Are optical power and electrical power directly comparable?

[Lawrenciumbc]

I am analysing an RF system that changes an incoming electrical signal to optical to be distributed by fibre optic and then back into electrical at the receiving end for coax cable to the receiver box. I am measuring the system and looking at its performance and notice that my optical power readings are around -6dBm and my electrical power readings after conversion are around 72dBuV.

By my calculations to convert the dBuV reading to dBm (correct me if I am wrong!), I am getting around -34dBm.

Obviously this is a fairly large drop and I wondered if optical power is directly comparable with electrical power in the same system or if they are are two separate entities? (or if I have done my conversion wrong!)

I want to be able to draw a conclusion to the power drop to the signal in the system during optical to electrical conversion and wasn't sure if it was a fair comparison.

 

[Sunnysky]

what emitter part number and efficiency?
laser efficiency is low.

 

[Lawrenciumbc]

well it's actually a satellite TV distribution system with an optical LNB rather than a standard coax LNB. The optical power at the LNB output is around 3dBm and it loses around 10dB going through an 8 way optical splitter which is why optical power after that point is around -6/-7dBm. This, according to the manufacturer is about right as per spec.

I am looking at the system as a whole from the LNB to the set top box and so it could be that the laser performance in the LNB could be improved as a result of my findings, but I don't have any information on the specific internals of the components.

The converter box between the optical splitter and the set top box is where this apparent huge power drop appears. It is a passive device so I expected some reasonable drop but wondered if it was something I could look into based on the power readings I have, or if comparing optical signal power to electrical signal power isn't exactly a fair 'apples for apples' comparison (for example with completely made up numbers, -34dBm in electrical power is equivalent to -10dBm in optical power or something like that).

The resulting signal to the set top box is fine and the picture is great for it's intended use (watching TV with) but improvements accross the board will obviously allow the technology to be used for other applications or extended transmission distances etc...

The components in question are:
LNB = http://www.gionlineshop.co.uk/Downloads/FibreMDU_LNB_H&C120-04[1].pdf
converter = http://www.gionlineshop.co.uk/Downloads/Mark III GTU Quad and Quatro -03.pdf

 

[Sunnysky]

If starting with 3dBm optical and splitter drops 11 dB ( to -8dBm) with cable loss say 12dB ( to -20 dBm = 0.01mW)

Detectors are usually 0.5A/mW with transimpedance amps.

Meanwhile 72dBμV is ~ 4mV in 75 Ω ~ -37dBm

hmm

 

[Lawrenciumbc]

yes, so essentially the power loss of the signal is the -7dBm after the optical splitter (the cable loss is negligible as the length is only a few meters and is spec'd at 0.3dB per km loss), down to -34dBm making a -27dBm loss through the conversion.

Maybe there could be some deliberate attenuation in there to bring the signal down for the set top box? (I know my fibre optic broadband has an attenuator on it before it goes into my router, as fitted by my ISP)

The idea of the system is to distribute the TV signal to many users off the same dish/LNB. With that in mind and speaking generally, a reasonable power needs to be transmitted to reach as many users as possible. This would then make sense why any required attenuation that may need to be applied would be done at this stage.