• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Powering 3-4 SATA HDDs off one Molex?

J

Jovec

Senior member
My Lian-Li HTPC case supports 4 HDDS. At the rear of the drive cage is a circuit board that has the SATA data and power connects. On the other side of the board, there are 4 SATA data connectors, but there is only one 4-pin Molex connector to which the power supply cable attaches.

In short, my case forces me to use one molex connector to power up to 4 Sata drives. I currently run 3 drives and am wondering if this setup will have any issues powering the drives.

Since upgrading to Win 7 (and the new case at the same time), I've experienced what appears to be long spin-up issues that occur post-boot. Media playback (regardless of the drive) would stall for a few seconds while the drives spin up. Upon reboot, the AHCI bios warned me of SMART errors on 2 drives. After a power-off, wait, reboot, the drives booted with no issues. I'm not inclined to believe that the 2 Samsung HDDS are having SMART issues at the same time, but I can believe that there is some other issue effecting the 2 Samsung drives but not the WD drive. I'm tempted to remove the HDD cage board and plug in each drive directly.

(AMD 780G with 700 SB - I suppose it could also be the AMD AHCI drivers too. Maybe the default MS ones would be better?)
 
Power may be an issue. 4 drives off a single line is a bit of a stretch. So that may be where the issue is coming from. Overall a single connector can support 4 drives but when they all try to spin up at the same time it pushes the limits. Once they are running things should be fine. Now i know there is a way to have the drives spin up at different times i just don't know how to do it. Hopefully someone here will chime in on it or you can probably google it. Putting about .5 seconds between each drive should do it.
 
At most each drive is using 1A on 12V and 1A on 5V , so worst case you are using 4A on 12V and 4A on 5v. A single connector with the standard 4 wires (2 power + 2 ground) can do that all day and then some.
 
Yes, I would think a single molex should provide enough power.

At this point, I'm narrowing it down to a combination of Windows 7-64, AMD SB700, the AMD AHCI driver, and my 3 drives. I know that doesn't seem like much narrowing, but here's my experimental evidence.

- This system ran Vista 64 with the AMD AHCI driver with no issues. The AMD AHCI is unchanged from 5/2009, thus it's the same driver AFAIK.
- Using the default MS AHCI drivers I do not experience this issue
- I don't experience this issue on two other Win7-64 comps with AMD SB750s, though those are single HDD comps
- Reinstalled the AMD AHCI driver and issue returned.
- Uninstalled the AMD AHCI driver and had boot time issue.
- Immediately rebooted and no boot time issue

I have another AMD SB700 system I plan to put Win7-64 on, and will see if this issue arises, though I doubt it since Win7 had thorough testing, there are a lot of SB700s in the wild, and the AMD AHCI hasn't been updated in 6 months. In other words, it's not purely an issue with one component, but my specific combination.

Of my three drives, my 2 Samsung drives show the boot-time AHCI error with the AMD AHCI driver installed. I understand that at that point in the boot process there is only the AHCI bios running and not the actual driver, but it seems that the AMD AHCI driver does trigger something in the drives that makes them have/show an error at reboot. This then gets cleared when the MS driver loads.

Smart monitoring with CrystalDiskInfo shows one Samsung drive with 1 reallocated sector warning but the other Samsung and WD are clear, so there is no correllation between the bios SMART error on the two Samsungs and the CrystalDiskInfo monitoring software SMART warning on the one Samsung that I can see.
 
Last edited:
most likely, at the other end, the hard drive connectors on the power supply are not isolated. i.e., they all tie in to the same circuit node. so the difference between using 2 cables & 1 cable to feed 4 drives is minimal.

if you use one cable, the current in the cable is greater. also, the cable presents a small inductance between the power supply and the hard drive.

i wouldn't worry about it. if your Power Supply can handle it, it'll handle it whether you use one or 2 cables to feed the hard drives.
 
I'm using GD2 with PHP 5.2.9.2 and I'm trying to create a dynamic compass type thing. What I'd like is simply an arrow image that will change where it's pointing based on the get variable I pass to it. Problem is that because the image is square, imagerotate (the function I'm using) makes the overall canvas bigger and bigger, to it's maximum size, at any of the 45 degree angles. This is not what I expected, nor what I want, what I'd much prefer is it to stay exactly in it's place and clipping to occur. I understand how I could do this manually, but to do so, I'd need the dimensions of the newly rotated image, in it's current form, rather than how it was.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top