- OvmsDev - Open Vehicles

Ampera details
by Lee Dalton 29 Jan '15

29 Jan '15

Hi folks, New here having just installed OVMS into my Ampera. Couple of questions, if I plug into the OVMS diag port can I read the Canbus messages ? Secondly, how can I send commands to the car via OVMS ? Is there a command I can send an and to trigger a specific bit of canbus message, or does that have to be hard coded into the OVMS firmware ? Thanks Lee dalton Sent from my iPhone

2 1

Fasttech production problems?
by Michael Balzer 22 Jan '15

22 Jan '15

Hi, according to a Twizy user the module is currently not available. The Fasttech support told him they've got problems with the module manufacturer (I thought that's Fasttech itself), and they're searching for a new manufacturer. There have been more reports on hardware failures lately, maybe that's what they mean? Any background info on this? Regards, Michael -- Michael Balzer * Paradestr. 8 * D-42107 Wuppertal Fon 0202 / 272 2201 * Handy 0176 / 206 989 26

6 12

Fwd: Re: OVMS App missing most charge points, detail info wrong
by Michael Balzer 15 Jan '15

15 Jan '15

FYI -------- Weitergeleitete Nachricht -------- Betreff: Re: OVMS App missing most charge points, detail info wrong Datum: Mon, 12 Jan 2015 18:30:38 +0800 Von: Christopher Cook (Open Charge Map) <christopher.cook(a)openchargemap.net> An: Michael Balzer <dexter(a)expeedo.de> Hi Michael, I'm also currently trying to get my android dev setup working again (for the OCM app). Not easy especially as they have now retired eclipse and moved to the new gradle build system in Android Studio, so there's lots of trial and error required to get things building again. The API prefix is also wrong in both cases, it should be: http://api.openchargemap.io/v2/referencedata/ (for pulling the reference data for dropdown lists etc) And the POI query should really be of the form: http://api.openchargemap.io/v2/poi/?output=json&verbose=false&<filter params> Max results filter can be much higher (2500 etc) but will potentially return more data (and take longer), my own preference would be to default the search to a smaller distance. API docs are here: http://openchargemap.org/site/develop/api It's also wrong on iOS, I did submit a pull request a while back but I don't think anything happened to it. Chris On 12/01/2015 18:13, Michael Balzer wrote: > Hi Christopher, > > thank you for your help! > > From home this morning ( 51.263201, 7.150157 ) I had 494/495 stations > on the map. > From office now ( 51.321685, 7.343394 ) I have 507 stations. The same > stations are missing. > > The search radius according to the App should already be switchable to > the vehicle range, but that switch has no effect on the number of > stations. > > So this definitely looks like the 500 results limit. > > Search: > https://github.com/openvehicles/Open-Vehicle-Android/search?utf8=%E2%9C%93&… > > > Seems to me it's this API call: > https://github.com/openvehicles/Open-Vehicle-Android/blob/81971beb45defd87e… > > > That also explains why the range switch has no effect, the distance is > fixed in the call. > > There's another call in ui/FragMap, but that's fixed (some > init/default maybe). > > I'll try to change the maxresults parameter, but I first need to get a > working Android development environment -- haven't developed Apps up > to now. > > Maybe the maxresults parameter should be user configurable. > > Regards, > Michael > > > Am 12.01.2015 um 01:57 schrieb Christopher Cook (Open Charge Map): >> Hi Michael, >> >> I'll download the code for the app from github and have a look, I >> haven't developed any of the OCM integration myself and I think >> (certainly in the case of iOS, not sure about Android) that 3rd party >> developers were paid to deliver the integration. Presumably then the >> hope is that the community will refine their efforts via github - I'm >> assuming the path is very much clear for enthusiastic developers to >> submit patches for these apps directly. I can help more on >> integration questions like calling the OCM APIs etc (such as your >> filtering question). I've held off contributing directly to the apps >> because I'm limited for time with OCM and other projects. >> >> If I find a simple fix for the API request to OCM I'll submit a pull >> request to github. >> >> For the example you've given, where is the center of your search >> (lat/long if possible)? By default the app is search from your >> current lat/long with a radius of 285 KM, max 500 results. Is it >> possible there are more than 500 charging locations in your search area? >> >> Best regards, >> Christopher Cook >> Open Charge Map >> >> On 12/01/2015 00:36, Michael Balzer wrote: >>> Hi Christopher, >>> >>> I first posted this on the OVMS developer list (I'm the Twizy >>> developer). J-C Saint-Po told me to forward this to you: >>> >>> I'm missing about 80% of all OCM charge stations on the OVMS map >>> display (Android). >>> >>> I cannot find any criteria in the station data for this effect. >>> Example: >>> visible: http://www.openchargemap.org/site/poi/details/32598 >>> missing: http://www.openchargemap.org/site/poi/details/32607 >>> >>> Both stations are from the same provider with identical capabilities. >>> >>> The connection filter does not affect this, neither does the range >>> filter or clustering. >>> >>> Also the detail info on visible points is wrong. >>> See attached screen shots for the map label and detail info on >>> #32598 as shown in the App. >>> >>> If I can help by more examples, tell me. >>> >>> This is no new issue, has been this way from the beginning. >>> I just decided to add some missing stations from my region and then >>> discovered they are not really missing in the OCM database. >>> >>> Regards, >>> Michael >>> >> >> > > -- Christopher Cook Open Charge Map

3 5

Free OVMS Hardware For UK EV Owners
by Kevin Sharpe 15 Jan '15

15 Jan '15

The following OVMS hardware is available for free in the UK (subject to ZCW donation); 3 x Ampera/Volt Kits; https://speakev.com/threads/open-vehicles-ampera-volt-hardware.6048/ 1 x Tesla Roadster Kit; https://speakev.com/threads/open-vehicles-tesla-roadster-hardware.6050/ Kevin Sharpe | Founder & Chair of Trustees Tel: +44 122 566 7544 ext: 800 | Skype: zerocarbonworld kevin.sharpe(a)zerocarbonworld.org <mailto:kevin.sharpe@zerocarbonworld.org> | www.zerocarbonworld.org <http://www.zerocarbonworld.org/> | twitter.com/zerocarbonworld <http://twitter.com/ZCWcharlie> Zero Carbon World is a UK Registered Charity #1141347

2 1

Re: [Ovmsdev] OVMS App missing most charge points, detail info wrong
by Michael Balzer 12 Jan '15

12 Jan '15

Hi Christopher, thank you for your help! From home this morning ( 51.263201, 7.150157 ) I had 494/495 stations on the map. From office now ( 51.321685, 7.343394 ) I have 507 stations. The same stations are missing. The search radius according to the App should already be switchable to the vehicle range, but that switch has no effect on the number of stations. So this definitely looks like the 500 results limit. Search: https://github.com/openvehicles/Open-Vehicle-Android/search?utf8=%E2%9C%93&… Seems to me it's this API call: https://github.com/openvehicles/Open-Vehicle-Android/blob/81971beb45defd87e… That also explains why the range switch has no effect, the distance is fixed in the call. There's another call in ui/FragMap, but that's fixed (some init/default maybe). I'll try to change the maxresults parameter, but I first need to get a working Android development environment -- haven't developed Apps up to now. Maybe the maxresults parameter should be user configurable. Regards, Michael Am 12.01.2015 um 01:57 schrieb Christopher Cook (Open Charge Map): > Hi Michael, > > I'll download the code for the app from github > and have a look, I haven't developed any of the > OCM integration myself and I think (certainly in > the case of iOS, not sure about Android) that > 3rd party developers were paid to deliver the > integration. Presumably then the hope is that > the community will refine their efforts via > github - I'm assuming the path is very much > clear for enthusiastic developers to submit > patches for these apps directly. I can help more > on integration questions like calling the OCM > APIs etc (such as your filtering question). I've > held off contributing directly to the apps > because I'm limited for time with OCM and other > projects. > > If I find a simple fix for the API request to > OCM I'll submit a pull request to github. > > For the example you've given, where is the > center of your search (lat/long if possible)? By > default the app is search from your current > lat/long with a radius of 285 KM, max 500 > results. Is it possible there are more than 500 > charging locations in your search area? > > Best regards, > Christopher Cook > Open Charge Map > > On 12/01/2015 00:36, Michael Balzer wrote: >> Hi Christopher, >> >> I first posted this on the OVMS developer list >> (I'm the Twizy developer). J-C Saint-Po told me >> to forward this to you: >> >> I'm missing about 80% of all OCM charge >> stations on the OVMS map display (Android). >> >> I cannot find any criteria in the station data >> for this effect. Example: >> visible: >> http://www.openchargemap.org/site/poi/details/32598 >> >> missing: >> http://www.openchargemap.org/site/poi/details/32607 >> >> >> Both stations are from the same provider with >> identical capabilities. >> >> The connection filter does not affect this, >> neither does the range filter or clustering. >> >> Also the detail info on visible points is wrong. >> See attached screen shots for the map label and >> detail info on #32598 as shown in the App. >> >> If I can help by more examples, tell me. >> >> This is no new issue, has been this way from >> the beginning. >> I just decided to add some missing stations >> from my region and then discovered they are not >> really missing in the OCM database. >> >> Regards, >> Michael >> > > -- Michael Balzer * Paradestr. 8 * D-42107 Wuppertal Fon 0202 / 272 2201 * Handy 0176 / 206 989 26

1 0

OCM most charge points missing, detail info wrong
by Michael Balzer 11 Jan '15

11 Jan '15

@anyone involved in the Openchargemap integration: I'm missing about 80% of all OCM charge stations on the OVMS map display (Android). I cannot find any criteria in the station data for this effect. Example: visible: http://www.openchargemap.org/site/poi/details/32598 missing: http://www.openchargemap.org/site/poi/details/32607 Both stations are from the same provider with identical capabilities. The connection filter does not affect this, neither does the range filter or clustering. Also the detail info on visible points is wrong. See attached screen shots for the map label and detail info on #32598 as shown in the App. If I can help by more examples, tell me. This is no new issue, has been this way from the beginning. I just decided to add some missing stations from my region and then discovered they are not really missing in the OCM database. Regards, Michael -- Michael Balzer * Paradestr. 8 * D-42107 Wuppertal Fon 0202 / 272 2201 * Handy 0176 / 206 989 26

2 1

Coding in vehicle_twizy - Memory section issue
by Julien NOZAIS 08 Jan '15

08 Jan '15

Hi everybody, best wishes for 2015. I came across a memory limit while working on my project, I hope you could help me with it. I am working on the vehicle_twizy.c file. So far I added about 300 lines of running code, and I started getting the compilation error "can not fit the section", which in general appears when you have too many modules enabled while compiling. Error - section '.idata_net.o_i' can not fit the section. Section '.idata_net.o_i' length=0x00000035 I identified that the memory issue is happening as I add more code, and disappearing when I comment the code, independently of the nature of the code. I first resorted to switching to less memory intensive variables such as INT8. I then tried to put as many of my variables as I could in a separate memory section. I used a pragma directive in the same fashion that is used for the battery monitoring system extension. #pragma udata overlay vehicle_overlay_data2 It helped for a while, but at some point as I was writing calls to the writesdo(..) function a few times, it started again. I guess writesdo's arguments are stored in a memory section that was already at almost full capacity. I thought there would have been some margin before running into this kind of memory problem. For now my workaround is to remove the battery monitoring from the compilation, as it leaves enough space for my code to run. Do you have any comments or ideas on this ? Best regards, Julien

2 1

Push notifications by email
by Michael Balzer 03 Jan '15

03 Jan '15

Happy new year everyone :-) I just added an email notification channel to the server code, following a request for a personal OVMS server. https://github.com/openvehicles/Open-Vehicle-Monitoring-System/commit/172a5… As the Android notifications need the hard coded C2DM channel and C2DM is no longer supported, this is primarily meant as an alternative for those using their own server and not wanting to build their own App. It might also be useful for other users & the main server -- I would like being able to receive my Twizy trip reports by mail. Regards, Michael -- Michael Balzer * Paradestr. 8 * D-42107 Wuppertal Fon 0202 / 272 2201 * Handy 0176 / 206 989 26

2 2

holidays = less driving = dead battery
by Lee Howard 02 Jan '15

02 Jan '15

Following the last long holiday break in the US I was prepared for this holiday... I have checked the car (2005 Chrysler Sebring) several hours before needing to drive it. That car doesn't get driven as often during holiday times, and so the OVMS module slowly eats at the battery. I've had to put a battery charger onto the 12V battery three times over the last two weeks to get enough charge into the battery to where the car could start. Yes, the weather is cold. That said, if I were to park the car at the airport for a few days while I was on a trip I would need to disconnect OVMS to ensure that there would be some battery to start the car when I returned. (And that would be a time when I would want to have OVMS working - making sure my car stayed where I left it - reminding me where I parked it when I returned.) Maybe I'm in a bad area for GSM reception and so the module eats more battery trying to get and maintain connectivity. That said, I truly feel like the OVMS module could do better. I know that the v3 module is being worked-on, and I suspect that battery-use issues are already being considered, but here's my 1-item wish-list for the module: 1) It needs to go to "sleep" when GPS position does not change much or when there is not some other indication that the car is being used, charged, or such where the data has changed. It needs to do this not only to conserve battery, but also to conserve GSM/GPRS usage fees with the SIM provider. (I've mentioned the GSM/GPRS usage fees matter in a previous thread.) Thanks, Lee. -- *Lee Howard* *Mainpine, Inc. Chief Technology Officer* Tel: +1 866 363 6680 | Fax: +1 360 462 8160 lee.howard(a)mainpine.com | www.mainpine.com

1 0

Tesla Model S CAN bus hacking (for fun, not necessarily profit)
by Mark Webb-Johnson 01 Jan '15

01 Jan '15

I'm setting up a mailing group for Tesla Model S CAN bus hacking. If you are interested, please let me have your eMail address and I'll add you in. Just starting work now on having a look at what is going on inside the Model S, decoding messages, and looking at possibilities for OVMS or other projects. Current status: https://github.com/openvehicles/CAN-RE-Tool/blob/master/rules/teslamodels Regards, Mark.

10 9