CRU rack mount USB3 readers and SSDs

[Leo Enticknap]

Has anyone else noticed that with a CRU 1RU rack mount USB3 reader (sorry - don't know the exact model number), if you insert a cartridge containing an SSD (as distinct from a spinning rust hard drive), the orange read/write activity light doesn't work?

This just happened to me. About a month ago, the spinning rust drive in my CRU drive of test and demo stuff bit the dust. I replaced it with a SSD (WD Blue) and re-wrote the same content to it. A little while after that, I used it at an install, and noticed no orange light. I dismissed it as just a one-off quirk and didn't worry about, especially as the customer was not there when I did the install and was unlikely to be bothered about it, anyways.

So just now, rinse and repeat. However, this time the customer was with me, and so imagine my embarrassment when I told her, "...and when you latch the drive, you should see the orange light blink. That means that the ICMP has seen the drive and knows that it exists." No orange light, but as with last time, it read the drive, and ingested DCPs from it, just fine.

I'm womdering if that LED is triggered by SATA power suck, and that because an SSD uses less current, it's not drawing enough to cause that LED to come on? A minor annoyance, but I must remember to mention that when doing customer training in future.

[Steve Guttag]

I hadn't noticed, honestly. So, I can neither confirm nor deny. I don't know what device/process is responsible for lighting the LED (is it the drive's responsibility to send that as data or is it up to the hardware to light it?).

[Ryan Gallagher]

I hate to trust an AI search result... but with the right question it appears maybe although pin11 is "activity" per the SATA standard... whether or not it is actually implemented and doing anything on SSDs may vary vendor to vendor? Which model SSD did you try?

I have at least a couple desktop SSD caddies, that although they have a light pipe for the LED, my Samsung, and Crucial drives do nothing with it.

[Leo Enticknap]

So in other words, the drive has to tell the SATA interface that it is "active" in order to trigger the light? Maybe. I'm 99.9% sure that the orange LED did flicker when I had that same drive in the 5.25" SATA internal bay in my home computer while I was writing to it. But that doesn't have the SATA to USB3 conversion electronics in it that those rack mount readers do: it just has SATA data and power jacks on the back of it. Maybe different models of CRU bay trigger the orange activity light in different ways?

[Steve Guttag]

I hate to trust an AI search result... but with the right question it appears maybe although pin11 is "activity" per the SATA standard... whether or not it is actually implemented and doing anything on SSDs may vary vendor to vendor? Which model SSD did you try?

I have at least a couple desktop SSD caddies, that although they have a light pipe for the LED, my Samsung, and Crucial drives do nothing with it.

So, just what is "pin-11" on a SATA connector (on the drive). The Data connector doesn't have 11 pins. The Power connector doesn't have a data connection.

[Caleb Williams]

So, just what is "pin-11" on a SATA connector (on the drive). The Data connector doesn't have 11 pins. The Power connector doesn't have a data connection.

The attached link is to a whitepaper from Western Digital discussing and comparing legacy and current SATA standard versions. Pin 11 on the power connector is reserved and can be used to provide an activity indicator, and that is what I would guess the standard CRU enclosures use for their activity lights.

However, not all SATA drive manufacturers will implement fuctionalities provided by the standard on pin 11, and that is likely why Leo is not seeing blinking lights with his SSD's.

[Ryan Gallagher]

Yeah I guess the AI result did not clarify it was referring to pin 11 on the sata power side, which may or may not be a separate cable/connector depending on the situation.

[Steve Guttag]

Even in their own document, pin-11 is "CMN" not Reserved :

image.png

So, IF the drive has the activity connector and IF the chassis and cable assemblies are set up to utilize it, then it could work? Sounds sketchy to me:

image.png

And this is for the Western Digital HC510 series.

[Ryan Gallagher]

Wikipedia appears to be more accurate and concise here, and goes into more detail on your second pinout. The first pinout ignoring it makes sense cause it is showing a molex style SATA power cable that is 5 conductors only (typically coming directly from a PSU), and so would have had to ignore those more backplane/management features, which seem to have been introduced in 2005 to the standard as optional.

via https://en.wikipedia.org/wiki/SATA

• Pin 11 might be used (often by chassis or backplane hardware independent from SATA host controller and its data connection) for staggered spinup, activity indication, emergency head parking, or other vendor defined functions in various combinations. It is an open-collector signal, which may be pulled down by the connector or the drive.
  • Host signaling: If pulled down at the connector (as it is on most cable-style SATA power connectors), the drive spins up as soon as power is applied. If left floating, the drive waits until it is spoken to. This prevents many drives from spinning up simultaneously, which might draw too much power.
  • Drive signaling: The pin is also pulled low by the drive to indicate drive activity. This may be used to give feedback to the user through an LED. Relevant definitions of pin operation have changed multiple times in published revisions of SATA standard, so the observed behavior may be dependent on device version, host version, firmware and software configuration.^[54][55][56] There is also a specification for transmission of drive temperature and other status values with activity signal pulses routinely used to make LED blink.^[57]

Since we know the caddies light up with a full size SATA drive, I would expect the backplane is set up to do something with it, so that leave the drive implementation as the suspect. SPINUP is certainly not needed on SSDs, so maybe they ignore the other features entirely, a bit of cost shaving.

[Stefan Scholz]

Actually I haven't checked, as long as it works. These drives have become rare, only one larger DCP distributor seems to them. Or what is left of these cartridge things after 15 years of abuse. Most with so so 3.5 in HDD.
A lot of distribution content these days is online delivery. Smaller distributors seem to love 2.5 in portable HDD or SSD sent in a mailing envelope.

[Ryan Gallagher]

Actually I haven't checked, as long as it works. These drives have become rare, only one larger DCP distributor seems to them. Or what is left of these cartridge things after 15 years of abuse. Most with so so 3.5 in HDD.
A lot of distribution content these days is online delivery. Smaller distributors seem to love 2.5 in portable HDD or SSD sent in a mailing envelope.

Our manager prefers the old ways, CRU drives (mostly from deluxe). I guess we are a classics house in more ways than one. Festivals is about the only time we see online delivery or alternative caddies. I would expect physical delivery is still quite common at all the part-time/mixed-use/education spaces? The infrastructure for online delivery is a bit of a leap if you are not showing film every day. (granted a very tiny leap if you are modest with your hardware choices).

I'd be all for it. Remoting into a booth machine to start ingests over climbing 6 flights of stairs on a non-film day just to boot up/ingest/shutdown would be a win. But for now anyway, the stairs are probably good for me (the off season workout routine compared to 70mm week).

[Leo Enticknap]

Just tried the WD Blue SSD in the 5.25" internal SATA CRU reader in my home PC, and it's as I remembered: the orange blinkies during read activity are present and correct, and the hard drive activity light on the PC's case that is connected to a header on the motherboard blinks in sync with it, too. So whether it's a signal on pin 11, reading current suck, or another method, I'm becoming increasingly convinced that whatever drives the orange LED differs significantly between SATA only CRU readers, and those that have SATA to USB conversion onboard. Maybe something to do with the fact that the PC's reader is powered by a SATA power connection from the PC's power supply, whereas the rack mount USB one has a dedicated internal power supply (you simply connect a regular C13 power cord to the back of it)?

Agreed with all that DCP distribution via CRU cartridges is rapidly fading into obsolescence. However, Europe seems to be ahead of us in moving to online distribution. I first noticed this during my first major install after covid, at a mixed use venue that is a French cultural center in LA, meaning that they intended to play almost entirely French movies supplied by distributors in France. Shortly after the install, which included a shiny new CRU reader, the artistic director started to receive emails from Paris containing FTP details, Aspera links, etc., and from some of them, a request for FTP details to upload content into from their end. I set up a NAS with port forwarding from the site's main gateway/router to enable that. Apparently when he asked one of these distributors if they could ship a physical drive, their response was incredulity, as if he'd just announced his intention to cross the Atlantic in a sailboat.

The Deluxe e-delivery thing is starting to gain traction here: I've nudged several customers into installing them over the last few years, and those that did have all been very happy with it. But there is definitely going to be a long tail for the CRU drive in the US.

[Carsten Kurz]

There is a current press release that DCDC is now performing a major effort to move from satellite to IP based terrestrial broadband:

https://dcdcdistribution.com/dcdc-announces-rollout-of-terrestrial-broadband-solution/

[Ryan Gallagher]

Just tried the WD Blue SSD in the 5.25" internal SATA CRU reader in my home PC, and it's as I remembered: the orange blinkies during read activity are present and correct, and the hard drive activity light on the PC's case that is connected to a header on the motherboard blinks in sync with it, too. So whether it's a signal on pin 11, reading current suck, or another method, I'm becoming increasingly convinced that whatever drives the orange LED differs significantly between SATA only CRU readers, and those that have SATA to USB conversion onboard. Maybe something to do with the fact that the PC's reader is powered by a SATA power connection from the PC's power supply, whereas the rack mount USB one has a dedicated internal power supply (you simply connect a regular C13 power cord to the back of it)?

If I remember to drag my usb3 CRU caddy home from work, i'll try the 3 or 4 brand SSDs I have handy at home, just out of curiosity.

[Harold Hallikainen]

There is a current press release that DCDC is now performing a major effort to move from satellite to IP based terrestrial broadband:

https://dcdcdistribution.com/dcdc-an...band-solution/

“It’s great that we’re able to offer an alternative digital delivery method to cinemas that can’t install our satellite solution.” noted Howard Kiedaisch, CEO of DCDC. “Our partnership with CineSend and Velocity is off to a great start as we’ve already delivered our first major studio title via broadband after only a couple of months”.

Is that indeed a "major effort to move from satellite to IP based terrestrial broadband" or just providing an alternate to satellite for the receive sites where satellite does not work?