By Manfred R. Will and Cindi L. Kemper
[Will is vice president of engineering and Kemper is vice president of international sales and marketing for Datatran, Denver, a manufacturer of break-out boxes.]
For years, the active vs. passive debate surrounding break-out box design has been a familiar topic among many top technicians and service managers in the data communications troubleshooting and interfacing industry.
These differences translate into definite advantages and sometimes disadvantages, which experienced technicians should know and be able to use to their own advantage. The added advantage that comes with using the appropriately designed break-out box easily can equate to greater efficiency and information-information that means time savings.
Simply put, the base argument still centers on the method by which LEDs (light emitting diodes) on the break-out box are powered. There are two basic approaches to powering a break-out box: One way is to power the unit with a battery; the other is to take power from the system being tested.
An active box, by definition, is one that uses a buffer or amplifier of some form to drive the LEDs. The power needed to do this generally is provided by one or more internal batteries.
A passive box, on the other hand, is one that draws its power from the signal line itself. This design, in turn, serves to power LEDs without using any outside power. There is no dependence on a battery to help the break-out box function.
Unfortunately, that's where the definition stops and the debate begins, because there's really no right answer. Supporters argue both sides, and pros and cons exist for both designs. Anyone contemplating a break-out box purchase should take time to understand both the advantages and disadvantages of each.
Active Design
One school of thought argues that test equipment should be transparent to the operation of the equipment being tested. False or ambiguous readings, which are caused by the testing unit loading down the system, are unacceptable, so for complex and intermittent problems, an active device should be used.
Classical engineering philosophy would support the concept that any device designed to measure the status of another device should not in any way affect the device being measured. This theory has its followers and would tend to support an actively designed box. (The technical explanation of this theory is expanded later in this article.)
On the other hand, a passively designed break-out box also has its supporters. Because a passive box is capable of reporting the relative signal strengths of each line in use, a great deal of added information suddenly is available at the technician's fingertips. Many technicians feel this additional information, along with its independence from an internal power source, far outweighs any question of impedance level. Besides this, new designs in break-out box construction are drawing as little as 2mA or less off the line.
To really understand the basis of this debate, a quick overview of the RS-232 protocol from a more technical point of view may be helpful.
RS-232 is probably the most common protocol for serial data communications between computers and peripherals.
This standard calls for both positive and negative voltage transmission, typically + 12V or 12V. A positive signal is + 3V to + 12V, a negative signal is 3V to 12V, and anything between +3V and-3V is a disallowed state. Integrated circuit manufacturers have responded to these standards with a pair of ICs (a transmitter and receiver) to handle the electrical characteristics of RS-232C. An RS-232 transmitter (i.e., quad line driver No. 1488) is designed to convert up to four TTL signals into + 12V or 12V levels with the output current limited to 10 mA. The receiver (i.e., quad line receiver No. 1489) converts up to four RS-232 signals back into TTL levels and typically exhibits an input impedance level between 3kO and 7kO.
----------
How they work
The connections between a computer and its peripherals or between a computer and data communications equipment are complex. When problems occur at these interfaces, it often is necessary to use some kind of test device to monitor the signals and determine the cause of the problem.
When this interface is a serial interface to which the EIA recommended standard RS-232 applies, a commonly used testing device for this purpose is the break-out box.
In operation, the break-out box is connected to both sides of the interface. For example, if 'he problem is with a printer interface, one side would be connected to the computer and the other side to the printer.
Connected in this manner, all communications lines would flow through the break-out box. Red and green LEDs on the box then show whether positive or negative signal voltage is present on each pin, on each side. (For a more detailed description of this interface, see "Testing Serial Communication With a Break-out Box" and Computer Corner in the February 1987 issue.)
------------------
These specifications describe the foundation that supports the classical engineering philosophy. The rational here implies that an input buffer with a relatively high input impedance level, say several 100k9, would not affect or load down the signal under test. These buffers are typically of a digital rather than linear (analog) design and are therefore set to turn on at a predefined threshold point (usually +3V for positive signals and-3V for negative signals). These buffers are capable of outputting 10mA to 30mA, so standard LEDs, which require 20mA for rated intensity and are more cost-effective from a manufacturing point of view, may be used. Because of the digital function of the buffer within an active-designed break-out box, all signals above the threshold voltage of the buffer will produce a bright, visible indication on its respective LED. Some technicians prefer an actively designed box because they require brightly lit LEDs. But today, a number of passive boxes use new high efficiency LEDs, which nearly equal the intensity of those in a buffered design even though no outside power source is being used.
Passive design
A passively designed box has its advantages as well. Returning for a moment to classical engineering philosophy, a device designed to test the performance of another device probably would affect or interact with the device under test. The rationale here is that a passive design (an LED combined with a series, current-limiting resistor that is placed across the circuit under test) derives its power from the circuit and should provide a reasonable load to the circuit.
Where the RS-232 protocol is concerned, a reasonable load would be equivalent to the input impedance of a 1489 receiver. The desired response in this case would be a reasonably bright LED indication. Because 1488 drivers are current-limited to 10mA and standard LEDs require about 20mA for rated intensity, an obvious problem exists.
Advances in LED technology throughout the last several years have led to the development of special high-efficiency LEDs. With only 2mA, these LEDs will demonstrate an equivalent brightness to conventional LEDs using 20mA. What all this means is that you can place a typical RS-232 receiver load on an RS-232 driver and see the result.
The advantage to this type of design is its ability to point out potential hardware problems or flaws. For example, a defective driver, which cannot supply rated current, can be quickly and easily detected. Another benefit inherent to a passive design is that the relationship between signal voltage and LED intensity is a linear one. As the signal voltage decreases, so does the LED's intensity, which provides a quick visual indication of relative signal strength. A 4V signal on an active box may be easier to see, but you would never know that a potential problem exists. The ability to measure while testing can be a real benefit in some cases.
Perhaps one of the most important benefits is a passively designed box's independence from batteries.
Untold numbers of technicians have found themselves in the field only to discover they are unable to use their break-out box because of worn out or low batteries. A passively designed box always is dependent solely on the power from the interface and thus eliminates any dependence on outside sources.
The active vs. passive debate continues to recruit new support on both sides. Perhaps now you have a better understanding of the advantages of each. Five years ago, when most passive break-out boxes were inefficient and active break-out boxes were just being introduced, technicians probably would have opted for an active design. But today, because the method of breakout box construction has changed, components are more efficient and technicians are open to new ideas.
The debate continues.
Also see: Logic probes: Troubleshooting to the component level