by 5B4AIY » 15 Jan 2015 16:46
Hi, Bob,
The short answer is no, it is of little consequence. When I re-designed the measurement system I decided to make maximum use of the 12-bit A-D converter. The full explanation is in Annex F, page 66 of the User Manual, but to answer your question, a change of +/-1 to the current calibration factor corresponds to approximately -/+ 1mA to the displayed value, and, equally, a change of +/-1 from the A-D converter corresponds to a change of about 1.1mA Since the current being measured is that of the PA, and its current is somewhat temperature sensitive, it's not unusual to find that you get slight differences in the calibration factor depending upon the ambient temperature, and the PA temperature.
Bear in mind that when calibrating the ammeter you will probably be measuring the total input current of the transceiver with a digital ammeter in series. You will need to measure the idle current first, then use the TUNE mode to get the PA to draw current and now measure the total input current. You then subtract the previously measured idle current to derive the PA current, and then adjust the calibration factor to suit. The net result of all this fiddling around is that in practice if you can repeat the calibration and get the calibration factor to be within about +/-10 on different days and temperatures, then you are doing well! What you are seeing is the raw value of the A-D converter count default value, and is scaled by the calibration factor. A change of +/- 1 count corresponds to a current change of +/- 1.1mA, so this is equally of little significance. To avoid lots of needless calculation, the alarm system actually uses the A-D converter raw value, and when you adjust the trip value what you are actually doing is altering the raw value at which to trip the alarm. As the calibration factor alters the ammeter reading, so the raw value for the trip setting has to be altered to suit. When you step to this page in the System Calibration menu, the equivalent ammeter trip reading corresponding to this raw value scaled by the calibration factor is what is displayed.
Because of this, I deliberately quantised the current measurement display to 10mA. I hope this explanation helps. The object of the exercise was to ensure that the overall resolution of the measurement system was as high as possible, and that the accuracy was then a hardware limitation.
Note that the latest version is 1.20a Build 7 which fixes a trivial bug in the reset defaults procedure. If you have already loaded Build 6, then you can safely load Build 7 over it without causing a checksum error. The only difference is that when resetting the defaults VFOs D, E, M and N were incorrectly set to LSB rather than USB. Build 7 fixes this as well as tidying up the reset defaults code into a more elegant form, which I had been meaning to do for some time.
Hope you are happy with the continuing updates.
73, Adrian, 5B4AIY