Archived Forum

Have you changed the RF Frequency in the set-up for both the emonTx and the Pi to 868 MHz? And are you using the correct aerial for 868 MHz?

Yeah, i belive so:

from emonhub.conf

[[RFM2Pi]]
    Type = EmonHubJeeInterfacer
    [[[init_settings]]]
        com_port = /dev/ttyAMA0
    [[[runtimesettings]]]
        group = 210
        frequency = 868
        baseid = 11

and from the arduino:

#define RF69_COMPAT 1    
#define freq RF12_868MHZ // Use the freq to match the module you have.const int nodeID = 10;  // emonTx RFM12B node ID
const int networkGroup = 210;  // emonTx RFM12B wireless network group - needs to be same as emonBase and emonGLCD 
const int UNO = 1;  // Set to 0 if you're not using the UNO bootloader (i.e using Duemilanove) 
                   

I am getting around -30 dB when the devices are very closed to each other. My frequency band is 433 MHz.

I have shared some data at http://emoncms.org/daturach. In that setup, distance is around 15 meters with 2 or 3 walls (15 cm. each) in between.

Have you actually got the 868 MHz radio Modules (see Building Blocks > RFM Radio Modules - Identifying different types).

What about the aerials? If a single wire, the length should be 87 mm.

This is my kit I ordered from openenergymonitor store:

Emontx v3

RFM12B / RFM69CW 433Mhz Low Power RF Module with SMA antenna

Raspberry pi 2, with

RFM69Pi 868Mhz Raspberry Pi Base Station

Receiver Board [RFM69Pi_868]

I tried changing frequency to 433 on both, Now the RSSI is about -60 to -50. I think its pretty low considering the devices are 2 meters from each other.

That range is pretty impressive considering the hardware is configured for 2 different frequencies, previous tests have found range as short as 10-20cm's when using RFM devices on the "other" frequency to which it was intended.

This discussion should have continued on your appropriatly titled "RSSI very low" thread rather than this "Optical Utility Meter LED Pulse Sensor an EmonTH" thread since the 2 have no common ground and will possibly get moved to avoid contaminating this thread.

Since the parts ordered are incompatible without modification you should drop the shop an email with a link to this thread (and the other?) as they are really helpful and will try to help you out, maybe you can exchange the 433 emonTx for a 868?

Failing that you will need to replace the antenna and the rfm sub board to optimize the range. the latter is not an easy task and should be avoided unless you are a dab hand at desoldering as damage is easily done.

There is another less obvious possibility that may work and that is to fit/remove the smd cap that is the physical difference between the rfm boards, fiddly but probably easier than removing the rfm board.

Paul

(moved from http://openenergymonitor.org/emon/node/10843)

I'd say that without professional rework tools, removing the RFM module without damage to the module, the motherboard or both is a near-impossible task. (Ask Robin Emley! I believe his trick is to make the PCB with the pads 'outboard' of the RFM module, and connect with short jumper wires, then removal is just a case of snipping the links and removing their remains one at a time.)

By far the easiest must be to convert the 433 MHz module to 868 MHz by removing the capacitor, though I have not tried this (and assumes all the other components are the same value for both frequencies - do we know this for certain?).

But it may be better to go the other way (add the capacitor, if we can determine the value) as the OEM 'standard' is now 433 MHz and I believe the shop doesn't hold large stocks of the 868 MHz version.

@Robert,

ALL of the antenna tuning SMD parts differ between the 433 and 868 MHz versions of the modules - it is just a handy side effect for a quick eyeball check that one set of cap pads is unused.  Five of them differ significantly - complete module swap sounds an easier path to go down.  With a heat gun, gravity assistance and braid to clean up, it is feasible, but not for the faint hearted.

Thanks for the clarification - it doesn't surprise me in the least. (I added those two paragraphs because Paul suggested it earlier - I assumed, obviously wrongly, that he knew.)

Would an illustrated write-up on removing a module be possible? It's a question that pops up from time to time.

Sorry no. I just threw the idea out there and then actually assumed as it didn't get discounted, it maybe feasible whilst waiting for a knowledgeable response to "do we know this for certain?".

Which we now have, thanks emjay, at least now we know. 

Paull