Dual PSU or bigger PSU?

[Stilez]

Thanks for that, the other pdf didn't show those, though at least on those graphs and if I'm reading them correctly they never go above startup current and also last for much less time.

I'm hampered by inexperience reading the outputs. Even assuming (as I think right) that the marker at the left side is the zero amps level and lead-in, I can't get the vertical scales from the 6 graphs to match the data in the tables very closely.

But I do notice this, which gives me pause. In fig.1 the 6TB drive 5v line is consistently above spinup - but transients way higher (see spike at the start). In fig.2 the 6TB 12v line hits almost spinup draw for either sustained periods, or dense bursts, lasting around 0.2 - 0.4 sec, even from such a brief sample. In fig.5 the 5v line pulls a surprise - a near solid 0.3 sec burst that looks way above even the starting transient (visible at left).

Some forum users such as @Bidule0hm , @cyberjock and @danb35 have measured HDD current in the past on this forum (admittedly focusing mostly on startup and at times not separating 5v from 12v); perhaps if one or more of our electrically skilled gods sees this subtle ping :p, they might interpret the Seagate graphs (p.13-15) for us, match them with the data tables (p.10-12) and suggest what can be gleaned from them regarding 5v and 12v in-use bursts and peaks, and their size and density?

Edit: Oh, I see they are here. Well, 2 anyway! Hi guys :) That was prescient! You arrived while I had this in draft and was trying to research the scope they used for hints on interpretation!

 

[Bidule0hm]

The left marker is the channel marker and also the 0 V (or 0 A if we use a current probe) reference level for this channel. The arrow on the right is the trigger level (set at 240 mA on fig 1 for example), it's not useful to us. At the bottom we can see the settings of the channel(s) and the timebase (for example, on fig 1, we can see the channel is 200 mA/div, DC coupled and BW limited (probably to 20 MHz), and the timebase is 2 s/div. There's also a software filtered trace (F2, in light blue/green) of channel 1 (C1) which seems to be a few kHz low-pass filter).

For us what's useful is the 200 mA/div and the 2 s/div infos and the filtered trace (because the original C1 trace is very noisy and the really high frequency spikes are not to be used in the current consumption estimation as they are smoothed by the output filters of the PSU so the rest of the PSU never see them) so just ignore the blue/violet trace.

From what I see I'm not 100 % sure the 5 and 12 V captures are aligned temporally speaking and I'm pretty sure they've not been captured at the same time but one after the other (so the drive may have done things a bit differently after the spin-up) so we don't want to try to correlate the 5 and 12 V traces (excepted for the spin-up part maybe).

So I'll interpret the fig 1 and 2 (the others are pretty much the same anyway): I guess the first big spike after the spin-up on the 12 V trace (in the middle of the 6th division, about 1.5 A) is the heads unparking and the cluster of smaller very short spikes (start of the 8th div, about 1.75 A) is the heads seeking to find the MBR.

I don't think any of the captures show a drive in write mode (AFAICS they just done the same I've done: power the drive not connected to a PC) so I guess it's pretty useless to you as you wanted to know what happens during drive "normal" use.