Sometimes only the right test equipment can determine the difference between ‘all systems go’ and ‘sorry, but no.’
Digital anything is supposed to be great: You plug it in, and it works. No messing with adjustments, no fiddling with knobs, no correcting for timing errors — it just works! Even better, it keeps working without all the fuss of analog. Then, one day it doesn't — now what?
Seasoned veterans know better. Digital is not the hassle-free solution everyone expected. What's worse, when digital equipment does go down, it usually goes down hard. What apparently worked fine yesterday doesn't work at all today. With digital, the light is either red or green. The yellow buffer zone is usually very small or non-existent. So what do you do when that familiar green light is suddenly bright red? Panic. Because it's too late.
The time to learn how the system works is while it's working, not after it has gone down. Because of this fact, test equipment is just as important in today's supposedly trouble-free digital world as it is in the analog realm. While those test lights are all a comfortable shade of green, let's take a close look at test equipment and troubleshooting.
Start small
When it comes to test equipment, there is enough available out there to exhaust even a large budget. And getting the most from test equipment is very dependent on the operator. What's crucial is his or her ability to use the equipment in a manner that provides solid answers to meaningful questions.
A few years back, not long after I took a job as chief engineer at a TV station, I was called in on a weekend. The CG-created logo was not working. Being new, I was unfamiliar with most of the equipment at the station, so my first question was, “Where does the logo come from?” The answer: “Off the disk.”
“Which disk?” I continued, and the response was a less-than-helpful shrug. The operator didn't know. I asked him to call up the logo and watched for drive lights. Sure enough, one drive lit up. After it failed again and stopped, I pulled the 8in. disk out. (For those too young to remember, 8in. disks were used before 5.25in. disks, which of course pre-date today's almost-extinct 3.5in. floppies.)
One look was all it took. About an inch from the outside was a groove worn clear through the magnetic material. You could see right through the disk! It was obvious that the logo was no longer there, having been accessed far too many times. No test equipment was needed, just a good set of eyes.
My point is this: Test equipment will not solve a problem; it will only provide you with an answer to a question about the problem. Knowing what questions to ask and in what order is the job of the technician attempting to solve the problem. One of the best testing resources is a knowledgeable operator who can accurately describe and duplicate a problem. Add to that a technician who can take meaningful observations and relate them to the described problem, and you have an efficient troubleshooting team, which will only be made better with the use of test equipment.
Many times the answers are so obvious we fail to ask the right questions. As a freelance engineer I work on a variety of equipment, including computers. One night I stayed late working on a client's computer. He decided to go home and asked me to shut down the system and lock up the office when I was finished.
The next morning he called saying the monitor was not working. Over the phone, I confirmed that the monitor was on — the power light shined — and that everything was normal, but I neglected to confirm that the CPU actually was on. After driving 20 miles to his office, I turned on the computer and everything worked. As it turned out, he normally turned everything on and off with a power strip. The night before, along with turning off the power strip, I also turned off the computer. No wonder nothing came up on the monitor. It was a frustrating situation, but it illustrates a point.
In the beginning, many problems can be solved through careful observation. Questions like “Is the power on?” can be answered simply by looking at the power light. But as you get better at solving technical problems, your questions will need answers that require test equipment.
In the above scenario, I would have gotten a more complete answer by checking the power supply output with a voltmeter. This would have allowed me to confirm that the +5VDC supply is indeed producing 5V (DC) with no AC ripple. Many devices have multiple voltage supplies and each must be checked to verify they are operating properly. A good analog or digital voltmeter can provide you with these answers for $100 or less.
The next piece of troubleshooting equipment is a general-purpose oscilloscope, which can be found in the $300 to $500 range. But to be useful, an oscilloscope needs at least two channels; it must be capable of triggering from an external source; and it should offer a bandwidth of at least 50MHz — and the higher, the better. For video work, the option of a TV horizontal and TV vertical trigger is quite helpful. A scope with these capabilities will likely cost two or three times the previous figures. Be aware that like the voltmeter, an oscilloscope is at best useless and at worst dangerous (and potentially lethal) in incompetent hands. If you are unsure of your troubleshooting skills and/or electrical knowledge, have a knowledgeable technician walk you through some simple steps and provide you with basic electrical safety instruction.
Signal-specific test equipment
Once you have some knowledge of basic test equipment, it makes sense to move on to more specific pieces. For instance, if you have an analog audio facility, get some quality test equipment specifically for analog audio. This might include an audio generator as well as a calibrated AC voltmeter, as audio is AC at a variety of frequencies and amplitudes. If, however, the majority of signals in your facility are digital audio, different equipment will be needed. Signal converters may allow the analog test equipment to be used, but the answers it will provide are not likely to solve the problems you encounter.
Like audio, video comes in several flavors, and so does video test equipment. Get the appropriate type (analog/digital, component/composite, SD/HD) for the signals in your facility. Facilities with more than one signal type will likely need more than one type of test equipment. Again, converters are helpful, but in most instances they won't help you answer the basic questions.
Today's facilities are more complex than ever. In addition to audio and video, numerous other signals can be found in most production facilities. Some of the most common signals include computer network signals (TCP/IP, Ethernet), control signals (RS-232, RS-422), compressed media (JPEG, MPEG), and others. In nearly all cases, signal-specific test equipment is available — usually for major money. If you deal with a specific signal regularly and find you are wasting considerable time troubleshooting problems, consider the appropriate test equipment. Otherwise, a good voltmeter and general-purpose oscilloscope can go a long way.
Something else that must be considered is the fact that much of today's equipment — especially computer equipment — is far cheaper and easier to replace than repair. Buying test equipment to troubleshoot a computer hard drive is probably a waste of money, unless you expect to be working on 10 to 20 drives every week. General-purpose test equipment such as a voltmeter and an oscilloscope may not allow you to repair items such as a hard drive, but they can help confirm that the hard drive is indeed the source of the problem. Often it is just as easy to plug the drive in question into another computer and see if it exhibits the same symptoms. But this must be done with care. More than once I have removed a shorted assembly from one device and plugged it into another only to find that the result is two broken pieces of equipment rather than one.
Troubleshooting tips
Now that we have discussed test equipment, let's talk about troubleshooting. The object of troubleshooting is to find the root cause of a problem. For me, the challenge is always to replace smaller and smaller parts. Let's say you have a production facility and the only editing room goes down. One way to fix that is to move across town to another facility — one with a working editing room, of course. That's about the largest part that you can replace.
Moving to a more reasonable level, you might be able to determine that the problem is in the production switcher. You could replace that, and that might be a good idea if you really need a better switcher. Even more reasonable (and economical): Find the board in the switcher that is bad and simply replace it.
The next step is where it gets tricky: finding the component or components on the board that have actually failed. This is called component-level repair. It is practiced far less than it used to be. One reason is simple economics: How much time do you want to pay a $100-an-hour technician to find a $5 part? If your technician is good, you might be up and running in an hour. If not, it may be three hours or more, especially if he or she has to obtain the replacement part. Sending the board back to the manufacturer might cost only $150, and the company will overnight a guaranteed replacement.
Another reason component-level repair is no longer common is that parts are often very difficult to obtain and replace. Many require special tools to mount and unmount. In addition, the majority of parts found in today's devices cannot be purchased. An example of this is the replacement lamp for a seven-year-old flatbed scanner. The lamp assembly mounts with two screws, contains two 12in. florescent bulbs, and plugs in with a connector — a simple replacement for a reasonably skilled technician. But this part is not sold by the manufacturer. The only way to replace it is by sending the unit in for repair.
The best way to find a problem's root cause is to divide the system in half. Determine a way to prove one half is working and the other is not. A simple problem might be a monitor connected to a camera. There are three parts: the monitor, the camera, and the wire that connects them. If you have another wire/camera/monitor combination that works, simply swap out the pieces one at a time until you identify the bad component. Once you do that, you can continue and divide that component in half and work through it again. Usually all it takes is three or four iterations to identify a broken component.
