I attempt to measure the power consumption of my PC at work with Fluksometer. For this I removed the shield on the power cable and attached a clamp to one of the wires. I set up a local server on my PC and connect to the Fluksometer through ethernet port. However, even after couple of days I haven't received any data. Even trying to get real time from a sensor I got the following output.
Anyone has any idea? Is it possible to measure a PC power consumption with Fluksometer?
root@Leila:/home# curl -v "http://192.168.255.1:8080/sensor/009413fb23b9be74db1e79c7c3843541?version=1.0&interval=minute&unit=watt&callback=realtime"
* About to connect() to 192.168.255.1 port 8080 (#0)
* Trying 192.168.255.1...
* connected
* Connected to 192.168.255.1 (192.168.255.1) port 8080 (#0)
> GET /sensor/009413fb23b9be74db1e79c7c3843541?version=1.0&interval=minute&unit=watt&callback=realtime HTTP/1.1
> User-Agent: curl/7.26.0
> Host: 192.168.255.1:8080
> Accept: */*
>
* additional stuff not fine transfer.c:1037: 0 0
* HTTP 1.1 or later with persistent connection, pipelining supported
< HTTP/1.1 200 OK
< Content-Type: text/plain
< Transfer-Encoding: chunked
<
* Connection #0 to host 192.168.255.1 left intact
* Closing connection #0
Can jou give us the output (printscreen or copy-paste) of the Syslog (tab 5) on the Fluksometer config screen ?
For local connect and easily receiving data better use MQTT --> see https://www.flukso.net/content/mqtt-mq-telemetry-transport
Perform the following steps to see what the FLM transmits:
Install mosquitto, usually via a corresponding binary build: http://mosquitto.org/download/
On installed mosquitto broker/client/server use the command line to start a client: Start --> Execute --> cmd.exe
Query what is transmitted from your FLM with following command: mosquitto_sub -h 'your local FLM's address' -v -t /sensor/#
Note that -p 1883 is "default", so usually unnecessary to specify.
You do not need to remove the "shield" ? on a cable.
You just need to put it around ONE single insulated cable .
It wont work if 2 are inside the insulation.
Also the power consumption may be too low to measure from the PC effectively.
1- Trying to connect via MQTT, I got connection refused error.
2- By "removing the shield", I meant the outer shield that covers three cables. The cables in the shield each is covered by an individual shield. I did so because I thought that the clamp is working based on induction. Thus, opposite currents in within the power cable affect the measurement and theoretically I should get 0 in that case. To clarify more, I put a photo of my clamp attachment here. Please let me know if it needs correction.
It seems that the html tag is not working here.
Please find the photo of clamp attachment here.
Clamp attachment
Also you can find a copy of device "syslog" here .
I meant tag.
The syslog looks strange with respect to supplied and used network addresses; there seems actually no link established and the FLM seems to be rebooting as if there is a conflict in address space - your LAN seems to work on 192.168.255.x - Leila PC on 104, the FLM tries to get 125, but seems to fail...
Jan 1 00:12:53 flukso-3f2fd6 daemon.warn dnsmasq[855]: not giving name Flukso.home to the DHCP
lease of 192.168.255.125 because the name exists in /etc/hosts with address 192.168.255.1
Please check your general network settings... How is the FLM connected to the internet?
Please see the manual https://www.flukso.net/files/flm02/manual.pdf - chapter 2.1
I haven't established internet connection for a couple of reasons.
If I want to set up internet connection, I should make it through Ethernet. I don't have such a wifi access that is easily configurable on fluksometer. Moreover, toggling to the Ethernet mode, the Ethernet IP address changed and I was not able to connect to device directly for ssh connection. Also I want to access the data on local server, because it is easier for me to manipulate them.
When I got the device fisrt, even the default IP did not work, so I reset it to manufacturer settings.
Also, I make configuration on the device via a VM on my machine, since the VM replicates the NIC, this warning may rooted there.
Anyhow, since ping, ssh, and several other ports on telnet except 1883 are responding, I think there is no connection problem.
Actually there is no connection, but a communication issue - in the default config like you seem to operate the FLM it is reporting via Wifi, not LAN; LAN here is used only to configure the device; the syslog shows that it actually is reporting nothing as it could not establish exactly the required reporting connection... Toggling to LAN indeed provides the FLM another ip address (as described in the manual) - you may read the assigned address from the DHCP server, thus the router - alternative would be you set up a DHCP server on your PC to supply the FLM with an address. You can SSH into the FLM or course with another address also... (I run my FLM on LAN only, with a supplied address be the router...) - closing reporting to internet is done by toggling the "send readings to Flukso platform" on the service tab; by the way, to get to know a PC's power consumption would be easier and especially cheaper with a plug-in energy monitor for around 10 bucks...
Now I toggled to the ethernet mode. But the thing is that DHCP server on flukso does not provide an IP for my PC, even after half an hour. What is the solution?
Is there any way to disable DHCP server on the device and provide the IP for Flukso by runnig a DHCP server on the PC or at list to be able to configure a static IP for the device?
I managed to toggle to ethernet mode via an static IP.
In the luci interface, before toggling to ethernet, I changed the internet connection in the network tab to manual. Now I have the ethernet mode with my desired IP. ALso I checked the wifi interfaces to make sure they are disabled. Although all the things are properly configured, I still get nothing from a real time sensor polling. Even from a web interface I got nothing, Now I want to know what the possible problem is.
There is no data to show or still it is a communication problem. If the power is too low to measure, shouldn't the sensor return (timestamp, "nan")?
You can find the updated "Syslog" here
If the power is to low, you could make some windings (say 2 to 10) and divide the result by the number of chosen windings.
You can also check if there is really an output on the current clamp by measure the dc voltage at the red and black wire.
If the curent clamp is a model 50 A than dc out is 5 Volt when 50 A is passing through if not connected to the FLM.
So with 100 W consumption, then the given current is 0,435 A and the dc voltage out is 0,0435 V
and with 500 W consumption, then the given current is 2,174 A and the dc voltage out is 0,217 V
and so on...at 230 V grid tension.
Since I don't have access to voltage meter, I cannot examine the clamp. But I have two clamps and the probability of having both clamps broken is very low. Today I tried both clamps by connecting them to the direct fuse cable, though, still no data appears even by attempting to retrieve realtime data.
I think there is a problem in communicating with sensor, but I don't have any idea what kind of problem it can be. Again I tried two sensors so presuming the sensors are broken is almost invalid. Any idea?
Does the syslog show a line like
Apr 20 06:19:49 flukso-6ccbfd daemon.info fluksod[668]: processed pulse 0b5cf0a60f22093fade6ae86e2b561f8:1397974789:7091063
Above link to the second syslog is broken... If not, the FLM has not been initialized correctly... And this is a matter of the IP setup, beliebe me or not; there is actually no need to fiddle around with SSHing into the FLM - my last post on this.
Actually this is what I got in the syslog. The device is toggled to the ethernet mode with a static IP assigned before toggling. Now, when I try to poll the sensors, it seems that it reboots. Syslog
In the last syslog, i see constant reboots ?? This is not normal.
Is this a FLM that you bought recently ? You have warranty i supposed...