Probably most hobbyists have no use and desire for, or access to this equipment; but there are times when meters do not provide enough information for thorough analysis of problems or designs. Both current and voltage waveforms yield much more data concerning wanted or unwanted pulse parameters. Often some very unexpected results are revealed, which may lead to problem solutions. This is particularly true in magneto=electric devices such as motors and switch machine solenoids. Electrical pulse noise is another culprit that often must be reduced to eliminate RF interference with TV's and radios.
Although originally bought for servicing computers, this equipment has proved very useful in profound evaluation of motors, powerpacks and other modelrailroad devices and circuits. The 100 MHZ band width digital readout socilloscope is over kill for these applications, since everything useful is well below 1 MHZ. Dual inputs permit synchronous comparisons of voltage and current waveforms. The digital readout and cursors are convenient for very accurate measurements of instantaneous voltage, current, time period, frequency and phase shift. From over 50 years of experience in scope usage, many learned techniques have enabled the viewing of some very difficult to obtain waveforms.
Note: Adjust brightness and contrast for optimum viewing.
S: Oscilloscope
D: Data Aquistion Unit
MON: Aniga 2000 monitor
MM: Digital Multimeter
Since many of the pulses and waveforms in modelrailroad devices have very low frequency components and long time periods, a storage scope would be required. Due to high cost, an alternative had to be found. An inexpensive, two channel, data acquisition unit with software and interface for the Amiga 2000 with two distinct functions was added. The first is a storage scope with time base from 10 µ s to 500 ms, amplifier range from 100 mv/div to 10 v/div, continuous and single sweep triggering and adjustable averaging to fill in between sample points. Once captured waveform veiws are controled by selecting a shorter segment which can be magnified by 5. The second function is a single shot plotter with a total time range from 1 second to 10 hours to capture transients and intermittant events. These permitted collecting and storing long time period waveforms by sampling methods. Display is on the computer's monitor and can be dumped for printout or stored on disk. Over a 12 year period many plots were stored for future reference.
Unfortunately the designer used the computer's master clock for synch and timing. When the computer was upgraded from 7 MHz to 30 MHz , the timing algorithm had to be rewritten in assembly language to correct the base reference timing. Later the the software was altered to use the real time clock, making the unit independent of the computer clock rate.
Armed with these two devices and in spite of my experience in electronics, many unexpected results occurred. Even advanced theory does not predict some of the strange, and often intermittent transients observed. Unfortunately, at the time of obsevation, not every display was considered wothy of saving, influenced by the original 40 MB storge capacity. Today with about 27 GB available, too much is saved, making sorting and notation difficult. Many of the seemingly nebulous comments, found throughout the articles, are based on observations using them.
One of these is that; on Kemtron, NJI Tenshodo and similar switch machines, the inrush current can exceed 40 amps for a few milliseconds. This explains why; stupidly and unsuccessfully, I spent over a half hour trying to adjust turnout linkage to one, using a 1 amp power pack that was handy. The internal pack resistance was high enough to prevent sufficient current for the surge needed. Further tests indicated that voltages as high as 40 v could be used safely with cutoff devices. This is a good argument for the use of CAPACITIVE SWITCH MACHINE POWER SUPPLIES.
With all the variations in design, motors present many surprises. Arcing between brushes and commutator segments varies drastically with brush pressure, size, shape or material along with segment material, slot size and trueness. Even shaft end play effects the amount of arcing, which is frequently different in forward or reverse. Arcing not only causes wear on both, but reduces efficiency due to resistance. RPM and torque can both suffer, but the worst effect is heat generation. Although not readily visible to the naked eye, the amount is translated into spike, pulse amplitude in both current and voltage waveforms. This is the major source of RF noise or interference. While on open frame motors some of the arcing can be readily corrected, can or other enclosed types may be extremely difficult. Many European motors have bypass capacitors across brush leads to reduce RF noise effects.
When powerpack PULSES are added, waveforms become very complex, almost defying analysis. Although an experienced eye may be able to separate the components, measurements of each requires very expensive scopes using sampling heads with fast Fourier transforms (FFT) and anti-aliasing software. Amiga FFT software is used for this type of spectrum analysis with reasonable results. The combination of added pulses and motor generated pulses produced some very short rise and fall times, which probably account for a large percentage of heat developed, when using pulse power.
BACK TO METERS
BACK TO ESOTERIC TEST EQUIPMENT
