1

I used to get poor performance from my PI 3 since I had a few heavy load programs running on it. After needing more than 128GB of MicroSD storage I got a USB to SATA cable and hooked up a 500GB SSD and soon realised that I was getting increased performance.

After a little looking research, I realised that my PI was getting extra power from the USB to SATA cable I had hooked up as the power lead of this adapter was plugged into a different USB.

If I supply my PI with extra power through the USB like this will I damage it in the long run?

  • Sounds implausible. How did you verify that the Pi was drawing current in through its USB sockets? – goobering Apr 29 '16 at 9:43
  • Using a USB volt meter on the USB port and you see increased power going through the USB. Also my PI when running a small IDE got bottlenecked to around 750mhz but not it's using all of its processor. – 0xen Apr 29 '16 at 9:49
  • It's a lot more likely that your performance improvements are due to the performance of your storage (microSD versus SSD), and not due to current coming in through the USB sockets. USB host devices (the Pi) absolutely should not draw current from client devices (the hard drive). Imagine what would happen if a mains powered device sent current into my desktop PC, for example. Sounds like a bad fire waiting to happen, no? Would it be possible for you to post a photo of the Pi connected to the voltmeter connected to the hard drive? It might help to explain what you're seeing. – goobering Apr 29 '16 at 9:58
  • You cannot power the Pi3 through USB. You have made so many changes and asked such a vague question noone can answer. – Milliways Apr 29 '16 at 9:58
1

It was entirely possible with the Pi2 (I realise you have a Pi3, but I haven't studied the schematics for that) to power the Pi from a "backpowering" USB hub.

You could power the Pi from a hub and disconnect the u-USB power connector completely. On principle, there is no problem with this except that it bypasses the inbuilt voltage surge protection circuit. Theoretically, if your HDD PSU squirts out a power surge for some reason, your RPi could fry.

As indicated by goobering, most USB voltage monitoring devices display the voltage and current being supplied BY the USB port. It's entirely possible that you observed an increase because your external USB PSU was allowing more current to be drawn through the port than the RPi could otherwise supply.

In answer to your question: it's not a matter of "the long run". It only takes microseconds to fry a microprocessor and you have bypassed the protection, so you are at risk.

2

You said you used a USB volt meter to monitor the connection between the Pi and the hard drive. I'm presuming that you're using something along the lines of the Charger Doctor from Adafruit. These devices measure the current flowing from a host device (in this case the Pi) to a client device (the hard drive).

Any values you read indicated the Pi supplying the hard drive, not the other way round.

A performance increase is plausible if you were using the SSD to store data used by whatever application you were running, rather than the microSD card. SSDs are much more sophisticated than microSD cards and allow far greater transfer speeds.

  • 1
    To clarify this, it is not just that an SSD is a more complex device than an SD card, but that the SD card controller that is part of the Pi's SoC is limited to 25 MB/s; you could see rates more than double that through the USB. – goldilocks Apr 29 '16 at 13:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.