No, I’ve not sold out to the ‘dark side’ and I’m not about to suggest you pay ten-times more than normal pricing for a/v and speaker cables that offer nothing more than regular cables, other than an overdose of manufacturer’s hype.
But during my recent very detailed testing of portable power packs, I came to realize something that has been lurking, unexplored, in the back of my mind for some time now : Some USB cables are better than others. And some are very very bad indeed.
This is simultaneously surprising and perhaps not quite so surprising. Some hardware companies (hello, Apple) go out of their way to make it ‘difficult’ to connect their devices to other devices, and/or to power supplies. What is described as ‘quality control’ is more likely to be ‘profit control’, enabling such companies to charge many times more for their cables than after-market accessory suppliers would otherwise charge.
In theory, a USB cable is a tremendously simple thing. In its simplest form (the USB version 2.0 specification), it has four wires and a shield. The newer version 3.0 and USB-C cables are slightly more sophisticated, and then additional magic sometimes appears in the form of some type of identifying electronics being added to the cable, basically to tell a connected device that it is speaking to an ‘expensive/official’ cable rather than to a generic cable. Plus, again in Apple’s case, instead of using the industry standard USB type connector, in any of its several different forms, there is a unique connector shape that Apple can therefore license and make more money from, even if someone else is making/selling the cable that has the connector on it.
It is difficult to test cables for their adequacy in carrying data because there are potentially other variables also impacting on data transfer rates. In addition, due to the nature of digital signals, often there is a simple yes/no transition where cables change from working to totally failing to work at all with data transfers.
It is easier to test cables for their ability to carry power for the purpose of recharging a portable device. So that’s what we did. Note that because the power is transported on separate wires within the complete cable bundle, reporting on power carrying capabilities does not imply any corresponding quality of data transfer abilities.
What Makes a Cable Good or Bad
When it comes to carrying data, the key issues are the amount of shielding on the data wires in the cable, and their twist rate. The wire size also has an impact, but not in the sense of its ability to carry current, but more to do with how the high frequency signals travel along the wire.
These issues – wire size, twist, and shielding – give rise to varying degrees of capacitance, resistance, and inductance per foot of cable length, all of which degrade the digital signal. Sooner or later, the degradation of signal gets to the point where the cable ceases to be able to effectively carry data.
There are some rules of thumb about maximum cable length for carrying digital signals that embody huge assumptions about cable quality, and also about the nature of the signal being injected at one end of the cable and the ability of the device at the other end to receive the signal. In general terms, the USB specification says that a cable should be capable of transporting a signal at least 5 m (16.5 ft) when using the USB 2.0 standard, or at least 3 m (10 ft) to conform to the USB 3.0 standard (the faster data transfer rate of USB 3.0 more quickly degrades, hence it requiring a shorter maximum length). But how do you know if a cable meets or exceeds this standard, and even if a claim is made, how do you know if you can trust it? You really can’t and don’t.
There are ‘active cables‘ and repeaters (basically another name for a USB hub) that can be used to extend these nominal ranges further, and depending on the quality of the cable, you might be able to get good data transfer rates even with much longer cables and without repeaters. Most of the time though, we only need a very few feet for our USB connections, and so the maximum length is not an issue.
When it comes to carrying power, to charge your device, the key issues are wire size and (heat) insulation. The larger the wire, the less its resistance for a given length, and with less resistance, more power gets transferred to your device, more quickly, and less heating occurs in the wire itself.
There is no such thing as maximum cable length for power transfer, other than to be aware that the longer the cable, the greater its resistance and so the more power loss that you’ll experience. Shorter is generally better than longer, but as noted below, good cables have such low resistance per foot that it is not a problem to greatly increase their length.
Typically the power wires in a USB cable range in size from about 20AWG to 28AWG. This, like – alas – so many other non-metric measuring units, is a bit of a confusing measure. AWG (American Wire Gauge) measures in a logarithmic fashion, and the measured size is opposite to its number. In other words, the larger the AWG number, the smaller the wire. A difference of three AWG units means the wire has halved (or doubled) in its cross-sectional area, and a difference of ten AWG units means it has increased or decreased ten-fold. This is the same sort of measuring system (ie logarithmic) as is used for sound pressure levels (in Decibels).
So there is a huge difference in size between a 20 AWG and a 28 AWG wire. The smaller the wire, the greater its resistance, and for a 20 AWG solid copper wire, its resistance is 0.01 ohms/foot. At the other end of the typical range of wire sizes, a 28 AWG wire has a resistance of 0.065 ohms/foot. A rule of thumb for current carrying is that 20 AWG can carry up to 11 amps of current, 28 AWG can carry up to 1.4 amps of current. You’re probably never going to want to run 11 amps of current through a USB cable, but modern fast chargers can often reach 2 amps, maybe even 2.5 or more amps, so the smaller cables (26 and 28 AWG) pose problems right from the get-go.
Here’s a helpful table of wire sizes and their resistance and current carrying capacities.
There are other issues in a cable – the quality of the plugs at each end and the connections between the plug terminals and the wires. This adds additional resistance to the total cable piece, of course. Remember also that a one foot cable has two feet of wire in it – one foot of the ‘positive’ wire and one foot of the ‘negative’ wire, so you need to double the resistance per foot of wire to get resistance per foot of (double wire) cable.
As a rule of thumb, it is fair to say that it is hard to have any sort of cable with much less than maybe 0.1 ohms of internal resistance, and some extra resistance will likely occur where the cable connectors connect with the external devices at each end.
The Harmful Effects of Resistance
What does this resistance mean? It means that your devices charge slower, and some of the power is lost in the cable, rather than transferred to the device itself. A slower charging rate is always a disappointment, but the loss of charging power doesn’t matter when you are simply charging from a mains powered charger. But if you’re using a portable battery recharger, and you’re losing potentially a third of your power to cable resistance, that is a very significant source of power loss.
Say you have a device that is being charged at a rate of 1.5A of power. If there is a 5V source, Ohm’s Law says that the device is the equivalent of a 3.3 ohm resistor. Now say that your cable and connections adds another 1 ohm of resistance. That means that 1/4.3, ie, 23% of your total power is being lost through cable resistance. It also means that instead of charging at 1.5A, it will instead charge at only 1.16A. So for each hour of theoretical charging, you now have to take almost an hour and twenty minutes.
This becomes even more impactful if you have a fast charge capable device – let’s say you have a tablet that can accept up to 2.5A of charge. Adding that bad cable with 1 ohm of resistance will now see 33% of your power lost, and your charging rate drop from 2.5A to 1.67A, in other words, a 50% increase in time to get the same amount of charge into your unit.
Some Real World Cable Tests
We measured a random collection of cables we had in our ‘spare cable box’ to see what amount of resistance they had and to get a sense for if there was much variation between cables or not. Is this really something to be tuned into, or is it a non-issue?
The very best cable we had – a 12″ Nomad cable with good connectors and 20 AWG wire inside, measured just under 0.1 ohms, a number so low that it strained our testing equipment to accurately measure it (see the section at the bottom on how we tested if you are interested in that).
The cable created a 0.165V drop in our 5V 2A test circuit, which implies a resistance of 0.165/2, or just under 0.1 ohms.
Our Nomad cable came with two additional adapters (see illustration above) that could adapt from the Micro-USB connection to either an Apple Lightning or a USB-C connection. We expected that adding another connector into the line would increase the resistance and tested this. It did indeed, but not to any great extent. The Lightning connector increased the resistance from 0.1 to 0.15 ohms, the USB-C connector made a much smaller increase in resistance, and too small to be significant – somewhere lost in the measurement error tolerances.
We then swapped the 1 ft Nomad cable for a longer 5 ft Nomad cable. This showed an increase in resistance of 0.06 ohms. This was not five times greater, although the cable was five times longer, and this is because most of the resistance is in the connectors, not in the lovely thick cable itself. Our conclusion when testing the two Nomad cables is that there is no appreciable penalty/power loss by choosing the longer cable over the shorter cable.
Now, moving on to other cables, we tested an official Apple cable, notionally 1 m (40″) in length. It caused an appreciable voltage drop of 0.46 volts, implying a resistance of about 0.23 ohms. We also had a ‘deluxe’ third party Apple cable, notionally 2 m in length, covered in fancy nylon braid. It created a 0.82V drop, or about 0.41 ohms. That was unsurprising because it had never worked well and often did not allow for data connections between an iPhone and computer. So much for ‘deluxe’ and fancy nylon braid!
We tested a nice ‘flat strip’ USB cable, about 2 ft in length. Although little more than half the length of the official Apple cable, it had a slightly greater 0.47 volt drop, equating to a resistance of 0.24 ohms. A shorter 1 ft flat cable had a 0.27 volt drop, 0.14 ohms.
We tested another nine cables, with voltage drops ranging from a low of 0.27 V (and therefore 0.15 ohms) for a one foot cable to a high of 1.1 V (ie 0.55 ohms) for a two different one meter (40″) cables, with different molded connectors and wire thickness. Plus one ‘ringer’ – a cable that had so much resistance that our test setup failed. Using our other test equipment suggested a resistance of about 6 ohms (our tester could only measure up to about 2.5 ohms.
One cable showed an astonishing fluctuation in voltage, so much so we ended up taking a short video to illustrate this. Clearly a bad cable! All other cables had smooth steady readings, as you’d expect.
We expect this points to either a cable break or a bad connection between the wiring and connectors. Soldered joints and regular joints can deteriorate over time, and repeated flexing can break a cable. It is important to appreciate that cables have finite lives.
In total, we tested two Nomad cables and 13 other cables. The ‘deluxe’ Apple cable was considered a reject, as were two of the regular cables, and a fourth was an outright disaster. That is a high 31% reject rate, particularly because most of the cables were only a couple of months old, and in ‘as new’ condition.
The three unsatisfactory cables were wasting over 20% of the charging power going through them due to substandard quality. That means they charge slower, and if you’re using a portable battery recharger, you’ve got less effective power to transfer to your device.
About the Winning Nomad Cables
We really liked the Nomad cables (see illustration at the top of the article). They had massively less resistance than any of the other cables we tested, and were even better than official Apple cables when used with their Apple adapter. They also claim to be very strong, able to withstand the typical stresses and strains of being tugged at, twisted, bent, and generally used and abused; a claim they back up with a five-year warranty on every cable.
They are not cheap. The two most general purpose cables – either one foot (0.3 m) or five foot (1.5 m) long – with a regular USB connector at one end and a choice of micro USB, USB-C and Apple at the other end, cost $29.95 and $34.95, either on Amazon (with Prime shipping) or their own website. But if you want ultra-reliable cables with lowest resistivity, allowing for fastest charging and the least amount of wasted power, there is no better choice.
Being a bit obsessive (and also having a ridiculous number of devices that always need charging on our travels) we have one of each cable, so no matter where we need to place our charger, we can conveniently have our electronics charged, and we’re wavering on possibly buying some more!
Noting the almost non-existent increase in resistance of the longer 5′ cable, but the much greater flexibility it gives, if you were to only choose one cable, then perhaps the longer cable would be the more universally deployable choice. It takes up very little more space, and weighs 2 oz rather than 0.7 oz, so from a travelling lightly perspective, the difference is minimal.
Summary
Cable quality varies widely from cable to cable. Some cables can result in your charge time increasing by as much as 50% – why spend money to get a special fast charger, and then waste its fast charge benefit by using it with an inappropriate cable.
Not only is your charge time increasing, but much of the amount of charge stored in portable battery rechargers is being wasted through unnecessarily high resistance in poor quality cables. Even good cables can go bad over time – they might suffer a partial (or complete) break, or a connection between the wires and plugs at each end can fail.
Spending $30 – $35 to get a highest quality cable will get you the fastest and most efficient charging for your portable devices. We recommend the Nomad range of high quality cables.
Supplement : How We Tested
It is difficult to accurately measure resistances as low as 0.05 ohms. The smallest unit on the scale of our lovely analog Avometer 8 is 1 ohm.
We also have several digital ohmmeters, and they will show down to 0.1 ohms on their scales. This is very precise, but misleadingly so – it implies more accuracy than we believe to be present. (What is the difference between accuracy and precision – two terms that are often interchangeably misused? Here’s a good explanation.)
If we assume the meter’s accuracy is plus or minus one or two units on the least significant digit displayed, then that means for a cable with a resistance of say 0.2 ohms, we might get a value anywhere between 0.0 to 0.4 ohms. With a variation in cable resistances less than the probable variation in measurements, this was clearly useless other than as an approximate validator of other methodologies.
So we instead created a test system that had a high current 5V power supply followed by a volt and ammeter, followed by a test cable, then a second volt and ammeter, and then a constant current dummy load device. We first ‘calibrated’ the test setup by connecting the two sets of measuring devices directly together without a cable, and established the amount of voltage loss that was unavoidably inherent in the connections and circuitry (a fairly consistently displayed drop of 0.08V), and also confirmed that the two devices were showing close to the identical amount of current flowing (of course the amp flow should be identical, and the minor differences observed simply showed the variation in accuracy of the two ammeters).
We then interposed various cables between the two devices, keeping the current at a constant 2A, and reading the voltages before and after the cable. Ohm’s Law tells us that the resistance of the cable equals the voltage drop divided by the current flow. A relatively high 2A current enabled the voltage drops to be somewhat significant (typically between 0.25 and 1 V on units that displayed in hundredths of a volt and probably accurate to five hundredths or so) making for acceptable calculations with at least some semblance of accuracy.
This allowed for us to get values down to as little as perhaps 0.1 ohms, plus or minus perhaps 0.1 ohms. The testing setup had a maximum upper limit of slightly less than 2.5 ohms – any more resistance than that and it would be impossible to maintain a 2A current. As it happened, while most cables were 0.5 ohms or less, one cable exceeded 2.5 ohms, and when measured with the digital ohmmeter, it appeared to be in the order of 6 ohms. The cable looked undamaged on the outside, but clearly something was very wrong inside.
We also repeated the testing and noted with approval that we had similar results each time we tested the same cable. We kept other variables such as the temperature of the test instruments and the cables within reasonably close tolerances. It might have been possible to have simply measured the voltage across the cable rather than the voltage difference at either end of the cable, but we didn’t have a convenient and very low resistance way of breaking into the circuit before and after the cable to measure across the cable, and we’d have needed to measure both the positive and negative sides separately rather than assume they were the same. So we’d still have cumulative errors, and a more complex test bed setup compared to measuring across both power lines before and after.
Short of investing a five-figure sum on better equipment, we feel the results are acceptable at least to categorize cables as good, average, or bad.
