RavCharge, a solution to charge timer and entune woes

[yblaser]

It worked like a charm last night starting the charging at 12:30 am. I did get three charge starting emails though, at 12:30 am, 12:31 am, and 12:50 am. Looking at my power consumption the car started charging at 12:30 am and continued without interruption.

One suggestion for future capability would be to have a pre-climate schedule like the charge schedule (similar to the car's internal capability) but have the option to only run pre-climate if the car is plugged in.

Again, great job!

 

[occ]

I extended charged last night. This morning, I logged into ravcharge.com for the first time and it reported I have approx 35kwh. Is is correct? I thought Tony's above post said you can differentiate between 100% "Normal" and "Extended" charge. (I've also press the "update now" button several times with the same 35.0kwh status).


Also, can the climate control override the 50% battery limitation?


Terrific app, BTW!

 

[fooljoe]

It worked like a charm last night starting the charging at 12:30 am. I did get three charge starting emails though, at 12:30 am, 12:31 am, and 12:50 am. Looking at my power consumption the car started charging at 12:30 am and continued without interruption.

The two emails at 12:30 and 12:31 seem to always happen when sending a charge start command through entune. The later one at 12:50 is because I've set RavCharge to automatically send a "backup" charge command 20 minutes after the nominal command, just in case. This is probably overly cautious, since the first one will try about 4 times until the charge start is confirmed, but it doesn't seem to hurt anything except for the extra email.

One suggestion for future capability would be to have a pre-climate schedule like the charge schedule (similar to the car's internal capability) but have the option to only run pre-climate if the car is plugged in.

Interesting idea, I'll check it out. There seems to be a lot of interesting things that can be done with the pre climate.

I extended charged last night. This morning, I logged into ravcharge.com for the first time and it reported I have approx 35kwh. Is is correct? I thought Tony's above post said you can differentiate between 100% "Normal" and "Extended" charge. (I've also press the "update now" button several times with the same 35.0kwh status).

Everything's built around normal mode for now. The app can tell if you're in extended or normal charge mode, but I think the SOC reading is always based on normal charge mode (like the bars in the car.) So 100% SOC in extended charge could really mean anything from 35-41.8kwh. I haven't really tested this to make sure though - if you get a chance do a status update after you drive a bit but before you lose any of the visible battery bars in the dash and we'll see what happens.

Also, can the climate control override the 50% battery limitation?

I actually didn't even know about this limitation, as I've never run into it. My guess would be RavCharge won't be able to do it if the entune app can't, but you're certainly welcome to try and report back!

Thanks for the feedback!

 

[Dsinned]

This may be a "dumb" question, but if you close (kill) your browser, while RAVCharge is waiting to commence a charge set by Timer1 left in the "On" mode, will this prevent it from working?

I'm not really sure what Timer2 is for, or how to use it. It seems redundant if you're already using Timer #1. Is #2 for scheduling a charge at a different time of day/night, say on different days of the week from #1? For example, can I set up Timer #1 for Saturday and Sundays starting charges at 21:00 (9pm), and TImer #2 for the rest of the week starting at 24:00 (midnight), where both pre-set scheduled charges always start during off peak rate periods?

Suggestion: Most of us probably have determined a running average miles/kWh that is suitable for our cars most of the time. To avoid confusion, RAVCharge should allow user input of this value instead of giving a selection of values at 2/3/4 miles/kWh. Inputting an average value characteristic of the user's car, which could be a 3 significant digits decimal number like "3.5", would then result in a more precise range estimate for the battery's current SoC.

 

[fooljoe]

This may be a "dumb" question, but if you close (kill) your browser, while RAVCharge is waiting to commence a charge set by the Timer #1 left in the "On" mode, will this prevent it from working?

No. Once you set a timer, and see that the time you selected is reflected on your screen, it's saved on a separate, always-running server that's independent of the server that handles the web page you see.

The same goes for the command buttons to update, charge, or start/stop climate control. Once you push the button, the command is sent to the server, where it will keep trying for about 8 minutes or until it works. You can wait around for the confirmation pop-up if you wish, but there's no need to. One idea for a future improvement in this area is to send email or SMS notifications to the user to confirm that a command worked or let you know if it failed so you can come back to RavCharge and try again.

I'm not really sure what Timer #2 is for, or how to use it. It seems redundant if you're already using Timer #1. Is #2 for scheduling a charge at a different time of day/night, say on different days of the week from #1? For example, can I set up Timer #1 for Saturday and Sundays starting charges at 21:00 (9pm), and TImer #2 for the rest of the week starting at 2400 (midnight), where both set scheduled charges to always start during off peak rate periods?

Yeah, that's the idea. Some people might want to charge at different times on the weekends, or charge twice per day, or use timer 2 for setting a temporary timer without having to change the usual timer, or whatever else anyone thinks of.

Suggestion: Most of us probably have determined a running average miles/kWh that is suitable for our cars most of the time. To avoid confusion, RAVCharge should allow user input of this value instead of giving a selection of values at 2/3/4 miles/kWh. Inputting an average value characteristic of the user's car, which could be a 3 significant digits decimal number like "3.5", would then result in a more precise range estimate for the battery's current SoC.

Good suggestion, and I have thought of doing this, but what you're describing is basically what the Guess-o-Meter already does, and I was trying to come up with something different. That said, the whole 2/3/4 thing was just a quick idea, and I certainly do plan on improving that display one way or another.

 

[TonyWilliams]

but what you're describing is basically what the Guess-o-Meter already does, and I was trying to come up with something different. That said, the whole 2/3/4 thing was just a quick idea, and I certainly do plan on improving that display one way or another.

The GOM is guessing based on some unknown formula based on past performance. Selecting 3.5 miles/kWh, for example, is my user selectable prediction of the future driving.

We already have this feature on the LEAF Spy app:

 

[fooljoe]

I tried out a new status page that lets you vary the miles/kwh and calculate the estimated range in real time. It'll reset to a default of 3 m/kwh each time you refresh for now, but I'll work on making the value you choose persistent.

EDIT: It's persistent now.

 

[Dsinned]

Last night I used Timer1 to do a scheduled (normal) charge. I set it to start at 9pm, and am very pleased to report that it WORKED exactly as advertised. The first attempt (according to Entunes Email Notification) failed at that actual time setting, then successful on the 2nd attempt (at 9:02pm). RavCharge sent another charge "start" command at 9:20pm for good measure.

The charge completed about 2 1/2 hours later (11:27pm), resulting in "100%" charge status on RavCharge and 105 miles ESTIMATED range (at default RavCharge setting of 3.0 mi/kWh).

At the end of last week, the last running average displayed on the center console inside my car was "3.5mi/kWh", atypical of my wife's usual driving style. I reset the display just before I drove the car ~40 more miles on Saturday, so the average eventually dropped to "3.3mi/kWh" by the end of the day. If my wife had been driving all day, no doubt the reading would have continued dropping even lower.

All of which brings me to say, I didn't change the default average for range estimation in RavCharge, because I don't really know which is more accurate, the GoM in the car or RavCharge. After charge completion last night, RavCharge displayed "105 miles" (still at the default of 3 mi/kWh). However, the actual range displayed on the GoM in the car (with A/C off) was "95 miles" for a mismatch of about 10 miles, or 0.3mi/kWh.

A lot of this is still essentially "guesswork". A range of 95 miles corresponds to a usable battery SoC of only ~32kWh, (not 35kWh as generally assumed after being fully charged in normal mode). However, calculating what 80% of a fully charged battery actually would be, results in 33.4kWh of usable capacity.

Which of these charge capacities is more accurate, Toyota's published 35kWh or 33.4kWh (80% of 41.8kWh)? To be more conservative, the obvious answer is the latter, which in my case just so happens to correlate closer to my actual GoM range.
Then there is battery capacity degradation over time as an influencing factor to be considered. I have only owner my car for the last 10 months. It has only been driven about 5000 miles, so I do not consider any cummulative battery degradation as yet very significant. However, RavCharge should probably still allow for degradation with some sort of time based multiplier to determine what corresponds to a 100% (normal) charge status. For example, -3% per year, or -0.1% per 500 miles, could be applied.

For the time being, my GoM's normalized reading based on a running average for my household's driving style of 3.0mi/kWh is probably a pretty good "number". The estimate can be skewed a little one way or the other if the estimation of range is more for one specific person's driving style or special circumstances.

The bottom line, I'd rather not rely on an overestimated range, so the RavCharge default of "3" is right for me, but just to be super conservative, 2.7 would be better to use as RavCharge's running average for my wife. :mrgreen:

In any case, I feel RavCharge is a complete SUCCESS!!!

Thank you!

 

[reeler]

I logged into RAVCharge.com and received the below dump in my browser. I have two RAV4s and the Entune apps don't even work well with the two cars. I hope yours will deal with this.

ReferenceError: /home/ubuntu/nodesrv/ui/views/timers.ejs:16
14| <h3>
15|
>> 16| <%- nick %>
17| </h3>
18| </div>
19| <div id="statuspage" data-role="content">

nick is not defined
at eval (eval at <anonymous> (/home/ubuntu/nodesrv/ui/node_modules/ejs/lib/ejs.js:236:14), <anonymous>:31:548)
at eval (eval at <anonymous> (/home/ubuntu/nodesrv/ui/node_modules/ejs/lib/ejs.js:236:14), <anonymous>:33:37)
at /home/ubuntu/nodesrv/ui/node_modules/ejs/lib/ejs.js:249:15
at Object.exports.render (/home/ubuntu/nodesrv/ui/node_modules/ejs/lib/ejs.js:287:13)
at View.exports.renderFile [as engine] (/home/ubuntu/nodesrv/ui/node_modules/ejs/lib/ejs.js:317:20)
at View.render (/home/ubuntu/nodesrv/ui/node_modules/express/lib/view.js:76:8)
at Function.app.render (/home/ubuntu/nodesrv/ui/node_modules/express/lib/application.js:505:10)
at ServerResponse.res.render (/home/ubuntu/nodesrv/ui/node_modules/express/lib/response.js:756:7)
at IncomingMessage.<anonymous> (/home/ubuntu/nodesrv/ui/app.js:436:13)
at IncomingMessage.EventEmitter.emit (events.js:117:20)

 

[fooljoe]

I logged into RAVCharge.com and received the below dump in my browser. I have two RAV4s and the Entune apps don't even work well with the two cars. I hope yours will deal with this.

Thanks for pointing this out. I haven't made any provision for someone with more than one car under the same entune account - although I thought it would just pick one, not crash :shock: . But now I have someone to do some testing and figure it out! Send me a PM with your entune email and I'll figure it out. Thanks!

EDIT: I figured out which email belongs to you, and I set it up so one of your cars works. The other one doesn't seem to work with entune, so I can't do anything about it. Maybe you still need to set it up with entune? Entune gives me the same error that I've seen with other users who tried RavCharge but hadn't completely set up entune yet.

 

[fooljoe]

A lot of this is still essentially "guesswork". A range of 95 miles corresponds to a usable battery SoC of only ~32kWh, (not 35kWh as generally assumed after being fully charged in normal mode). However, calculating what 80% of a fully charged battery actually would be, results in 33.4kWh of usable capacity.

Yeah, there's definitely a lot of guesswork here, and I plan on eventually letting the user set what the full charge capacity is (or at least the % degradation applied to the nominal 35kwh.)

Which of these charge capacities is more accurate, Toyota's published 35kWh or 33.4kWh (80% of 41.8kWh)? To be more conservative, the obvious answer is the latter, which in my case just so happens to correlate closer to my actual GoM range.

I'm assuming the 35kwh is the right answer here, but it's certainly up for debate. Coming from the LEAF world, it's reasonable to assume the normal charge is 80%, but I don't think Toyota has given any indication that it's supposed to be an 80% charge.

Then there is battery capacity degradation over time as an influencing factor to be considered. I have only owned my car for the last 10 months. It has only been driven about 5000 miles, so I do not consider any cummulative battery degradation as yet very significant. However, RavCharge should probably still allow for degradation with some sort of time based multiplier to determine what corresponds to a 100% (normal) charge status. For example, -3% per year, or -0.1% per 500 miles, could be applied.

Yup, that's on my to-do list. Just gotta make sure all the core functionality is stable before I go breaking things...

In any case, I feel RavCharge is a complete SUCCESS!!!

Thanks for the compliment. I feel like I've still got a ways to go to make sure I've ironed out all the bugs and added in a few more useful features before I'll declare victory, but I'm glad that it's mostly working out great so far!

 

[boredcleaner]

I just logged on and tried out RavCharge. Nice. Entunes is really a big disappointment. If only there was a way to extract more information. Particularly, it would be nice to be able to create a driving log: miles driven, kwh used, time elapsed, etc. Similarly, it would be nice to be able to create a charge log.

Would you be interested in help for possible mobile app development?

 

[TonyWilliams]

Which of these charge capacities is more accurate, Toyota's published 35kWh or 33.4kWh (80% of 41.8kWh)? To be more conservative, the obvious answer is the latter, which in my case just so happens to correlate closer to my actual GoM range.

I'm assuming the 35kwh is the right answer here, but it's certainly up for debate. Coming from the LEAF world, it's reasonable to assume the normal charge is 80%, but I don't think Toyota has given any indication that it's supposed to be an 80% charge.

The EPA decided for you! They average 80% and 100% charges on only two cars; the Nissan LEAF and the Toyota Rav4 EV.

Plus, I have actually driven the car down to turtle and arrived with 80% burned of 41.8kWh. Unfortunately, I can't do that anymore with 6.5% loss.

 

[Dsinned]

Here's a thought, can RavCharge incorporate data from Entune's ECO Dashboard regarding EV Efficiency?

This history data in the ECO Dashboard is essential a tally of "running averages" of each EV's driving efficiency in miles/kWh from month to month and day to day. The latest update in the ECO Dashboard seems to correspond to the "learned" efficiency up to a particular point in time on a daily basis.

Perhaps this should be the basis for a "default" in Ravcharge, either as a preset by the program or selectable by the end user if it can be displayed as a reference on the status screen. I realize even this "number" is hard to understand how it was actually determined; some sort of "secret" that only Toyota (or Tesla) knows. However, it does seem to take into account "driving styles" and other real world factors that influence driving range in terms of miles per kWh.

 

[fooljoe]

Here's a thought, can RavCharge incorporate data from Entune's ECO Dashboard regarding EV Efficiency?

Early on I aimed to pull data from the ECO Dashboard area as well as the main status/charge/climate areas, but for whatever reason I couldn't get it to work. It seems that stuff uses a different system. I intend to revisit that at some point, but I've got a lot of other stuff on my list already!

 

[Tadol]

Just logged in on my iPad - looks great, worked flawlessly getting in, can't wait to understand how it all works.

I'll read about it tomorrow. Thanks very much for the effort!!

 

[fooljoe]

I just logged on and tried out RavCharge. Nice. Entunes is really a big disappointment. If only there was a way to extract more information. Particularly, it would be nice to be able to create a driving log: miles driven, kwh used, time elapsed, etc. Similarly, it would be nice to be able to create a charge log.

I have some ideas about the log-type features, but I think it would be pretty limited in accuracy with the information available through entune.

Would you be interested in help for possible mobile app development?

It is a mobile app! I just prefer the website approach over developing and maintaining separate iOS, Android, yada-yada-yada versions of the same thing, but maybe once things have settled down a bit I'll consider that.

By the way, to really get the "app-like" experience, at least on an iPhone or iPad, navigate Safari to the site, login with 'remember me' checked, then touch the little forward arrow / rectangle thing at the bottom and choose 'Add to Home Screen". Then you'll get a nice little app icon on your springboard that you can touch to bring you straight to RavCharge, plus it will open the page in full-screen mode, without any pesky address bar at the top or navigation bar at the bottom.

My graphic design team (aka wife) just made us some fancy new graphics, including the springboard app icon, favicon, and logo in the header bar. Also in the header bar there's a new menu button in place of the logout button, and the menu now has the logout button and will eventually have a 'settings' area that can be used to set options like battery capacity degradation, time zone, and so on.

 

[Dsinned]

TERRIFIC!!! Love the new icon!

Please make RavCharge check to see if the car is "Unplugged" first before sending any preset charge commands. This should prevent nuisance Entunes email notifications from being sent that say "Failure to charge . . . due to not plugged in", when leaving the car unplugged was intentional. When unplugged, and scheduled charges are still set inside the car, Entunes does not normally send such messages. It just skips the scheduled charges and assumes the owner left the car unplugged intentionally. I think RavCharge should behave the same way for its scheduled charges when the Timer is left ON. Of course, "Charge Now" manual charging is a different story!

My RavCharge Timer schedule consists of a setting to initiate charging very early Saturday and Sunday mornings starting at 12:01am, which is when I would normally want automated charges to occur while the car is parked overnight. I also have the departure time set inside the car to end at 08:50am on the same days of the week (or 6:50am on week days). This is so that active bulk charge completion (known to Entunes), plus the additional idle period for so-called cell balancing and top off charging (unknown to Entunes) will complete by 9:00am (or 7:00am on weekdays).

I have found that this sequence of scheduled charging events in the RAV4 EV's overall charging process, generally provides sufficient time for all sequential events (including cell balancing) to occur within these time parameters. Thus, resulting in final completion of everything in that process precisely at 8:57am (or 6:57am on weekdays). I don't know why the last 27 minutes is the fixed amount of time required to complete post bulk charging for cell balancing purposes, but that is positively what actually happens.

 

[miimura]

I love the App-Like setup on iOS. The new efficiency slider to calculate range is a huge improvement over the 2/3/4 mi/kWh at launch. Keep up the great work.

 

[fooljoe]

Please make RavCharge check to see if the car is "Unplugged" first before sending any preset charge commands. This should prevent nuisance Entunes email notifications from being sent that say "Failure to charge . . . due to not plugged in", when leaving the car unplugged was intentional. When unplugged, and scheduled charges are still set inside the car, Entunes does not normally send such messages. It just skips the scheduled charges and assumes the owner left the car unplugged intentionally. I think RavCharge should behave the same way for its scheduled charges when the Timer is left ON. Of course, "Charge Now" manual charging is a different story!

Great suggestion, and this has occurred to me as well, but I have some trepidation about doing this. I've noticed (with my car at least) that the car does not always report its plugged in status - possibly depending on how long of a time span there is between when you park and when you plug in - so there's a risk of skipped charges if we rely on that. Of course I could have RavCharge send a status update command some time before a scheduled charge, but that would complicate things, and status updates aren't 100%.

I don't know about you, but I'd rather suffer a bit of extra entune spam rather than risk a charge not happening because the car didn't report that it was plugged in. But it's something that I still plan to explore.