-
-
Notifications
You must be signed in to change notification settings - Fork 1.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Z-Stack 3 on CC2652R/CC1352P adventures #1429
Comments
@Koenkk can this board be bought from inside of Europe? I need to pay extra taxes if bought outside. |
@JLFN mine shipped from the Netherlands. |
mine too |
Can you please give me link of the page where you bought it from that shipped from the NL. |
@JLFN just from TI. |
The Xiaomi end device undefined model ID issue had nothing to do with the CC2652R, it was actually a zigbee2mqtt bug, fixed in latest dev branch. |
I've got mine today morning (from NL as well). Today, there will be a big migration project :) |
Before I also "jump on the adventure train", I have 2 questions.
Thx |
Hello, does that part have an antenna? What is the range of that? and pays off the switch from cc2531 to cc2652r ???? |
@salopette it has a pcb antenna and an unconnected connector, see https://e2e.ti.com/support/wireless-connectivity/bluetooth/f/538/t/723337 |
I only have osram plugs and aqara devices, aqara temp/humidity sensors are mostly ok to rebind, but it's not as straightforward for the other ones. |
I'm done with re-pairing... Looks good... Re-pairing done with QBKG11LM, QBKG04LM, RTCGQ11LM, GL-C-008, E1746. Link qualities are better than CC2531. |
i've been able to pair MCCGQ01LM, MFKZQ01LM and DJT11LM, but still there's something very wrong with aqara devices. |
@dh-harald do you see big improvements in link quality? |
@Koenkk with CC2531, I had 50-70 link quality, not I've 100-120. |
This morning it looked like my cc2652 had crashed. z2m wasn't able to use its serial port anymore. Had to restart rpi. Most of the pairings have been lost. Link quality looks also better than what I had with cheap cc2530+cc2591. |
@tunip range looks better, antenna should be possible as there's a connector. |
I've something similar ATM... All devices are offline, can't ping by the controller. I had it yesterday, I needed to disconnect it from usb (poweroff)... Now I'm not at home, so I need to figure out, how I can disconect it from remote :( |
@dh-harald if it was connected to a PC, I would try to use uniflash to reset the board. |
It's an arm64 box, so I'm afraid, I can't run uniflash on it :( |
@tunip I already tried them, but it didn't work. |
Mine also crashed after 2 days (Request timeout errors) |
the mqtt messages with empty action are not related to cc2652, are they ?
|
another funny issue with a door window sensor, and I haven't touched the pair button, I was just playing with a magnet. In the end the device is removed from the network
|
looks like it happens very often :-(
|
@lolorc the empty actions are probably some converter bug. I haven't seen the dev leaving yet (and I have many Xiaomi devices) Mine crashed again this morning, but I was able to get it working again by only replugging the CC2652R (no restart of zigbee2mqtt). Trying to catch a |
mine crashed as well this morning, it can't stay more than 12h without crashing. Is there a way to debug or get a crash stack from the second serial port for example ? replugging cc2652r is not a solution when not at home, but hey it's the early alpha tests with cc2652r. I've been running with 'DEBUG=* | tee /tmp/something' for a while now, but it's lost when rebooting, I'm now logging to a nfs server. |
the leaveind issues might be related to fairecasoimeme/ZiGate#38 (comment) it wasn't happening with cc2530+firmware from end of 2018. if you look at my logs, the leaveInd message has rejoin:1, is it normal to consider it gone ? |
I saw a post that it was possible to connect to the cc1352p-2 via an ESP8266 or an raspberry using serial2ethernet. Maybe is that the best solution for you? It is possible to let all containers on your NUC and let them communicate with the zigbee over ethernet. another solution maybe to install the zigbee2mqtt on another device which is connected over MQTT. |
Mhh I could be do that. What is the main problem in fact ? Is it the CC1352P-2 board connected via USB ? Is it Z2M which runs on x86 or x64 architecture (with Docker) ? Is it my Mosquitto MQTT broker (without Docker in my case) running on x86 / x64 architecture (I don't think so that is the problem) I have some difficulties to understand why he could be more stable on Raspberry pi than NUC. ARM architecture ? |
@kevincaradant my production setup is Ubuntu 20.04 in Docker on a NUC6CAYH with the ZZH adapter, never had these issues (also ran a long time on with a CC2652 launchpad, no problems) |
Ok thank you. it's Interesting maybe I just should upgrade my ubuntu version. I'm going to create a backup of it and try this in the case where I got again adapter issue 6000 ms. if others runs on NUC and with a CC1352P, I'm interesting about your feedback on it |
Just to add my 2 cents - I've been running z2m with LAUNCHXL-CC26X2R1 first and now with a zzh! stick (also with CC2652) on a PCEngines apu2c4 (with an AMD Jaguar x86_64 CPU) for a few months and fortunately never faced the "adapter error" issue. So it's not a general issue on x86 PCs... |
Thanks :) Just a last question, on CC26XX and CC13XX
Thanks I will tell you in some days / weeks if eveything works well :) |
Not sure I'd call it a manual, but here's what I've done: Parts (I hope it's ok to post ali links):
Load the latest and greatest release of zigbee2mqtt firmware on the CC1352P2. Then disconnect the 2 jumpers in the picture - I just put them on one of the pins to not lose them: I used 4 cables to establish the serial connection to the S2E adapter. The S2E adapter is powered from the CC1352P2 3.3V pin, the other pins are ground, DIO12 (RX) and DIO13 (TX) : On the S2E side, make sure to attach the RX pin of the CC1352P2 (DIO12) to the TX pin of the S2E adapter, and the TX pin to the RX. The CC1352P2 gets powered through the micro-USB from the POE splitter, whereas the S2E adapter should be connected to the splitter with the RJ45 connector. To configure the S2E adapter, in my case I had to go to 192.168.1.7 (not 192.168.0.7 as stated on aliexpress...), factory setting is user admin, pwd admin. I configured it to use DHCP, then configured a static IP on the DHCP server. The settings of the device I set to: The actual ip and ports are not important of course, I highlighted the ip and port which have to be used in the configuration.yaml of zigbee2mqtt:
Don't forget to change the pan_id when migrating from a CC2531 as I did. As said before, this is very stable, never went down, only once for updating the firmware :) . That's about all I can think of, hope it helps someone. |
From what I could see on the pinout, the pins are labelled DIO2 (RX) and DIO3 (TX) on the cc2652r, and they are at they same place as the ones on the cc1352. But of course using the jumper pins is the same, + you get the labels 👍 so that's even better in order to be sure when connecting it. |
I've been running a week now with XDS firmware Another possible solution for other 1352P2 users would be to use a simple FTDI adapter and bypass the XDS debugger, using @markuspeter 's directions above. |
Has anybody tested the new firmware with enhanced transmit power? Would you share the experience and what impact can be observed on LQI? I would expect that receiver sensitivity is more important here, but I'm really curious :) |
I'm running with the latest firmware, the LQI is better around of 20/30. In this moment I have a lot of timeout wierd issues but I don't think it's related to the latest firmware (with the increased range), the previous one had also this issue from my side. LQI is very sensible, sometimes I have 90 and sometimes I have 25 :). It's a little bit tricky to understand this value because a lot of factors can have a bad behavior on it during few seconds ( interferences, walls, wind, computers, phone, wifi ect ect). In an other way, you can have 90 as LQI with a lot issues (interferences) not stable and a 10 LQI stable (like my bulbs, 6 meters away of the coordinator) but 0 issue with them. But personnally, I think it's not bad to have the most high value possible for the range increasing the signal :). |
@kevincaradant Thanks Kevin for the feedback. I fully agree with you regarding LQI. But anyway it is still one of the few quantitative and measurable factors when talking about zigbee network reliability and stability.. I'm glad to hear that transmit power had at least some impact :) |
Finally had two more crashes on the CC1352P2 running XDS |
Which Zigbee 3.0 firmware would work on the CC2652P (with amplifier)? CC1352P has amp (amp enable registers and sub-1 GHz, not used) but the CC2652R has no amp and no sub-1 GHz registers. One of the modules I ordered has a built-in amp: https://www.aliexpress.com/item/1005001265532720.html My plan is to connect then thru a CP2104 or FT232H serial to USB and use it as a coordinator. Thank you |
What is the NUC that you're using ? Under what OS ? Do you use a USB 2.0 or 3.0 ? I'm using the latest firmware available for CC1352-P2 with Z2M 1.16.2. For the moment everything works as expected. 2 days ago, I had 3 Xiaomi which left the network without reason ... I needed to re-pair them but maybe it's a problem of battery even if I have more than 3080mV (100%). So I still keep on eye on this issue. |
Regarding Xiaomi sensors leaving the network. I noticed that two of my Aqara sensors are pretty sensitive to coordinator availability. E.g. when they cannot reach coordinator for some time - it is out of reach, or offline, those devices go to deep sleep and stop sending reports. I need to wake them by pressing pairing button. Strange thing is that this happen more likely with only two of my sensors and I have many more which remain awake.. Either it depends on the firmware version of the sensor, or maybe it depends on the frequency of messages being sent. Maybe there is a counter which count unsuccessfully delivered messages and if the threshold is reached, device falls asleep. That's my observation. |
My ZZH stopped working today again. Running the latest firmware with the increased antenna gain. Just unplugging and plugging in the ZZH and everything was up and running again. |
Yes this is interesting your point of view. I don't know if it was useful or not but I enabled
Logs ? somehing which can explain it ? How many time it works before to fall down ? |
See: |
@ProASV thank you for the tip. |
Sorry if this is wrong place for my question but can i use CC1352P all versions for zigbee coordinator? I mean i have the CC1352P - 1 and i have used it for almost a year now but in the documentation is written that it has to be the CC1352P - 2. Am i missing here something or can i use all the boards 1,2 and 4? |
Since this is not an issue anymore I've created a discussion for it: #5266 . Let's continue there. |
NOTE 2020-12-11 locked, please continue here: #5266
Texas Instruments has recently released a new high performance Zigbee chip: the CC2652R. This chip is much faster and contains much more memory than the CC2530/CC2531.
This issue is created in order to gather feedback of users using the CC2652R AND Z-Stack 3.0 (= Zigbee 3.0)
Notes
data/coordinator_backup.json
. This backup is automatically restored when starting with a new flashed CC2652R.The text was updated successfully, but these errors were encountered: