Wanted to report back - So I after that initial big improvement things are back to showing 9.6 on some of my MS6's in the basement - seems like the network is all over the place in terms of speed according to the zniffer. Could this be due to the location of the UZB device itself?
Also realized that "disabling" a device in HE does not temporarily remove it from the mesh so my initial thought about the original Living Room Nano being the issue may have been incorrect..
This stuff is not as straightforward as I hoped - don't know why I thought it would be.
Some diagnostic and low level traffic is always sent at 9.6kb - even on z-wave plus. It is that way to ensure backwards compatibility all the way back to 100 series z-wave devices that could only talk at 9.6kb speeds..
If it is core command traffic (basic set, basic get, multilevel, etc) then yes, that isn't great.
Very cool will take a look at that thank you. I kinda suspected as much.
My issues go deeper it seems but am okay with digging around a little. One thing I may have inadvertently done in trying to simplify things is reduce the strength of my z-wave mesh.
Now that I have some additional free time thanks to the pandemic ( ) I have a bit more time to focus on this stuff.
Yep that's what I should have realized.. given my (limited) understanding of how z-wave works. Disabling the device in HE does stop HE processes for that device though so maybe there is something interesting going on with that as well.
I figured it out, it always amazes me that this stuff works when you see how many hoops there are to jump through to get to the end results. Doing the firmware upgrade now, wow is it slow for me, but the numbers are moving.
Well this is the hard way of doing it really. The easy way is to buy the Dev Kit which includes everything including a sniffer or to buy a Zwave ToolBox which is also a sniffer or buy a CIT if you can.
Just got my UZB in the mail today. This thing is going to cost me some money. Seeing how some of my non-plus GE switches are causing some grief. Far too many packets are at 9.6k.
I went down that path last year all to get my mailbox contact sensor to keep a good connection. Been solid since installing it. I don’t have it hooked up to a computer to view the map, it’s by by hub to help with mesh. Plus I have replaced most non plus GEs with Zigbee when Lowes was dumping them. I still have a few non plus left including a water valve so curious what my Zwave network looks like hence the purchase.
I did get this all going from following thread post 1. I've got all non plus z-wave devices in the house, and all already work flawless for me. (Got the z-wave poller app running for them - that is probably my sniffer traffic I saw) I ran the sniffer and didn't spend a lot of time on it, i only saw a few red crc packets, and everything was reporting 40k speed. I don't know if I'll find much with this, but figured for about $40, it is so worth having this in the toolbag!
Finally got the software installed and now zniffing! I have been changing out my older zwave devices with zigbee devices when Lowe's was dumping those. I have a total of 77 Zwave devices and only 6 are non-plus:
GE Fan Switch (will be replacing)
GE Switch (will be replacing)
Water Valve (keeping)
Contact Sensor (can replace)
Siren (keeping)
Schlage Lock (keeping)
In my zniffer many most traffic is 100k, but there is still quite a bit at 9.6 and very few at 40k. Is there a way for me to identify why some traffic is slower? I ask because I would expect to see the devices mentioned above in the list with traffic routing through them and I don't. How do I know what impact I will have placing some of these without seeing traffic flow through them?
What is odd is a zwave plus GE Motion switch very near my hub is only getting traffic at 9.6. GE Motion is 161, 39 is Aeon plus Multisensor, 23 is a GE plus in wall receptacle.
Yes I noticed that with mine as well. All my MS6's are usb powered. I hope that is not the issue...hmmm.
Note: I just upgraded the MS6's them to the latest firmware but am still getting that behavior. I wonder if there is a way to tell it to pretend to be on battery power..
