Isn't that a cute little gadget. A small in-line cable with a 2000mAh battery.
If I were going the battery route, for the price I might consider a more general purpose powerbank with a larger battery, such as the 10000mAh unit mentioned by @snell above. Just need to make sure it is an always-on design.
well most USB batteries will not charge at the same time as output, so that is the general problem, also my Swithes are on the Generator and battery so getting power to my HE when power is out is what I am looking for. this just cleans the power..
it's just an idea/theory. if I could get a battery that does charge at the same time, I would go that route, I am sure it's out there, just not sure offhand.
I agree with @dennypage on the speculative RFI issue 100%. It seems extremely unlikely that a low power POE splitter would have ANY significant EMI that would affect Z-Wave.
Interestingly, I have 4 UCTronics POE adapters powering my 4 hubs, and 3 are in close proximity to one another ( 6 - 12 " separation) and have noted no significant Z-Wave issues. The 3 colocated are a C8, C7 & C5, and the separated unit is a C8.
ALL are being powered by Ubiquiti as the PoE Source.
USW-24-POE (250 Watts).
Scott
PS -- Those PROCETS look really nice though, and that screw mount flange is pretty nice -- perhaps I'll order some -- more stable voltage is good, and screw mount is a ++.
The Mission Cables unit is what is referred to as an "always-on" battery design. I.E. it powers the output even when changing, and does not have a low current shutoff.
I believe the battery referenced by @snell is an always-on design. Another good always-on solution is the Voltaic V25/V50/V75 IoT batteries. They are specifically designed for this.
There are several threads that have discussed USB battery banks made by Talentcell, available on Amazon, since they meet this requirement.
@dennypage since you have the equipment and setup already would you be able to test two different "Phone Charger" power blocks? There has been some info floating around that they may not supply steady voltage unless it is specifically designed to do so, which I think is not true if they are USB-A style adaptors.
So anyway.... I was thinking maybe test a 1A and a 2A+ phone adaptor, just to see how they compare to the supplied power block. I assume you have some laying around as most of us do 
. I am currently using an old iPhone 1A adaptor on my hub and it seems to be fine.
Just my take... the previous issues I had with Z-Wave were with the PoE supplying power directly to it and the splitter in close proximity (>12"). ALL of my hubs are now on the USB batteries packs I mentioned with PoE splitters supplying them and the splitters are still in close proximity (<12").
So it does not appear to be a "proximity to splitter" issue. Also note, I have my 5 hubs all in a general cluster ~36" across on top of my server cabinet in my basement. All the PoE is supplied by a Ubiquiti Pro PoE switch. 2 of the hubs have Z-Wave disabled though.
Sounds like hubitat needed an extra capacitor or two on the power supply side of the board to steady out the voltage.
I don't think it's related to the physical connector, but what the block is designed for. If it's just designed to charge a battery, it doesn't need to be stable voltage. I know the original Apple 5W iPhone blocks definitely had unstable voltage output. I may take a recording when time permits, and I have another reason to be on the ladder. 
I'd also like to see the voltage changes using different POE switches and how they contribute to the c8/z-wave problem. I'm using one of the cheap ones just fine but the moment the unifi comes into play everything goes tits up for z-wave. Put the cisco or mokerlink or tplink back into play and it's fine.
I think Amazon might get a little annoyed if I started ordering and returning large numbers of switches. I'm probably already on a watch list.
Rick should order them for you and drive them over... 
hey @dennypage so the Procent PT-PTC-AT worked? I'd love to get my hub back on PoE, it's been sitting over on my network rack since PoE didn't work, and that's at one edge of my mesh (rather than in the middle of my house), which is less than ideal (it works, but not great). And I'm willing to drop the money to try it again. =)
In hind sight, like I posted in my thread, I've had issues with Zwave on C7 and C5 even that I thought were just flakey devices, but it now tracks they were due to PoE. I thought it was noise... but this does seem to indicate it's voltage stability. Seems like throwing some capacitance in line would help for these situations, but that would be another project. =) Wonder if throwing a usb power bank, or the UPS project in line would help smooth things out.
Thanks!
What Unifi switch are you using?
I've noticed problems with current throughput with some of my cameras on the Flex 5 port switches, but I'm using the 250W USW 24 for my hubs, and haven't seen any issues whatsoever.
Scott
[edited to correct typo]
Yes. I tested 2x of the Procet PT-PTC-AT, 1x of the PT-ATC-5V, and 3x of the PTC-AF-5V. Some had slightly more voltage variance than others, but all of the units worked.
[In case someone is going to ask, the graphs above all correspond to the first unit of each type I received.]
FWIW, I kept one of the PT-PTC-AT, and all three of the PTC-AF-5V.
You mean PT-PTC-AT?
Yes, thanks. Typo.
Perfect, you're ready for the Zombie Apocalypse, and the Voltage Apocalypse.
No worries! just bought one (Amazon has them 50% off right now). Here's hoping!
I have a small 8 port unifi, but i see it happening with all of them. I small tplink, mokerlink, and cisco catalyst don't have issues.
Download the Hubitat app
