Buying Used – Electronics in Particular
This is the second part of a two-part article about buying used gear. The first part discussed general considerations when buying any type of used gear, and how it is possible to be mislead and to end up buying junk you don’t need, while all the time thinking you’ve bought something useful at a bargain price, and closed with some suggestions and tips on how to bargain for the best possible price.
Now it is time to focus in on the special considerations relating to electronic equipment.
Age and Reliability
Just about everything has a finite life. The older a thing is when you buy it, the less remaining life it has. Even ‘your grandfather’s axe’ has a finite life – you have to keep replacing its head and handle every decade or two.
To make things more complex, ‘age’ is measured two different ways. The first aspect of age is the simple passing of time, no matter if the unit is being used or not. Some components age even when not being used. Rubber cracks, plastics lose their plasticizers, springs lose their tension, seals leak, wet things dry out (and dry things get wet) and so on. Rust never sleeps, right?
The second aspect of age is the number of hours the unit has actually been powered on and in use. Just about everything has a finite total number of hours of life – sure, sometimes that number might be very high, but it is still a number, and every minute anything is turned on, you’re steadily rolling the dice as the minute hand moves, each time hoping you don’t get an unlucky result and your equipment randomly failing.
There is a third element of age as well – the type of operation and environment. Something that has been run ‘hard’ at 110% of rated power will age massively more quickly than something that has been run carefully/gently at 90% of rated power. Something with a ‘dirty’ power supply is going to be stressed more than something with a nice clean stable power supply. Something that has been in a very hot environment will age very much more quickly than something that has been kept cool. Heat is the universal enemy of all electronic circuitry, and as a rule of thumb, for every ten degrees hotter that something is operated at, you are halving its life. Run it 20 degrees hotter, and you’ve reduced its life to one-quarter, and so on.
When you see something, you have no way of guessing about how the unit has been used in the past, and a product that is about to fail seldom gives you any sort of indication that is about to happen.
The simple passing of time affects electronics as much as it does anything else. A surprising but key aging element is corrosion – the oxidation of the leads on electronic components and the increasing difficulty of soldering them effectively to other devices, and the slow failure of existing soldered joins. The key factors here are not to store components in regular plastic bags, and to keep them cool and dry.
Regular plastic bags are thought to ‘outgas’ and ‘leach’ out chemicals that accelerate corrosion and harm the materials stored inside them. Barrier bags (a fancy way of saying ‘nylon’) are okay, and archival plastics probably are too, but regular PE type materials – best to avoid them. This is a good article about component storage.
Of course, moisture is an enemy and corrosion accelerant, so keep things as dry and humidity as low as possible. A related thing – bad news if you (or previous owners) live near the sea, with the higher salt levels brought in from sea spray.
Some things have obvious life-limiting factors as part of their design, and may be impractical to maintain. Other things have very long lives and are practical to maintain.
In the case of electronics, some items will wear out and fail semi-randomly, some will do so moderately predictably, and some will last almost forever. And it is all overlaid with an element of random chance. You’ve no way of knowing if your particular piece of gear will end up with a long life or a short life. The only thing you do know is that the more it is used, the closer it is getting to its ultimate failure.
Semi-conductor and Component Aging
If you have really old gear with tubes inside, then you need to plan for occasional replacements of the tubes. Just like incandescent light bulbs burn out, so too do tubes. But, unlike a lightbulb which burns more or less the same every day until suddenly failing, with a tube, it isn’t only just having it fail by sudden total burning out of the filament. Tubes also have several other factors that influence their longevity, and their performance steadily declines, every hour they are being used. The more they are used, the less remaining life they have, and the more poorly they will perform.
Most tubes will have a rated life stated on their specification sheet, and you’ll see this life expectancy can vary wildly. One make/model of tube might be rated for 1,000 hours, another for 10,000 hours. And, in all these cases, please be sure to understand what this rating means. A rating of 1,000 hours doesn’t mean ‘all these tubes will work perfectly for 1,000 hours and then fail some time after’. It means ‘some tubes will fail quickly, some will fail slowly, and on average, you’ll get about 1,000 hours overall’. That is a very different scenario, isn’t it – some tubes will start failing immediately.
Back in the days of early computers that used thousands of vacuum tubes for their logic, people were employed as a full-time job just going through the racks and non-stop replacing tubes.
There’s another factor with tubes, too. Unfortunately, ‘new old stock’ tubes might already have some percentage of their working life used up, just by sitting on the shelf (for example, if the vacuum seal is less than perfect). If you are able to test your tubes on a tube tester, that will show you where on the spectrum each tube lies as between brand new and into the failure zone.
The solid-state components – the transistors and integrated circuits – will fail semi-randomly. There is an initial period that is sometimes termed ‘infant mortality’ where new components might fail, then there’s a long period of reliability where failures are random and rare, then beyond that, the failure rate starts to inch up again. The steady declining performance of vacuum tubes is not as pronounced with solid state devices, although it is present, but to a much more subtle degree.
Interestingly, and sadly, the newer and more modern the solid state devices, the faster they will fail. Increasing miniaturization makes even individual atoms and molecules significant elements in an integrated circuit, and gradual effects such as the migration of materials across substrate barriers are much more significant in miniaturized components than in earlier items that were hundreds of times larger. The tolerance range between operating voltages and maximum voltages is also greatly reduced in modern semiconductors.
There’s not a lot you can do about that, though, but just because a neighbor has a 50-year-old radio working perfectly absolutely does not guarantee that the new radio you buy tomorrow will outlast it. Similarly, as you read around the internet and prudently do your research, be sure to understand that the people who confidently point to very long life with their electronics are probably, and by definition, talking about old gear. When you think about it, it is impossible to say ‘my radio has worked perfectly for 50 years’ when you are talking about a radio you bought just last year, isn’t it! But if you search out some of the technical papers about semi-conductor longevity, you’ll see the ugly truth that the smaller the componentry, the more delicate it is and the shorter its life. This is also a factor in terms of EMP vulnerability – an electronic component designed for 2 volt logic and within a thousandth of an inch of the next component on the same chip is going to be fried by an induced 5 volt power surge which can also pass across the tiny gap between it and the next component, whereas an old transistor, 100 times larger in size, and working off 12 volts, will laugh at a 5V surge and not even notice it.
Resistors have a long life, particularly if they are newer metal film rather than older carbon film. Older style carbon film resistors are cylindrical in shape and usually have a dark brown body; newer metal film resistors are a lighter brown color and typically are ‘bar bell’ shaped – cylindrical but of larger diameter at each end.
And inductors – coils of wire – last about as close to ‘for ever’ as anything ever does.
The ‘Achilles Heel’ of most electronics are the electrolytic capacitors. These can dry out (or leak) and generally have a life somewhere between 10 – 40 years. Some people replace all their electrolytic capacitors every 10 – 20 years, whether they need to be replaced or not, particularly because if a capacitor fails, it can cause cascading problems throughout the circuit including the failure of power transistors and other costly/hard to replace components.
Batteries – Both Obvious and Obscured
Perhaps the most regularly replaced item in any piece of electronics are the batteries. And, in considering this, don’t forget that many electronic items have both the obvious/main battery, but possibly also some other obscured/hidden batteries that are used to do things such as storing the device’s settings in memory so that it doesn’t reset every time you turn it off. These tiny batteries are invariably overlooked until they fail, so if you are buying any used equipment, it is a good idea to ascertain if they have such secondary batteries and then to replace them if they exist.
A regular rechargeable battery has a very variable life, depending on the conditions in which it has been used, stored, and recharged. Unless you have special test equipment, and also know the expected values for an optimum conditional representative unit, you can’t really check to see what state of health the batteries are – better to play it safe and treat all batteries as nearly at the end of their lives.
On the other hand, the good news about most rechargeable batteries is that their ‘life’ isn’t defined by a sudden total failure. Instead, it is just a steady decline, with each subsequent recharge storing less power than the one before, until you get to a point where it is no longer convenient to keep recharging a battery that takes two hours to recharge and then runs for 30 minutes (or whatever). So in a Level 3 situation, you’d keep using batteries long past the point you’d swap them.
On the other hand, rechargeable batteries can indeed also fail completely, and not hold any charge at all. All the more reason to replace all rechargeable batteries (or at least to buy a spare set and store them until needed) when you buy used electronic equipment.
So maybe something fails in the electronic item you purchased. Can you trouble-shoot to find the failure, and can you then fix it? Years ago, the answer was ‘yes’ – large-scale discrete components, individually soldered to spacious printed circuit boards, were easy to replace, and when something failed, you only needed to replace that one thing.
But now, with tiny SMD components less than half the size of a grain of rice, the ability to troubleshoot and replace is much more complicated than it used to be, and requires a steady and skilled hand and soldering iron. ICs might contain millions of transistors, so a single transistor failure might take the entire IC out of service. That is okay if you have spare ICs, but pretty soon, you’ll find you are spending more to build an inventory of spare parts than you are on the gear to start with. Another difference is that whereas before, it was practical to have a spare parts inventory of all the different resistor, inductor and capacitor values, plus a smattering of the most common diodes and transistors, these days, most products have their own unique integrated circuits so you must have spares for each item rather than a common inventory of spares for everything.
This is also a reason to standardize. Rather than having a dozen different walkie-talkie radios, for example, buy twelve (or thirteen or more) the same. If one fails, you then use it as spare parts for the others. This also makes it enormously easier from an operational perspective if everyone has the same gear. So that ‘really nice’ and inexpensive piece of gear you’re drooling over has to be considered in the context of the practicality of requiring people to learn how to use a different interface, and the need to keep a separate inventory of spares for it. Better to use the money to buy another one of the standard model units you already have.
The Future Lifespan You Need
An interesting consideration is to decide how long you need something to last in order to have received fair value from it. Of course, ideally, everything would last for ever, and equally ideally, we’ll never experience an event that forces us to resort to our emergency equipment and supplies. The world would just continue on, the same as it has been until now.
But both those scenarios are sadly unrealistic!
Should we consider it as good news or bad news that many of us have been prepping, to some degree or another, for more than two decades? Is your glass half-full or half-empty? Are you upset if at the end of a year, you’ve not filed any claims on your car insurance policy? Do ten or twenty years with no need to activate your preps make next year riskier or safer?
The point behind these (largely unanswerable) questions is that much of the stuff we’ve bought to have for a possible future WTSHTF situation are things we’ve now owned for ten years or longer. If we bought something that was already 10 – 20 years old when it was purchased, it would therefore of course be up to 30 years old now, and at the point where it would be prudent to think about replacing (or at least giving it a thorough overhaul and replacing many of its components) – or possibly supplementing it with a new set too.
In other words, if you are buying something today that you need to use tomorrow and which will have paid for itself in terms of value and use within a year or so, it doesn’t really matter if its life is two years or twenty years (although of course twenty years would be nicer!). Your main focus is on an item that will work reliably today and for a long enough period so that it has paid for itself by the time it fails and you replace it.
But if you’re buying something that you won’t even touch for a decade or two, you need something that will be able to sit in storage for that period of time, and then be activated and work for another decade or two or however long you expect to need to rely on it.
This issue – that if you buy something old today, it might be very old by the time you finally get to use it – is another reason to avoid buying old stuff unless you know how to maintain and repair it and have an inventory of the necessary spares to do exactly that.
What We Do
We don’t buy anything essential that was made before 2000 (mainly because it is a nice round number to use as a cut-off) and we generally prefer to buy things that are less than ten years old, unless there is a special reason to choose something older.
And whenever we can cost-justify it (which is most of the time – not because we have lots of money, but simply because the cost/value/benefit equation supports it) we prefer to buy brand new gear, run it for a few months to check for any ‘infant mortality’ and then put it in extended storage until needed.
Part One of This Article
This is the second part of a two-part article. If you’ve not already done so, you might like to also read the first part, which talks about buying used/second-hand gear in more general terms.