I stumpled upon the Haseman PRM-10 Z-wave 10-channel true RMS power meter, which I find very interesting:
https://haseman-electric.com/home/index.php/component/edocman/prm-10-user-manual/
I have not found anyone discussing it used with OH2. Where would I start getting such a device working with OH? Since it is rather expensive I don’t want to buy it unless I am quite convinced it can be made to work with OH. Any pointers? Anyone else interested in this device?
(Edit: updated link to manual)
In order to add it to the community maintained database, you could look at the reference guide.
Otherwise, if you have the device and provide the xml file created by OH, one of the other maintainers could add it for you.
OK, ordered one 
Please keep us posted. I have looked at this device as well but never got to order one.
I have done a little bit of testing with the device, and based on that I have decided to send it back.
The primary reason for it not being usable in my setup/application is that it can only measure the power on a single phase, because it measures the voltage on the supply terminals. It cannot measure the voltage on other phases, nor use the single phase and “guess” (or be configured for) the phase shift, to estimate power consumption on other phases. It cannot even report the Apparent power, which I think would have been more valuable instead of showing a totally wrong power and power factor.
I guess my expectations for the device were too high 
A second thing to be aware of with regards to integration with OH is, that like the Haseman RS-10PM, it is based on the Z-uno development module from Z-wave.me, so a database entry cannot be created (I do have the XML file, and can confirm the manufacturer ID matches Z-wave.me):
Haseman RS-10PM 10 Channel relay module
The seller says that they have verified it working with OpenHAB through the Z-way binding.
Regarding hooking up the device, there are a lot of screw terminals on the device, all numbered, but not labeled regarding where to connect mains and current clamps. The user manual has a drawing regarding which connections to use, but that does not refer to the numbers on the device, so one has to count the screw terminals. Not the most user friendly design, I would say.
Also, a note of caution: The device I received did not have a CE mark. There was a printed copy of the manual (same as the one linked above), but although that states “Conforms to EU regulations: EN55022 EN610006”, the manual does not have a CE mark.
I contacted the seller, who assured me that device was compliant. He said Haseman responded that the CE mark was not on the product “to save the good design”.
I asked the seller since he was in contact with Haseman, if he could request a declaration of conformity, and he actually did, and forwarded it to me. All in all, the seller has been very helpful.
Thats sad… I´d look forward to hear how it goes.
Isnt this a matter of calibrating the device? (I have no read the manual, but I would assume it would need some kind of calibrating).
Regarding this single phase only… Thats just a simple design decision I guess. It has not been build for measuring three phases devices.
Regarding it can´t build database entry, I find it quite strange. I have a Hasemann dimmer device, (which basicly is a Fibraro dimmer), and it works just fine with openhab and my UZB1 Zwave dongle.
Which is why it could be place on the side or back of the device…
What an odd answer! 
No, there is nothing to calibrate. The device records true RMS power and power factor, which means it must measure both the line voltage (using the supply terminals) and the current (using the current clamp). When I supplied the PRM-10 from one phase and plugged a kettle (with the current clamp) into another phase, the reading was 982W, PF=0.51, compared to the readings 1682W, PF=0.99 when they were connected to the same phase. If there had been a separate channel for the “Apparent power”, I would have expected it to show approximately 1682W in both cases, but unfortunately there is no such reading.
I am not trying to measure a three-phase device, but I wanted to put the PRM-10 in the circuit breaker panel and monitor power consumption on different circuits (“sikringsgrupper”), but that is not possible because the circuits are distributed on the three phases.
Then I guess it is not made with the Z-uno, and therefore does not use the Z-uno device ID. The problem is that the Z-uno is a development board and depending on the software installed on it, it has different definition of endpoints, but the manufacturer and device ID is the same.
It cant measure RMS without knowing the exact voltages (V) and current (A) or energy (W).
Its a huge surprise to me, how come calibration isn´t possible. Make no sense!!
Okay then its useless
Hmm just read the manual…
It measure the current by the use of the clamps. And I would assume it measures the voltage from the AC connections. Thats enough for RMS.
But as you say, it cant work with different circutbreakers, unless they´re connected to the same phase. Which means, in your situation its useless because your circutbreakers are most probably splitted and a mix between the three main phases.
I wonder why they chosen this kind of design, insted of making it capable of using more than one phase. It has to be very speciel setup for the need of 10 inputs.