anyone to get involved with a DIY ecu?
 

I know it's a bit ambitious and I know there are a couple of open source ECU projects around, but...

I am a software developer and cars are just a hobby. Looking at the existing aftermarket engine control units I believe time has come for a simple, clear and extendable EFI project. We have boards under $20 running at speeds above 100MHz, we have some great real real-time operation systems - I believe these things combined should make a very simple EFI possible.

Actually I have my test car running with my code, with hardware on the clumsy side:
Here are some HOWTOs to give you an idea:
how to get RPM value from stock ECU
how to feed data to TunerStudio
stock ECU stimulation

https://www.miataturbo.net/attachmen...ine=1378601378

A breakout board for early Miata ECU connector http://rusefi.com/hardwave/176122-6/

Most of us here run a DIY ecu.

Right, I should correct myself - I am talking about a design-it-yourself.

You're not going to find many of these here, for much the same reason that most of us do not physically raise our own cattle or stitch together our own clothing. In the end, it's usually cheaper and easier to pay someone else to do this, and yields more satisfactory results.

There have been a couple of projects like this which have spring up in the past, I'm not sure that any have yielded results. One is here: https://www.miataturbo.net/ecus-tuni...ino-ecu-50695/

Certainly has the potential to be a fun project, but the reality is that something like this is outside of the grasp of the vast majority of car hobbyists, even those who are handy with a soldering iron.

I assume you have a chauffeur to drive you around the track, and a cook to mix your cereal in the morning. Lucky you, you do not have to worry about anything :)

I do not need to find many, a couple of guys with some time to spare would be great :)

I will definitely get in touch with the guy, but his choice to use arduino was a choice to walk in cement boots. stm32f4 is _much_ more powerful, so hopefully I would not need to bend backwards on the micro controller side.

yawn.

I am really sorry that I do exist.

I'm really sorry that I ended this sentence in a preposition of.

Hello Techsalvager, how have you been?

It must be some inside joke I guess? :)

You think way too highly of Tech. He'd never be able to pull off something like what the OP osted.

I really don't think you appreciate how much money you WONT save by doing this. You will still be at least a couple hundred bux into it with hardware alone, and even if you're a genius you'll still have to invest huge amounts of time into setting it all up and getting it to work.

So if you have hours, days, weeks, and possibly months to waste on this, and don't value your time for any amount of money, I guess you'll save a couple hundred bux.

Good luck though. I hope you pull it off.

Yep. You don't even wanna know lol

It is, he is a/the joke.

Good luck with your project, but don't be surprised at everyone's skepticism, there have been a few of these and have never really come to fruition - or anything for that matter

One of the relatively successful ones (DIYEFi) like to continually bash the MegaSquirt system for it's perceived weaknessess, but the truth is, it's massively more successful than DIYEFi's effort, which has a grand total of maybe 10 cars running it's system. Smacks of sour grapes.

Seeing as there's a strong MegaSquirt following on this forum (and MX5/miata forums in general), you can probably see why this would be tiresome.

Good luck with your endeavor, but I think you'll be quite some time before you are able to offer anything close to even the most basic and cheap MeaSquirt builds.

Ok, let's turn this into an intervention thread :) I have the money and I have the time, it's also going to snow soon and my racing season would be over. How on earth should I waste my evenings? Start ice racing?

And another thing - my race rulebook does not allow a $500 MS, but it allows $20 microcontroller plus 1000 hours of my time. And since you will tell me that a $20 microcontroller would not be enough I would tell you that in fact it would be: I will take the huge DIP chip out of my stock ECU and solder my own $20 board in and use the stock ECU pcb for the hardware.

I'm not going to talk you out of it. You seem to be pretty motivated to do it.

So do it lol

I'm just going to sit back and enjoy my MS3x, and check on this thread every once in a while. If it ever even gets anywhere.

Didn't Joe build a MSI with about $20 in parts?

https://www.miataturbo.net/megasquir...-easily-63040/

Haha. No, I'm not that rich. I only have one servant- he both drives me to the track and cooks my oatmeal.




It wasn't an MS1. That would have been bootlegging, and that would be wrong.

It was just an ECU that was EXTREMELY similar to an MS1, and was compatible with the same codebase.



Russian, this was the last ECU I had, before the MS3Pro:

https://www.miataturbo.net/attachmen...ine=1327996323

https://www.miataturbo.net/attachmen...ine=1327996323


I think that qualifies me to speak with some authority on the subject of building your own ECU completely from scratch. Having done so once, I do not intend to do so again. It was an amusing project which kept me busy for a while, but the end product wasn't really all that much better to what I could have purchased for $200. It had a few tweaks and improvements over a base MS1, but not enough to justify doing it again. I have a job which pays me more than the money I'd save building these vs. buying MS3Pros off the shelf.


In your case, I understand why you want / need to do this, and it sounds like fun. I'm just explaining why you're not going to get a huge amount of support / info from most of the folks here.

lol iswydt :giggle:

I'm all for any other ECU other than megasquirt, just because megasquirt source code is shit and makes me angry

seriously, reading that code is like jerking off via a beehive full of bees.