From: Franck GALVIN [mailto:[email protected]] Sent: Tuesday, May 06, 2003 8:36 AM To: [email protected] Subject: Correction factors
Dear all, I perform conducted or radiated emission test with HP8594EM spectrum analyser or HP8656A receiver. To not have measurements errors I must enter correction factors (antenna or LISN, cable and other ...). My questions are: 1/ When you enter data to antenna factors for example, you write ''frequency'', ''amplitude'' and what about ''frequency scale'' (log or lin) ? 2/ Does frequency scale depend of correction type (cable, amplifier, antenna, LISN, transient limiter...)? Thanks in advance for your answers. Franck GALVIN eLAbs (France) Franck: I have somewhat older HP hardware & software than your setup, but here's how the HP software works for me. When you are creating an equipment calibration table, you are asked to declare the number of data pairs (frequency & amplitude) that define a curve which represents the response of the equipment. For example, an external attenuator might be defined by only two pairs, say, 0.000001 / 20 & 18000 / 20 (data units are MHz & dB). However, for something like a current probe or antenna, you might have a file of 50 pairs or so. You also choose LIN or LOG, which specifies the interpolation method used between the data points. I always choose LOG, since the completed emission plots usually use a logarithmic frequency scale. I don't think this makes any practical difference, but then I always enter enough data pairs so that the delta dB from one pair to another is 1 dB or less. After setting up a calibration file (you need a transducer, cable and limit set of files), be sure to test the expected accuracy by injecting a known signal into the coax cable "head-end" to verify that the factors are being added correctly to the raw spectrum analyzer data. IIRC, the HP convention was to always subtract the correction factor value >from the raw data. Thus, you have to make the attenuator data a negative number, so that subtracting a negative yields adding a positive! (Hope you don't think I'm joking!!! Read you software manual very carefully.) Regards, Ed Ed Price [email protected] NARTE Certified EMC Engineer & Technician Electromagnetic Compatibility Lab Cubic Defense Systems San Diego, CA USA 858-505-2780 (Voice) 858-505-1583 (Fax) Military & Avionics EMC Is Our Specialty
