I've got a pair of Pi2Bs using externally-powered USB mechanical hard disks.

Both are self-hosting FreeBSD-current with varying levels of success and one variation might be the power supplies used with the powered hubs to run the disks.

I've placed a readme and an image of the power supply voltage trace taken directly from the powered USB hub used to connect the disks to the Pis at

http://www.zefox.net/~fbsd/rpi2/disktrouble/

The image labeled "badpower.png" is from the machine with disk corruption problems, that labeled "goodpower.png" is from a second Pi2B that has so
far exhibited no trouble of that sort.

Never having made such a measurement before I don't have a very clear notion
of that's acceptable or normal. Neither trace is magnificent, in fact they're both rather noisy, with the troublesome machine slightly worse.

If anybody's made a similar measurement and can offer an opinion as to what's normal or at least tolerable I'd be grateful for a reply.

Thanks for reading,

bob prohaska


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

bp@www.zefox.net wrote:

I've got a pair of Pi2Bs using externally-powered USB mechanical hard disks.

Both are self-hosting FreeBSD-current with varying levels of success and one variation might be the power supplies used with the powered hubs to run the disks.

I've placed a readme and an image of the power supply voltage trace taken directly from the powered USB hub used to connect the disks to the Pis at

http://www.zefox.net/~fbsd/rpi2/disktrouble/

The image labeled "badpower.png" is from the machine with disk corruption problems, that labeled "goodpower.png" is from a second Pi2B that has so
far exhibited no trouble of that sort.

Never having made such a measurement before I don't have a very clear notion of that's acceptable or normal. Neither trace is magnificent, in fact they're both rather noisy, with the troublesome machine slightly worse.

The spec is I think 5Vñ5%. So that's 4.75V to 5.25V. 'goodpower' has spikes
of 200mV and 'badpower' of 50mV, so goodpower is actually worse. But they're both AC coupled waveforms so there's no clue what the DC voltage is. eg badpower could be 4.78V -50mV and goodpower 5.10V -200mV and the first
would be out of spec and the second not. Can you measure the voltage with a meter?

Theo

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Theo <theom+news@chiark.greenend.org.uk> wrote:

bp@www.zefox.net wrote:

I've got a pair of Pi2Bs using externally-powered USB mechanical hard disks. >>
Both are self-hosting FreeBSD-current with varying levels of success and one >> variation might be the power supplies used with the powered hubs to run the >> disks.

I've placed a readme and an image of the power supply voltage trace taken
directly from the powered USB hub used to connect the disks to the Pis at >>
http://www.zefox.net/~fbsd/rpi2/disktrouble/

The image labeled "badpower.png" is from the machine with disk corruption
problems, that labeled "goodpower.png" is from a second Pi2B that has so
far exhibited no trouble of that sort.

Never having made such a measurement before I don't have a very clear notion >> of that's acceptable or normal. Neither trace is magnificent, in fact they're
both rather noisy, with the troublesome machine slightly worse.

The spec is I think 5Vñ5%. So that's 4.75V to 5.25V. 'goodpower' has spikes of 200mV and 'badpower' of 50mV, so goodpower is actually worse. But they're both AC coupled waveforms so there's no clue what the DC voltage is. eg badpower could be 4.78V -50mV and goodpower 5.10V -200mV and the first
would be out of spec and the second not. Can you measure the voltage with a meter?

In the readme:

Goodpower reports 5.09 volts DC level

Badpower reports 5.23 volts DC level


Both numbers came from a voltmeter on each supply.

Thanks for writing!

bob prohaska

Theo

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)