GM Volt Forum banner

1 - 20 of 23 Posts

·
Registered
Joined
·
1,018 Posts
Discussion Starter · #1 · (Edited)
I'm trying to find the true miles per energy input (as kWh or gallons of gas) efficiency data vor my new Volt. Can you help checking whether this assessment is correct? It does not seem to match the info displayed on voltstats.net or myvolt.com

The Volt itself prominently shows the MPG measure, which is totally meaningless since it ignores the kWh of electrical energy that we pay for to propel the Volt. Most data on volstat.net and myvolt.com is borderline useless for the same reason.

Since I drive mainly electric, I think the best measure is to divide the number of electrical miles driven by the number kilowatt hours that came out of the electrical plug. This 'plug-to-wheel efficiency' is the number of miles that my Volt drives on a kWh of energy including all losses, climate control and brake re-generation. The bigger the better, since I pay about 2x the national average for a kWh here in the Bay Area. The EPA sticker for my 2013 Volt lists 98 MPGe, which at 33.7 kWh/gallon equates to 2.9 Miles per kWh plug-to-wheel.

I once measured 14.2 kWh for a full level-1 charge using a kill-a-watt meter, and my electrical range in the current weather is about 43 miles. So that makes 3.03 miles per kWh plug-to-wheel, slightly better than sticker, it seems about right.

Most of the times, however, the battery is not empty, or I disconnect the charge cord before its full. Also, the level-2 240V charger does not have a power consumption meter. So where can I get proper miles/kWh data for daily use?

The Volt itself displays both the number of electrical miles driven and the amount of battery kWh used. Unfortunately it doesn't divide that into a mile/kWh (or MPGe) rating for the last charge. Also, this is 'battery-to-wheel' data that does not include the charging and battery losses.

Myvolt.com doesn't display this data directly either. It shows a 'lifetime electric ecomomy' which in my case just shows '-' (apparently, my Volt doesn't have a life ;) ). It shows lifetime gas MPG at 175, which is BS since it obviously counted electrically powered miles to get to that inflated number. Volstats.net is even less informative, since it does not read the kWh charged into the battery at all (but instead derives it indirectly).

The good news is that myvolt.com does show my charging history in great detail, including the number of kWh charged each time. It appears that this number might be trustworthy: a full charge of the empty battery takes about 12.2 kWh, and I can see the correct time and length of each charge. The battery produces some 10.3 kWh, so about 20% loss seems reasonable for a 240V charge. Does anybody know whether this is correct trustworthy data indeed? And why does the Volt not display this data if it knows it?

I downloaded last week's data charging details from myvolt.com into an excel spreadsheet. I also downloaded the trip EV mileage data for the same period. For the past 5 days I drove 162.85 electric miles, and it took 51.88 kWh to top the batteries off after this driving. That means I averaged 3.14 miles/kWh in a rather warm week that required constant airconditioning. That is 106 MPGe. Even the off-peak PG&E rate here in Silicon Valley is a whopping 20c/kWh, so I'm doing 6.4 cents/mile, about 3x less than the 22 MPG car that the Volt replaced at current gas prices.

This measured data seems to have no relation to the voltstats.net page of my Volt, which lists my MPGe at only 67.40. What data can be trusted?
 

·
Registered
Joined
·
805 Posts
Wow, you have a 20c per kWh rate? That's hurts. Do you get a flat rate?

I think the MPGe from the stats is based on $1.50 per full charge up (12c/kWh x 12.5kWh per charge) and only 35 miles per Battery charge (EPA rating), and based on $3.95/Gallon of gas. And base on the national average price per kWh of 12 cents.

SO, on your personal stats:

576.38 EV Miles
103.51 Gas Miles
MPGcs 26.61

You'd have to take EV Miles divided by 35, which equals 16.468 Charge cycles, then times $1.50, which equals $24.70 for cost of EV miles.
THEN,...
103.51 Gas miles divided by 26.61 MPG equals 3.89 times a gallon cost of $3.95 equals $15.37 for cost of Gas miles.

Add $24.70 plus $15.37 equals $40.07 (Total cost for all miles)

THEN,...

Divide $40.07 by $3.95 would equal a total gallons equivalent of 10.14 Gallons.

679.89 divided by 10.14 Gallons is around 67.05 MPGe.

I didn't go down to 1/1000ths accuracy on all computations, so it differs slightly from what the Voltstats says of 67.4 MPGe

But this seems close to their method.
 

·
Registered
Joined
·
6,464 Posts
This measured data seems to have no relation to the voltstats.net page of my Volt, which lists my MPGe at only 67.40. What data can be trusted?
I always default to the data that I calculate myself, so I would trust your 106 MPGe.

I wonder if your Voltstats numbers are calculating your MPGe based both on your electric miles (miles per kWh) and your ICE miles (miles per gal * 33.7 kWh). If they are including your gas mileage in your MPGe, then it would lower those numbers significantly.
 

·
Registered
Joined
·
309 Posts
PG&E does offer "special" rates for EV owners. The VOLT is elgible. I think you only need to have a "smart" meter. I applied a few weeks ago, but have yet to see a change to my monthly bill. PG&E told me it will take a few billing cycles to take effect.

Btw, I have a PowerSave "EnviR" home energy monitor for keeping track of my 240V charging power consumption. It's not that accurate though, like I think you would need it to be.
 

·
Premium Member
Joined
·
14,156 Posts
The good news is that myvolt.com does show my charging history in great detail, including the number of kWh charged each time. It appears that this number might be trustworthy: a full charge of the empty battery takes about 12.2 kWh, and I can see the correct time and length of each charge. The battery produces some 10.3 kWh, so about 20% loss seems reasonable for a 240V charge. Does anybody know whether this is correct trustworthy data indeed? And why does the Volt not display this data if it knows it?

I downloaded last week's data charging details from myvolt.com into an excel spreadsheet. I also downloaded the trip EV mileage data for the same period. For the past 5 days I drove 162.85 electric miles, and it took 51.88 kWh to top the batteries off after this driving. That means I averaged 3.14 miles/kWh in a rather warm week that required constant airconditioning. That is 106 MPGe. Even the off-peak PG&E rate here in Silicon Valley is a whopping 20c/kWh, so I'm doing 6.4 cents/mile, about 3x less than the 22 MPG car that the Volt replaced at current gas prices.
The 3.14 miles/kWh from the wall sound right, assuming you have a light foot. Don't you get a monthly report that gives you the kWh/100 miles number? (That's wall to wheels BTW). Anyway, a 20% loss seems high for 3.3 kW charging. I'd think 15% would be more normal. However, if it was warm during charging the need for more cooling might account for the higher than expected losses.
 

·
Moderator
Joined
·
8,680 Posts
Voltstats knows that it doesn't know the electrical side - and so it calculates MPGe assuming EPA miles per kWh, with your real gas/electric split and gas mileage. Since you figured out a way to get your actual electric performance, go with that over Voltstats.
 

·
Registered
Joined
·
3,165 Posts
As other said, voltstats MPGe is just a simple, and wrong, estimate.

I have a spreadsheet that I use with the voltstats.net data to estimate my daily use.. It uses battery SOC and an external calibration (wall kWh for full charge) and it presumes you don't charge midday. I used to work quite well for me but with the recent change that on-star forced on voltstats.net (only polling 2 times per day), its not near as useful as the car is often charging by the time it takes its reading so you get less frequent recomputation of SOC usage per EV mile). Still a reasonable model.

You can get a copy of the computations at http://vast.uccs.edu/~tboult/VOLT/voltdata_withMPGe.xls
 

·
Registered
Joined
·
284 Posts
I would hope that your numbers would take into account any TMS power penalty and cabin preconditioning in computing your efficiency. Not accounting for charging efficiency, I know I'm getting anywhere from 3.6 to 4.2 M/Kwh which would work out to between 121 and 142 mpge. I live in a co-op in an incorporated village that generates their own electricity. I plug into one of the 120 volt outlets in my garage and since it's in the building common area, i charge at a commercial rate on 5.93 cents/kwh as opposed to the 9.2 cents/kwh I get charged for my apartment. Now Long Island is basically flat, I also don't have to deal with the hilly terrain that exist in San Fransisco, but i think your scenario represents maybe close to a worst case scenario from a terrain and power cost point of view. Also I think a 12% penalty for charging and TMS loss is more realistic.
 

·
Registered
Joined
·
48 Posts
I had my Volt for about a month and have been doing my best to tally these numbers (with the complete and utter frustration with the OnStar numbers since OnStar seems to arbitrarily lose charge session in the statistics it returns (not to mention the random delays in how long it takes for their site to receive those charge listings that it chooses to display).

Anyway, a few of the entries from my spreadsheet (for which I do have the charging info from OnStar) is:

Code:
Miles     KWused     OnStar     Measured    Utility
9.9           2.5         3.09          3.47       3.68
11.8         3.2         4.13          4.56
19.5         4.6         5.94          6.54
23.7         5.9         7.41          8.17
19.5                       5.7           6.33
13.9         3.9         4.98          5.51
So, I am finding a consistent ratio of about 1.1 between the kWh reported by OnStar and the measured energy through my watt meter. I am also finding a consistent ratio of about 1.4 between the energy reported by the console display on the Volt at the end of the day, and the measured energy through my watt meter. The "Utility" number, of which I have only one entry listed above since I have not done the manual calculation yet for the other dates, is my usage as reported by my smart meter for the period midnight to 6AM after subtracting for my best estimate of baseline usage during this period (this is my minimal use for the lowest hour in the midnight to 6AM range, multiplied by 6 - note that my Volt ends charging usually by 4AM). I would not expect this number to be as accurate as the others, yet I do find it to be consistently higher than what the watt meter reports as used, and I attribute that to a combination of underestimating my baseline, and (hopefully not much) heating of the electrical wiring.
 

·
Registered
Joined
·
1,018 Posts
Discussion Starter · #10 · (Edited)
Dinosaur blood mile vs electrical mile on the Volt.

Thanks for the many valuable data. It is mind boggling why GM and volstat.net focus so much on that irrelevant (and easily manipulatable) MPG number. At the same time they don't show the more accurate efficiency numbers. Even the MPGe number is useless in comparing the actual relative operating cost of electric driving versus driving on gas, because it assumes incorrect electrical and gas costs. Rather than MPG or MPGe, I want to find out:


  • My true electric operating cost per mile. This is the utility cost multiplied by the mile/kWh wall-to-wheel efficiency. The earlier analysis shows I'm at about 6.4 cents/mile in the SF bay area.

  • My true gas operating cost per mile for the Volt. This is the gas cost ($4.39 premium unleaded here in the Bay Area) times the MPGcs (38MPG ?), so roughly 12 cents/mile. I have not measured this, but at current rates, I expect that a dinosaur blood mile is 2x more expensive than an electric mile in the Volt. Obviously, this $ comparison does not take the environmental and geo-political impact of the energy sources into account.

  • How to manipulate the true cost by any means: either by getting better electric rates, or by driving or charging more efficiently. For the latter, this thread is interesting.

To do all that we must know what data can be trusted and how accurate they are. And that is surprisingly hard despite the fancy remote monitoring that Onstar provides.

BCNeuman's input is great, since it shows a 10% undercounting of kWh by myvolt. I suspect that the Volt does not actually measure the charging current, but simply multiplies charging time by a factor. And it seems that that factor is 10% optimistic. It is good to know that as a rule of thumb, the battery-to-wheel number reported by the Volt is 1.4 times lower than the true utility usage. The error will vary depending on the battery cooling/heating requirements.

As for utility costs, PG&E's time-of-use and multiple tier plans make it very hard to assess the true cost and the best rate plan. The killer is the low monthly baseline. I my case, the baseline is only 319 kWh/month, while our family typically uses about 900 kWh/month before buying the Volt. That means that any extra usage for the Volt always comes from the most expensive tier: $0.34 in E-1, $0.31 in E-6 TOU off-peak, or $0.198 in the E-9a TOU off-peak. The ways to drop the cost is to go solar (which I'm considering) or to charge at work (which I do).
 

·
Registered
Joined
·
3,165 Posts
Thanks for the many valuable data. It is mind boggling why GM and volstat.net focus so much on that irrelevant (and easily manipulatable) MPG number. At the same time they don't show the more accurate efficiency numbers. Even the MPGe number is useless in comparing the actual relative operating cost of electric driving versus driving on gas, because it assumes incorrect electrical and gas costs. Rather than MPG or MPGe, I want to find out:


  • My true electric operating cost per mile. This is the utility cost multiplied by the kWh/mile wall-to-wheel efficiency. The earlier analysis shows I'm at about 6.4 cents/mile in the SF bay area.

SNIP
.

Its not mike's fault that volstats.net does not compute those things.. its really driven by what onstar pulls from the car..

My spreadsheet above computes cost per mile, if the power is well estiamted.. I have an idea on how to improve that with the new voltstats.net data. Note my model does not account for TMS very well... I just use a different factor in winter (i.e. I recalibrate wall to car a few times a year).
 

·
Registered
Joined
·
1,018 Posts
Discussion Starter · #12 · (Edited)
I think the MPGe from the stats is based on $1.50 per full charge up (12c/kWh x 12.5kWh per charge) and only 35 miles per Battery charge (EPA rating), and based on $3.95/Gallon of gas. And base on the national average price per kWh of 12 cents.

SO, on your personal stats:

576.38 EV Miles
103.51 Gas Miles
MPGcs 26.61
....
Wow, thanks, That is a very convoluted method that volstats.net needs to use due to the limited data from onstar. Where does volstat get the MPGcs number come from? Onstar? It seems low and is likely wildly inaccurate (probably because the car has burnt only 4 gallons since its birth in Hamtramck).

I any case, that number seems a poor representation for the actual efficiency (and my driving style), nor does it seem useful for the true operating cost of the Volt.
 

·
Registered
Joined
·
3,165 Posts
Wow, thanks, That is a very convoluted method that volstats.net needs to use due to the limited data from onstar. Where does volstat get the MPGcs number come from? Onstar? It seems low and is likely wildly inaccurate (probably because the car has burnt only 4 gallons since its birth in Hamtramck).

I any case, that number seems a poor representation for the actual efficiency (and my driving style), nor does it seem useful for the true operation cost of the Volt.

I'm not the developer of voltstats.net but I've chatted with him about this.


Its even crazier when your MPG_CS seems to change when you have used absolutely no gas or driven in CS Mode.


The car reports lifetime MPG, lifetime EV mile and lifetime total miles from which one can compute approximate gallons used (LTM / lifetime MPG), then compute MPG_CS as (LTM-LEVM)/Gallons used.

Its also convoluted based on the data one can get from the onstar interface.
Note you can get some odd results with the above when the milage is low, see
http://gm-volt.com/forum/archive/index.php/t-9059.html

where it explained the car's MPG can be reset before delivery as part fo the dealers PDI, but the odometer (lifetime total miles) is not reset.

So the lifetime MPG is using a number of gallons and number of miles that you cannot see, which is why MPG_CS varies slightly even when there is no gas usage.. the approximation errors are changing. It was driving me nuts that I could not make them align with my measurements.
 

·
Registered
Joined
·
476 Posts
BCNeuman's input is great, since it shows a 10% undercounting of kWh by myvolt. I suspect that the Volt does not actually measure the charging current, but simply multiplies charging time by a factor. And it seems that that factor is 10% optimistic. It is good to know that as a rule of thumb, the battery-to-wheel number reported by the Volt is 1.4 times lower than the true utility usage. The error will vary depending on the battery cooling/heating requirements.
1.4 (I guess you are saying 28+% loss) seems rather high to me. My Coulomb 240v charger reports about 15-16% loss, e.g. car reports 9.7 kw-hr, full recharge takes 11.5 kw-hr. 120v charger loses more but I doubt that much. EPA says 20% loss, 12.9kwhr for the 2011/2012 models with slightly less capacity than the 2013s. It uses somewhat more if it gets really hot and the car has to run TMS a lot. BCNeuman maybe underestimated his household baseline usage during the time to get the higher figures.
 

·
Registered
Joined
·
1,018 Posts
Discussion Starter · #15 ·
1.4 (I guess you are saying 28+% loss) seems rather high to me. My Coulomb 240v charger reports about 15-16% loss, e.g. car reports 9.7 kw-hr, full recharge takes 11.5 kw-hr. 120v charger loses more but I doubt that much. EPA says 20% loss, 12.9kwhr for the 2011/2012 models with slightly less capacity than the 2013s. It uses somewhat more if it gets really hot and the car has to run TMS a lot. BCNeuman maybe underestimated his household baseline usage during the time to get the higher figures.
Thanks Stephen. So your data shows a 15% between measured wall-to-wheel and Volt-reported battery-to-wheel consumption. That is more in line with the myvolt charging data, and would suggest that I can trust that data.
 

·
Registered
Joined
·
1,018 Posts
Discussion Starter · #16 ·
Is the weird MPG reporting a deliberate marketing ploy?

I'm not the developer of voltstats.net but I've chatted with him about this.

Its even crazier when your MPG_CS seems to change when you have used absolutely no gas or driven in CS Mode.
...
I wonder what GM was thinking by only providing the convoluted MPG number in the Volt. I suspect that it was a deliberate marketing trick to come up with a number that always looks great compared to gas cars and that is not so easily comparable with other (electric) cars. So rather than giving meaningful data, some clever marketing expert opted to have the Volt's display show a bogus MPG number in a big font, and make actual operating efficiency data hard to get. Am I too cynical? There does not seem so much to hide: its pretty good and competitive.
 

·
Registered
Joined
·
476 Posts
Wow, thanks, That is a very convoluted method that volstats.net needs to use due to the limited data from onstar. Where does volstat get the MPGcs number come from? Onstar? It seems low and is likely wildly inaccurate (probably because the car has burnt only 4 gallons since its birth in Hamtramck).
BTW, for the voltstats figures:
1. MPGe is calculated using the EPA estimate of the 2011/2012 models, which is 36 kw-hr per 100 miles, and a figure of 33.7 kw-hr per "electric gallon" = ~ 94 MPGe. Not like how VikAiRious stated; it has nothing to do with the cost of electricity or the cost of gas, it's completely independent of those factors. It just happens that with the costs he used it works out to be about the same. But if gas went up to 4.50 per gallon, his calculation would change, but the MPGe would not. Many of us beat EPA; my actual usage is fluctuating between 30-31 kw-hr per 100 depending on summer/winter here in mild San Jose. As stated before the site can't get kw-hr usage data from Onstar (yet, he's in contact with them to maybe get the data eventually), you pretty much have to calculate this yourself if you want accuracy.

2. MPGcs is derived from the "lifetime MPG" reported by onstar. It figures out gallons used by total miles / MPG. Then total miles - ev miles = your gas miles, and MPGcs = gas miles / gallons used. The problem with this, is that the lifetime MPG reported by Onstar is kind of weirdly inaccurate, compared with the actual gallons in the tank that is also in the voltstats/onstar data. If you download your car's data spreadsheet from Voltstats, you'll see that it is showing gallons burned very slowly increasing, even in days/weeks you are 100% EV and not using gas at all. So this introduces a small error that is greatly magnified for cars that haven't burned a lot of gas. For me it was about an extra .067 gallons per month or so than reality. If you drive more, the absolute error seems to be about the same, so the percentage it is off by decreases. So by the end of one year, my real MPGcs was ~41.5 mpg on 20.5 gallons (based on fillups, and what was left in the tank vs. the fullish tank I started with), while voltstats/onstar was reporting 39.9 mpg on 21.3 gallons. But early on in my ownership, the error was greater, since it would be showing like 4.4 gallons burned when I had only really used 4 gallons.


I agree that "MPG" as the Volt uses a lot is rather stupid. I wish the car had stats for lifetime and trip meter MPG-CS, and kw-hr/100 miles, separately. OK if the car didn't measure charging losses as I can always increase the latter figure by 15-20% to estimate.
 

·
Registered
Joined
·
48 Posts
The ratio between measured energy from the wall, and the reported energy used to charge the battery on MyVolt is 1.1. The 1.4 ratio is the energy reportedly used to drive the vehicle the previous day (as reported on the console at the end of the days driving), and the energy from the wall to perform a full recharge. Thus the 1.4 figure includes battery inefficiency.
 

·
Registered
Joined
·
48 Posts
It would be useful to understand what "losses" account for the differences in energy at different points of measurement. So, lets try to understand this one linkage at a time. Based on the numbers I posted above (and which continue to b e confirmed as I enter each days data) I am seeing a difference of 10% (e.g. a possible loss of 9%) in the energy measured using my KiloWatt meter to which the charging unit is connected, and the kWh reported for charging by Onstar.

What is being measured and reported by Onstar?

Is it supposed to be the energy taken from the wall (in which case it seems to under-report by 9%), or is it the energy added to the battery (meaning that 9% of what is coming from the wall is going to losses in the charging circuit, energy intentionally used to warm or cool the batteries to their optimal charge conditions, and other energy usage by the vehicle during the time that the volt is plugged in and charging.
 

·
Registered
Joined
·
1,018 Posts
Discussion Starter · #20 ·
Indeed, it would be good to know what charge data is measured by Onstar (which gets it from our Volt): energy sticked in the battery or total energy. I suppose that it cannot measure the losses in the charge cable.

I observed that the onstar-reported charge data were easily constant for all the 'big charges' I saw. Here is my Level2 charge data:
250 minutes 12.19kWh = 2.9256 kWh/h
251 minutes 12.23kWh = 2.9235 kWh/h
256 minutes 12.48kWh = 2.9250 kWh/h
225 minutes 10.98kWh = 2.9280 kWh/h
163 minutes 08.72kWh = 3.2098 kWh/h
053 minutes 02.69kWh = 3.0452 kWh/h

The long 'empty-to-full' charges have a suspiciously similar charge rate. The shorter charges have different rates. So it does not seem that the Volt multiplies charge time by a constant current to get kWh, but it might still be that it does not actually measure the incoming current (but instead interrogaties the battery controllers). Thisteardown shows the battery controller board where it likely gets that data from.
 
1 - 20 of 23 Posts
Top