Jump to content

DIY Fuel injection thread.


yoeddynz

Recommended Posts

Why is it rubbish/shit? Don't bag someones hard work without justifying yourself. 

 

1300loops/second and under 100uS timing accuracy at 10k rpm should be fine for most setups. It is still in development but I fully expect it to cut into the MS market in the future when it is a bit more user friendly.

sorry, i'll elaborate,

100uS is not enough for 10 000 RMP though. It will run, but it's not good enough (if you ask me) because 10K RPM = 166 2/3 RMS. 166 2/3 x 0.0001S = 0.0166667 rotations per loop. 0.01666667 * 360 degrees = 6 degrees timing accuracy. personally i think it should be accurate to at least 1 degree, but would really expect 0.5 deg.

Not to mention the fact that the micro on there is not meant for automotive use, neither are any of the other parts, and parts start behaving REALLY weird when you have them in a car with elevated temperatures... plus that micro is just not the micro of choice. You can get a better micro, for less money. (like a 32 bit Cortex M0 as a minimum, potentially with a floating point unit for super fast math).

It's a cool exercise in "hmm, i wonder if i can make this atmega run a car?". I've been wanting to do it myself actually, to see what the smallest micro is i could use to run a car, but thats all it is, a fun experiment. I wouldnt run an actual car on it that i wanted to be able to get in and drive every day without question.

but thats just me. Most of the code looks quite cool, and i sure do like the project and hope they improve upon it because i'm pretty sure they can improve upon those loop times.

Link to comment
Share on other sites

  • Replies 2.4k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Actually just read that was an old implementation - they now have 2uS accuracy at 10k RPM which is 0.6degrees. 

 

I agree he could have chosen a more suitable microcontroller but I think that is not the point of the exercise. Straight away the cost and development time would go up. The project started to mainly cater for small engines (lawnmowers etc) but has ballooned from there. It will be interesting to see if they have any reliability issues as they get tested more. You could also claim the MS is hardly perfect for automotive use either with DB37 connectors failing often etc. 

  • Like 3
Link to comment
Share on other sites

Hey Im all for trying out the speeduino on the basis that its fun building this stuff when the price is cheap, I learn more and its is proper proper 'BUILDING my own car'

For me its like MS fun and shinanigans but on an even cheaper level.

 

Remember guys that this thread is called "DIY fuel injection".... not the  "I just splashed out lots of money on a posh ecu and it works perfect out of the box" thread.  I cant afford that thread..

 

For those that gripe about it not being perfect, things might break, its not as good as my computer that cost me $1500 then I reckon you are looking at it wrong. Its about experimenting and fun. If more people get into the new things like this and are proactive in fixing things, suggesting mods, adapting ideas then hey.. the product will get better.

 

Ned- you of all people I thought would be well into this? I certainly didn't think you'd just slam it outright straight away. Why not jump on board and help make it better!

 

...or build your own ECU which is better and we can all buy that :-)

  • Like 3
Link to comment
Share on other sites

Cool! Did you have to import any of the components?

The board came from Josh (the guy behind all of this) in Australia. I see he is currently sold out at the moment. 

You could get some made from the likes of dirtypcb, seeedstudio etc but you might need to buy 10

All the components have been purchased from nz.rs-online.com apart from R25 and R27 - these are 2 watt 160ohm resistors that only come in a pack of 1000. I think i'll get these from element14. 

If your able use hall effect sensors for cam and or crank you can skip getting a vr conditioner. This is currently the only really expensive part of the board. 

You can also get the max9926 chips from nz.rs and continue the diyness and make up a vr conditioner as well

The arduino mega came from aliexpress. Check the speeduino forums for which one to get

 

  • Like 1
Link to comment
Share on other sites

I just started browsing on ali for a mega 2560. Damn there's a lot of options!!!  But yeah- cheap! Ive been warned by them on speeduino to avoid ones with ch340 chips, i think because they can cause problems with the USB connection.

 

Still lots to learn but having fun :-)

 

Josh (noisymime) is hoping to have the first of his latest speeduino boards available by end of November.

Link to comment
Share on other sites

I like that it has crank trigger support for the great jeep 4.0 high output. I've never figured out why no one has made megasquirt support for that factory crank trigger output,  seems like there would be plenty of people keen since they made 10 million of them and people do mince about with them enough to want to put an ecu on easily.

Link to comment
Share on other sites

Ned- you of all people I thought would be well into this? I certainly didn't think you'd just slam it outright straight away. Why not jump on board and help make it better!

I love the idea, for sure, dont get me wrong and it sure does have its up sides, and i see it started as a fun little project for small engines etc etc and it just grew bigger into what it is, and i appreciate that. It's a cool thing for wanting to tinker and thats what this thread is all about, but they could have build something better for similar money. In saying that, i'm opting with putting a toyota 7K in my car when for less money i can buy a 4AGE that makes 3 x the power so i guess i'm all about polishing that turd until it shines like a diamond in a goats ass! 

who was expecting a photo of a goats behind in here? :P

 

...or build your own ECU which is better and we can all buy that :-)

eaRCI.jpg

not far away now ;)

  • Like 4
Link to comment
Share on other sites

It's not as exciting as it looks in fairness.

I need an ECU for the starlet, and with the exchange rate, buying a microsquirt is too spendy, so for the price of getting just the loom shipped to my door, i can build a whole ECU.

It's based on the FreeEMS stuff Fred wrote/is writing but is pin compatible with microsquirt, so if i ever get sick of it or if my design is rubbish, i can just plug a microsquirt in and use that haha.

So same as microsquirt, 2 fuel, 2 spark, 4 spare low side drives. Some analogue inputs, 2 VR/HAL inputs. Has USB on board too, which is the only real difference functionality wise.

FreeEMS is definitely not ready for this to be a thing lots of people use. It currently has no good way to tune it, which is the main down side of it all but i'm happy to make do with whats there (for now anyway) for me its as much an exercise in designing the thing as it is having an end product so even if its shit, i'm still happy.

  • Like 3
Link to comment
Share on other sites

Awesome. You have built one already?! Have you built some for fred and his team as well? I was talking with zac about how cheap it is to get boards printed off but designing that sort of stuff is a whole new level for myself!

 

Back to speeduino...

I love the idea, for sure, dont get me wrong and it sure does have its up sides, and i see it started as a fun little project for small engines etc etc and it just grew bigger into what it is, and i appreciate that. It's a cool thing for wanting to tinker and thats what this thread is all about, but they could have build something better for similar money.

 

If you think they can do better for similar money then why don't you jump on board with them and help/suggest ideas. Its all open source and from only the one day or so that I have been browsing the Speeduino forum they all seem very friendly and open to suggestions. Why not help out? You obviously have the skills.

 

For me the idea of possibly getting my Viva Wagons 1200 engine or a Datsun engine running a simple tunable ECU, that's tuned via Tunerstudio which I know well enough now, all for under $100 is just too much of a fun thing not to try.

Link to comment
Share on other sites

Remember guys that this thread is called "DIY fuel injection".... not the  "I just splashed out lots of money on a posh ecu and it works perfect out of the box" thread.  I cant afford that thread..

 

I'll start a "Blew all my money on a fancy ECU but now no money for shoes and bread" thread
  • Like 4
Link to comment
Share on other sites

Now look what you've done Alex! :P

I havent made them yet, finishing touches ATM and sending off this weekend so will have them in a week or 2 probably. In time to run the KP for drag day i hope :D

and i've jumped on the forum, i had a look around at the schematics and they seem to be very stongly influenced by the FreeEMS ones. A lot of similarities there, which is cool. Means that other than the micro, there isnt much different hardware wise between my board and theirs because mine is also strongly influenced by FreeEMS (because it has to be) so if my thing works, it would be easy to spin up a slightly different board with the same interfacing hardware, just a different micro/brain so that i can run his code, as it seems more complete than FreeEMS...

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...