• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

Planning on Megasquirt - Want to do it Right

Xovaan

research flat earth
Joined
Jul 31, 2008
Location
SF, CA
Hey guys,

I really want to Megasquirt my car for educational purposes and because it seems fun as hell. At the same time, I want to do it right: This means no shortcuts, good materials, and a satisfying end result that leaves the car better than when I started. Currently, I have no deeper academic knowledge than nomenclature and a basic understanding of what each component does, hypothetically and/or anecdotally. I understand VE, AFR, dwell, and a lot of other basic principles. But the primary goal of this project is fulfilling this knowledge so I'll be proficient in the inner workings of EMS, or at least my own vehicle.

The vehicle's drivetrain that I'll be megasquirting is a 1994 B230FT with LH2.4, EZ116K, TLAO chips, and a 012 MAF. Eventually this will lead up to me swapping in another built 16vt. I plan on having this vehicle be my daily driver (once it's daily driveable ;-) ). I've heard great things about MS3-Pro, but most of my information comes from recommendations and reading online. The car is registered in Tillamook (primary residence) so I don't have to worry about smog / alternative ECUs with this particular vehicle.


Some questions:


1. MS3-Pro with fresh wiring harness or MS3-Pro with module in ECU case and a donor LH2.4 harness: Which one would you go for and why? What are your choice materials for building either harness to at least OEM quality?

2. MAP vs. MAF. What are the advantages and disadvantages of each? I'm leaning toward a MAP sensor since it seems like a far better way to tune, but I'm open to any suggestions.

3. Controlling crank sensing. I've heard great things about Yoshifab's DSM CAS, and if I'm controlling spark with the ECU, this seems like a good way to fill the distributor hole with something useful and ditch the CPS on the flywheel. Thoughts?

4. What resources helped you guys the most when going through the same processes I'm attempting?

5. Are there any software suites worth looking into? Why or why not? What are the benefits of tuner suites initially and in the long run?



Feel free to make fun of me or add any insight that would help me with such a project. You guys have been a great help to me going from not knowing how to change oil to doing several engine swaps, wiring my car from K-Jet to LH2.4, and my also helping in my budget 16vt project (which I miss thoroughly). I feel this is the next step in learning how cars really work.

Thanks in advance! :)
 
1. MS3-Pro can be bought as a complete set including wiring looms and Tuner-Pro license. This way you could place the ECU in a chosen spot and wire from there.
The module needs a lot of work to make it fit inside an LH 2.4 enclosure but, in theory, the stock engine loom can be used for the most part. Some hurdles can be if you want to relocate a crank sensor or add COP ignition. This is when you need to work outside the stock loom and sometimes there are no wires available.
I do not have experience with the older Volvo's but in some cases some relays need different grounding or a relay added to keep the ECU from holding power through a powered port.

2. MAP is easier and gets rid of a restriction in the intake path. Using MAF you are free to choose any diameter MAF you can find though. I do not have experience with MAF so can't tell if it will work OK and if it adds complexity.

3. IMO if possible use the stock 60-2 trigger ring on the flywheel with the stock sensor in the stock sensor location if present on the engine. Will be the easiest install. Sensor should work with MS, correct me if wrong. But this will only give you a crank signal. If you want to add a cam signal for sequential use I think the older engine's don't have a physical provision for it except inside the distributer. This is where the DSM-CAS would be a nice option. I guess it is a proven setup.

4. I guess Google can be of great help. Lots have been covered on this forum already.
Make sure the wiring is 100%. Might prevent lots of troubles.

5. I guess the software provided with the MS3-Pro is your best bet to get started.
 
Running an MS3 in a stock ecu case with an old crappy harness is a horrible idea. Not only have you eliminated 95% of the usable I/O, but you've also created the potential for a thousand more problems associated with trying to use old crappy stock harnesses.

I've found MAF to be way better for daily driven applications, and I've tuned hundreds of turn key setups on speed density and alpha N. It easily compensates for changes in engine performance, modifications, and climate changes all by itself. To you, none of this matters if you can't properly scale injector latency. Simply copying someone else's data doesn't work, even if the source is reputable.

Crank sensing produces the most stable output, period. That said, nobody (that I'm aware of) has yet to replicate the numbers I made on Yoshifab's junkyard 16v setup (compared to a similar setup) with a CAS, and that's probably worst case scenario of spark scatter next to using an LH2.2 distributor. The DSM CAS solution was never meant to be something that held a tenth of degree of precision, it provides an easy reliable means of adding cam and crank signals to an engine using ordinary hand tools and little electrical knowledge. And for the vast majority, it works better than anything else they can come up with.

If you're looking into MS, there's no software suite. You use the software for the ecu you're running. EFI Analytics is the only software you'll use for MS.
 
Excellent responses, Johann and gross polluter. Thank you for the insight!

So what I'm reading, and please feel free to correct me if I'm wrong: MS3-Pro with a brand new harness and new connectors, MAF instead of MAP, and keeping the stock crank sensor in my M47 bellhousing is the way to go?

Gross, I see you're in San Diego. I don't suppose you'd ever consider meeting and discussing your thoughts? I'd buy you lunch for your time and you could laugh at my lemon. :)
 
Just about done with the install on my wagon, pretty much everything above nailed it. Haven't gone the MAF route yet (Nick, coming to WM and wanna set mine up on MAF...box is already ready and wired for it). Crank sensor works fine (at least on MS2V3). I actually did reuse the stock harness, added a couple wires to the ECU connector for the IAT and TPS return, so far minimal issues, but my harness is actually in good shape.

Nice thing is I can swap it back to LH for whatever reason in about 20-30 minutes, though underhood looks completely stock right now.
 
my recommendation would be to build you own harness using most of you stock sensors on the motor. since you engine already has a 60-2 crank trigger there is no need to modify a distributor and ms2 is capable of running your motor with it, with one ignition module built in to run the stock coil through the existing distributor.

you will want to add an intake air temp sensor vw 1.8t work well and are easy to install you just drill 2 holes in the intake manifold one hole needs tapped for a screw and one seals the sensor with an o-ring if you have a lh manifold with a cold start injector the sensor actually fits in that location with no mods.

ditch the maf, map sensing works great after all anyway and you will simplify all the intake plumbing and have more room in the bay.

you will need a wideband o2 sensor for tuning and tuner studio is well worth the $40 for tuning. its auto tune works great you can feel the car start running better as you drive.

wiring and connectors. get an 8' harness from diyautotune.com all the wires have good quality insulation rated for automotive use. and the wires are labeled so you know where they go

for connectors go to pick and pull and cut some off a newer car like an audi a4 for all your injectors, tps, iat, crank trigger. they have nice spring clips and seal to make water proof connections. get a big bag of 3m heat shrink and seal crimp connectors lots of pink some blue and a few large yellow these make easy reliable water proof connections. heat shrink everything slide 2 pieces of heat shrink down the bundle of wires before crimp connecting so you can just slide the heat shrink back up to the end of the wires for a nice clean look.

injectors get new, not "modified" injectors like bosch design 3 or newer seimens deka are great too. don't bother running the old low z type there are plenty of newer high z or impedance injectors that flow plenty for your future 16v swap.


also if you really want to learn build the board yourself you will be much more familiar with the components and what they do, at least read the whole build manual.

last tip go to msextra.com forums for advice, or find someone in your area who also has megasquirt to help you when you get stuck.
 
I have yet to run one with MAF functional, but based on the results of others I would think it's the way to go. Nick has done some damn nice tuning work on his cars, his Volvo with the B230FT, and with the 2JZ, and his Mustang with the 2JZ, so I would tend to look toward his advice as gospel in this situation.
 
Thanks for the info, guys!

Speaking of air temp senors... are there any sensors that fit into the stock locations that are considered of better quality than stock and better for Megasquirt? Since I'm moving away from LH, are there any better ways of monitoring engine components?
 
On the sensor side, I moved from Volvo sensors to GM sensors. Megasquirt is setup for them by default, yes you can calibrate the sensors but it is a bit of a chore in my experience. With the Intake temp sensor be sure to place it between the throttle body and intercooler you will want to position it closer to the intercooler to prevent heat soak. As for the coolant temp sensor depending on the bosses available in the head and or the block you will want to find one that accepts the sensor or you can drill out the existing hole and tap it for the gm sensor. If you can while building your harness avoid solder pigtails and get the connectors with terminals that you crimp on, it will make for a cleaner harness and less prone to issues.
 
Sounds good! Is there a general tool you'd recommend for crimping automotive connectors? Well, particularly one that doesn't cost upward of $200 like Paladin Tools or whatnot. :lol:
 
Sounds like a day out then. :)

Just put quick steer on my car so I'll need an alignment before I head anywhere, but lemme know what works best for you!
 
alternative to DSM CAS but same location?

would something like this solve the problem of the insufficient 24 pulses per rev on the DSM CAS?

http://www.automationdirect.com/adc...Medium_Duty_Incremental_Encoders_(Quadrature)
Lots of options up to something like 3600 pulses per rev

They have an A and B in quadrature but probably just use one of them, some of them also have an index which could be used as the cam pulse. A suitable adapter like the one Yoshifab makes for the DSM CAS could make it fit I suppose.

bit pricey but maybe if you got one with differential signalling that could help justify the cost?
 

They are beautifully built and have great bearings. The problem is they are not built to take vibration. The encoder is glass, there are bundles of un-secured wires and the components on the PCB's are often not glued down. Bolting that to an internal combustion engine is just asking for pain and eventual expensive failure.
 
24 pulses (12 pulses per 360 degrees) isn't insufficient by any means. The other user is making a bigger issue out of a different issue than it really is.
 
Running an MS3 in a stock ecu case with an old crappy harness is a horrible idea. Not only have you eliminated 95% of the usable I/O, but you've also created the potential for a thousand more problems associated with trying to use old crappy stock harnesses.

I've found MAF to be way better for daily driven applications, and I've tuned hundreds of turn key setups on speed density and alpha N. It easily compensates for changes in engine performance, modifications, and climate changes all by itself. To you, none of this matters if you can't properly scale injector latency. Simply copying someone else's data doesn't work, even if the source is reputable.

Crank sensing produces the most stable output, period. That said, nobody (that I'm aware of) has yet to replicate the numbers I made on Yoshifab's junkyard 16v setup (compared to a similar setup) with a CAS, and that's probably worst case scenario of spark scatter next to using an LH2.2 distributor. The DSM CAS solution was never meant to be something that held a tenth of degree of precision, it provides an easy reliable means of adding cam and crank signals to an engine using ordinary hand tools and little electrical knowledge. And for the vast majority, it works better than anything else they can come up with.

If you're looking into MS, there's no software suite. You use the software for the ecu you're running. EFI Analytics is the only software you'll use for MS.

I misinterpreted the comparison of crank sensing and the DSM CAS. If my flywheel doesn't have the teeth for VR and I'm currently using an LH2.2 distributor for sensing is it worth the extra effort to do a trigger wheel on the front pulley? I have a DSM CAS with Yoshi's adapter and MS3+MS3X v3.0, just waiting until I can have downtime on the car again to move up from MS1 extra v2.2. Just trying to get the current best practice, apologies for the thread hijack
 
Back
Top