I posted about this in another thread, but thought I'd repeat it again for general information and those interested.
A few weeks ago, I took the 100gb 7200 RPM Hitachi Travelstar out of my DLSD Powerbook to be replaced with an mSATA SSD.
I had a couple of plans for this drive after removal-I wanted to clone it to the new SSD, and also put it in my TiBook(which still has its OEM 4400 RPM drive).
I have commented before on here about how power-hungry this drive is, and is a noticeable drain on the laptop when running. Trying to clone it proved to be no exception to this-most laptop drives I've encountered-whether IDE or SATA-can be powered from a single USB port. This 7200 RPM drive was an exception-power from a single USB port was not enough to spin the drive up. I have some "split" USB cables that came with my enclosures that address this situation(basically they have a second USB plug just to provide power) but couldn't locate one when I was doing this work.
Instead, I used a laptop to desktop IDE adapter, which requires the use of a separate power supply. I had a cheap Chinese-made brick that plugged directly into the wall for just such situations-it outputs 5v and 12v to a standard Molex plug.
I hooked everything up and got Carbon Copy Cloner started, only to have it stop about 300mb in. I started investigating, and got a very strong odor of burning electronics. It seems as though the power brick had decided to fry itself. I cracked it open, and found that it had actually melted through the plastic casing(hidden under the label).
Unfortunately, it also took the hard drive along with it-I could not get the drive to spin up or otherwise give any signs of life.
The situation wasn't a complete catastrophe-the DLSD is sort of a work laptop for me. I have the same programs installed on it as my work G5, and also regularly(manually) sync important files between the two. I cloned the drive in my work computer. That still left me missing one file, though-a presentation that represented a LOT of work, and unfortunately my only back-up of it had been on a flash drive that I left behind in a hotel room. I wanted to resurrect the drive, both for use, and to get the one file off of it, although didn't care about it enough to spend $100+ at a data recovery company for the one file.
A trip to Ebay found me an exact replacement logic board for the drive. I have swapped logic boards before on hard drives, and my past experience has been that it effectively a plug-and-play operation. All my experience, however, was with older drives. So, I figured that for the few bucks the replacement board cost, it was worth a shot.
I installed the replacement board and could get the drive to spin up. It would not, however, mount, or be otherwise recognized by any OS I tried(OS X, Linux, and Windows). I set the project aside.
Last night, I did some reading on the internet, and found out that most "modern" drives have an NVRAM chip on the logic board that must be transplanted to a replacement board. Unfortunately, this chip is soldered, and I did not feel up to digging into this quite yet(although, fortunately, it is only 8 contact points and they are relatively large). I also read that powering up a drive with the wrong NVRAM chip installed could potentially corrupt the data.
The same article I was reading all of this in actually was dealing with Travelstar drives in particular(although it's generally applicable to most modern drives) and mentioned that the most common failure points on these drives are either a 2A surface mount fuse(F1), or if that was good diodes D1 and/or D2 would sometimes fail. A quick check with my VOM confirmed that fuse F1 was indeed open.
My first thought-just to get my data-was to simply put a solder bridge across the fuse to short it and power it up long enough to pull my data off(plus see if that actually was the problem). Unfortunately, surface tension can be a real pain-even at sizes this small-and all I succeeded in doing was sticking the fuse to the end of my soldering iron
So, I went to plan B. I grabbed a piece of 24 gauge stranded wire out of my junk box, stripped the insulation, and picked off one strand of wire. I soldered this across the two contact points where the fuse was attached, confirmed conductivity, and then reinstalled the board.
I held my breath while I plugged the drive in-and just about jumped for joy when it mounted on the desktop of my Macbook Pro(using the split USB cable to connect it to the enclosure). I had it powered up for just about 3 minutes, copied my one file off it, and then unmounted it and put the drive away. Fortunately, the replacement PCB hadn't corrupted my data as I had been warned could happen.
I started shopping for 2A surface mount fuses, and a thought occurred to me. I measured the diameter of the copper wire strands with my micrometer, then pulled out my CRC Handbook of Chemistry and Physics. It seems as though for pure copper wire, the diameter of wire I used had a fuse current(or, in other words, current at which it would melt) of around 2.5-3A-or nearly the exact correct value for the fuse I'd removed.
Just to be on the safe side, I pulled the PCB again, and using a loupe, strong light, and my pocket knife made a carefully eyeballed "notch" in the wire I'd soldered in place. Without knowing exactly how deep I made it, this should bring the fuse current down to around 2A or maybe even a little lower(I can't imagine the drive-in normal operation-drawing anywhere close to 2A).
So, with that in mind, I'm going to call the drive fixed and proceed with installing it in my TiBook. I couldn't be happier with the outcome-I'm just sorry that I blew $15 on a PCB that turned out to be useless!
And, yes, this may be a lot of work, but it was still less time total than I had in the presentation I recovered off the drive. Plus, these high speed, larger capacity 2.5" IDE hard drives don't exactly grow on trees anymore and I think it was well worth the effort to save this one. I think my TiBook will certainly benefit!
A few weeks ago, I took the 100gb 7200 RPM Hitachi Travelstar out of my DLSD Powerbook to be replaced with an mSATA SSD.
I had a couple of plans for this drive after removal-I wanted to clone it to the new SSD, and also put it in my TiBook(which still has its OEM 4400 RPM drive).
I have commented before on here about how power-hungry this drive is, and is a noticeable drain on the laptop when running. Trying to clone it proved to be no exception to this-most laptop drives I've encountered-whether IDE or SATA-can be powered from a single USB port. This 7200 RPM drive was an exception-power from a single USB port was not enough to spin the drive up. I have some "split" USB cables that came with my enclosures that address this situation(basically they have a second USB plug just to provide power) but couldn't locate one when I was doing this work.
Instead, I used a laptop to desktop IDE adapter, which requires the use of a separate power supply. I had a cheap Chinese-made brick that plugged directly into the wall for just such situations-it outputs 5v and 12v to a standard Molex plug.
I hooked everything up and got Carbon Copy Cloner started, only to have it stop about 300mb in. I started investigating, and got a very strong odor of burning electronics. It seems as though the power brick had decided to fry itself. I cracked it open, and found that it had actually melted through the plastic casing(hidden under the label).
Unfortunately, it also took the hard drive along with it-I could not get the drive to spin up or otherwise give any signs of life.
The situation wasn't a complete catastrophe-the DLSD is sort of a work laptop for me. I have the same programs installed on it as my work G5, and also regularly(manually) sync important files between the two. I cloned the drive in my work computer. That still left me missing one file, though-a presentation that represented a LOT of work, and unfortunately my only back-up of it had been on a flash drive that I left behind in a hotel room. I wanted to resurrect the drive, both for use, and to get the one file off of it, although didn't care about it enough to spend $100+ at a data recovery company for the one file.
A trip to Ebay found me an exact replacement logic board for the drive. I have swapped logic boards before on hard drives, and my past experience has been that it effectively a plug-and-play operation. All my experience, however, was with older drives. So, I figured that for the few bucks the replacement board cost, it was worth a shot.
I installed the replacement board and could get the drive to spin up. It would not, however, mount, or be otherwise recognized by any OS I tried(OS X, Linux, and Windows). I set the project aside.
Last night, I did some reading on the internet, and found out that most "modern" drives have an NVRAM chip on the logic board that must be transplanted to a replacement board. Unfortunately, this chip is soldered, and I did not feel up to digging into this quite yet(although, fortunately, it is only 8 contact points and they are relatively large). I also read that powering up a drive with the wrong NVRAM chip installed could potentially corrupt the data.
The same article I was reading all of this in actually was dealing with Travelstar drives in particular(although it's generally applicable to most modern drives) and mentioned that the most common failure points on these drives are either a 2A surface mount fuse(F1), or if that was good diodes D1 and/or D2 would sometimes fail. A quick check with my VOM confirmed that fuse F1 was indeed open.
My first thought-just to get my data-was to simply put a solder bridge across the fuse to short it and power it up long enough to pull my data off(plus see if that actually was the problem). Unfortunately, surface tension can be a real pain-even at sizes this small-and all I succeeded in doing was sticking the fuse to the end of my soldering iron
So, I went to plan B. I grabbed a piece of 24 gauge stranded wire out of my junk box, stripped the insulation, and picked off one strand of wire. I soldered this across the two contact points where the fuse was attached, confirmed conductivity, and then reinstalled the board.
I held my breath while I plugged the drive in-and just about jumped for joy when it mounted on the desktop of my Macbook Pro(using the split USB cable to connect it to the enclosure). I had it powered up for just about 3 minutes, copied my one file off it, and then unmounted it and put the drive away. Fortunately, the replacement PCB hadn't corrupted my data as I had been warned could happen.
I started shopping for 2A surface mount fuses, and a thought occurred to me. I measured the diameter of the copper wire strands with my micrometer, then pulled out my CRC Handbook of Chemistry and Physics. It seems as though for pure copper wire, the diameter of wire I used had a fuse current(or, in other words, current at which it would melt) of around 2.5-3A-or nearly the exact correct value for the fuse I'd removed.
Just to be on the safe side, I pulled the PCB again, and using a loupe, strong light, and my pocket knife made a carefully eyeballed "notch" in the wire I'd soldered in place. Without knowing exactly how deep I made it, this should bring the fuse current down to around 2A or maybe even a little lower(I can't imagine the drive-in normal operation-drawing anywhere close to 2A).
So, with that in mind, I'm going to call the drive fixed and proceed with installing it in my TiBook. I couldn't be happier with the outcome-I'm just sorry that I blew $15 on a PCB that turned out to be useless!
And, yes, this may be a lot of work, but it was still less time total than I had in the presentation I recovered off the drive. Plus, these high speed, larger capacity 2.5" IDE hard drives don't exactly grow on trees anymore and I think it was well worth the effort to save this one. I think my TiBook will certainly benefit!
Last edited:
