MPGuino settings for 49 state Honda Civic VX.
 

Hi guys. I have MPGuino installed ver 0.86 and there is plenty of other settings besides two listed in wiki page. Can someone that had verified accuracy of Guino in real life driving provide those settings below please.
I tried default value for microsec/gal and it seems it provides MPGS too high.
After further tests I found that that 175 tires that are one size over 165 tires affect VSSP/M reading and possible usec so doing testing for both parameters.

My car. 1992 Honda Civic VX (federal, non-cali version, lean burn). Stock wheels and all stock configuration besides 175 instead of 165 tires due to no 165 availability.

Contrast = 52. Looks most clear to me. Brightness = 3.

VSS PULSES/MINUTE = 8208 per wiki page for 165 tires? What is yours in VX?
Testing mine by GPS for 175 tires as tire size effects odo and speedo.

Microsec/Gallon = please provide yours in VX and why is yours different?

Pulses/2Rev = ???? Please provide yours in VX? Mine is set to 1 per wiki but it's not currently in use per wiki

Timeout = 600000000 ul. 10 min for hopping into convenience store

Tank Gal = 10000 (mine tank always seems like 10 gal). Most VX came with 12 gal. Some like mine came with 10 gal.

Injector DelayuS = ???????????????????? PROBABLY IMPORTANT
Currently running default 500 us

Weight (lbs) = 2094 VX stock curb weight

Scratchpad (odo?) = notepad.

VSS DELAY MS = ???????????????????? PROBABLY IMPORTANT. Currently set to 2 us.

INJTRG 0-DN 1-UP =???????????????... 0 currently set.

Hi, I'm using Canadian metric version but it's similar so I'll answer you with my best knowledge.

Timeout = time before MPGuino goes to sleep ;) and it resets current trip counters
Injector delay = time that guino waits before it stop counting time of injector open state. (after current is turned off on the inj it stays open for a while and inj delay is that while)
Scratchpad = not used in v0.86 (as I remember)
VSS delay - time that guino waits before it can read another vss signal
Inj Trigger - If you're reading +12V or ground signal

Current is an average reading from the current trip, instant is a reading in the given moment.

Andrzej czesc Rodaku. Did you leave those below settings at default or did you change them? Here is what i was able to gather from Mpguino 0.86 code

Thank you. That helped a lot. I still would like to find clarification on few parameters like

Pulses/2 revs = 3ul default but Per wiki jap cars are 1. Confused here!????

Injector DelayuS = 500 ul ...that's default but wifi specific for VX is blank. Does that mean 0 or 500...???

VSS Delay ms = 2 ul .....not sure what VX specific would be ???

Her is what i extracted from my 0.86 code and what you provided if anyone whats to use it.

MPGuino Default settings

Cut out of officeal MPGuino 0.86 code
unsigned long parms[] = { 55ul, 8208ul, 500000000ul, 3ul, 420000000ul, 10300ul,
500ul, 2400ul, 0ul, 2ul, 0ul, 0ul };//default values

char * parmLabels[] = { "Contrast", "VSS Pulses/Mile", "MicroSec/Gallon",
"Pulses/2 revs", "Timout(microSec)", "Tank Gal * 1000",
"Injector DelayuS", "Weight (lbs)", "Scratchpad(odo?)", "VSS Delay ms",
"InjTrg 0-Dn 1-Up", "Metric (1=yes)" };

Contrast = 55 ul
VSS Pulses/Mile = 8208 ul Civic VX value
MicroSec/Gallon(Custom Value to car) = 269113500. Yoshi SuperMID designer value custom to Civic VX. Not this is Supermid CPU number not guino.
Pulses/2 revs = 3ul default but Per wiki jap cars are 1. Confused here!????
Timout(microSec) = 420000000 ul ...dispaly timeout. Can be anything.
Tank Gal * 1000 = 10300 ul ...assign custom value to your car = 10000 = 10 gal tank
Injector DelayuS = 500 ul default
Weight (lbs) = 2400 ul default ....change to 2094 lbs
Scratchpad(odo?) = 0 ul
VSS Delay ms = 2 ul
InjTrg 0-Dn 1-Up = 0 ul
Metric (1=yes) = 0 ul

Czesc! :)

Pulses per rev = it depends on the car, the simplest way to be sure that the number you've entered is correct is to check RPMs showed by MPGuino if they corresponding to the tachometer. Basically it means how many injector pulses your engine have per one rev.
After you enter everything you should calibrate it anyway. Just check MPGuino numbers with the pump and correct settings. It's very easy. But remember to check VSS pulses first. You may use GPS speedo to check if speed on MPGuino corresponds to GPS. That's the most accurate IMHO. Then you may calibrate uSec/Gal settings.

One more thing. I'm sure 500us inj delay will be good. But it has to be entered using settings screen, regardless default settings, you need to confirm that. You shouldn't leave this field empty.

Also check out the MPGuino wiki at MPGuino - EcoModder

Andrzej thank you very much. There are few discreptencies between few threads on settings. In wiki i believe it has been established that
1994 Honda Civic VX 49-state 1.5L lean burn 5 spd manual has

vss/m = 8208 MicroSec/Gallon = 0279235000

yet designer of SuperMID, Yoshi claims mics/gal is different. I find Yoshi to be more accurate based on what i know my car does at given time in terms of mpgs. wiki numbers are BIG and seem WAY off.

VSS Pulses/Mile = 8208 ul Civic VX value
MicroSec/Gallon(Custom Value to car) = 269113500. Yoshi SuperMID designer value custom to Civic VX.

Andrzej can you provide more details on how to test this. I think i'm lost a little bit. I will try to re-read this few times and do this but not sure i'm getting it all. Also what concerns me are those few settings that are left at default in MPGuino and for example delay is listed blank so that means zero?, where in mpguino vss delay default is set to 500ms. So is it zero or 500ms. Wiki has blank value so one would assume zero yet default is 500ms. Also does VSS delay matter and what does that mean for results? Those 3 values are my concern below as they are not documented anywhere what it should be set ??? Any ideas?

Pulses/2 revs = 3ul default but Per wiki jap cars are 1. Confused here!????

Injector DelayuS = 500 ul ...that's default but wifi specific for VX is blank. Does that mean 0 or 500...???

VSS Delay ms = 2 ul .....not sure what VX specific would be ???

NachtRitter are you using wiki value for MicroSec/Gallon = 0279235000 or Yoshi SuperMID designer of MicroSec/Gallon(Custom Value to car) = 269113500 ??? By simple eye test looking at MPG while driving Yoshi seems more accurate as it gives me 52 mpgs while driving in 5th gear while Wiki was showing me 80mpgs...no way this car is doing 80mpgs, however i do realize that those numbers can go up to 120 if you let gas go but this is simple eye test :))) ...btw i really appreciate everyone's help. thank you.

Stewie,

You shouldn't bother so much about initial settings of MPGuino. Really it doesn't matter. I was starting wit MPGuino in Previa then in BMW, and I didn't had a clue about what I should put into initial settings. So I've entered just anything and started to drive. First I've accelerated to 100km/h and checked speed showed on MPGuino screen. I can't remember the exact value but let's say it was 112km/h and 8000 VSS pulses per km for example. So the value was too low. To correct VSS settings I've used a simple equation:

vss pulases in guino * (speed from guino / vehicle speed) = corrected VSS pulses to enter into guino settings

8000 * (112/100) = 8960
After entering correct VSS pulses I started to drive again and I've checked guino speed vs speedometer. It was fairly accurate. You may calibrate it further by comparing long distance guino miles vs miles from odo.

Then I started FE calibration. First you have to fill-up the tank and drive normally for few hundred km/miles, then you fill up and compare numbers from the pump with those from mpguino.

Equation to correct us per galon/litre is similar to that for VSS.

us per galon from guino settings * (MPG from the guino / MPG from the pump) = correct us per galon

example:
MPG from the pump = 62MPG
MPG from guino = 58MPG
us per galon = 8932000 us

8932000 * ( 58 / 62) = 8355742

Other your questions:

Don't over-think that! Can't remember if v0.86 have rev gauge but if it has switch to that screen and check if revs are corresponding to tachometer. If not correct settings accordingly to the difference. Let's say rev meter in guino shows 2000revs/min when tacho shows 1000revs/min and you have 2 pulses per rev in settings. Change it to 1 pulse per rev and voila!

Leave it with 500us and you'll be fine.


If you have stable reading of instant speed don't change anything there. If not you may increase a bit the value.

If after distance calibration you'll notice big difference between odo and guino you may check that settings too.


I'm hoping that explain everything to you :)


If you have any other questions don't hesitate to ask.

That's great explanation. Thank you. Sorry for so many questions but i'm trying to understand this device better and without set values i just don't know. My tank is just running out now, another day or two so i would like to get settings correct so i can test with full tank. It takes 2 weeks to empty my tank so if i won't get it right it will probably take few tires so few months.

I did GPS speed test today and MPGuino showed slightly slower speed then GPS app on my phone. I believe GPS on my phone showed same speed as speedometer in car HOWEVER my speedometer in car is incorrectly slightly as my tires are one size bigger 175 instead of 165 as you that's how i bought the car. So if i understand you correctly my vss/m and MicroSec/Gallon will be different then default of (vss/m = 8208 MicroSec/Gallon = 0279235000) due to slightly bigger tire size?

Should i adjust VSS delay to get speed right or VSS value? Also mpguino does not have rpm guage. I do notice instant mpg changes as soon as i change any values including REVS and Delays.

stewie, the wiki 0279235000 is way too high heres what I use

Contrast = 55
VSS Pulses/Mile = 8208
MicroSec/Gallon= 0257763367
Pulses/2 revs = 0
Timout(microSec) = 20000000
Tank Gal * 1000
Injector DelayuS = 500
Weight (lbs) = 2021
Scratchpad = 0
VSS Delay ms = 2
InjTrg 0-Dn 1-Up = 0
Metric (1=yes) = 0

I have used these #'s for over 55,000 miles
the mileage dead on
the MPG is off a bit trying new # 0257072561 I'll let you know how close that is in a tank or 2

WD40 ...thank you. I bow to you now! Andrzej thank you very much first and foremost. I knew wiki numbers were not accurate. I know this car but i don't know mpguino so numbers were off. BTW does tire size matter. I have 175 instead of 165? Are the mpgs you getting off to the upside or downside?

one setting is missing.......this one? what you got for this one?

Pulses/2 revs = ????

stewie: I have 175 tires as well.
Pulses/2 revs = 0 on mine.
The mpg on the mpguino were to high by 1.14 mpg on a 426 mile tank so the new # I'm testing for Microsec/Gallon = 0257072561 instead of 0257763367

Cool cool many thanks!!! This is exactly what i need. Just curious you have 12 or 10 gal tank. Mine is 10 gal yet most ppl have 12 gal tank. Also are you calculating bigger size tire when calculating final mileage? This is important if you tweaking guino. How are you changing calculations on Microsec/Gallon when you change those values?

Tire size calculator

stock size 165/70/13
+1 size 175/70/13 so

if you odo reads 426 miles one needs to multiply x2.5% = 426x2.5%= 426+10.65= 436.65 odo final miles divided by total gals.
With bigger wheels tires roll too slow as tire is bigger therefore lower milage and 2.5% corrections needs to be applied. You probably know that.

stewie:
my calculations are hit an miss by trial and error over the years, i have adjusted many times in the beginning and just today I redid it again based on Andrzej methods of calculating, however i tried his method and it increased the error so I reversed his calculations and its going the correct direction now however still way to much of a correction so I went again with my hit and miss method which seems closer.

Cool but what i mean is that your ODO with 175 tires is inaccurate in first place as yours and mine for that matter bigger 175 tires roll slower then 165 tires as they roll faster. What that does is on 175 your ODO is short by x2.5% of total mileage on miles. So if your ODO shows 426 miles it's NOT 426 but really 436.65 so then you divide by total gallons burned. If not adjusted/accounted your manual calculation will be off. So i guess what i'm trying to say that your figure might be correct if you are NOT adjusting manual mileage to tweak guino. If you had stock tires you would not need to do this. Hope we can collectively figure this out here :)))) You're getting pretty good MPGs bro. What psi on tires you running. Andrzej if you can help us figure this out it would be great. I just started this so right now your calculations are overwhelming me a bit. Maybe in time i will re-read and reapply. I'm hoping we can get correct settings for the this device.

stewie:
gotcha .. never worried about the tire size % difference before, but taking that into account the mpguino was only off by 0.31 mpg this time by to little at the 0257763367, so I need to readjust it in the opposite direction then .. I will recalclate this and get back to you, thanks for your persistence.

lol...sorry bro that i'm being too persistent. I'm guess i'm anxious to get this right.
Yes please let me know when you get more info. I will be testing both of your figures. I did little test run on the way back from gym and nubmers displayed look damn accurate to me. I'm happy to be in the ball park so i think you again. I knew that wiki posted numbers are way off.

Pity v0.86 doesn't have rpm gauge. My "canadian" metric version has it and it's great for diagnostics. But if you mounted MPGuino on your own you can change the code and set up rpm gauge. I can look into the code and provide you all the changes that you need to do that.

Anyway - you've mentioned that your speedometer can be off due to different than oem tire size. Speedos are off very often by 5 to 10% so don't worry about that. You may calibrate it (mentally) with GPS and that will give you how big is the difference.
But if speedo is off then probably your odo is off too. You should check the difference with GPS or mile markers (10 miles or more), to be sure what exact distance you're doing on the tank. Then you can calibrate MPGuino for the exact value and relay on that to get your FE numbers.

Code changes for v0.86 to have Big RPM gauge

Just find the code and add the code marked in red

Define display func:
Define display func name:
Display func body:

Andrzej you're the man! Thank you. I found ISP port on the board. Is that the serial port in need to use to get into chip? How do i do that? Is there a guide for doing this? Are there any other functions that can be added to 0.86 guino as well. What would i use RPS for in terms of calibration?

Ok so i tested speed with GPS and here is what i found. I think, i'm not sure but i think speed on guino is accurate. I did find GPS to change the speed little quicker .30mph then guino however isn't GPS delayed as its got all the way to satellite and back. Wouldn't latency of GPS cause inaccuracy in speed reading on GPS? The numbers i'm getting now are more believable then wiki. On weekend i will have to rewire guino as its only connected to key on power so resets all settings when key off. Need to find constant +12V. Thinking i will run it off car stereo. The clock which is missing btw is too hard to get to. Also there is option of tapping wires close to fuse box as well.

I found little info on calibration. Will try that when i get my constant +12v fixed

"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."

What i'm interested in is whether 2.5% tire adjustment Tire size calculator will have to be added taken into account in #7 on odo. It is plausible that manual calculation are in fact in synch with mpguino when taking into account 2.5% adjustment.

stewie:
the latest # I have tried is 258562433 it seems to be the closest I have seen, I have only run it for 100 miles tho so a full tank may be off a bit.
In the mean time I have altered all my gaslogs to adjust for the tires size.

WD40 sweet bro. You're great man! How does your ODO mileage compare to Guino mileage after tire adjustment? the gal/sec number you want is probably that matches or is closest to those ODO match after tire mileage adj with Guino mileage. I'll put in some input as well after i get it installed properly this weekend. We'll figure this out. I really do appreciate your and Andrzej input. I have a starting point with your figures to tweak it now where wiki number was so off i would spent a month or more on getting it right so for that WD40 i thank you much. BTW what is our tires PSI if i may ask. I just bumped mine to 45psi. Max i can do is 50. In winter i ran 34psi. My goal is to run 50mpgs or plus constant which i can only do occasionally now.

There are many things you can add to MPGuino, right now I'm working on temp gauge, range gauge (how many km/miles you can ride with current FE and given tank size), and coast down time and distance meter.

If temp gauge will work as it should I think I'll include temperature management system into it. My guino will close & open grill accordingly to engine temp. Or I'm hoping it will :)

Anyway when I get my code working I'll post it here and maybe I'll add it to MPGuino wiki. Problem is that I'm working with "canadian" version, but code is very similar so it shouldn't be hard to include those changes into v0.86 "official" code. Maybe I'll make two versions of my code changes for canadian and official version. If there will be someone interested :)

stewie:
I ran my tires as high as 65 psi but there is no noticeable difference in MPG between that and the 50 that I run now.
The big difference is less road noise and bone jarring.

I forgot to answer your question. Yes you can upload changed code using ISP, and simple serial or parallel to ISP cable. Or using any Atmel compatible ISP programmer.
My MPGUino I've build on my own using Arduino. Good thing is that I can make any changes I need and it's much easier with all the ports available.

Small note on your rpm code:

The "RPM" and "rpm":"rpm" is redundant. An improvement would be:


void doDisplayBigRPM() {bigNum(instantrpm(), "RPM", "");}

or

void doDisplayBigRPM() {bigNum(instantrpm(), "", "RPM");}

Andrzej thank you. Still will need to research into how to do that. Found some cool info.

Mpguino_calibration
http://ecomodder.com/wiki/index.php/Mpguino_calibration

Injectors explantion...middle of the page
http://combustory.com/wiki/index.php/MPGuino#Enclosures

Code hacks
http://ecomodder.com/wiki/index.php/Code_hacks

Programmer
http://www.ladyada.net/make/usbtinyisp/index.html

Guys i did test drive today (with 500 miles numbers from WD40) and downloaded app with decimal speedometer with multiple sats. So Guino reports speed under (within .30-.50mph) and miles (within .40-.60m) under GPS so i guess i'll have to tweak those numbers as they appear off which would explain the mpgs being off.

stewie: please let me know what #'s you come up with that I can test on mine.
I've just gotten used to close enough for settings, but closer to perfect is always welcome.
also please join team honda http://ecomodder.com/forum/team.php?id=1

WD40 i am extremely thankful for you providing me those numbers. They are close. Main thing is to get miles right to GPS miles. I'm trying to figure this out.

Andrzej how do we tweak speed. I think that should be accurate as guino taps into Vehicle speed sensor. Does VSS gets adjusted from those two variables we provide or mph is provided from VSS as static value?

Ok i found something else. I thin its similar to our situation. I think PPM value is also incorrect as we are getting same results as him so i will try his value and check speed then miles
http://ecomodder.com/forum/showthrea...tml#post178967

Ahhhh...this is hard. Sooooo...I did VSS P/M test and the closest i could get is 8086 and 8090. 8100 was little too high for 175 tires on our VX. I suspect that on 165 tires wiki number for VSS P/M is correct or close. Our analog speedo and odo is off by 2.5% due to tire size so you can not use those instruments for tweaking guino. Also wiki number is based on 165 tires so can't use that. Based on GPS i was only able to get within xx.2x mph number. The latency on GPS affects the number and numbers on guino are instant and on gps there is latancy so thats hard. Im assuming with km it's easier as its more accurate nubmer. Also i can bearly notice changes xx.2x so how can i record that. With that VSS P/M miles were off too a bit but that's normal when you change parameter. Next i will be tweaking Microsec/Gallon. Microsec i was testing was 0257763367. I will change this now however to 258562433 as miles were off a bit and will adjust Andrzej's method based on that.

stewie:
I have tried 8022 with 258562433 and it was the closest i have come to the correct mileage, but I cannot confirm the mpg, it seems quite high, so i would imagine that we need to go to the original 0279235000 and retest with new #'s for the 8208 and see if we can get correct mileage #'s.
I will play with it again today.

WD40 holla brother. Ok here is what i've done last night and today. First and formost. Anyone with 175 tires can not use or rely on analog insturments to tweak guino in our cars. If one has 165 tire then maybe but still analog speed really? ...anyway. The only way to get correct measurements is to use GPS. Like i said i used GPS with xx.xx decimals and closest number to xx.2x degree of error i found was 8090 VSS P/M. Today i drove to work 12 miles and measured miles like steffan did in his post on GPS and did this.

8090 VSS x 11.80 guino miles = 95,462/11.794 gps miles = 8094 VSSP/M.
This is based on 12 miles those 12 miles where very close as you can see. I will test 8094 VSS new figure for speed and miles on the way back but like i said earlier 8100 was by little too high so this seems like a bulls eye. This would be best on longer continuance trip. 8208 default number is too high as speed is off and even 8100 speed was off so 8094 for 175 tires might do the trick. If this figure turns out ok i will then test us per gal using Andrzej method. 8094 is really close or dead on. If 8094 wont be dead on i will try to test it on longer trip. Whole tank would be ideal but my GPS is phone based and i'm not leaving me phone in car. Maybe if i get really picky about i will buy for 30 days GPS and do whole tank test. I think i'm getting this now.

stewie:
try this site as your gps Map Bike Rides with Elevation Profiles, Analyze Cycling Performance, Train Better. Ride With GPS its very accurate and takes into account hills.
I will try your #'s as mine are still off.

Thanks bro. I will try this app. I was using android speedometer and this has decimals which was very helpful. I'm positive that 8208 is OFF for VX with 175 tires. Andrzej was right. Test your numbers. I think 8094 is correct number. BTW do not change your PSI tire pressure when doing any of those test. I'm sure that affect MPGS as well. I will test on the way back the mileage and mpg and let you know.

https://play.google.com/store/apps/d...id.tacho&hl=en

https://lh3.ggpht.com/RLe73Lc9YSWJbV...VZ2-4SnoA=w705

stewie:
I just tested 8099 0258562433 and it came out 100% with gps, we still need to see how the MPG works out when compared to reality.

Was that for speed or distance and after what distance? Do you get decimal on your GPS for speed like xx.xx? We'll get that right my friend.

Yeah first we needed that VSSP/M correct first. I think it's 8094 but i will test on way back now. Difference between 8094 and 8099 will be small anyway.

Now we can do FE calibration for MPG. If above parameter is incorrect like speed and distance then FE will be bad too so now we can do second one. For MPGs we will have to figure out usec/gal ...this number 0258562433...
BTW i would think do not adjust 2.5% for tire size when doing ODO from your ODO analog in car. Just use ODO and GUINO numbers. We will then get correct value as we'll offset the error. GPS total distance would be best with guino numbers. Not sure how to do that besides buying car GPS? Any idea?


Equation to correct us per galon/litre is similar to that for VSS.

us per galon from guino settings * (MPG from the guino / MPG from the pump) = correct us per galon

example:
MPG from the pump = 62MPG
MPG from guino = 58MPG
us per galon = 8932000 us

8932000 * ( 58 / 62) = 8355742

stewie:
trip testing 8088 0258562433
gps = 56.70 klm / 35.30 miles = actual mileage
mpguino = 56.89 klm / 35.35 miles = +0.19 klm / +0.05 miles

==================
trip testing 8099 0258562433
gps = 56.70 klm / 35.30 miles = actual mileage
mpguino = 56.70 klm / 35.30 miles = +0.00 klm / +0.00 miles

I was thinking that maybe the wiki # is worth a try with the new VSS #
might test that out today.

Hmmmmmm....interesting. Cool. At 8100 the speedometer was off so not sure that number is ok for speedo vs GPS speed. That's what i was testing basically. Then did miles today. I will drive back 12 miles on 8094 and if it's off then try your number. Best thing to do would be to have GPS in car and drive on whole tank of gas. That would be most accurate but that's not possible with phones as you take them out of car and you will forget to turn it on and have to turn it on every time for every trip. Annoying. lol

Yeah give a shot? You don't trust your usec #? Thing is wiki number for MPGS was too high, way high. Try that one and i know Yoshi number maybe a worth a try too. Current MPG number i get on your number seem very close to what my car get. I think yours is in the ball park and new VSSP/m will throw it off now but with 2 tanks we'll get correct value. But like Andrzej said it really does NOT matter what initial usec numbers we use as when we do calculation at end of the tank by dividing you will get correct number. Then after that you can do 2 more tanks of calibration and you exact numbers you can trust.

MicroSec/Gallon(Custom Value to car) = 269113500. Yoshi SuperMID designer value custom to Civic VX.

I was thinking that GPS total distance for tank would be best with guino numbers. Not sure how to do that besides buying car GPS? Any idea?