Is 60mV a normal voltage loss for a power supply cable ?

[hibol]

Hello,
I measured voltage I have on my Pi, because I wasn't sure if it was enough.
The result was about 4.79V.
Then I asked myself what was the loss in the USB / micro USB cable (80 cm), and I found it is 60mV.

Is it normal or is the cable bad ?

Thanks for the support

[Mortimer]

60mV is not a lot, I can quite believe that amount of loss over a supply cable.

[computerpie]

So if you are losing 60mV accross the wire and get 4.79v at the Raspi then your supply is only putting out 4.85V. That sounds low to start with.
The cable resistance would be E=I*R or R=E/I or R=.06/.7 (assuming Raspi drawing 700mA ) So R= .086 Ohms. Not bad for an 80 cm cable.

If you are planning on connecting anything to the Raspi I'd look into a better power supply.

[hibol]

Thank you for the quick answers.
I tried two different Apple power supplies, one rated 5V 1A, and the other 5V 2.1A.

Actually, my Pi is working, I just have problems with my wifi network, and I don't know if it could be better with a higher voltage.

As my voltmeter is quite old, I will try the same measurements with another one as soon as I can.
And for the moment, I have no other power supply to test.

[computerpie]

Take your measurements across TP1 and TP2. You want that to be as close to 5v as possible with everything plugged in (wifi dongle, kb, ms etc... . The closer to 5 v the better. The 2 A supply should be ok, but if it's measuring 4.79 v at TP1 and TP2 I'd say it's marginal.

[Grumpy Mike]

I would check your meter because the Apple power supply are normally much closer to 5V than that.

[JeremyF]

I would check your meter because the Apple power supply are normally much closer to 5V than that.

Apple PSUs are good but they don't come with a MicroUSB cable and most cheap cables online quite frankly are horrible.

[hibol]

The cable I use is a Samsung one (provided with a smartphone), and the 4.79V is measured between TP1 and TP2.
I will tell you tonight the results with another voltmeter.
It's good to hear that Apple's PSUs are good as I only have those.

[hibol]

So I tried with another voltmeter, and I got 4.89V. Results are similar wwith the two PSUs...
Still not so good.
It is the same if I use a powered USB hub or no hub at all...
And I have no other power supply to try

[mahjongg]

Measure the resistance of the polyfuse with the multimeter, it should be a very low voltage, sub one ohm. To measure that set the meter to the lowest ohm setting (if it has multiple), then measure what you see if you short the testpins together, it might be something like 0.3 Ohm. Write that down, now measure between the pins of F3 (on the back near the power input). Obviously with nothing connected to the PI. Now you might measure something like say 0.7 Ohm, subtract the 0.3 Ohm, and you end up with an approximation of the resistance of F3 (0.4 Ohm).

Post it here.
Now turn on, and boot the PI, with a keyboard and mouse attached, but nothing else.
Set the multimeter to the lowest DC voltage setting.
Now measure the voltage across F3 (same points as before when measuring its resistance).
Post the result here too.

This will give us both an indication if the fuse is OK, but also an indication of whether the PI is drawing more current than normal.

You can also connect HDMI and see if the voltage across F3 increases, this is an indication that the HDMI is using current, it shouldn't (much).

In the end disconnect all, and measure again the resistance of F3, as described above, so see if it has changed, it should not.

[hibol]

Thank you for your advice, mahjongg.
The resistance of F3 is 0.3 Ohm

With a keyboard and a mouse, the voltage on F3 is 85.8mV.
If I plug HDMI too, the voltage is 89.9V (but sometimes reaches 100mV)
Then if I measure the resistance of F3 again I get 0.2 Ohm.

[mahjongg]

well, the resistance of the fuse seems to be OK, nothing out of the ordinary, and 85mV (0.085V) loss is also not really worrisome

Im assuming the second reading (0.2 Ohm) is the right one, if the current would have been very big the fuse might have heated up, and increased its resistance, but this has clearly not happened, as the calculated current I=U/R is just 430mA, which is more or less normal.

Together with the 60mV cable loss, the total loss is 145mV (0.145 V), when subtracted from an "ideal" 5.000 Volt would result in 4.855 Volt, which is good enough.

[hibol]

Ok, thank you
So it is normal that I don't have 5V between TP1 and TP2, and it doesn't mean my power supply is bad, is it true ?

[mahjongg]

electronic devices never need a mathematically precise value of their operating voltage or none of them would work.

My experience derived by reading these fora is that a lower bound of 4.8 Volt and an upper bound of 5.2V is mostly safe, but some mention the PI working with a lower value, your mileage may vary.

[bredman]

The RPi CPU is actually very tolerant of low voltages. You are more likely to have problems with USB devices and Ethernet if you drop below 4.85 volts.

[hibol]

Ok. I heard there are a lot of problems with USB devices, and I am also experiencing them (like the system becoming slower when using low speed and high speed USB devices at the same time).
So I think it doesn't help me to make my wifi dongle work well...
I hope that future updates will improve this.

Anyway, thanks for the help.
Working or not really, it is a lot fo fun to play with the Pi

[timothy3592]

I'm surprised nobody suggested using a powered hub, it seems to be the standard default response to any problem with high current draw USB devices (like wifi dongles). I hate to break up a deep and intense conversation over internal resistances and such, but their is a pair of polyfuses that keep things like wifi dongles and external hard drives from powering themselves comfortably.

[hibol]

The results are the same if I use a powered USB hub or not.
That's my main problem. And whatever configuration I try, I am not able to have a stable wifi connection.
That's why I started to wonder about power supply, voltage and so one.

[mahjongg]

Well, I know that there is very much work being done on the USB software, and very recently we have seen a version that shows very major improvements in stability and such. This is still not in the "pipeline" for distributions, as work is still ongoing, but I hope the USB situation will improve a lot soon.

[hibol]

I hope too
That's what I install updates each time I am able to connect to a network, and this event is random
It would be nice if you can tell when you hear about a major update which might help my Pi to work !
Or is there a website which describes every raspbian updates (I mean in a comprehensible language for someone who is not a developer) ?

[mahjongg]

Or is there a website which describes every raspbian updates (I mean in a comprehensible language for someone who is not a developer) ?

The former yeah, but I'm afraid the latter no.
In any case these fixes should show up in a firmware update, or a new distro update sooner or later. I'm afraid that if you want the fixes before that you must be kind of a hacker, but don't fret I think most distro creators will implement these upgrades rather fast. even when its not the latest improvement that is being worked on. We do know there are many more possible!

In the meantime you could read the "USB, the elephant in our room" thread, which discusses many of these improvements.