Joe Perez

iTrader: (8)

Honestly, I could design it (schematic, board layout, and documentation) but I don't have time to do manufacturing and support. That is a task better suited for someone who is already established in the MS community. I'm just trying to provoke thought and maybe plant the seeds of an idea.

True. The problem is that the signals to/from the CPU required to interact with the daughterboard (trigger in, ignition out, fans, EBC, VTEC, etc) do not come to the DB37 in the 3.0 schematic. They're all just pads, and you'd still have to solder a bunch of jumpers in. Additionally, a daughterboard would not give the injector drivers (and other noisy, high-current circuits) their own dedicates ground, separate from the logic ground.

I haven't seen any discussion there on form factor yet. Guess I haven't looked hard enough.

First, let me re-state that daughterboards should not be necessary. I'd like to see at least 75% of users running without them.

That said, daughterboards can be made perfectly reliable. The key is to use a proper pass-through connector designed specifically for stacking, and to provide standoffs for securing the board, rather than relying upon the connector(s) to support the load of the board.


Something caught my eye over at DIYEFI, in the "Letting go of the "MegaSquirt mindset" (hardware)" thread:
Absolutely 100% correct. MS R2.2 and R3 provide just enough hardware to almost satisfy everyone, but not enough to actually satisfy anyone. If a board could be built that supports 75% of all four-cylinder engines that came with factory EFI sensors, without needing to run a single jumper wire, I'd consider it a smashing success. Even the Honda guys, who were still using distributors well into the 21'st century (B18 series engines) would be supported.

The problem that DIYEFI is going to have is that there are too many chefs stirring the soup. MS got off the ground because B&G designed it, standardized it, and guided its evolution. I don't agree with every decision they made, but they did a good job of keeping it on track.

Matt Cramer

We've been thinking about a way to get a daughterboard on existing Megasquirts without the pick-up point issues, and it's close to some of the approaches mentioned here. Right now it's just an idea being kicked around, but it sure sounds like there may be a demand for it.

Form factor hasn't been discussed on the MS-III forum as much as feature sets, but there's been some mention of it. The MS-III is likely to be B&G's official device for people who want a LOT of outputs, but there may be room for something between that and the existing V3.0.

yertnamreg1218

I don't seem to understand the legal issue regarding modifying the MS-II design. Just because they use a certain voltage divider ratio to read a sensor doesn't mean you can't use that same ratio. There are only so many circuits that will do what we need in terms of breaking out CPU I/O to usable automotive signals. I don't see any reason why we can't adapt their board to do out bidding.

I think of it kinda like patents. You can't go and directly copy a device and sell it as your own, but you are free to improve upon current devices. I don't see why we can't do the same thing. (I'm probably wrong though, I'm a naive 21 year old)

rb26dett

Sorry for the late bump, just came across this looking through some statistics...

I disagree, the OEM's are using VASTLY more powerful cores than that. The extra IO is only just enough for a fully setup car. Logging all sensors etc and controlling all relevant systems. It's not excessive I assure you. Both ms1 and ms2 extra are heavily optimised to do what they do as well as they do it. I don't want to resort to unsightly optimisations unless I absolutely have to. I'll let your "hack" comment wait until there is some hardware to describe :-) (or if you mean the firmware, you are sorely mistaken)

LOL, fixed it for ya :

Stable and clean are primary goals, not an after thought.

Fred.

FoundSoul

iTrader: (5)

Joe-- What you're looking for, from what I can see, will be there in the MS3. These are all good points though, and a preliminary feature list should be posted soon based on user input and feasibility (but it's a healthy list). We'd love your input on what's missing, not just in features, but in bringing in all together and making sure nothing's missed in the details...