How do you test a USB cable for suitability?

[piglet]

Hi All,

I have a large number of USB cables used for pi's and charging things. I bought a nice powerful 5v supply and still got the lightening bolt showing on my pi screen from time to time until I replaced the leads with branded expensive ones.

However I have some other expensive ones which appear to be utterly useless.

What's the best way to test a usb cable? I want to go through the lot and chuck out all the ones with stuff little better than wet string on the inside. How do I test them? I have a multimeter but I can't see any measurable resistance difference between a known good and a known bad one....

Go on...how do I do it?

[karrika]

One way could be to measure the voltage at Pi3 5V while it is running a performance test suite.
At least this test would be relevant to what you are interested in.

[jahboater]

Do any of the cables have writing on them? AWGnn perhaps.
The official Raspberry Pi3 PSU cable is 18 AWG for example.
Perhaps discard the ones with larger numbers (thinner wires) or those with no spec.

[Burngate]

Given that the potential current could be as much as 2A, and with 5.0v going into the cable and at least 4.6v coming out, we're looking at 0.2 ohms maximum.
That's 0.1 ohms for each leg of the cable, including the connections at each end.
Few multimeters can measure resistance that low, and still fewer can do it reliably.

A better way would be to put a known voltage in and a known low resistance - about 2 ohms - on the output, and measure the voltage across the resistance.
Which is what karrika's method does.

[B.Goode]

If the Real Life test is the behaviour of the RPi, you could power the board with each cable in turn - driven from a robust known good power supply - run for a short period with any 'stress test' software applications, then use the

Code: Select all

vcgencmd get_throttled

command to see if any under-voltage events have been detected.

See viewtopic.php?f=63&t=147781&start=50#p972790 for details of how to interpret the output.

[Milliways]

Testing a µUSB cable is a non-trivial task.

The first challenge is finding a µUSB socket to terminate the cable, but assuming you can find one solder a 5Ω 5W (or greater) resistor to the power pins.

Apply 5V to the cable, and measure the voltage across the resistor. (This assumes you have a quality 5V supply, if not you need to measure the supply voltage ON LOAD, again a suitable adapter is required.)

ALTERNATIVELY, if you can source a µUSB plug, just make a cable with proper wire, which is what I do for my testing.

[rpdom]

The first challenge is finding a µUSB socket to terminate the cable, but assuming you can find one solder a 5Ω 5W (or greater) resistor to the power pins.

For testing I use microUSB breakout boards. They can easily be found in various forms on ebay and similar sites for not a lot of money.

ALTERNATIVELY, if you can source a µUSB plug, just make a cable with proper wire, which is what I do for my testing.

That's also what I do for powering some of my systems

[hippy]

One way of determining a low resistance value is to measure the time it takes to charge a capacitor through it. A micro or even a Pi can do that and probably without needing a lot of external components.

A simple 0V-to-Vusb shorting socket would do the job at the 'far end', which is easy to construct, putting both 0V and Vusb lines in series. The control and measuring circuit being at the 'near end'.

[HawaiianPi]

Many USB cables have wire gauge markings on them.

Look for markings similar to this:

And toss anything with number higher than 24 for the power lines (higher number = thinner wires).

Also read this: viewtopic.php?f=63&t=165393

[piglet]

The first challenge is finding a µUSB socket to terminate the cable, but assuming you can find one solder a 5Ω 5W (or greater) resistor to the power pins.

Apply 5V to the cable, and measure the voltage across the resistor.)

Ok - so I went home last night via Maplins and have the closest I could find: 5.6Ω 10W wirewound, which together with the 5.31v supply I'll use should draw about 0.95A - which is pretty close to the 1A of 5Ω at 5V

I also asked for a 4.7Ω (4R7) one...which when I got home I discovered was a 4.7KΩ (4K7)...not so useful...

1) Solder the resistor across the micro-usb socket "+" & ground (I have a break-out board I can do this on)
2) Plug USB lead into PSU
2) Measure voltage across the resistor under load.

Results from initial lunchtime test using the 5.3v (unloaded) 3A Liteon branded supply:

1) Resistor starts to get warm fairly fast. Not surprising as it's trying to dissipate quite a lot of power
2) Voltage across the load on cheap and nasty lead is identical to when using my nice thick Volutz cables.

Hmmm. Will try again with a range of cables later.

I wonder if the traces on the micro-usb break-out board could be the limiting factor here. They look pretty small.