t vago's MPGuino workspace thread - Page 18 - Fuel Economy, Hypermiling, EcoModding News and Forum

t vago · 10-09-2013, 09:21 PM

Quote:

Originally Posted by josemapiro

I've tried but it gives the following error:
mpguino_1_86_tav_c: 5607: error: 'idxCorrFactor' was not declared in this scope
mpguino_1_86_tav_c: 5630: error: 'idxCorrFactor' was not declared in this scope

Thank you
José Rodrigues

Ah... that was the other thing.

Go to about line 1758, and there should be something that looks like:

Code:

#ifdef useChryslerMAPCorrection
const uint8_t idxCorrFactor = nextAllowedValue + 1;
#define nextAllowedValue idxCorrFactor
#endif

Then, make this change:

Code:

#ifdef useIsqrt
const uint8_t idxCorrFactor = nextAllowedValue + 1;
#define nextAllowedValue idxCorrFactor
#endif

Also, go to line 870, and change this line from:

Code:

const uint8_t pInjectorCountIdx = 				nextAllowedValue + 1;

to this:

Code:

const uint8_t pInjectorCountIdx = 				pRefFuelPressureIdx + 1;

Failure to do that will result in code, compiling with the "useCalculatedFuelFactor" option enabled, to display wildly inaccurate fuel rates and speeds, as I learned to my great surprise the other day. Ugh...

I can't post the latest code yet, because I just got done reverting the fuel injector interrupt code back to using two separate pins, and I haven't yet tested it yet to see if it works.

AndrzejM · 10-10-2013, 04:05 AM

t_vago, In next couple of weeks I'm taking a round trip across central Europe, with a plan to make ~4kkm and the plan is to test your code. So my question is... Any recent updates that you could post here for me to upload the latest version? Finally I'll be able to test it and I'm so glad to be able to finally see all those nice features working.

Ardent · 10-10-2013, 08:45 AM

An observation/question or "am I missing something", not a complaint:

I added a VSS sensor to vehicle "B" in order to be MPGuino capable. Known is that in "direct" 1:1 transmission gear that ~2450 RPM = 60 MPH. Thus 1 minute = 1 mile = 2450 driveshaft rotations with a 4 toothed sensor wheel = 9800 teeth or pulses/mile.

However, it seems as though both a rising and falling edge are registering as a pulse since displayed MPH was double actual. PPM needed to be set to 19600 to display a realistic value (not yet calibrated -- just ballpark).

t vago · 10-10-2013, 02:39 PM

Quote:

Originally Posted by AndrzejM

t_vago, In next couple of weeks I'm taking a round trip across central Europe, with a plan to make ~4kkm and the plan is to test your code. So my question is... Any recent updates that you could post here for me to upload the latest version? Finally I'll be able to test it and I'm so glad to be able to finally see all those nice features working.

a 4k km test run? That would truly be awesome! Are you going to be able to reprogram your MPGuino during this trip? Reason I'm asking, is because the code itself is still changing, although no longer on a daily basis. I'm currently chasing down a potentially serious bug in the fuel injector pulse reading code, and I'd like to know that you would be able to apply the possible bug fix, when one comes out.

Also, I'd like to finalize the A-B graphical display, and I'd like to get the coastdown test code working (there are several defined EEPROM parameters so far, the RAM variables have been defined, and the coastdown data gathering routine is up and running, but the actual computation and display code has not yet been coded).

I'll post the latest version, later on today.

Quote:

Originally Posted by Ardent

it seems as though both a rising and falling edge are registering as a pulse since displayed MPH was double actual. PPM needed to be set to 19600 to display a realistic value (not yet calibrated -- just ballpark).

Yah, both the 1.86 code and the original MPGuino code count both the rising and the falling edges of the VSS pulse train. At first glance, that doesn't appear to make a whole lot of sense. However, if you consider that some VSS pulse trains are mechanically generated via reed switches (instead of Hall effect sensors), then it makes a little bit of sense, because there is then a need to be able to debounce said switches.

It's a fancy way of saying "I kept it in the 1.86 code because it's always been there."

josemapiro · 10-10-2013, 04:12 PM

Quote:

Originally Posted by t vago

Ah... that was the other thing.

Go to about line 1758, and there should be something that looks like:

Code:

#ifdef useChryslerMAPCorrection
const uint8_t idxCorrFactor = nextAllowedValue + 1;
#define nextAllowedValue idxCorrFactor
#endif

Then, make this change:

Code:

#ifdef useIsqrt
const uint8_t idxCorrFactor = nextAllowedValue + 1;
#define nextAllowedValue idxCorrFactor
#endif

Also, go to line 870, and change this line from:

Code:

const uint8_t pInjectorCountIdx = 				nextAllowedValue + 1;

to this:

Code:

const uint8_t pInjectorCountIdx = 				pRefFuelPressureIdx + 1;

Failure to do that will result in code, compiling with the "useCalculatedFuelFactor" option enabled, to display wildly inaccurate fuel rates and speeds, as I learned to my great surprise the other day. Ugh...

I can't post the latest code yet, because I just got done reverting the fuel injector interrupt code back to using two separate pins, and I haven't yet tested it yet to see if it works.

Ok I have done to change the code, so missing test.

Thank you
José Rodrigues

AndrzejM · 10-10-2013, 05:05 PM

Quote:

Originally Posted by t vago

a 4k km test run? That would truly be awesome! Are you going to be able to reprogram your MPGuino during this trip? Reason I'm asking, is because the code itself is still changing, although no longer on a daily basis. I'm currently chasing down a potentially serious bug in the fuel injector pulse reading code, and I'd like to know that you would be able to apply the possible bug fix, when one comes out.

Also, I'd like to finalize the A-B graphical display, and I'd like to get the coastdown test code working (there are several defined EEPROM parameters so far, the RAM variables have been defined, and the coastdown data gathering routine is up and running, but the actual computation and display code has not yet been coded).

Yes that's the plan - 4kkm test run

Two parts of it first will be trip to Prague, next week and the week after I'm going to Sofia. I'll be able to upload new code without any problem. If I'll only get access to EM

So you can post updates and I'll upload the code immediately. And of course I'll post the results

Ardent · 10-10-2013, 07:57 PM

Quote:

Originally Posted by t vago

It's a fancy way of saying "I kept it in the 1.86 code because it's always been there."

Ha! How about some planned obscolescence?

t vago · 10-11-2013, 01:40 AM

Quote:

Originally Posted by Ardent

Ha! How about some planned obscolescence?

Heh... It's possible... Gotta do a massive edit on the wiki page, though, to correct all them VSS pulses/mile values there... and gotta let everyone know not to use previous versions of MPGuino...

Believe me, I'd love to do nothing more than to add one little line to the code that would correct all of this. However, my hands are tied, so to speak.

In any case, I tested the latest and greatest version of the code earlier today, and I can say with high confidence that it is mostly bug free! (At least, as far as I could tell...) So, here it is (at the first post of this thread).

AndrzejM · 10-11-2013, 03:42 AM

t vago,

You've done a tremendous job and hats off gentlemen! I'll uppload the code today after configuring it to my MPGuino setup, like legacy buttons 16MHz etc. Hopefully I'll have a feedback for you next week after first part of my planned trip.

Ardent · 10-11-2013, 10:07 AM

Arduino UNO with remote display enclosure: