The Project Pad

Has anyone concerted from megasquirt 1 to Megasquirt 3 or 3x? Buddy of mine offered me an ms1 he never used, and I've been thinking about converting from 034 IIB efi to MS, do to numerous reasons. I see there's a ms3 daughter board that is used to do the conversion on diy autotune. Is that all that's needed? What else is needed to complete the conversion? Thanks for any input

MS3X is all of the extra outputs that the MS3's processor can control internally. IIRC it *will* require a double-height case to fit the daughterboard.

My RX-7 has been Megasquirted since 2009-ish on one of the original "group buy" 1.01 units, it's on MS2/Extra now. I have no experience with MS3X but I've done a few cars with MS3-Pro. They all tune the same, you just get more native features as you go up the processor food chain. The unit that I was just playing with today has closed-loop idle control (both air and timing) as well as control over the fans, A/C compressor, torque converter clutch (2D load/speed map), fuel/timing/boost map blending between E0 and E85, a bunch of other features I can't even remember right now because I set it up in the spring. And it needed a couple tweaks to start after it sat outside when it was 5 degrees F. I had the start tables rmarkable close, actually, the fuel was good but I needed to add a little more cranking IAC.

The MS3-Pro unit, in my opinion, is way more better than FAST XFI, BigStuff3, or Accel DFI, the "main" other aftermarket systems that I'm familiar with. (Mostly the FAST system, as we mostly work with GM product and FAST is very popular with these guys) It's my understanding that MS3X is basically MS3-Pro without the nice case/wiring harness or the ability to change inputs/outputs entirely in software.

Noisy Cricket wrote:MS3X is all of the extra outputs that the MS3's processor can control internally. IIRC it *will* require a double-height case to fit the daughterboard.

My RX-7 has been Megasquirted since 2009-ish on one of the original "group buy" 1.01 units, it's on MS2/Extra now. I have no experience with MS3X but I've done a few cars with MS3-Pro. They all tune the same, you just get more native features as you go up the processor food chain. The unit that I was just playing with today has closed-loop idle control (both air and timing) as well as control over the fans, A/C compressor, torque converter clutch (2D load/speed map), fuel/timing/boost map blending between E0 and E85, a bunch of other features I can't even remember right now because I set it up in the spring. And it needed a couple tweaks to start after it sat outside when it was 5 degrees F. I had the start tables rmarkable close, actually, the fuel was good but I needed to add a little more cranking IAC.

The MS3-Pro unit, in my opinion, is way more better than FAST XFI, BigStuff3, or Accel DFI, the "main" other aftermarket systems that I'm familiar with. (Mostly the FAST system, as we mostly work with GM product and FAST is very popular with these guys) It's my understanding that MS3X is basically MS3-Pro without the nice case/wiring harness or the ability to change inputs/outputs entirely in software.

Did you convert your board from MS1 to MS2? In the description of the daughterboard for the MS3, (https://www.diyautotune.com/product/meg ... board-kit/), it says youll need to add the MS2 jumpers. Do you know anything about that?

I went from MS2 V3.0 board to MS3X V3.0. one main thing to look at is the color of the board inside the case. If it's blue, you are good to go, but if its green or red, you will have a lot of trouble with electrical noise.

If you are starting with an original MS1, there will be other things needed to upgrade not just the processor. Starting with MS2, like I did, there isn't nearly as much to change. Just remember, there is a lot of things the MS can control, but you have to set up the electronics for it. Hopefully Elaw will chime in here, because he is the one that built mine.

DE80q wrote:I went from MS2 V3.0 board to MS3X V3.0. one main thing to look at is the color of the board inside the case. If it's blue, you are good to go, but if its green or red, you will have a lot of trouble with electrical noise.

If you are starting with an original MS1, there will be other things needed to upgrade not just the processor. Starting with MS2, like I did, there isn't nearly as much to change. Just remember, there is a lot of things the MS can control, but you have to set up the electronics for it. Hopefully Elaw will chime in here, because he is the one that built mine.

Maybe if I were to send him my MS1, he could convert for me? Wonder if anyone out there could. I'm good at soldering and I know my way around electronics, but I'd prefer someone who has done it before do it to make sure it'll work

You will want to make sure that you have a good board first.

When I switched, I started with a new board because I fried some components on the 1.01 and there is basically zero support for it. But it's the same board for everything, really.

I bought a 3.57 board specifically because I wanted the reliability of surface mount components. Not that I ever had a problem with the kit based components, but the price difference was negligible.

DE80q wrote:You will want to make sure that you have a good board first.

^^ What he said! The v3.0 and v3.57 mainboards are good and fairly easy to convert for use with the MS3 processor and optional (and highly recommended) MS3X I/O card. The earlier mainboards, not so much. And what another poster said is correct... with the MS3 you'll need a taller case which like all the other parts is available from DIYAutotune.

Going from an MS1 to MS3 tune is a big jump too. It's not *completely* different but is by no means completely the same. Among other things, the fuel and timing maps are 16x16 vs. 12x12 for MS1. I did a conversion for someone from MS1 to MS2 and it ended up well but did take some work.

On the other hand, if you've ever felt limited in any way by the MS1, the MS3+MS3X combo (or MS3 pro) will blow your mind. It has so much more functionality it's amazing.

Oh and if you're going to go with the MS3, consider getting the RTC and knock boards too. The RTC (real-time clock) will allow time-stamping of SD card datalogs so you'll always know when a log was taken. The knock board is pretty much essential if you're going to use a knock sensor (or sensors: it can accommodate two). I'm not 100% sure but I think the MS3 pro has both of those built in.

To be pedantic, MS1 was 8x8 maps

This is the main thing that annoys me about MS2/Extra. I can't use an 8x8 map. There's an option for 12x12 instead of 16x16, so I went with that and invented a bunch of BS cells that I'll never actually use. But the log viewing software doesn't work with 12x12, it sort of throws up and tries to fit the 12x12 map to a 16x16 grid. When I report this as a problem, i get the typical MS community response of "WHY U NO USE ALL POSSIBLE FEATURES BLAH".

My engine could probably get away with a 2x2 map and run really well on a 4x4 map, it only runs between 50kpa and 95kpa and the torque curve is so broad that it's mostly the same cells from 3000 to 9000rpm, why should I be forced to run a 16x16 map if I don't have to?

Ah... you're probably right, for the MS1 B&G code.

I've been running the "extra" code for so long that whenever I think of MS1 or MS2 I'm really thinking of MS1/extra or MS2/extra.

And to be pedantic back at ya, suppose for example you tried to convert the VE table you posted to 4x4. That means the 3x3 area at the lower left would be converted to one cell. Looking at that area (from 600 to 1200 RPM and 20 to 32 kPa), the average of the values in those cells is 20. But at 1200 RPM and 32 kPa your VE is 24... that means that just using a single VE value of 20, the engine would be getting 17% less fuel than it needs at that point. An engine that's getting 17% less fuel than it needs may run, but won't run "well" IMHO.

Also keep in mind that while you may have an engine with a super-smooth VE map, not everyone does. I've never had it on a dyno but I bet if you carefully mapped the VE on my engine from 1200 down to say 700 RPM, the VE numbers would be all over the place. With a 4x4 VE map, my car would probably be undriveable.

And I'll make one more argument. Your problem is easy to solve: if you really want a 4x4 map, just select 9 cells at a time and set them the same, or adjust them all at the same time. Yeah the map may be 12x12 but you can work with it as if it was 4x4. The opposite problem, not having enough cells to meet your engine's requirements, is nowhere near as easy to solve... in fact it's basically impossible.

elaw wrote:Ah... you're probably right, for the MS1 B&G code.

I've been running the "extra" code for so long that whenever I think of MS1 or MS2 I'm really thinking of MS1/extra or MS2/extra.

And to be pedantic back at ya, suppose for example you tried to convert the VE table you posted to 4x4. That means the 3x3 area at the lower left would be converted to one cell. Looking at that area (from 600 to 1200 RPM and 20 to 32 kPa), the average of the values in those cells is 20. But at 1200 RPM and 32 kPa your VE is 24... that means that just using a single VE value of 20, the engine would be getting 17% less fuel than it needs at that point. An engine that's getting 17% less fuel than it needs may run, but won't run "well" IMHO.

It idles at 1200 if I set the idle really low. I can set it to idle at 900, technically, but at that point it is running at about 90kpa and it stalls easily, so I usually have the idle up around 1400-1500.

There are a LOT of unused cells in that map! I think at closed throttle at 8000rpm it might pull down to 30kpa. And the 10k map is completely redundant since I have a hard limiter at 10,5 and it's set to start dumping mass quantities of fuel at 9000rpm, like 25% extra or something. It's kinda neat, i had it on a dyno and the torque instantly drops by a significant percentage when that happens.

Of course there is a really easy solution... don't use farking speed density for airflow estimation and use a MAF like real computers do. That's next on the list of things I might do someday if I ever get around to it. New problem: You really want to use a frequency-type sensor like GM uses for noise/reliability reasons. MS2 only works with voltage type MAFs, not frequency type. MS3 will work with frequency. So upgrading to MAF sensing means going to MS3.

My Quantum, if I ever get back to working on it, will be running MS3-Pro and it damn well IS going to be running with a MAF sensor. I would much much rather tweak a MAF-transfer curve and an A/F table than all this speed-density nonsense Especially since it's probably going to need every kilopascal of a 3-bar MAP sensor.

Yeah, MAF can be sweet, especially for a n/a engine. It can make tuning a ton easier!

I actually put one on my car and life was great except when the compressor bypass valve on the turbo would open and cause the MAF to wig out. I tried to tune around it for a while and finally gave up. The sensor's still on there partly for datalogging purposes and partly because I'm too lazy to remove it.

You know there's a solution for that

I've only ever worked on one Buick that had a BOV and it caused way more problems than it solved. Really the only purpose of a bypass valve is to eliminate noises that customers might not want to hear in their nice expensive new car. Me, I think those noises sound pretty damn good. Hell, turbo chatter is half of what made Audi S1s so sonorously awesome...