|
|---|
| Updated: July 25, '07
The following applications apply to all models of the AIM.
DC power and communication cable isolation: Some measurements can be improved by isolating the AIM from the DC power line and also isolating the RS232 data cable. This picture, courtesy of Danny, K6MHE shows how he has isolated his AIM4160 (the 430 and the 4170 are electrically the same as the 4160) with good results for his balun measurements in the HF bands, up to 30 MHz. This isolation is probably more beneficial when the measured impedance level is in the Kohm range.
DC Power cable: 6 Turns, Fair-Rite # 2631102002 Mix 31 RS232/USB: 9 Turns, Fair-Rite # 26311803802 Mix 31 For even better isolation see his Bluetooth interface below.
Data averaging, smoothing and trend: The AIM software has several features to improve the quality of the measurement data. Which of these is better depends on the situation.
Averaging is the most fundamental. This causes several data readings to be taken at each test frequency and these readings are averaged. If there is random noise present, the noise is reduced by the square root of the number of readings. For example, an average of 4 readings will cut the random noise in half compared to no averaging at all. It's important to distinguish random noise (like noise in the output of a receiver) from systematic noise which may be due to some measurement inconsistency or some external noise source. Systematic noise will not be reduced by averaging.
Up to 16 readings can be included in the average. Of course, taking more reading does make the scan slower.
Smoothing is an extension of averaging. It allows more readings at each measurement point to be included in the averaging. This is very effective on random noise but it takes a long time for the result to reach the final average value. For a smoothing factor of 10, it may take 40-50 scans to reach a good result. You can watch the results in recycle mode. After the final results has stabilized, stop recycle mode and do a single scan. The data used for smoothing is retained in memory until the graph scan limits are changed or smoothing is turned off.
Trend uses previous data points to predict the best value for the next data point. That is, if the data had an upward trend, it's expected this upward trend will continue. This makes the plot look much better in the presence of random or systematic noise. The disadvantage is that data points are shifted slightly on the frequency axis, so phase zero crossings may be affected. The amount of the shift depends on the frequency step size used during the scan. A smaller step size results in a smaller phase shift. The effect of trend can be evaluated by doing a scan and then turning trend on (or off) and observing the results. The original raw data will be re-plotted with the new trend value. Compare the results both ways and decide if trend is applicable for a particular situation.
The bottom part of the diagram shows how the Remote Power is routed through the RS232 connector using pin 6 of J5. This circuit is normally completely open, so a jumper labeled jmp1 has to be installed if this option is used. The same voltage limits apply for Remote Power as for the Battery or the regular DC power: Minimum=6V, Maximum=14V When using the RS232 cable to supply DC power to the AIM, use a wire on pin 6 of the DB9 connector. My understanding is that this corresponds to an input on the PC RS232 connector, so it should not have any voltage on it coming from the PC. Conversely, inputs to the PC's RS232 interface can stand at least 15 volts, although I don't suggest testing this. If you do run DC power to the AIM, make a special cable just for this purpose and label it conspicuously. Connect the DC power only to the AIM and not to the PC on pin 6. The PC interface only needs wires on pins 2, 3 and 5. Then install the optional jumper 1 in the AIM, as shown as a blue wire in the previous pictures. The jumper can be soldered on the top side of the board so you don't have to remove the PCB from the case.
Another jumper must be installed as shown in this photo to bypass a resistor that is in series with the RS232 ground wire.
Tuning Stub Adjustment: The length of a coax tuning stub can be checked quickly using the Distance to Fault function:
 This gives the total length of an open or shorted stub and the equivalent quarter wave frequency:
To make a new stub, another function will help you trim the coax quickly for any desired electrical length:
 Enter the quarter wave target frequency and the initial length of the coax. The starting length should be accurately measured because this will be used to calculate the velocity factor of the coax.
 After the data is entered, the program measures the present quarter wave frequency of the coax and calculates how much to trim off to get to the target frequency. This excess length is displayed at the bottom of the graph. As you trim the stub, this value will be updated. When the length gets close to the target value, the graph will zoom in for a closer look.
 To make a stub with a different electrical length, multiply the target frequency by 0.25 and divide by the desired electrical length (fraction of a wavelength). For example, to make a line that is 3/8 wavelength at 7.15MHz, multiply 7.15 by 0.25 and then divide by 3/8 (=0.375) to get the effective target frequency of 4.766MHz (7.15*0.25/0.375 = 4.766). Enter 4.766MHz in the dialog box. Now when the stub is trimmed to be one-quarter wavelength at 4.766MHz, it will be 3/8 wavelength at 7.15MHz.
Piezo Electric Transducer: This graph illustrates how the AIM can be used to measure components besides R,L,C circuits: Devices like this have several resonant points below 1MHz and they can be viewed in one scan.
The calibration device used for a short circuit (green label) was originally made with a piece of wire connecting the center conductor and the ground tab. This causes about 6 or 7 nanohenries (nH) of stray inductance that may affect some measurements. In particular when measuring coax stubs, it makes the resonant frequency appear to be a little higher than it should be. This picture shows the original short circuit cal load and a better model that is made by bending the ground tab over and soldering it directly to the center conductor. Before bending the tab, tin the inside surface with solder. The tab is a little springy and it won't lay flat against the center contact, but the narrow gap can be filled with solder. The rubber boot is not glued. It can be removed by applying some force and wiggling it. | ||||||||||||||||||||||||||||||||||||||||||||
Remote Operation:
This diagram shows how the battery can be wired so it will be completely disconnected
when the plug for the DC power supply is inserted in its socket. It is not possible to charge the battery even when the DC voltage is much higher than the battery voltage. If you need to charge a battery, use the original terminal labeled BAT - on the printed circuit board.
Short Calibration: