Ceases to read after a while, hangs

[bkenn011]

Anyone else experienced this? I was able to get it to run fine on a Particle Photon using both hardware serial and software serial, the latter as I needed the one free hw serial to do something else a little more intensive.

Seems like after somewhere between a minute and 5 minutes it'll just cease to read and my output to the arduino IDE serial monitor quits entirely.

[cterwilliger]

Yes, I have seen that. I have two devices, a Victron BMV-712 and a Victron SmartSolar MPPT-150/85. The BMV-712 has worked flawlessly with an ESP8266. The MPPT saw the same issues you saw with an ESP8266 and the exact same code. I changed it to an ESP32 and the issue went away. I use only HW serial. SW serial has problems keeping up.

…

Sent from my iPhone

On Jan 29, 2021, at 13:32, Brandon Kennedy ***@***.***> wrote:  Anyone else experienced this? I was able to get it to run fine on a Particle Photon using both hardware serial and software serial, the latter as I needed the one free hw serial to do something else a little more intensive. Seems like after somewhere between a minute and 5 minutes it'll just cease to read and my output to the arduino IDE serial monitor quits entirely. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[bkenn011]

Interesting, but glad to know I'm not the only one. This library manages to provide reliable data (kudos) until it goes kaput. Not the case with the one I cobbled together a few years ago, where rogue data appears over hardware and software serial alike. I solved the issue on that project by incorporating a nano and a particle photon, where the nano can process and validate the MPPT data, repackage to something more easily parsed, and then sent to the photon at a slower rate (and read via software serial). Clunky, but hasn't given me any problems. I'm guessing you used a logic shifter for the MPPT->ESP32 to handle the 5v?

[cterwilliger]

I use the library to grab the data, JSONize it, and send it to a MQTT broker. So, the ESP8266 or ESP32 don’t spend much time handling the data. Not sure why 8266 can’t handle the stream from MPPT, but ESP32 works. Yes, using a simple resistor voltage divider to convert 5V to 3V. From: Brandon Kennedy <notifications@github.com> Sent: 2 February, 2021 11:37 To: cterwilliger/VeDirectFrameHandler <VeDirectFrameHandler@noreply.github.com> Cc: Chris Terwilliger <chris@terwilliger.org>; Comment <comment@noreply.github.com> Subject: Re: [cterwilliger/VeDirectFrameHandler] Ceases to read after a while, hangs (#6) Interesting, but glad to know I'm not the only one. This library manages to provide reliable data (kudos) until it goes kaput. Not the case with the one I cobbled together a few years ago, where rogue data appears over hardware and software serial alike. I solved the issue on that project by incorporating a nano and a particle photon, where the nano can process and validate the MPPT data, repackage to something more easily parsed, and then sent to the photon at a slower rate (and read via software serial). Clunky, but hasn't given me any problems. I'm guessing you used a logic shifter for the MPPT->ESP32 to handle the 5v? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#6 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/AE4PBECS3PKVFX4OSWDJAQLS5BA5DANCNFSM4WZPJNCA> . <https://github.com/notifications/beacon/AE4PBEDSFY2BJ7JDMF5I2EDS5BA5DA5CNFSM4WZPJNCKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOFYA6DUA.gif>

[cterwilliger]

Please try with with new VeDirectFrameHandler.h. frameLen changed from 18 to 22 per new protocol doc.

[I-Connect]

not sure if your issue is fixed, I had same sort of issue (also using software serial) and was able to resolve it

#11

[DannyAt]

not sure if your issue is fixed, I had same sort of issue (also using software serial) and was able to resolve it

#11

How did you resolve it, I have same issue?

[giacinti]

Not sure if it may be related but hex protocol can somehow interfere as unhandled packets are regularly sent which causes the main loop to break with a bad crc... I have proposed a PR for implementing hex protocol #7 and using this version flawlessly since then

[cterwilliger]

I've been on the road all summer with spotty internet. Let me look at PR #7 and I'll see if we can get that merged as well as some other possible fixes. Philippe, how did you test the PR?

[giacinti]

I'm currently using the version with hex protocol on an ESP32 device connected to a Victron MPPT 75/15 with a very old firmware (139). I have implemented an ESPHOME component leveraging this library and use MQTT for gathering data and sending hex command (see here: https://github.com/giacinti/vedirect)