OT or not? Buying a USB caddy for 2.5" drive

For the new laptop, the new laptop should suffice.

The Intel application note on USB, says that the designer should

*not* cut off power at the USB.org limit. That's why a USB2 port with 500mA limit, the fuse is 1.1 amperes. On a USB3 port, the 900ma limit has a 2.0 ampere fuse or so. These are Polyfuses, used on desktop computers.

Laptops offer a different thermal environment, and the limited air circulation makes Polyfuses problematic. A silicon fuse, an eight pin DIP with MOSFETs in it, is used to limit current to two plugs. And the tradition (implementation error), is a much more precise cutoff at 500mA or 900mA.

Whether the laptop designers have learned any lessons by the year 2022, I don't know.

But on the surface of it, your new laptop should be able to source 900mA on a USB3 port.

Paul

Reply to
Paul
Loading thread data ...

I've measured my Samsung SSDs (850, 860 kind of thing), and they generally range around 2.5W . It's hard to tell whether there is any "spiking" going on. I'd need a scope or other faster instrumentation, to be sure.

With regard to SATA SSDs, you have to watch out for older Kingston Sandforce based drives. Those use hardware-based data compression to make the flash go further, and the spike current on those is 7W on a write. Sandforce has pretty well disappeared as near as I can determine, as I don't see the chipset name mentioned any more in adverts.

For 2.5" spinning drives, the 5V @ 1 ampere requirement only lasts for 10 seconds, and then the power requirement drops back to more reasonable values. If the drive parks itself, then another ten second spinup may be the result later on.

The current flow on spinning drives, has a degree of "regulation" going on. The motor is running in a kind of current limited mode. It's a bit of an indirect current limit though. Like on the 3.5" drives, if the manufacturer wanted, they could turn a knob on the drive, and the motor would draw an extra ampere and would get up to speed in about four seconds. On 3.5" drives, there is actually a "policy" in place - large capacity drives and small capacity drives, have their motors set differently. The spinup time is not a constant, on 3.5" drives. The slowest ever, might have been a Hitachi at around 28 seconds. The BIOS timeout is set at 35 seconds.

The datasheet in the past, used to have a scope waveform of current draw, to demonstrate to the engineer reading it, that the design was current limiting. The motors use three phase drive, the motor controller chip synthesizes a waveform, and also enforces an acceleration profile (by some means). And the ripple-coated waveform at a roughly fixed current, is the result.

If the spindle is "stuck", the firmware on the drive reprograms the motor controller for some sort of "noisy" modulation that makes a peculiar sound. Whether this is a side effect of current limiting, or is a purpose-built modulation pattern, I can't tell. FDB motors, when the lube is gone, they seize up quite nicely. The motor will never spin again after that.

Paul

Reply to
Paul

Yes, it powers a 2.5in 5400rpm drive from a single USB3 cable. For

3.5in drives you need something with a power supply for the 12V rail

You can tell rsync just to copy local/ to remote/ and it copies everything. If you then delete files in local/ then they will remain in remote/

You can tell rsync to delete as well. You copy local/ to remote/, job done. Then you delete files in local/ and next time you run rsync it will delete the files from remote/ you deleted from local/ and sync the remaining files.

Reply to
mm0fmf

In message <tklbg3$r8ed$ snipped-for-privacy@dont-email.me, at 11:25:48 on Fri, 11 Nov

2022, NY snipped-for-privacy@privacy.>>> Op 09/11/2022 om 17:08 schreef Roland Perry:

Almost. Like that, but powered via the USB cable not a wallwart. And with the circuit board encapsulated with just the connectors (often two or three different pinouts) exposed.

Reply to
Roland Perry

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.