wmcbrine
Posts: 385
Joined: Wed Apr 29, 2015 10:26 am
Location: Laurel, MD
Contact: Website

Guess-o-meter

Tue Apr 23, 2019 6:07 pm

I don't usually have an issue with the range remaining display*, but it did a couple of amusing things yesterday:

1. On a four-mile trip, I lost ten estimated miles (rather than 8) on the round trip... OK, whatever. But I lost 8 of those on the way out, and only 2 on the way back (same path).

2. One more trip after that, and I came home and plugged in. Usually I run the car down to blinking (~8 miles RR or less), and then charge it to full. But yesterday, I realized there was something I needed to do before that, so I pulled the plug after three hours. The RR said 48... I drove 7.5 miles round trip... and it said 49 when I was done. :) I did lose a bar or two, though.

* That's a lie. My car regularly reports ~62-64 miles RR after a full charge, but by the time I get to flashing, miles driven + RR tends to be only in the 50's. But, this doesn't seem too serious.

wmcbrine
Posts: 385
Joined: Wed Apr 29, 2015 10:26 am
Location: Laurel, MD
Contact: Website

Re: Guess-o-meter

Wed Apr 24, 2019 5:48 pm

wmcbrine wrote:1. On a four-mile trip, I lost ten estimated miles (rather than 8) on the round trip... OK, whatever. But I lost 8 of those on the way out, and only 2 on the way back (same path).

This turns out to be more or less reproducible. I didn't realize it at first, but the outgoing trip is about 95% uphill.

JoeS
Site Moderator
Posts: 3740
Joined: Thu Dec 15, 2011 5:47 am
Location: Silicon Valley, California

Re: Guess-o-meter

Thu Apr 25, 2019 8:09 am

wmcbrine, thank you for your post and subsequent update.

We've had discussions over the years about the accuracy of our Range Remaining (RR) display. In our case, we DON'T call it a GOM (Guess-O-Meter - like in a Leaf) because the i-MiEV's algorithm has now been proven to be very consistent, and which I believe to be as follows:

1. The prediction is very simply based on the moving average of the consumption of the preceding 15 miles (24km).
2. The RR display is decremented by roughly 20% when the heater is turned on (also dependent on fan speed)
3. The RR display is decremented by very roughly 10% when the aircon is turned on.

Turning on the HVAC does not affect the internally-stored motive energy consumption moving average figure.

Yes, as you noticed, changes in elevation can make a significant difference. For example, I live approximately 500' above the town I drive down to every day and invariably gain a few RR miles in the process. Thus, for me going home, getting off the Interstate driving 2-1/2 miles and climbing approximately 300', means I need to compensate for my RR reading and always leave a bit of reserve - been lucky so far, as in seven years neither my wife nor I have ever seen turtle.

Especially on highway driving, by far the greatest variable is airspeed and an 80mph run can rapidly skew things dramatically - as can bucking a serious headwind.

I've found altitude to not be too much of a problem as long as you get to go back down the hill during the trip as it seems the net loss is not much more than in level driving.

Until manufacturers include the Nav destination (which will contain the elevation changes and anticipated vehicle speeds), internal temperature, external temperature, windspeed prediction over the route, and driver's characteristics (e.g., featherfoot vs. leadfoot), the Range Remaining on any car will be just a rough estimate. In my experience, Tesla's Trip graph is very accurate because it encompasses most of those variables and thus significantly contributes to my long-distance-driving peace of mind.
EVs: 2 Wht/Blu SE Prem., '13 Tesla MS85, 3 156v CorbinSparrows (2 Li-ion), 24v EcoScoot(LiFePO4)
EV Conv: 156v '86 Ram PU, 144v '65 Saab 96
Hybrids: 48v1kW bike
ICE: '88 Isuzu Trooper. Mothballed: '67 Saab (orig.owner), '76 MBZ L206D RHD RV

Don
Site Moderator
Posts: 2869
Joined: Thu May 10, 2012 3:55 pm
Location: Biloxi MS

Re: Guess-o-meter

Thu Apr 25, 2019 10:01 am

wmcbrine wrote:This turns out to be more or less reproducible. I didn't realize it at first, but the outgoing trip is about 95% uphill.

Yes, as you (finally) discovered, the RR calculation is actually quite accurate, given the parameters is uses to base it's calculations on. Now that you know this, *and* you know how and where the car was driven for the previous 15 miles, what it's showing you won't come as a surprise very often

Don
2012 iMiEV SE Premium, White
2012 iMiEV SE, White
2017 Chevy Volt Premier
2014 Ford Transit Connect XLT SWB wagon, 14,000 miles
1979 Honda CBX six into six

Return to “General / Main Owners Forum”