A question about a USB2 to IDE adapter-cable

[BonzaiDuck]

The subject of a recent mildly lengthy thread of mine -- several of the ending post coming from me -- the OP -- was a desire to extend the usefulness of a StarTech DRW115ATABK hot-swap bay and caddies.

The original investment -- probably in 2007 -- was about $360 for bay, spare caddies and 3x Hitachi 500GB IDE drives. They were never used for more than backup, and the SMART "odometer" on one or more shows cumulative usage of less than two weeks.

The Startech DRW came with 32-bit software for hot-swapping IDE's, but only operable under older OS's. I've deployed the caddy to a WHS-2011 system. I originally thought it would "hot-swap" with an SATA-to-IDE adapter connected to my SATA-controller, but "no cigar."

There is a possibility of a wiring-mod soldering project that powers that adapter from the bay's PCB. When the key-switch is turned off, it would power off the adapter as well as the drive. But there's always a chance I'll bork the Startech drive-bay for making a mistake in choosing the solder points for a 5V wire and a ground. So that's on the back burner.

Instead, it occurred to me I can afford to lose some throughput or transfer speed if I treat the caddy and bays as an "external USB2" device. The backup purpose of the IDE drives is not significantly hampered by loss of full ATA speed. Those devices have a power toggle switch to be flipped off after "Safely remove" provides the appropriate message from the system tray. Or, one could simply remove the USB cable (after safe removal) before powering down such an external assembly.

So I ordered the pricier option of a USB2-to-IDE adapter to use ONLY the cable:

http://www.newegg.com/Product/Produc...-712-_-Product

This was a bit hasty, and I might have saved half the outlay for ordering a different C2G model. But I'm going forward to test it.

I look at the "manual/guide" to the kit, and it says:

" . . . Also, make sure the drive is attached and powered before connecting the adapter to the computer's USB port."

In one way, this makes sense, since the power-brick with Molex-to-IDE isn't switched. But on my external USB2 to IDE mobile drive boxes, I could either "Safely remove" and then pull the USB cable, or "Safely remove" and then power down the box with its "0/I" toggle switch.

I DON'T THINK THE BOLDED QUOTE FROM C2G'S DOCUMENTATION APPLIES TO THIS, SINCE THE KEYSWITCH ON THE CADDY FUNCTIONS THE SAME WAY AS THE TOGGLE SWITCH ON AN EXTERNAL USB2 BOX.

WHAT DO OTHER FORUM MEMBERS THINK?

The purpose of this is to leave the simple USB2-to-IDE cable connected to my motherboard, so I can hot-swap the IDE caddies. Swapping a caddy into the bay would require me to turn the keyswitch, which powers both the drive and the IDE interface of two PCB cards (caddy and bay).

Please -- enlighten me about (1) risk to the motherboard USB port, or (2) the C2G cable for which I paid too much. I just cannot see how there would be such a risk.

 

[BonzaiDuck]

Probably nobody is much interested in this, but it appears that USB can salvage an old IDE hot-swap bay with caddies, offer full hot-swap -- except for one factor.

My WHS "client backups" seem to daily modify all files on that folder of the server. If I back it up to a 500GB HDD (just enough capacity) with a USB-to-IDE-drive approach, it looks like a 15-hour process. On the up side, it's a background process that hardly makes a dent in server capacity. On the down side, the length of time it takes cuts into routinely schedule client backups that will change files that are being backed up.

Just for anyone interested -- who might have a good quality IDE hot-swap bay with caddies and drives that have plenty of life left. USB2 may have been OK for smaller capacity IDEs used occasionally. Don't waste your money on USB2 technology for this. Get a USB3 PCI-E x1 controller; find an available PCI-e x1 (or greater) slot in the box, even if you have to change out a PCI_E G-bit NIC for a PCI G-bit NIC to free up the slot. Get the 20-pin-to-female-USB3 cable and a highly respected USB3-to-IDE adapter.

Total cost could be $50, and the parts can be redeployed in future use. Then, your backups to old higher-capacity IDE drives are going to go quickly.

 

[rsutoratosu]

I wouldn't put too much into this.. I dont even think IDE drives are being made anymore.

I have those usb 2 to ide cables.. i have them because some of the clients i deal with still have ide drives..

 

[BonzaiDuck]

I wouldn't put too much into this.. I dont even think IDE drives are being made anymore.

I have those usb 2 to ide cables.. i have them because some of the clients i deal with still have ide drives..

I agree. I also discovered that USB2 is just too . . . freakin' . . . slow. I can "get more" out of the low-mileage IDE/PATA drives even without hot-swapping, but it is an inconvenience.

And somehow I was mistaken: with such a cable, one has to disconnect it from the USB connection to the motherboard, or you cannot swap another drive into the system and have it recognized. The device must be power-cycled for that to happen. So USB3 will be faster, but none of it resolves the problem. I also saw a You-Tube where a guy installed a toggle switch in a 5.25" bay-cover that switches off the USB "device" of these cables.

Take a look at this:

http://www.addonics.com/products/saturn_hot-swap-kit.php

There is a company named "Beantech" which apparently manufactures the PCB's on the bay side of these hot-swap IDE/SATA assemblies. Whether Beantech or other, I've now seen three of those PCBs -- not specific to origin or manufacture, but I'd wager it's Beantech -- all with the little white plug next to the IDE plug.

"Dave" in UK had determined which solder-points on the PCB were correct for wiring a PATA-to-SATA adapter's 5V power so that turning the keyswitch (either caddy or bay, by design) will power down the adapter as well as the drive. That would make it truly "hot-swap-able."

I could only guess that Dave overlooked that white two-pin plug. I'll just have to pull out the old multi-meter and see if I can make it work, with risk of borking the bay and its PCB.