Why are there so many pins on a SATA power connector?

Question

I was just looking at the power connector for a SATA optical drive and noticed that it had fifteen pins!

I then marveled at the data connector next to it, which has a measly seven pins.

I know that there are only physically three pins needed for the power connection:

• +12v DC
• +5v DC
• Ground

Why did they decide to use a fifteen-pin power connector that's twice the width of the data connector?

Solution 1

Here's the SATA data & power pinouts.

Remember, SATA is a serial bus. This means data transfer only needs two paths -- TX (transmit) and RX (receive). In the case of SATA, there are actually two pins for each (a TX+ and TX-, and a RX+ and RX-); this is called twisted pair and (just like in twisted pair Ethernet) allows for longer wire runs with less noise from other wires. The other data pins are for ground, which also assist with noise elimination. So SATA doesn't need more data pins.

Power, on the other hand, supplies 3.3 V, 5 V, 12 V, and ground. Not to mention extra pins (not present in all connectors) for hotplugging, activity indication, and staggered spinup. As to why so many? Wikipedia again:

Each voltage transmits through three pins ganged together, because the small contacts by themselves cannot supply sufficient current for some devices. (Each pin should be able to provide 1.5 A.)

Solution 2

I am not sure this has been sufficiently answered.

I don't have an answer per-se, but I can share what I've learned in search of an answer:

The best, given answer- and really the only actual answer is the quote by way of @quack quixote from Wikipedia.

Each voltage transmits through three pins ganged together because the small contacts by themselves cannot supply sufficient current for some devices. (Each pin should be able to provide 1.5 A.)

But why 3 of each. They can't be used for signaling, you can't pull one V+ low, or pull a ground-up because it's serially connected. Why not make a larger contact and use just one, If you look, each power pin is immediately adjacent to its friends-

12V-12V-12V-Gnd-Gnd-Gnd-5V-5V-5V-Gnd-Gnd-Gnd-3.3-3.3-3.3

So, why not remove the spacing and make a larger contact area, if needed. Ta-Da! Now the connector is 1/2 as wide. 12V-Gnd-5V-Gnd-3.3 (Or 2/3rds more precisely). In most cases, it's not redundant either, it isn't broken out to 3 until the adapter.

Interesting point 1-

That 3.3v line is apparently 3.3 no longer- at least for 2 of 3...

To sum up, for products supporting the optional SATA 3.3 power disable (PWDIS) function, the third pin (P3) of the SATA connector is now assigned as the Power Disable Control pin. If P3 is driven HIGH (2.1V-3.6V), the power to the drive circuitry will be cut. All drives with this optional feature will not power up if a legacy SATA connector is used. This is because P3 driven HIGH will prevent the drive from powering up. The easy, and not so elegant, solution is to use a 4-pin Molex to SATA connector or a power supply equipped with SATA connectors that follow the SATA 3.3 specification.

Source-Toms Hardware

Western Digital Whitepaper

But wait, there's more-

In addition, according to the SATA-IO Standards group, the 11th pin (the second ground from left in my written version.) provides staggered spinup AND an activity indicator- presumably for flashing LEDs and such.

From a SATA-IO Press Release (SATA-IO is the International Organization that owns and manages Serial ATA specifications as open industry standards.):

Additional advancements in the revision 3.3 specification include:

• Power Disable: Allows for remote power cycling of SATA drives to help ease maintenance in the data center.

• Single-Pin Activity Indicator and Spin-Up Control: An activity indicator and staggered spin-up can be controlled by the same pin, adding flexibility and providing users with more choices.

• Transmitter Emphasis Specification: A new transmitter specification increases interoperability and reliability in electrically demanding environments. SATA-IO Press Release

You would think the answer would be found, in depth on the SATA-IO website, but I have been unable to find it. Much of it is behind a paywall, unfortunately. Perhaps someone with this arcane answer can provide more info as to the thought process behind it. It would be in the 1.0 specifications I imagine.

A discussion on the arrangement:

...from a once rival, now sister site:

A closely related discussion also has some excellent ideas, if they aren't sufficiently cited. Wikipedia perhaps...?

The new SATA power connector contains many more pins for several reasons:

3.3 V is supplied along with the traditional 5 V and 12 V supplies. To reduce impedance and increase current capability, each voltage is supplied by three pins in parallel, though one pin in each group is intended for precharging.

Five parallel pins provide a low-impedance ground connection.**

Two ground pins, and one pin for each supplied voltage, support hot-plug precharging. Ground pins 4 and 12 in a hot-swap cable are the longest, so they make contact first when the connectors are mated. Drive power connector pins 3, 7, and 13 are longer than the others, so they make contact next. The drive uses them to charge its internal bypass capacitors through current-limiting resistances. Finally, the remaining power pins make contact, bypassing the resistances and providing a low-impedance source of each voltage.

This two-step mating process avoids glitches to other loads and possible arcing or erosion of the SATA power connector contacts.

Pin 11 can function for staggered spinup, activity indication, both, or nothing.

Here are the pins, as a reference for above:

**This explains some other interesting aspects, though I am unsure what they mean by 5 parallel pins. 5 parallel grounds I suppose (6- pin 11).

Conclusion

In conclusion, the only reasonable assumption is that these additional pins were left, in some cases, to provide additional functionality in the future. And perhaps its the idea of a better connection, through redundancy maybe. It seems in 2 places, its been stated that the parallel connections are preferred. This I guess ensures that at least one of the pins are making contact, to combat corrosion and other effects. When these pins contact a surface, the actual contact area is relatively small, this was perhaps a way to improve that. However, this is not the case for data, but perhaps its less of an issue. I notice that one particular pin of my iPhone connector always collects black grime, while the remainder are unmarred. I think this shows the corrosion that can build up on particular power pins- I think this is the ground pin. Its the V+ pin, that corrodes.

Addendum- the 3.3V/Power Disable Problem

I figure most people end up here after the following scenario-

Since shucking hard drives has become a thing- and was a thing I was unaware of long before the time of this answer, the 3.3V pin/ and power disable issue is particularly relevant. A lot of users are shucking drives and finding they simply don't work in their computers, while they do work in a external bay.- There are other possible reasons for for this that should be excluded, principally, so I read, is encryption.

But instead of using a molex to sata (or more likely a sata to molex to sata connector) you can also do one of the following:

• Put tape over these three pins (on the drive).
• Simply clip the 3.3 wire altogether. (tape over the ends to keep from causing a potentially damaging short to ground [the chassis] by being careless)
• Convert a molex connector to sata, as Molex 4 pin connectors only have 12V and 5V and 2 Ground. (repeated here for those just scanning this long post).

You don't need them. Hard drives never use them, nearly nothing uses the 3.3V rail. The impact of wire resistance to cause Voltage droop grows as voltage falls - as a percent of original voltage, so if a device wants 3.3V, and it usually everything does use 3.3V or 5V (and in larger chips, where heat matters- 1.8, 1.33 and 1.0) in their digital circuitry- just not from the main PSU. But its more reliable and pretty easy to make it locally onboard. In the near future, we may be using 12V only (like videocards/PCIe) which has been proposed and already exists in Servers. The devices will just step down the voltage to whatever they want.

Solution 3

Power connectors (Wikipedia) - specifically

A third voltage is supplied, 3.3 V, in addition to the traditional 5 V and 12 V.

and

Each voltage transmits through three pins ganged together, because the small contacts by themselves cannot supply sufficient current for some devices. (Each pin should be able to provide 1.5 A.)

So that is nine pins needed for power, though very few (if any) drives use the 3.3 V lines, plus a few for ground.

Comments

[Broam]

...and yet all of the power adapters I see don't handle the 3.3v pin. It must be considered de facto optional.

[quack quixote]

@Broam: there's 3x 3.3v pins. are they all missing? they might be (should be) missing on molex-to-SATA power adapters; those won't have a 3.3v lead, just 5v/12v

[Broam]

I was referring to the Molex(TM) adapters, yes. The only ones I see that handle 3.3v come straight from a PSU.

[quack quixote]

@Broam: yep. you'd need additional hardware to provide 3.3v from a 5v & 12v source. for most of what you'd need the molex adapters for (SATA optical drives or hard drives for an older computer), the devices can get away with not using the 3.3v. newer low-power devices may not work properly with those power plugs.

[user3144803]

3.3v isn't really a requirement yet I have only seen it on a 1.8" SSD so far although obviously we are moving that way as component sizes decrease. You only won't have it if you are using a Molex adapter.

[RomanSt]

They should have given us two power connectors: one for a limited current 5V and 12V (total of 4 pins, with a separate ground for each), and a separate one for all the server/enterprise extras like 3v3, staggered spin-up, more current, etc. 99.9% of all HDDs would come with a tiny 4 pin power connector only and nothing else. Alas.

[user]

@romkyns As it stands, SATA drives can physically use SAS connectors (specifically SFF-8482). A limit of 1.5A also likely would not be enough; you'd probably want at least twice that on both 5V and 12V, and there's no denying that 3.3V can be useful (yes, you can have a DC-to-DC converter on the drive if you need it, but the PSU can already supply 3.3V, so why not use that and save the money?). You could leave some connectors dangling (exposed or simply unconnected), but then what's the point? A single standard helps everyone.

[RomanSt]

@MichaelKjörling But it's not a single standard... Consumer PSU makers don't connect 3V3, so it only looks compatible, but it actually isn't. The PSU connector would ideally reflect this, preventing use on drives that require 3V3 without sacrificing compatibility of consumer drives with SFF-8482.

[Hashbrown]

Regarding the 3.3v power disable problem (and the "nearly nothing uses the 3.3V rail") I actually have the opposite problem. I retrieved an mSATA SSD from a dead laptop and put it in a 2.5" SATA adapter. It only works in a real machine; external 2.5" caddy, multi-bay aggregators, and I assume sata-to-molex in these too if I had any on me. On the disk itself it says "3.3V 0.95A" so I assume the adapter exclusively uses 3.3v for mSATA and doesn't step down the 12 or 5 volt rails :( It's strange 'cause there's a lot of tiny resistors and circuitry on the adapter..idk what it's adapting..