fzabkar

Member
  • Content Count

    32
  • Joined

  • Last visited

Community Reputation

0 Neutral

About fzabkar

  • Rank
    Member
  1. I have a general observation/request. I find that the photos of the PCBs are frustrating because they have insufficient resolution to enable someone to identify the ICs. Would reviewers consider supplementing their articles with higher quality photos or CCD scans? I often find that manufacturers do not always promote all the features of their products, but these features are often evident on the PCB. If the reviewer were able to identify these "hidden" features, then that would add value to the article. Also, I am active in the professional data recovery forums and am always looking for resources to help others in the troubleshooting of failed PCBs. A database of hi-res photos would be of great benefit to the wider community. One other request is in regard to HDD reviews. Would reviewers consider running a "5GB or 10GB short stroke test" in HD Tune? The graph produced by such a test usually shows the real number of physical heads, whereas HDD manufacturers nowadays shy away from this info. The access time graph also shows the real RPM of the drive, but the drive may in fact report a different speed. For example, I suspect that one particular WD "IntelliPower" drive is spinning at 7200 RPM but is reporting a speed of 5400 RPM. One could prove this by analysing the sound spectrum of the spinning drive.
  2. It appears that one enclosure is configured with a sector size of 4096 bytes while the other is configured for 512-byte sectors. When you move from one enclosure to the other, the capacity changes by a factor of 8. 7.68 terabytes / 8 = 960 gigabytes The reason for this anomaly is that the partition table reports the number of LBAs in each partition, so the capacity of the partition is ... (number of LBAs in partition table) x (sector size)
  3. So the 1TB model must have two 500GB platters, not a single 1TB platter as claimed in "the platter breakdown"? And how can WD claim that the new 4TB models have a 48% performance increase over the earlier versions? I see only 11%. 171 / 154 = 1.11 A figure of 11% is consistent with the bits-per-track increase for AF models over their non-AF counterparts.
  4. I fail to see where the enhancements are. If we were to take a Seagate 1TB-per-platter AF drive, its maximum sustained data rate at 7200RPM is 210MB/s. Now let's assume that we reduce the platter density to 500GB and allow all other things to remain equal. This means that the capacity would be halved, and that there would be a reduction in the number of bits per track by a factor of 1/sqrt(2), and a similar reduction in the number of tracks per inch. Since the maximum transfer rate is directly related to the number of bits per track, then we would expect an equivalent Seagate 500GB-per-platter AF drive to have a transfer rate of ... 210 / sqrt(2) = 148.5 MB/s Therefore ISTM that WD's current models perform only marginally better than last generation's Seagate drives. Furthermore, if we were to reduce a 1TB-per-platter Seagate drive to 800GB-per-platter, then its transfer rate would be ... 210 / sqrt(1TB / 800GB) = 210 / sqrt(1.25) = 188MB/s. This is still more than 10% faster than WD's drives.
  5. The labels on the drives indicate that they were manufactured on 21 June 2013. However, the YYWW (Year / Week) date codes for the chips on the PCB are 1306, 1313, and 1308. This would suggest that the PCB may have been in stock for 3 months before the drive was assembled. In general I find that other HDD manufacturers adopt a "just-in-time" philosophy when manufacturing their drives, so I'm wondering whether these models are slow moving items.
  6. http://www.diskinternals.com/download/zip_repair.exe
  7. @kingslytaylor, did you even bother to read this thread before dumping your spam? How does your software miraculously recover data from a head crash?
  8. I'm not a data recovery professional, and I've never encountered the subject problem personally, so I can't give you any valid suggestions. However, the drive would spin down after 20 seconds if it cannot access the firmware area on the platters, or if it cannot detect a track servo signal. This "System Area" (SA) is where the bulk of the firmware is stored. The procedure involves repairing damaged or corrupted modules in the SA. If your drive has a bad head, then perhaps this may be the reason why the procedure doesn't work as expected.
  9. AIUI, the 3V battery merely provides enough current to turn on the "enable" signal for the USB-RS232 bridge IC. It doesn't really provide "power" for the CA-42 electronics. In fact I have seen a case where the 4th pin in the jumper block was successfully used to enable the CA-42 rather than using a battery. It all depends on the design of the CA-42 interface. In fact there are numerous different OEM implementations of the CA-42 cable. In your case, however, it appears that there is nothing wrong with the "adapter". Instead there appears to be a firmware problem with the drive, or perhaps some problem with your methodology.
  10. Swap the 25X40L002 and 25FW408A chips. W25X40L, Winbond, 4Mbit, 2.3V - 3.6V serial flash memory with 4KB sectors and dual output SPI: http://www.winbond.com/NR/rdonlyres/0971C40C-F202-49CA-90AF-0F0268ECF0E5/0/W25X10L_W25X20L_W25X40L_W25X80L.pdf LE25FW408A, Sanyo, NOR type serial flash memory, MSOP8, 2.7 to 3.6 V, 4MB, SPIO/3: http://www.sanyosemi.com/en/memory/topics/highspeed-nor.php The others chips are FETKYs: http://www.users.on.net/~fzabkar/HDD/HDD_Semis_Regs.html FDFS2P102A, Integrated P-Channel PowerTrench® MOSFET and Schottky Diode, Fairchild, 20V, 3.3A: http://pdf.datasheetcatalog.com/datasheet/fairchild/FDFS2P102A.pdf A hot air gun is easier but more dangerous than a soldering iron in some respects. I suggest that you practice on the replacement board first. BTW, if the drive spins up, then the board is probably OK.
  11. AISI, all you would need to do would be to transfer the 8-pin serial flash memory chip. Be very careful, though. If you damage this chip, then data recovery will become extremely expensive, if not impossible. The best way to transfer the chip's contents is with an SOIC/SOP chip clip plus a chip programmer. This database should help you identify the chip: http://www.users.on.net/~fzabkar/HDD/HDD_EEPROM_Flash.html
  12. Seagate's old model numbering system would tell you the number of platters in the second-to-last numeric character. Nowadays Seagate tout their "New, Simplified Model Numbers" in a document which they unfortunately refer to as their "Cheat Sheet". http://www.seagate.com/files/staticfiles/docs/pdf/marketing/st-model-number-cheat-sheet-sc504-1-1102us.pdf
  13. To be fair, Seagate have until the last couple of years been one of the best companies when it comes to documentation. Twenty years ago they gave me detailed paper technical manuals (books) for free, and they've always had detailed product information in their Product Manuals. Sadly, times have changed ...
  14. fzabkar

    WD Protoge WD400EB

    Can you upload a HD Tune read benchmark graph? That will help us determine whether it is a single platter or dual platter drive.
  15. BTW, the uF value is printed on the capacitor. A quick voltage measurement would be very enlightening, too. It would tell us a lot about the backup scheme now being employed in enterprise SSDs, and how it differs from consumer grade SSDs. If you don't have a digital multimeter, one can be purchased for US$5. If you don't know how to take a voltage measurement, then read the accompanying instruction booklet. I've taught plenty of novices how to do this. Here is a cheap DMM (aka Digital Multimeter, US$5): http://www.harborfreight.com/7-function-multimeter-98025.html http://manuals.harborfreight.com/manuals/98000-98999/98025.pdf (Product Manual) If you don't know how to read the numbers off the label of a capacitor, then see this datasheet: http://www.nichicon.co.jp/english/products/pdfs/e-he.pdf Notice the "100V 680uF" markings on the image in the top left corner of page #1.