Scanguage II Calibration: 'Adjustment Factor on Fillup'
 

As per the manual:

"The screen will show how much fuel the ScanGauge has computed to have been used since the last time a FILLUP was done. Use the value displayed by the pump or on the receipt to adjust this value up or down (using the upper buttons) to match it as close as possible. The lower line shows the ?adjustment factor? in percent that the ScanGauge will use to compute fuel use. Press DONE after you have finished adjusting the indicated fuel to match the actual pumped fuel. Pressing DONE enters this correction factor into flash memory"


I'm thinking that the adjustment factor on the fillup is something I will ignore from here on...

When I first got the unit, I did a fill, and then told the Scanguage I had done a fill, but did not put in the real amount.

I then drove for a day, and did another fill - adjusting the fillup in Scanguage to match what I had actually used. The 'adjustment factor' was 19.4%

My next tank was 4.0 L/100 (7.7 litres / 192 kms), but Scanguage showed 4.7

If that 19.4% had actually shown zero, I'm thinking the fuel economy in Scanguage for this tank would have matched... ? Am I correct?

I have verified and adjusted the Scanguage speed against a GPS, and verified the distance travelled on a longer trip, and I am within 2/10 of 1%... (I wish you could enter more than just whole numbers for this adjustment - I could make it exact...)

Is this likely to be the only calibration required? It seems the fillup amount 'adjustment factor' introduces too many variables... did you use the same pump, facing the same way, filled to the exact same level... and add to that exaggerated differences on partial fills...

How have others done it? Any examples of how the fillup 'adjustment factor' has helped? Have you averaged out 10 fills FE at the same pump against the Scanguage FE for the same?

I have the SGI and mine was consitantly within .1 gallons on every fill after the first adjustment. I only had to adjust mine 1%. Check to be sure you have the engine parameters set for your engine size on the setup menu and that it is set for diesel fuel.

I don't know how far you drive in a day, but I'm guessing it wasn't far enough to use enough fuel to reduce the significance of the variation from fill to fill. The ScanGauge manual recommends using at least 10 gallons or 3/4 of a tank (something like that) before attempting this procedure. Best practice would be to fill the tank, reset the ScanGauge with 0% correction factor, and drive through a whole tank before trying to match the ScanGauge reading to the next pump reading. Go through that once or twice and you should be in good shape as long as the rest of the parameters are set correctly. Just keep in mind that it's not a perfect science, and the ScanGauge may be a little over or under depending on how much gas the pump feels like giving you on a particular day.

Yeah and don't forget to allow for pumping error - if you pump slow it puts more gas in than it says it pumps and you will fill up with less gallons and show higher gas mileage.

Have you tested this theory rigorously?

I only fill up every three weeks or so - trip to Maine this weekend so that should be good for two tanks of fuel - will check it more then.

Cool. How are you going to test it?

Precisely - that is to be expected.

I do have 'Diesela' and .8 litre engine size selected.

After that, it seems to me there are primarily two calibrations that effect the FE number - the speed - and the fillup 'adjustment factor'.

Being that I have calibrated the speed with a GPS, I'm leaning towards ignoring the fillup 'adjustment factor', unless after 5000 miles, there is a percentage point or two difference between the cumulative S II FE, and the FE number obtained through the total volume of fuel purchased, and total (GPS calibrated) distance travelled.

Perhaps that is where I will start - simply start keeping track of the Scanguage 'tank' data along side my existing data, and then compare it every 1000 miles to ensure it is within 1% overall. Obviously some folks really trust the numbers from the Scanguage - and I'm merely trying to get to that level objectively.

There is one advantage I have - in that I can fill right to the top of the filler tube in my smart - visually seeing the level of the fuel... and I can ensure there is no foam... this has allowed me to calculate very accurate tank by tank FE numbers, even on partial fills, simply based on GPS distance and fuel volume filled... BUT - it would be so much better to have a second source for a FE number, hence Scanguage.

i already filled and waiting to refill to check the accuracy too. sg has been showing my trips to work arounf 36-38 mpg but when i checked mileage was only 32.7. my wqay to compute mileage was that when low fuel light came on i put $20 in and next time light camer on i ised the number of gallons and miles drive3d to figure mileage. e-mailed ron at linear logics and he told me that i needed to do the corect fill and refill for best accuracy.

Yes, enquiring minds would like to know. I have wondered the same thing...

Here's how I fill my diesel smart normally:

First, I fill at full pump speed until I can hear the fuel starting to come up to the bottom of the filler neck... this is typically when the auto shut off would happen.

Secondly, I typically STOP at this point, and let the 'foam' settle for 10-15 seconds, which probably also allows for more air to escape, and creates more room for fuel.

Thirdly, I fill at a slower rate until the fuel level is right at the top of the filler tube... making sure there is no foam.

Anyway - all of this WAS in an effort to try and be consistent with fills - so I could use the data to show FE of specific 'runs', 'tanks', and the associated gains/losses of tire PSI, tire sizes etc...

At the end of the day, and hence my initial query, I'm looking to use Scanguage to narrow live performance down further... I have a host of possible changes I could make to the car, and I do want to know the effect of each.

There is a bizarre logic in there... it makes sense if
a) the price per gallon is the same, and
b) you could know for certain the 'low fuel' light came on with EXACTLY the same volume of fuel remaining in the tank, and that you filled IMMEDIATELY upon seeing it light up.

Being that that is not realistic at all, what you really need to be doing is

a) fill to shut-off at a specific pump - and note your odometer reading...
b) then drive until you have used about 3/4 of a tank
c) refill to shutoff, at the SAME pump, and note your odometer reading

At this point you can determine the distance travelled, and since you now have a fuel volume to put against this, you can work out your mpg.

And you should be reasonably close to reality if you filled to shut-off at the same pump and filled at least 3/4 of a tank...

Thats what i'm doing now but going out of town this weekend and will need gas on the trip home. i've only used 1/4 tank, so it's kinda early to do a refill. whixch mean when i do refill won't be at some station/pump. i could refill right before my trip but wouldn't take more than 3-4 gallons. might just do that for a short accuracy check and when back from trip (300 miles) check again.

That would be a good idea. You also should begin tracking your FE over a loger period of time as well, so you know what you are averaging overall.

If you do your fills as you suggest you might, you should also be able to get a quick reference against the Scanguage. Just ignore the fillup 'adjustment factor'... simply hit 'done' when you do a fillup so it starts tracking a new tank (or partial tank).

Well I record the reading of the pump and the ScanGauge calculated use and know that the tank only holds so many gallons and it has in the past taken 10 gallons to fill when on the low light so if it consistantly takes less gallons to fill than expected then it should be pretty clear something is going on. Cold weather for the weekend and I will be gassing up in Newport Maine HA HA so I should get a cold tank full yielding even better gas mileage. Other than that it would take a 1 gallon container or a bigger calibrated container. Anyone want to purchase 1 gallon of slow pumped gas and then measure it later in a calibrated container?

Ya know on the low fuel light it actually may be pretty accurate because filling can take half a gallon or more or less depending upon how long you keep pumping in that last squirt to top it off whereas running down to reserve is probably going to be pretty consistant and not off by half a gallon. Of course it does depend upon how accurate they measure the botton of the tank. I guess you could put less gas in when it lights up and test it several times by only adding a gallon or two each time and keep running it down to empty.

I am wondering about the SG not measuring the fuel used accurately maybe something is going on with the engine braking when it measures fuel being injected but it really is not injecting fuel and that causes it to think you have used more than you actually have used. May have to do a bunch of down shifting on a tank of gas and see if it really throws off the measurement.

The only way to get it close IMO is to do full top offs at the same pump all the time. In my case it still shows to be off here nad there by a 1/10th or so.

Another thing Ive found is tire wear starts playing a part.

Ive never had mine be 100% spot on. Theres always a small diff. between the SG tank readings and the hand calc's for fuel ued or the total miles can be off a bit and fuel right,,,, Yada yada yada.

For me its not worth the fiddling with to try and keep it on the bubble. But yet accept it for what it is and use it as a vauble tool to fine tune, adjust or test diff. in me and my tatics.

Its a great tool! But it aint no lab insterment. :)

I'm gonan fiddle with mine a lil and see what it shows for a few fillups, but i agree it'll probly never be spot on and will always reguire a diff corrrection

Where I'm at with it now is:

a) Calibrated with GPS to match as close as possible the real distance travelled. I ended up adding 5% to make it match.
b) I am NOT adjusting any fills within Scanguage. I am however using the same pump, same direction, filled exactly to the same level (in so much as that is possible - I have a diesel - so that makes it easier)
c) I will compare tank by tank fills against Scanguage, and keep an overall number going... this will confirm the accuracy...

It seems to me that if you DO NOT adjust the speed to a GPS, then you end up adjusting the fillup, and accomplish the same thing (the FE number). I guess I'm just trying to make a perceived ambiguity in the user guide something concrete (without trying to create a lab instrument).

Nope the fillup adjusts the MPG by tweeking the amount of fuel you are burning per injector pulse. The speed calibration with a GPS will correct for the speedo error derived from the car ECU speed sensor. Both will affect the indicated and average MPG that it calculates.

That's what I meant... both ultimately affect the final FE number that the guage shows.

Simply put - I find it confusing to try and make adjustments on both, at least with my car. The GPS seems to be a sure fire way to calibrate it in my application.

Anyway, safe to say I think Scanguage rocks.

Well did a topoff today to check accuracy and scanguage was showing a lil under and had to correct it. was showing 3.6 gallons used and i pumped 3.8. mileage figured 34.23 mpg.

going to nostalgia car show/race tommorrow so will put about 200-250 mile on and since i'll be back in town will wait for low fuel light to check a full fillup. i don't have a gps so i'll have to just use the proper refill and correct it. correction worked out to 5.5% which is about inline with mileage i figured.

The scanguage speed and odometer continues to be bang on compared to GPS - 3% adjustment based on current PSI/tires.

Now I am starting to trend the 'fillup variable'. I'll need to verify the trend at a second station, so that will take another week or so. I am still running the Scanguage with 0% fillup adjustment.

I'm having a very difficult time doing this calibration, mainly due to the fact that the scangauge keeps "locking up" every so often... this requires me to unplug it and then plug it back in. I've yet to go through a tank without this happening, so I lose all tank data.

If this continues, I'll have to send it back for a new one.

Matt - so yours is locking up too huh - I find that if you shut off the ignition after being on for a time too short for the SG to communicate then it hangs and sometimes when changing the menu and then turning off the ignition screws it up too. Then it just hangs for no reason but I don't think it affects the fuel used total calculation just that it indicates that it was disconnected.

I stopped playing with the fill up adjustment after the first 9% correction was made because it kept messing with the instant MPG reading plus with the pump error we all know about who is to say what we actually are getting. The speed adjustment is still bang on with the GPS so that will stay at 3% until my tires wear down more but it probably could go up a little since it doesn't jump to the next MPH digit at .5mph over. So I am loosing a little of the fractional mph by rounding down all the time.

JanGeo -

ScanGauge is amazing because of conformance to the OBDII specification, but I'll bet that all these computers have "personalities" of implementation that lead to these quirks. From the POV of a programmer, I would say that alot of the things done at this site would void the warranty, or at least give the telephone support person fits. Also, I would think that the SG would be least reliable with newer cars like the Fit, because it's doubtfull that the SG has been *tested* against the newer cars.

For me the SG will always be a relative barometer of performance. Pick a set of rules for using your SG and stick with them.

CarloSW2

The only reliable way to use the fillup adjustment factor IMO is to:

a) calibrate the speed - you want this to match
b) leave the fillup adjustment at zero for about 5 fills/partial fills.
c) fill at the same pump/same direction, yada yada
d) keep a record of each tank as Scanguage shows it, distance, fuel used and mpg
e) after 5 fills - compare the actual receipts against the Scanguage totals - and if required, adjust as required to match the 'percentage of variance'

When I first posted this thread, it was partly due to the frustration of having adjusted the fillup based on a single result, and then seeing it needing to be adjusted again the next time, etc...

Now after 5 fills - I can see a trend - a real difference - and I will make a one time adjustment. I've also since learned that OBDII verus CANBUS in my car yield different results. OBDII is doing it off of sensors, so it is a reasonable estimation... most importantly - it is consistent...

I think the manual could be expanded a bit to help articulate these things clearer... none-the-less, I'm very pleased with it.

Interesting . . . Any thoughts on the injector operation when engine braking - some say the fuel is cut off and the SG will show it still injecting fuel . . . I see at higher revs as much at 0.5 - 0.6gph being injected with no throttle.

Well my fillup a couple days ago showed that slow pumping at this particular station Somerset pump #3 yielded very close gallons to what the SG said I used . . . SG 9.6 gallons - pump 9.678 gallons full. I need to get back to the Dartmouth pump #8 again. Be nice to have a clear filler hose going to the tank to see the fuel level as it comes up to the filler nozzle.