A contractor recently called me to help him out of a troublesome situation. He was working at a service provider on a fiber-to-the-home project, testing the optical power at the head end, splitters and subscriber. When measuring the optical power at one location, his power meter indicated +1.1 decibel-milliwatts (dBm), but a tech from another company measured –14.7. He wanted to know what was going on.

Initial questions 

My first question was if both were measuring in dBm. Most power meters measure in decibel-milliwatts for the actual optical power output of a transmitter or input of a receiver and decibels (dB), a relative scale used for loss measurement. 

If one of the meters was set on dBm and the other on dB, the measurements would certainly not agree. He had checked that already, and both were set on the dBm scale for power measurement.

Then I asked if both were set to the same wavelength. Fiber optic power meters have detectors sensitive to the wavelength of light being measured, so the calibration is different at each wavelength. He had checked this, and they were the same.

The next thing I asked was which instruments (including brand and type) were used. His was a well-known brand of optical power meter, while the other tech’s power meter was a bright orange and green inexpensive imported power meter bought online.

You are probably expecting me to say you cannot trust one of those cheap imports, but, no, I have tested a number of them, and they work quite well. However, they do have a design flaw that can cause a problem like this one.

Calibration station

The problem is calibration. All of the inexpensive meters I tested have an option to allow the owner to calibrate the instrument themselves. Push a couple of buttons and you can make the dBm scale read anything you want it to. That violates every law of calibration!

Calibration means that an instrument has been tested against a standard and set up to make measurements traceable to that standard. Within the limits of instrument and measurement uncertainty, your instrument should measure with the same value as the standard and every other instrument calibrated to the same one. 

An instrument should be calibrated regularly by a certified calibration lab with a transfer standard traceable to a national standard. The calibration lab should put a sticker on the instrument indicating the date of calibration that will show tampering if the instrument is opened up and the calibration invalidated.

I’m very aware of the details of calibrating optical power for fiber optics. When I started a fiber optic test company during the early days of this industry, there was no standard for optical power calibration. When we ran into calibration problems like this on an early military project, I had the military customer help me convince the National Bureau of Standards (NBS), now called the National Institute of Standards and Technology, to start a program for this purpose. I worked with NBS to create a system that is still used today to provide international calibration of optical power meters. 

The calibration of optical power meters is simple. You have a transfer standard, which is a laboratory-grade optical power meter to compare to the instrument being calibrated. Since the calibration of power meters is wavelength-sensitive, you also have sources of known wavelengths to use for calibrating power. Measure the source with the standard meter, then with the one being calibrated, compare the results and adjust the meter being calibrated to read the same as the standard meter.

Every instrument, especially optical power meters, should be periodically calibrated. These inexpensive power meters provide no information on their traceable calibration, so they are questionable for making power measurements. If you have one of them and a name brand meter that is calibrated properly, you can set the other to read the same as the calibrated meter. However, you need to check it against a calibrated meter periodically to ensure no one has hit the wrong buttons and changed the calibration.

Most labs can calibrate fiber optic power meters. The optical time-domain reflectometer (OTDR) also needs calibration. Calibrating OTDRs is more complicated than power meters because you have several parameters to calibrate and several setup parameters in the OTDR. 

In my column for ECmag.com, “Calibration of Fiber Optic Instruments,” I discuss calibration techniques and give more detail on power meter and OTDR calibration.

jim hayes