8 views today  ·  8 views since Jan 2026

Why Your Shiny New Enterprise SATA Drive Refuses to Power Up

TL;DR

Modern enterprise SATA drives (WD Ultrastar, HGST Helium, Seagate Exos, Toshiba MG) use pin 3 of the SATA power connector as a "power disable" signal. Standard desktop PSUs supply 3.3 V on this pin, which tells the drive to stay powered off. The fix is to prevent that 3.3 V reaching pin 3.

The symptom

You have just upgraded your NAS or homelab storage with a lovely pair of enterprise-grade refurbished drives. You physically install them, power on the machine, and... nothing. The drives do not spin up. They do not appear in the BIOS. Your NAS operating system (TrueNAS, Unraid, Proxmox, etc.) does not see them at all. lsblk and smartctl --scan come up empty. It is as if the drives are not even connected.

Meanwhile, the drives themselves are silent. No spin-up noise. No warmth. No clicking. No LED activity from the drive caddy (if you have one).

Your first instinct is probably to suspect a dead drive, a bad SATA cable, or a faulty port. All reasonable suspicions — but if you bought enterprise drives manufactured after roughly 2015, the cause is almost certainly something else entirely.

The cause: SATA 3.3 and the power disable feature

The SATA power connector has 15 pins, grouped into three rails plus some ground and signal pins. The relevant ones here are the three 3.3 V pins at one end of the connector — pins 1, 2, and 3.

On older drives, all three of these pins simply provided 3.3 V power to the drive electronics. Nothing fancy.

Then along came the SATA 3.3 specification. This repurposed pin 3 as a new signal called "PWDIS" — Power Disable. The idea was clever: in a large data centre with hundreds of drives per rack, administrators wanted a way to individually power-cycle a misbehaving drive without physically pulling it from the chassis. The new spec defined that if 3.3 V is present on pin 3, the drive should remain powered off. Remove the 3.3 V, and the drive boots up normally.

This is brilliant in a data centre context. Your storage chassis has intelligent backplanes that can assert or de-assert pin 3 on command via a management interface. A support technician in a different country can toggle a specific drive off and on again without anyone walking to the rack.

The problem? Your consumer PSU does not have an intelligent backplane. It just dumps 3.3 V onto pin 3 continuously, because that is what pin 3 used to mean. And the drive, dutifully following the SATA 3.3 spec, sees that 3.3 V and refuses to spin up.

How to identify affected drives

Look at the drive's power requirements label. Affected drives will show only 5 V and 12 V ratings, like this:

RATED: 5V 400mA 12V 550mA DC

Note the absence of 3.3 V. If the label shows no 3.3 V rating, pin 3 is not used for power on that drive — it is almost certainly configured as the PWDIS signal.

For reference, these drive families (and many others) are known to implement PWDIS:

  • WD Ultrastar DC HC3xx, HC5xx, HC5xx series (including HGST-branded predecessors like He10, He12)
  • Seagate Exos 7E8, 7E10, X12, X14, X16, X18, X20, X22 series
  • Toshiba MG07, MG08, MG09, MG10 series

Consumer drives — WD Red, Seagate IronWolf, Toshiba N300 — generally do not implement PWDIS and will work with any standard SATA power cable.

Three ways to fix it

All three solutions achieve the same thing: prevent 3.3 V from reaching pin 3 of the drive's SATA power connector. Pick whichever you are most comfortable with.

Option 1: Kapton tape or electrical tape on pin 3

Cover pin 3 of the SATA power connector with an incredibly small piece of tape so it does not make electrical contact with the drive (good luck with that).

Pin 3 is the third pin from the end with the 3.3 V / 5 V / 12 V rails (the wider end of the L-shaped connector). Count: pin 1, pin 2, pin 3. A tiny strip of tape is all you need. If you want to double-check the pin numbering, PinoutGuide.com has a clear visual diagram of the SATA power connector.

Pros: Completely reversible. Costs nothing. Easy to undo if you ever move the drive to a system that properly implements PWDIS.

Cons: Slightly fiddly to apply precisely. Tape can come loose over time or shift during installation, though in practice this is rare.

Option 2: Molex-to-SATA power adapter

Molex connectors (the old 4-pin PATA-era power connectors) carry only 5 V and 12 V — they predate the SATA 3.3 V rail. A Molex-to-SATA adapter simply has nothing to supply on pin 3, which means no 3.3 V, which means the drive spins up.

Pros: No modifications needed. Clean installation.

Cons: Molex-to-SATA adapters have a somewhat patchy reputation — cheap ones have been implicated in fires over the years due to poorly crimped pins. Buy a quality one from a reputable brand if you go this route.

Option 3: Cut or remove the orange 3.3 V wire from the SATA power cable

Physically remove the orange wire from the SATA power connector. You can either cut it flush, or if the connector allows, pop the pin out and reinsulate it.

Pros: Permanent. Cannot come loose. Clean solution for a dedicated build.

Cons: Permanent. You have modified the cable and cannot easily revert. Affects all drives connected to that cable, not just the enterprise one.

My preferred approach

I used a SATA power splitter (one source, two destinations). The first output of the splitter already had the 3.3 V pin exposed as a separate wire rather than being fully enclosed in the connector, so I was able to remove that pin from the first connector entirely and wrap it in insulation. Both drives on the splitter now run without 3.3 V on pin 3, and the original PSU cable is untouched and can still be used for anything else if I rearrange the build later. Get yourself a pinout of a SATA connector to confirm. Generally, it's the first 3 pins next to the notch, all wired in parallel.

Verifying the fix

After applying whichever fix you chose, power the machine back on. The drives should spin up immediately — you will hear them, and there should be a characteristic few seconds of initialisation clicks. In your operating system:

sudo smartctl --scan

Both drives should now appear. You can then check their health:

sudo smartctl -a /dev/sdX

Pay particular attention to Power_On_Hours if you bought the drives refurbished — this tells you how much runtime they have already done. And check Reallocated_Sector_Ct (ID 5), Current_Pending_Sector (ID 197), and Offline_Uncorrectable (ID 198) — all three should be zero on a healthy drive.

Why this is not better documented

Honestly, it should be. The spec change happened years ago. But the information tends to be scattered across forum posts, Reddit threads, and data centre hardware documentation that most home users never encounter. Sellers of refurbished enterprise drives rarely mention it either, because in their primary market (other data centres buying in bulk), nobody hits the problem — the buyers have chassis designed for these drives.

If you are just an enthusiast setting up a home NAS or homelab, the first you hear about PWDIS is usually when your brand new drives refuse to turn on and you spend an hour panicking that you bought duds.

Hopefully this post saves someone that hour.

Further reading

Previous Post