MPGuino release one workspace

[cmags]

So this project seems to be coming along quite nicely. I see a very reasonable schematic and good code. How close do you think you are to actually testing this in car and having a functioning MPG meter? I'm trying to decide on waiting for this and building one myself, or hopping in on that $135 ScanGuageII group buy going on in the forum. I have until Tuesday to decide. (Though I may pick up the SG and sell it later if this turns out to be sufficient for my needs).

One final question - in regards to the calibration techniques mentioned - they all seem to start out with a "guess". Are they suggesting that the guess is for initial calibration and then finer calibration is required later? I may not have read closely enough (and I haven't yet read the code), but I didn't see where the fine calibration comes in.

Thanks again for all the work being put into this. It's certainly looking good!

[larrydag]

cmags, I don't know for certain but there is still some considerable testing to be done. Getting the hardware configured is just step one. I don't know if anyone has fired it up in their car yet. Troubleshooting the code and working with the car is going to be step two. If you need to decide by Tuesday than the Opengauge is probably not going to be done by then.

Other than that we could always use a another hand in helping with the project!

[i_am_socket]

I'm loving the progress on this :-) The SG has way more options than I need at a price I don't need as well, and this project is the barebones I'm looking for. What is the general guestimation of beta hardware completion? Is there a list of milestones?

I'd love to help but my only experience with electronics is splicing wires and a failed DIY walkman kit. I'll look through the code, but the language is a bit foreign to me (I do mostly web/database work in PHP & Javascript)

[dcb]

cmags, re: calibration:
After the initial guess, the version we have here will benefit from a "true up" when you refill your car. Basically you
1. top off the tank
2. note your milage on the odo
3. reset the guino
4. drive for hundreds of miles
5. go to the same pump at the same station pointing in the same direction (same time of day/temperature would be nice)
6. fill up the tank to the same spot/number of clicks
7. note the change in mileage on the odo, see if it matches the miles on the guino
8. note the gallons it took to fillup the car, see if it matches the gallons on the guino

If 7 or 8 aren't reasonably close, then note the percentage that the reading is off and adjust the vssPulsesPerMile (for 7) or injectorGPH (for 8) at the top of the source code (at the top of this thread) by that percentage and reload it onto the guino.



Eventually we hope to have enough data from enough different types of cars that people can get a pretty good first guess based on their vehicle, but even a bad guess will give you a relative reading that you can use to see if you are improving or not.

Also we hope to utilize these buttons and add menus and such with future software upgrades so you can calibrate it in the car, but we are gunning for a simple working mpg gauge at the moment (with maybe a few nice features):
http://forum.ecomodder.com/showthrea...9.html?p=22751

[cmags]

Thanks for the update, dcb. I guess the good "guess" will definitely still be useful as a relative gauge, and tuning it up via tanks will be fine. (By the way, is there any knowledge out there about how consistent even the same pump is at turning off at the same point?) It'll definitely be close enough.

In terms of having this ready by early next week, I understand that wouldn't have been do-able. I just need to decide whether to wait on this or jump on the SG by then. I've got a trip from Atlanta to Boston and back coming up in June and would like to have some sort of MPG meter by then to try to keep my fuel costs under the projected $500 (~$50/tank, baseline of 300mi/tank, 1200mi each way, plus driving around for two weeks visiting friends and family). I'm taking steps to improve my mileage, but the feedback will be key in that.

I'll prob. jump on the SG anyway, but I love DIY's so something like this is right up my alley. The SG has a number of other features that I'm interested in, such as the code reader, and intake temps etc (useful in gauging whether a grill block is overheating the car). If the *duino platform matures enough to be able to be able to string peripherals off it, I'm all for it.

Thanks again, and keep up the good work. I'll be keeping an eye on this project.

[MetroMPG]

Originally Posted by dcb

05/08/08 Couple of code tweaks and a test run in the car (97 saturn s2) was promising. The miles reading is in the ballpark, but the gallons was at least an order of magnitude off. Also everything was in whole numbers, will need to include floating point routines or device an integer based strategy.

Woohoo for first test run! Congrats.

[dcb]

Here's a couple pics.
I scrounged up a quick and dirty bit of hookup parts from the garage and hooked it up today:


The mileage reading was suprizingly not horribly far off, gallons needs work, as does fractional numbers.


At night that LED is a killer, I think we will be glad to have brightness control

[larrydag]

Here is my working prototype. This was the best picture I could get. dcb, you're going to have to tell me how you got to hookup with your ECU. Did you go to the external ODB connector or did you splice into the main connector.

by the way, if anyone has experience with Honda ECU's let me know. My ECU is 37820-P0B-A51

[MetroMPG]

I may be mistaken, but Schultz may have access to a db with ECU pinouts...

[larrydag]

Thanks Metro but I can get the pinouts. I just don't know how to connect to the thing. It's a closed box except for the big 3 connectors to the pinouts. Any ideas?