On the 12v filtering, all I can think is to average it, perhaps over a minute, or for a set number of readings. Or perhaps require 2 readings higher than the current reading before increasing it (or generally 2 readings before it is changed). This could be done in the function to handle reading the value.

For example, say we kept a total, a count and the last value. Then, input_voltage() would read the value, add it to the total and increment the count. If count > 5 (say), set last value = total / count, then reset total,count=0. It would return last value as the result.

Do you have any historical list of the values reported from the function? Does it go 11.5, 11.6, 11.5, 11.5, 13.2, 11.5, 11.6, etc? Of stay falsely high for multiple readings?

Regards, Mark.

On 28 Apr, 2013, at 11:06 PM, Michael Balzer wrote:

I just checked in my fix for the USSD reply handler. My tests with streaming enabled did not produce any more crashes, so I think that was the problem.

My tests show there still is an issue with streaming enabled: if I let a client open (perl / App), very few GPS points now get through, both to the App and to the GPS log. Could be due to the modem not getting the per second GPS requests or due to some msg queue issue. I'll debug that ASAP.

I also reactivated the 12V input filtering as the false peaks still occur and I've got no more ideas for the A/D conversion. Any hint on this is appreciated.

Regards,
Michael


Am 14.04.2013 23:02, schrieb Michael Balzer:
I also had the perl client running in a loop to log all data.

I found the bug occured both times just after a perl client restart. That could mean the "peer connected" reaction (net_msg.c line 660) could somehow be involved, or -- as my perl client copy also sends a command 41 to query the SIM account balance -- the USSD handling could be the source of the bug.

I'm pretty sure it's the USSD reply handling. It's not completely well coded yet (sorry!), and it could behave bad in a situation with much regular data traffic going on... and that again matches.

I'll fix this ASAP, but I can't fix it now (too late).

As triggering this bug seems to be unlikely, I would not consider this a problem for the release.

Regards,
Michael

PS: I checked in the A/D converter change though, that seems to solve the 12V readings!


Am 14.04.2013 18:04, schrieb Michael Balzer:

I've had GPS streaming enabled during the last days, and neither GPS nor GSM connectivity are good at the moment. May be related, as streaming means more data communication. But still that should not lead to RAM corruption.


Am 14.04.2013 17:54, schrieb Michael Balzer:
Mark,

I've got the last but one version in my car, just lacking the new VEHICLE commands.

I've got two open issues, the minor one being to raise the 12V auto acquisition time even more, as the faults have become better but it's not yet perfect. I'm trying 20 TAD now, as we don't need to be fast on that A/D conversion.

But a major issue might have turned up: I've just had the second RAM corruption within three days. Effect:

MP-0 c32,0,106,119,*-OVM-DebugCrash,2013-04-14 14:06:40,0,2.2.7/RT2.6.4/V2,0,0000,20
MP-0 c32,0,107,119,*-OVM-DebugCrash,2013-04-14 14:25:13,0,2.2.7/RT2.6.4/V2,80,0020,61

As there is no checkpoint 61, this means at least the debug variables have been overwritten, was the same with the first occurence (but other values).

As the module behaves weird afterwards I suppose more variables have been corrupted.

I've looked through the last changes several times, I'm pretty sure none could have introduced this kind of bug... if I'm the only one experiencing this, maybe I've got some hardware issue as well?

I'll continue checking the source for possible buffer overruns...

Regards,
Michael


Am 14.04.2013 16:29, schrieb Mark Webb-Johnson:

I've just built from latest sources, v2.3.1. This is a release candidate, and I have 24 hours to test before i need to send it to China.

It is in my car now, and seems ok. I'll do more testing tomorrow.

I'd be grateful if others could grab it and put it in their cars to ensure there are no major problems.

Regards, Mark.

On 13 Apr, 2013, at 10:04 PM, Mark Webb-Johnson <mark@webb-johnson.net> wrote:

All seems ok.

I've committed and pushed my changes (just the stp_ltf_h() function, and change to report cac as 999.99 to server).

I've just flashed and it will go in my car tomorrow.

Regards, Mark

On 13 Apr, 2013, at 2:51 AM, Michael Balzer <dexter@expeedo.de> wrote:

Mark,

I just checked in my latest changes.

In net.c I added the pending send check also to the new net_notify_errorcode send, I think that should be correct, but please have a look.

I saw the new stp_f() function, but I don't understand it's purpose yet. What's the difference to my stp_l2f() function?
Btw: l2f = "long to float" -- I think "_f" should be reserved for an stp call with a real float type parameter, if we once really need one.

Regards,
Michael


Am 12.04.2013 20:16, schrieb Michael Balzer:
Nikki,

yes, see attachment.

I could not make out any possible source of that strange error I had yesterday (completely garbled data).

I made a clean build with a freshly started MPLAB and got no errors up to now.

May have been something completely different, not related to my last changes.

Regards,
Michael


Am 12.04.2013 17:01, schrieb Nikki Gordon-Bloomfield:
Michael, 

Any new firmware for me to test this weekend? 

Nikki. 

On 11 Apr 2013, at 18:33, Michael Balzer <dexter@expeedo.de> wrote:

I think I was a bit too fast on this, Nikki, don't use that version, it seems to have some new bug causing frequent crashes... :-(

Regards,
Michael


Am 11.04.2013 19:22, schrieb Michael Balzer:
Mark, Nikki,

I'm currently trying to fix a last bug with the power statistics (the distance counter still went wrong), and I'm trying to get the 12V A/D conversion to work without these strange "peaks" by raising the automatic acquisition time. I will finish testing this ASAP or tell you in time before sunday.

Nikki, I attached the latest version in case you'd like to test as well.

Regards,
Michael


Am 11.04.2013 16:41, schrieb Mark Webb-Johnson:
China is bugging me for final firmware for the next batch of hardware. I need to get them this early next week.

I've committed what I need for the Tesla Roadster (and framework). Last outstanding small change to make is to send CAC as 999.99, rather than the current 99999, using Tom's new library function.

Michael B & J: can you check Twizzy and Volt/Ampera, to see if everything is ok with current firmware or there is anything you need changed?

I plan to cut release candidate for 2.3.1 on Sunday.

Regards, Mark.

_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev 



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26
<dexter.vcf>_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26
<dexter.vcf>
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26



_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev


-- 
Michael Balzer * Paradestr. 8 * D-42107 Wuppertal
Fon 0202 / 272 2201 * Handy 0176 / 206 989 26
<dexter.vcf>_______________________________________________
OvmsDev mailing list
OvmsDev@lists.teslaclub.hk
http://lists.teslaclub.hk/mailman/listinfo/ovmsdev