Intermittents and Typical Repairs

A while back, a friend “gave” me a Mackie 12” TH-12A powered stage monitor when he was clearing out his shop of things he either didn’t want or didn’t want to spend time trying to fix. I have no reason to want a powered stage monitor, but for whatever reason I decided I wanted to take a crack at seeing if I could fix the monitor. I like Mackie gear and when it breaks unreasonably easily, I get curious. The symptom was “no power indicator or output.” Supposedly, someone renting the speaker had dropped it and it didn’t power up after that incident. My friend had already made a quick attempt at troubleshooting the speaker and confirmed the reported symptom. 

Because of the previous repair, the amp panel had already been removed from the speaker box and the wiring disconnected. My first suspicion is always “connectors and solder connections.” To remove the first likely candidate, I always check the connectors, physically, and clean the connections. A smarter person would re-connect the amp to the speakers and confirm the initial diagnosis, but I wasn’t doing this for money and I didn’t really care if the solution turned out to be overly simple. I was back in my teenage mode of just wanting to see how the product was assembled and looking at the thing out of curiosity, first, and problem-solving, second. There are advantages to being less-than-meticulous about documenting an organized repair. The first of those advantages is that I spend more effort exploring, examining, and learning than I do ticking off boxes and accounting for my time. 

A disadvantage to this process is that it is entirely possible to trip an intermittent fault back into operation and I wouldn’t know that I’d even passed over the intermittent. In my experience, at least 50% of the electronic problems I’ve been asked to troubleshoot are intermittent, which is the nature of connectors and solder connection failures. They rarely outright fail, unless there is enough current in the area of the fault to cause complete failure by burning, arcing, or in-rush current causing damage to other related components. 

What I found during my inspection was that several of the Molex connectors (see red circled connectors in the picture at right) had received enough tension from their attached wires to pull the male base up along the length of the pins; possibly removing contact from the female socket. That isn’t a problem you’d expect to see in a typical application, but Mackie had cut the speaker wires short enough that there was no strain relief possible and the weight of the wire put some strain on the connectors. One of the first things I noticed in inspecting the power amp board was that the woofer male base had slid more than half-way up the pins. Looking around the boards, I found at least two more connectors with a similar appearance. 

With that background, I disconnected, one-at-a-time, all of the Molex connectors, pressed the male header firmly against the PCB, cleaned the connectors with Caig DeoxIT, reassembled the male and female connectors, and glued the connectors to each other and the PCB with a latex adhesive. After cleaning and reattaching all of the Mackie’s connectors, I power up the speaker and found that it worked. I let the adhesive cure for a few hours and gave the amp some signal while tapping all over the amp and power supply assemble looking for intermittent operation. Not finding any evidence of intermittent operation, I “burned it in” for several hours at a variety of volume levels. With some confidence that I’d found the problem, I gave the speaker back to its owner. 

This kind of repair reminds me of the many things that were always frustrating as a repair tech. First, troubleshooting intermittent failures is miserable work. When you are "finished," at best you feel you've done everything you could to be certain you've fixed the problem(s), but you are never even close to 100% confident. Intermittent faults have a Murphy's Law mind of their own. I used to joke that about half of the problems I'd been called in to repair "fixed themselves the moment they saw a toolbox." Of course, half of that half were user error issues that often cleared themselves up before or immediately after I arrived. That kind of repair is really difficult to convince customers that my time still needed to be paid for. 

Early on in the last version of Wirebender Audio Services, I quit taking telephone calls for service work because a few customers would pretend they didn't know who called me, when the problem "fixed itself." A few of those customers learned that I never forget being stiffed and whined like spoiled children when they learned that under no circumstances would I ever do work for them again. In the case of the Mackie speaker repair, I have a couple other ideas to chase down if the intermittent operation comes back to haunt me, but since I didn't have an associated bill with the original repair I can be fairly relaxed about the whole affair. 

How Much Will This Cost?

One of the many things students don’t get much of in school is practical economics. As a result, the hourly rate lots of freshly-hatched audio technicians earn is embarrassingly unlivable. One of many reasons the career-longevity of people in the music and entertainment business is notoriously short. While I was working on one of those projects that usually inspires the notion “I could be doing better than this working at a convenience store,” I decided to write something about my own business experience in the music and audio business.

Until the day I retired, I had a price list for the various services I supplied to a variety of; from musicians and facilities who hired me for location recording gigs to studio and audio equipment repair services for musicians and recording studios to building contractors (acoustic design consulting) to law firms (audio forensics). Some of my price list was determined by the liability insurance required which was mostly determined by the customer (law firms) and the risk (to me, financially) involved in making a court statement about whatever the law suit involved. Acoustic consulting work for commercial construction was almost as risky and required similar insurance protection.

My rates ran from $75/hour to $275/hour. Those are “billable” rates, however, which are not necessarily the same as the hours I actually spent on a project. For example, a lot of recording work requires prep time and effort that the customer doesn’t see or appreciate. So, it just gets folded in to the billed rate. Recording projects are particularly expensive, time-wise, from my perspective. It’s possible at the high end of the “audio engineering” world that people actually make a decent income, but I always suspect a trust fund is involved when someone can afford to spend a career in entertainment.

Oddly and classically American, my experience with the high-ticket projects came dangerously close to being volunteer work because my liability insurance was so expensive. When I started planning for retirement and was winding down my businesses, the legal work went first. That saved me about $3500/year in insurance costs. The only way doing that work makes sense is if you do enough of it to make up for the insurance cost; at least 20 billable hours. It isn’t reliable work and lawyers, as you might guess, don’t pay their bills quickly and are hard to collect from. It was easy work to give up.

For that same reason, the acoustic consulting work went next. That saved me about $1500/year in insurance costs. Again, it is work you can’t count on and it’s usually not much fun. Although the engineer for the last contractor I worked with was terrific. He did everything he could to take responsibility from me for liability and customer interaction. Designing a quiet staircase in a multi-use building or plugging all of the sound passage routes between a first floor pediatric dental office and a lower-level computer programming company is just not exciting work. Recording studio work is fun, but not particularly profitable. Most of your customers are cheap and not particularly sophisticated and your competition all has a trust fund to spend.

Mostly because several of my ex-Studio Maintenance students were doing a lot of studio maintenance and looking for more work, that business was mostly easy to quit. I just started directing inquiries to them. In the last two years, I’d raised my prices to $225/hour hoping that would drive the business to cheaper techs. When that didn’t work, I had to tell a few customers that I was absolutely no longer interested in crawling around their studios or scouring the food, drink, and drugs out of their wreaked equipment. At least one of those characters had alienated every decent tech I knew of and, I suspect, they simply shut down the studio after the last piece of gear quit working.

Since the 70’s, I haven’t done music equipment repair for anyone other than myself, mostly buying broken gear and selling it in original or hot-rodded condition. In March, 2018, I had about a half-dozen pieces of equipment left in my shop and I am slowly working my way through the pile. When it’s done, I’ll probably convert the electronics shop to a guitar building and repair shop. Or maybe I’ll hand the bench over to my wife for her art projects.

I still do occasional location recording gigs, but with really limited capabilities: no more than 8 simultaneous channels and my interest in working on overdub or large-track-count projects is close to zero. My simple rule for recording projects is, “If I like what you’re doing and the project is fun (meaning, you are not an asshole), I do the job for my cost. If I don’t like you or the project, you can’t afford me.” Most people can’t afford me. I’ve quit pretending like people or projects that I’d just as soon avoid.

I’m not being a snob, I would just rather be retired and working on my own projects than worrying about pleasing someone else. I spent more than 50 years of my life trying to make other people happy. For the few years I have left to live, the only person I’m interested in satisfying (avocationally and vocationally) is me. I’ve paid all of the multi-tasking, menial labor, service-job dues I’m planning on paying until I croak. For a long while, I was fairly good at pretending to care what marginally-talented, egotistical, humorless, joyless people thought of my work. Those days are done and, unless I find a way to go broke between now and my last breath, they will never return.

The High Cost of Doubt

When I was a working engineer and, later, an engineering manager, my least favorite answer to any question was always, “Because everyone does it that way.” For the last 50 years, a mantra of mine, and my current email signature, has been a Bertrand Russell quote, “The fact that an opinion has been widely held is no evidence whatever that it is not utterly absurd; indeed in view of the silliness of the majority of mankind, a widespread belief is more often likely to be foolish than sensible." More than often this is true, but occasionally my kneejerk reaction to convention takes me, the hard way, through the processes that others have already explored and I discover sometimes things are always done pretty much the same way for a reason. Sometimes that reason is bullshit, but there is often a core assumption to be evaluated that with my limited-to-non-existent visual skills I can not see until I experiment.

I am making an electric bass this semester in my Red Wing Southeast Community Tech Guitar Repair and Building program. There are imposed limitations on the kind of instrument we can build as first year students. (I know, I’ve been at it for two years and I’m still taking 1st year classes. Get over it. I am in no hurry.)

My baseline bass is a Steinberg L-Series headless and almost body-less bass. I’ve owned a couple of these instruments in the last thirty years and, currently, own and play a Hohner knock-off of this bass. I love everything about my Hohner from the low weight to the perfect balance to the sound. The only thing I don’t love is the look, which is always a heavily enameled solid color body and neck with a rosewood fingerboard. If the guitar is going to be made from wood, which my Hohner is and my Steinbergers were not. The headless design was not an option for the first year guitar design course. One of the failures from my acoustic guitar build was my original intention to make my instrument from all-domestic wood. I screwed up my first two walnut fingerboards and my one-piece walnut neck block and had to start over quickly, so I snagged a slab of quarter-sawn rosewood and a mahogany neck block from a guitar builder who was dumping material stock cheap and those two critical parts of the instrument ended up coming from imported wood.

This time, I wanted to avoid that and my design was for all walnut; partially because I love the look of walnut and partially because I wanted to take a second look at a design I started almost 40 years ago. That design was based on the B.C. Rich Mockingbird body and incorporated a fretless Fender Precision neck. Considering my rookie woodworking status, that instrument came out pretty well, except that it weighed close to 50 pounds. I started with a 2” thick piece of old growth walnut that I’d milled myself from rough walnut stock I’d found in a friend’s barn. For my current instrument, I wanted everything I’d accomplished with that original instrument without the physical stress. Literally, my bandmates called my old bass “the refrigerator” because between the bass and the Star roadcase, hauling that thing up a flight of stairs was a lot like carrying a refrigerator with one hand. It currently resides in my daughter’s music room where no one has played it for decades. Honestly, it was a terrific bass, sonically, but it will cut a groove in your shoulder that will bisect the player if given enough time.

Once the bass was assembled and finished, I had to decide where to install the strap pins. My working theory was “balanced,” So, I fooled around with clamps, string, and a strap until I found a location that appeared to be evenly balanced. I wrestled with this measurement for quite a while and mentally wrestled with my memory of instrument feel, strap constriction or flexibility (tightly restrained or hyper loose like the classical acoustic guitar strap attachment), and the concept of “balance” for hours. Finally, I decided to blow off convention and attach the strap in the two places shown in these two pictures. Punching holes in my recently sanded, finished, and polished instrument wasn’t an easy move and going for something less convention for the neck position pin was a mental stretch.

The picture at left shows what balanced looks like. The guitar hangs perfectly neutral when the neck is parallel to the ground. This sort of defeats about 50% of the reason I designed in that “handle” at the top of the instrument, but I thought balance was more important than justifying my design geekiness. Turns out, balanced is not particularly comfortable in a bass, at least for me. While my Steinberger/Hohner L-Series instruments are, in fact, balanced, a slight increase in headstock weight makes a body-heavy strap positon a lot more desireable. For one, I have short limbs (and fingers) and I end up pulling the neck up and slightly to my right when I play the bass. So, I repositioned the neck pin as far up that “handle” as possible, which is pretty much exactly where everyone else on the planet puts a neck strap pin. The end result was a much more comfortable balance and a more relaxed playing position. Which, of course, everyone who has built an electric guitar in the last 75 years knew before I decided to test traditional thinking.

ARTICLE: Putting It All Together

All Rights Reserved © 1999 Thomas W. Day
(Originally published in Recording Magazine, June 1999 Volume 12, Number 9, P20-31, under the title "An Ounce of Cure.")

In the Beginning?


Usually, we collect a pile of equipment and, over time, figure out how to put it all together. More often than not, there isn't a way around this scenario. The only exception I can think of would be that you've just won the lottery and are out to buy everything you've ever dreamed of owning; and will never dream again when that big shopping trip is finished. So, we buy the bits as we find them, find the money, or both.

Then, we find a space in the rack and a pair of empty jacks on the patchbay and on we go. Just like we'd always planned to put things together that way. The bigger the rig gets, the more likely we are to be retroactive about how it's all assembled.

Having money or being famous doesn't seem to keep us from creating audio disaster areas. One of the biggest-buck studios I've visited had some of the worst laid-out equipment I've ever seen. And I've stumbled through dinky, one-bedroom studios with three or four pieces of equipment that were nothing but a tangled rat's nest of cables. Obviously, really pristine layouts exist in either environment, too. Planning and maintenance are what separate the mediocre from the superb.


Planning for Success

Even if you've been growing your equipment collection in the same room since the 60's, it won't hurt to wade your way through a planning exercise. It may identify a spot in your signal path that has been bugging you since you moved into your studio.

While it probably seems like a waste of time, especially if your system only includes a computer and a few MIDI pieces and a mixer, I recommend drawing a schematic of the equipment connection plan. Before you ever connect a cable or plug in a power cord, draw out the equipment you own and determine the best place in the signal path for each piece. A good example of a planned signal path is the block diagram that you'll probably find in your mixer's owners' manual.

If you aren't going to do this layout work on CAD, consider stealing an old Industrial Engineering trick. Make a small, rough drawing of each piece of equipment, piece of furniture, and rack cabinet that you'll have in the room. Cut out the drawings and put them aside for a moment. On scale with your furniture, draw your room, including the lighting and power sources. Now, using a temporary office adhesive, do a few dozen "what ifs" with the equipment and furniture cutouts on your scale studio facility. Eventually, you'll find the best place for everything and make the most use of your available space.

If you can design your rack and instrument layout to absolutely minimize cable lengths, do that too. Electrons like short paths. The shorter your cables, the less likely your wiring will pick up noise and interference from undesirable sources. Short cables also introduce fewer losses in your signal path, due to reactive cabling components (inductance, capacitance, and resistance).


Things that Go "humm" in the Night


An important reason for doing all this planning is to make your system quiet. When you start recording you can add all the noise you want, but a great recording studio has a very low noise floor. You can save yourself a few decibels in signal-to-noise by identifying the best physical locations for your equipment. If you can avoid running low level signal cabling over power transformers, do it. This is the reason that knowledgeable people isolate their power amplifiers from their low level signal equipment. Power amps have big, nasty supply transformers than can couple 60Hz harmonics into just about anything placed nearby.

While we're talking about undesirable noise sources, I should mention a few. Light dimmers, for example, are one of the most insidious things left over from the Evil Empire. There are dimmers specially made to reduce EMI, but they don't always work as well as advertised. Personally, I think it's smarter to design your lighting system so that you can illuminate what you need illuminated. Turn off lights when you don't need them.

Old style florescent lights are almost as evil. The ballasts (transformers) are terrific sources of power line harmonics and they can plague you when they get near mikes, instrument pickups, tape heads, and any other high-gain, low level signal source. Some mixers seem incapable of shielding their internal circuits from florescent fixtures that live anywhere in the vicinity. The florescent bulbs that screw into normal incandescent bulb sockets don't seem to be a problem. I understand low noise shielded florescent tube fixtures are also available.

I mentioned power supply transformers, earlier. There is an AC-to-DC power supply in nearly every piece of equipment you'll buy. Some manufacturers save themselves from power supply EMI by using wall-warts. That helps them in the specs-manship wars, but it can create all new problems for you in your studio. I downgrade my interest in any piece of equipment that uses a wall-wart. The product has to be out-of-this-world to push me past that design flaw.

Traditional power supplies end up almost anywhere in the cases of equipment and, while their location is usually optimal for that specific piece of gear, the transformer orientation can generate hum and noise in nearby equipment. Most of the time, you're going to stack a half dozen effects and other gear in a rack, piling power supply upon power supply. With some of the supply transformers at just about every location possible, relative to the other equipment's' sensitive circuitry or input connectors, it's possible that the wrong rack organization can produce a lot of noise at the output of your system.

One simple and fairly reliable method of checking for inter-equipment interference begins with terminating the input connectors (using a balanced or unbalanced connector shunted with a 100-600 ohm resistor).

Turn off all of the pieces of equipment in the rack but the one you are testing. Crank the gain of the DUT (Device Under Test) to maximum. If you have a DMM (Digital Multi-Meter), connect its leads to an output terminal and measure the output noise. If you don't have any equipment, connect the DUT's output to your board and listen to the noise output of this piece through the board's headphone output. Once you have the base noise level, turn on the two surrounding pieces of equipment in the rack. If the noise stays constant, there is no power supply interference. If it increases, turn off each of the surrounding pieces, one at a time, and see if the interference is caused by one or both. If one of the two is the culprit, try moving it. If it's both, the DUT may be poorly shielded or you may have a ground loop problem.

With all EMI problems, distance is your friend. If you can afford to give up a rack space between each piece, you'll most likely have a quieter system than if you have to pack every space in your rack. It's a good idea to leave a couple of spaces between any piece of equipment that uses a lot of power (like power amplifiers and mixer power supplies) and small signal equipment. You should consider doing this just for the advantage you will get in heat dissipation, if not for the signal-to-noise precaution.


Be Cool


Heat is the enemy of all electrical equipment. Hot air travels upward, so it's a good idea to plan for this. If you're going to be driving power amplifiers hard enough that they will get hot (or if you have tube equipment), it's a good idea to put that kind of equipment at the top of the rack. It's a better idea to force ventilate the rack when you're producing a lot of heat in that enclosed space. Do not put high temperature equipment under equipment that has moving parts or rubber components, like tape machines. The heat will accelerate the deterioration of lubrication and flexible items like pinch rollers will quickly harden.


Making the Best Connections


Eventually, you get to start wiring stuff up, but not yet. Having spent a good portion of my life finding and repairing poor connections, I may be superstitious about wiring and connectors. While I'm not particularly fond of soldering phono (RCA) cables, I still roll my own. There are several companies who specialize in audio cabling and they do a good job at it. You can buy well assembled, high quality interconnects for not a lot more than what the cable and connectors might cost you. Still, if you are good with a soldering iron and can do a proper cable prep, do the work yourself. If you aren't, you're probably better off buying cables. I think the most significant downside to not being able to your own cables is that you need to keep plenty of spares on hand if you can't do repairs. The most significant upsides to making your own interconnects are that the cables can be exactly the right length for the job and you know who to blame for any failures.

Usually, you don't get many chances to "optimize" once you have your cables and equipment in place. But if you buy professional equipment, you may have an opportunity to use "hard" connections; screw-type barrier strips. If you have a choice between a screwed-down connection and a plug-in connector, always pick the first. Crimp a spade connector (properly sized for the wire used) onto bare cable (do not tin the wire) and torque the #6 screw to 8 in/lb. OK, I'm exaggerating about using a torque driver, but I do believe in the value of hard connections.

If you're really paranoid, which I am, go an extra mile before you haul out the screwdriver. I have used CAIG Labs products since the mid-70's. In several of my career incarnations, I've used some of their products in case lots and by the gallon. I'll be among the first to admit that product names like ProGold, DeoxIT, and CaiLube sound pretty New Agey, but the stuff works. I swear by it. I even spend my own money on it for my home gear. I spray the bare wires and the terminal connectors with DeoxIT before I crimp the two together. I spray the barrier strips before I insert the terminal spade and tighten the screws down. Unless they are plated, I spray every tin-lead connector in my system with DeoxIT. If they're plated, I use ProGold. It doesn't take much and it adds a new application to the "ounce of prevention" adage. I also use those two products on other connectors, too.


Grounds Should Not Loop


Now we've got stuff in the rack and we're ready to get connected? Not exactly. We're not through planning. The evil curse of AC-powered analog electronics are ground-loops. Once you have a ground loop problem, you can take dozens of troubleshooting hours to find it. Taking a few preventative steps in the beginning will make later troubleshooting a lot easier. And, you may have noticed that I haven't actually put a screwdriver to rack screws yet.

Ground-loops are probably the most common place where a studio's signal-to-noise ratio gets degraded. A ground-loop is what happens when there are at least two paths for current that appears on the signal or power ground. For instance, a pair of audio signal processing pieces are connected to two different power outlets but are also connected to the same patchbay, through their input and output jacks. If there is any potential voltage between the ground connections on those two power outlets, that voltage will find a common path through the signal ground wiring. That wiring is tied together at the patchbay. The current produced by the ground wiring voltage appears on the shielding of the input and output cables. The electrical components of the cables and the common-mode amplification characteristics of the signal processing equipment will allow the signal on the ground wiring into the equipment's amplifiers. If the two pieces of equipment are semi-pro and unbalanced, you have no common mode rejection and you may have lots of hum.

Obviously, one tactic for breaking this closed loop is to cut off the third prong of at least one of the power cords. That's also a great way to design up a system that may someday provide your wife with a lump sum payment from your life insurance carrier. Dumb, dumb, dumb. You do not want to set yourself up to be the return path for ground current when a power supply transformer fails and several amps of current are looking for a route back to Mother Earth.

Of course, some equipment is sold with two-prong power wiring. To get away without the safety precaution of a secure case ground path, the external case of the equipment must be "double insulated" from power ground to prevent ground current from finding its way to the case of the product. This is common in equipment designed for home use (stereo stuff). This equipment can be very difficult to integrate into a recording studio because of grounding issues.

One much better solution to power line ground loops is to use an "isolated ground" for your entire system. This means that you actually take the precaution of making sure there is a real, earth ground connected to your equipment racks and other gear. This ground will return all the way back to the main building ground at the circuit panel. This system is called "star grounding." To make the job complete, you also make sure that a big, fat solid copper wire runs from the building ground to a copper rod (your local codes will determine the length of the rod) that is driven into the earth. The rod ought to be as close to the circuit breaker box as possible. A clamp to a cold water tap is nice, but the copper rod is better. This is very practical for a home studio and ought to be one of the first things you do when you start laying out your studio design. This is especially critical in locations that see regular lightening storms.

Once you have this honest-to-NEC (National Electrical Code) ground, don't connect anything outside of your studio to the ground wire. If all of your audio gear is connected to a single, isolated ground circuit, your power ground loops ought to be minimized. It's a great theory, anyway. Connecting the pieces without ground-loops can involve magic and luck.

If everything you own is pro standard, the problem is made considerably more simple with balanced connections. Usually pro gear has balanced inputs and outputs. Balanced circuits are less susceptible than unbalanced.circuits to ground loops and other noise sources. It's possible that some ground loop current will find a way into, even, the best designed equipment. Some precautions with your interconnecting wiring will help prevent that from happening.

You will need to be prepared to disconnect occasional ground connections in the signal path to break up loops between the signal ground to the power ground. Ground exists at both ends of the cable, from the case grounds, so completing the shield ground connection is often unnecessary. I color code my XLR cables (red for single-ended ground, green for connected ground) to make ground-loop troubleshooting a little easier. I have found that I use about four times as many single-ended ground cables as I do the connected versions.

Direct boxes usually have ground-lift switches. If those switches are going to do you any good, you need to wire your XLR connectors so that pin 1 or the XLR case is not connected at one end of your cables. Again, pin 1 is already connected to circuit and case ground at the mixer or effect, so a continuous ground connection is not necessary for shielding ground.

Unbalanced connections can also benefit from wiring tactics. You can create a simulated balanced connection with a couple of lightweight tricks. If you're connecting a balanced unit to an unbalanced unit, adding a 50-100 ohm resistor to the ground terminal at the unbalanced end of the cable may provide enough resistance to the ground path to divert the ground loop current away from the connection. If the connection is unbalanced to unbalanced, the same 50-100 ohm resistor on one of the pin 3 connections may do the same trick. In both cases, use two-conductor shielded cable (like microphone cable) and only terminate the shield wire on one end of the cable.

Usually, if you are careful with the power grounding and the interconnecting shield grounding, you can avoid ground loops in your equipment. But the more equipment you have, the more likely it is that you're going to have problems. If nothing else will make you consider all-in-one multiple-effect boxes, this might. In fact, I think that analog connection hassles ought to make a lot of home studio owners think really hard about going the full digital route.

There are two tactics commonly used for case grounding in equipment racks. A fairly uncommon tactic is to run a buss-bar (a very large gauge wire) connected to all of the equipment cases. The other tactic is to isolate the equipment cases. This is the approach I've found to be most useful. (I have a parts bin full of #10 nylon shoulder washers for this purpose.) Put a nylon washer on each side of the rack ear and add a metal flatwasher on the outside, so the screw doesn't tear up the nylon insulator. Four screws, four flat washers, and 8 nylon washers per piece of mounted equipment. This makes the power cord ground the only source of power grounding for each piece of gear, resulting in an effective "star grounding" system.

One other item to watch out for is case to case contact. In an effort to pack 2" of circuitry into a 1 3/4" box, some manufacturers chew up all of the allowed vertical space in the rack, allowing contact with the equipment above and/or below their gear. If you have to, put some kind of insulating shield between pieces to prevent case-to-case contact in the rack.


Preventing Dead Babies


OK, now we're all wired up and screwed down. There are just a few other tips I'd offer before you get serious about using your equipment. The primary failure mode of almost all electronic equipment is "infant mortality." That usually means the first 100 hours of use. The fastest way to get that period behind you is to continuously run your gear for a few days. For most equipment, this simply means turn it on and leave it on. Doing this offers you two advantages. You eliminate a large percentage of possible failures that might, otherwise, happen at the worst possible time and you make fairly certain that, if a failure occurs, it's going to happen inside the equipment's warranty period. This kind of burn-in used to be done by manufacturers, but, in the interests of reducing manufacturing expenses and process time, now they leave it up to you. So do it. It will only cost you a few watt/hours of power and it might save you a lot of hassle when you can least afford to be hassled.

With electro-mechanical equipment, just running the gear probably isn't going to provide a decent burn-in. Switches need to be switched. Motors need to spin. Doors need to open. And so on. It won't hurt your new tape machine to cycle through a tape or two. Some equipment weak-knees will show up just by powering the equipment off and on a few times. Power amplifiers, for example, have large supply surges on power-up. Sometimes all it takes to overstress a poor connection or a weak component is a couple of cycles through the power-on cycle.

Keeping the equipment's function in mind, it's in your best interest to start using the equipment you buy as soon as possible. Don't pack your room full of gear that you intend to use . . . someday. Unless repairs costs are no problem for you, buy it when you need it and use it as soon as you buy it.


Be Your Own Power Company


This next precaution may complicate the already complicated ground loop planning we've done. However, if your power line is not as solid as the power company claims, you may want to protect some parts of your system with an Uninterruptible Power Supply (UPS). An UPS will do two very valuable things for critical components in your system: it will provide a constant, filtered source of AC power and it will give you time to safely power down your system during a blackout. If you've been around computers for any length of time, you know that power company brown-outs can wreak havoc with computer systems. A fluctuating power supply can cause random memory bit-flips that will contaminate your data or destabilize your system. The UPS transformer and filter circuitry prevents that from happening. A tripped circuit breaker or lightning strike on a nearby power pole transformer, will flip every bit that isn't already at zero. The UPS battery backup and inverter circuitry provide you with a warning and a reasonable time to save what you're doing and bail out before the battery goes dead.

The only trick in selecting an UPS is determining what size supply you need. The UPS specification sheet will help you with that. First, decide what needs to be protected and write down the volt-amperage (VA, which is roughly the input wattage rating) requirements for each component. Some equipment specs only tell you the input current requirement, not the VA rating. Simply multiply the current draw times 120V for the VA rating. Add up the VA ratings of the equipment you've decided to protect. Compare that total to the chart provided by the UPS manufacturer and decide how much protection you need and can afford.

For example, you have a computer system that requires 250VA, a computer monitor that uses 125VA, and a hard disk recording system that needs another 250VA. That adds up to 625VA required for the system. Looking at the chart for the 600VA UPS, you see that unit will provide you with about 5 minutes of backup power. The 1kVA system will keep you running for 10 minutes during a blackout. You, now, need to calculate how much time you will need to save what you're doing and power your system down during an emergency. You also have to factor in what you can afford, bigger is more expensive.

If, after all that, you still remember how to do music, I think we're ready to turn this stuff on and see what it sounds like.