give to many false signals too. If I was to install one, I don't think I would let it do anything more than light a bulb on the dash.

[Luce]

Oh, Colorado. I see why you would want the second barro. Or you could stop and start the motor every 1000' of elevation.

[bobcowan]

Yeah, I think what I'm looking for is not that difficult or complicated. But, the nature of MS makes it seem that way. You have to figure out the wiring, and figure out how to do the set to get the things you want. A little more complicated than something like Holley, where it's almost idiot proof. But it's a heck of a lot cheaper.

[Luce]

And idiot proof makes it frustrating and limiting to an intelligent person.

Once you get the MS in your hands, run it on the simulator while hooked to your laptop, you'll really see the beauty of it. My concern is having to tune it myself. From what I've read, all of the dyno tuning shops know how to tune a ford or chevy computer, and maybe fast or holley, but don't even know MS exists. And since it's under $1500 you and it are ****e.

Download Tuner studio and watch the first startup video. It's not interactive without the MS up and communicating, but you can see all of the things to mess with.

[bobcowan]

So true. The idiot proof set ups are nice, until you build an engine that's out of their parameters. Which is the whole reason I'm upgrading.

Finding the right tuner is really important. Now I'm using a Haltech F10 ECU. At first, i rally struggled getting it set up right. Then I found a dyno shop with a tuner that knew the system. He had it tuned perfectly in no time.

I'm going to start by creating the new fuel maps by manually copying the old fuel maps. After that I'll let the computer automatically adjust and update the maps with a wide band sensor.

Then I'll set the ignition curve fairly conservative. After the fuel maps are fine tuned, then I can fine tune the ignition map. I'm tempted to leave the Mallory distributer in place until the fuel maps are tuned, then install the coil packs and crank trigger. That way, if the engine doesn't start, I'll only have to trouble shoot one system.

[Luce]

I'm getting really close to starting mine. I have the engine harness complete, and the megasquirt/MDS under dash part wired. I included a 4 position weatherpack connector so I can unplug the MDS box and let megasquirt directly fire the coil. I've read too many stories about the MDS boxes randomly crapping out to not have some suspenders with my belt. But I will say, the MDS box is a effin frlame thrower. Instead of a little tick, the sparkplug sounds like morris code with a test tone. If actually makes the spark plug warm to the touch after a a few minutes.

I have my home made stainless tank installed and plumbed, and no leaks when I energize the pump. My temporary low buck (under $200) side pipes are welded up( for the most part)

Anyway, the dizzy is plugged into the harness and spun with a drill. The injectors tick (pump is disconnected for now) when I add their power, and all of the sensors are giving reasonable and stable readings. I have my home made stainless tank installed and plumbed, and no leaks when I energize the pump. I need 6 more hours of welding and grinding on my side pipes. The body is only on with 2 bolts right now, so I'll pull it off, add fluids and fire it up in the next 2 weeks.

So far, the MS project has been fairly smooth. I think if I can get it to start and idle, it's all over but the crying. I think the tuning will be the easy part. The laptop and an o'scope lets you easily find all of the problems. I did have to chase down one problem with the spark control setup, but it was trivial. The output needs to be set the opposite what the docs recommend to trigger the MDS and I installed the pots backwards on the board, so you turn them the opposite way to adjust the VR trigger. I do have a little more noise than I hoped for, so I have to run a little higher hysteresis than I would like, thus requiring a faster cranking speed before I get a tach signal, but if it's a problem, I'll add a HEI module on the engine to amplify the dizzy output before it goes into the harness. We'll see after I get the oil pump primed, dizzy on and starter bolted up and wired to the batt.

I'm dieing to hear this beast!!

Edit. I added pics of the engine with harness and accessories. I made all of the brackets for the Alt, A/C compressor, idler and coil.

[bobcowan]

Glad to hear you're progressing well. So am I.

I go all he parts from DIY a couple of weeks ago. They were pretty easy to deal with, no surprises, fast service. And Matt has been very helpfull with all my silly questions. I would recommend them to anyone.

I ordered a MS-II Extra, with 3 additional coil drivers, and dual on board MAP sensors. I had DIY assemble the ECU, using a v3.57 main board. I have no experiance in that sort of thing, and didn't want to screw it up.

I bought their coil packs, with a standard HEI tips; I also ordered a spare for the trunk. And they sent me a few other odds and ends that I needed.

I bought an assembled JimStim board. That gives me the oppurtunity to plug the ECU into the lap top, and play around for a while in the comfort of my living room. I paid for the registered verrsion of TunerStudio - no sense spending all this time and money and going cheap on the software. It really has been enlightening. Learning about the software long before engine start is pretty important. Programming the spare outputs for fan, shift light, and tachometer was pretty easy - just a couple mouse clicks. No external circuits are required.

I spent the last couple of weeks poring over the software and documentation. Before doing anything, I wanted to make sure I understand it well. I have a pile of notes and schematics for referance. I want all this to be well documented, in case I have a problem, or decide to sell the car.

I finished fabing the trigger wheel and Hall sensor mount last week. They came out pretty good, I think. I started with 3/16" steel for the base plate, mounted it at three points, and then double braced it in two directions. Probably a lot heavier than it needs to be - but that helps stabilize the timing. I bought the Hall sensor with a threaded body; setting the air gap should be pretty easy.

Today I finished fabbing the mount for the coil packs. That took a long time. I tried a couple of things, and then decided I didn't like them, and started over a couple of times. It had to be easy to remove for repairs and maint. And I also wanted to incorporate the mount for the airil seperater. Three bolts, three hoses, and two plugs, and the whole thing comes off as one unit - seperater, hoses, coils, bracket, and base plate.

Tonight I started building the wiring harnesses. My initial intention was to mount the MSII in the same place as the Haltech, and then incorporate the existing harness into the new one. But I think it will be easier to just build a whole new harness from scratch. Then I can mount the ECU in a better spot. I'm going to put Weatherpack connecters in strategic locations to make engine removal a lot easier

I also have an LM-1 I want to use for initial tuning. TunerStudio can read the signal from the LM-1, and make changes to the fuel maps to meet the target AFR. Once that's done, a simple narrow band sensor should work fine for day to day driving. I'm going to put a Molex connecter under under the dash for the LM-1, making it a simple plug in to switch from one to the other.

I'm probably 2-3 weeks away from engine start. No hurrey.

[vector1]

how much money so far or how much do you plan to spend? this is just the ignition correct?

[bobcowan]

So far, I'm for about $1,200'ish. That's the ECU custom built by DIY AutoTune, Coil packs, Simulater board, and few other parts. I also bought a Weatherpack connecter kit and crimping tool from them. Most of the wires for the harness I bought at a surplus store. Total cost there was about $15 for a whole bag full of wire in differant colors and sizes.

$1,200 sounds like a lot of money. But that's about 1/3 the cost of a similar system from other companies.

I spent last week end building the wiring harness, and re-doing some of the dash wiring. I'm about 1/2 way done. Building this harness is a lot of work. Soldering the wires to the connecter was not difficult, and only took a couple of hours. But keeping everything organized, building in some special connecters, etc, makes it a big deal. It's not just building a harness, it's adapting the new harness to the old components.

I have not yet installed the computer in the car. I need to make some mounting plates to make this fit where I want it to.

It's a lot of work, but so is building a car. I'm not in a rush, and I want to do it right the first time.