It's most likely a supply issue, with possibly some timing problems thrown in for good measure. The larger, more expensive cards have more than one flash chip internally, and read and write multiple chips in parallel to increase the total throughput (ie, RAID-like striping). Further, faster flash chips increase the writing current to get their ideal high speed, so even if there's only one flash chip internally, the current draw is greater than slower cards. The greater the storage space, the faster the chip simply because people would notice and complain if it took longer to copy their 1G card to their computer than it took for 256MB, even though they know they're transferring more information. For this reason, even cheap large flash devices are faster and consume more current than the older devices. The supply issue is tricky. The SD standard has been updated over time, but the actual current ratings are not crystal clear, and are complicated by SDIO. The older manufacturers may have only prepared for 100mA, while many newer cards peak in excess of several times that. You might try soldering a huge capacitor across the supply pins of the SD card to see if it helps. The alkalines don't supply much current, while the nimh don't supply much voltage, but plenty of current. So I'd look into that first. Take an old card, and solder very fine wires to the contacts so you can measure voltage with an oscilliscope. This may verify it. Otherwise, the issue may be timing and simple compatability. As nice as a standard is, there are so many ways to follow it. Newer high capacity cards, IIRC, are not required to support some of the older access methods. If your camera also accepts mmc cards, then it may use one of the methods that the newer cards simply dispense with, assuming that most older devices couldn't handle the capacity anyway. This is why manufacturers specify using "good" companies for the card - sandisk and others seem to support the older standards better, while the cut-rate cards produced and branded by the millions shrink the die as much as possible to increase the profit in a low margin business. But given that it's not a cut-rate camera, and cut rate cards (kodak and kodak or sandisk) then I expect it's a power issue. You should try lithium AA batteries. It's an expensive test, but they can supply both the voltage and the current required. Or use your bench supply. The format option probably draws a lot of current over a long period of time. Format it on the computer, and then try it out with pictures, it might work as-is if you don't have to write to the card for a minute at a time. -Adam On 6/2/07, Dr Skip wrote: > I'm not conversant with SD card requirements from a circuit design point > yet, but in order to understand more, I'm looking at 2 devices that seem > to have flaws. The first is a Kodak camera. It's several years old, with > no spec on max capacity, but Kodak sells up to a 2GB card on its site > for it. While I haven't tested all combinations, it seems that 128 MB > and 256 MB cards format fine, but a Sandisk or Kodak 1GB card, as well > as a no-label store brand all cause the camera to power down partway > through the format without a clue. An interesting note is that they say > to use their cards or 'bigger name' cards as they have tighter > tolerances and cheaper ones may not work. Well, the 1GB ones from > anywhere don't work. > > Next is a tablet/pad that specs no limit on SD card cap. I've seen > several units that don't operate with a 1GB card (Sandisk or house > brand) but will do fine with a smaller card. Some other units seem fine. > On further testing, it appears that working units may fail to work with > the bigger cards if the batteries are not very freshly charged. However, > the smaller cards are fine anytime. They too say to only use quality > cards and recommend Sandisk and they may not work with off-brands. > > The camera uses 2 AA cells for 3v nom (2.4v for typical NimH), and the > tablet uses 4 AAA NimH cells, possibly in series or as 2 sets of 3v > (I've not taken it apart and they install as 2 banks). Using alkalines > to increase the voltage doesn't do it for the camera, but can help the > marginal tablets. > > Other, more expensive devices I've seen using SD cards don't seem to > have this problem. It seems like a supply design problem to me, and > smells like maybe there isn't enough overhead on whatever regulator they > use, and if bigger cards draw more, the supply drops below spec. > Assuming the interface is on a chip and dropped into the design, are > there some interface chips that are known to have this problem? Do > bigger cards draw more in use? And how do the cheap cards differ from > "better" ones and how does one test this? > > Any ideas from experience on what's going on? I'd like to not design in > these problems when my time comes... > > Thanks, > Skip > > > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Moving in southeast Michigan? Buy my house: http://ubasics.com/house/ Interested in electronics? Check out the projects at http://ubasics.com Building your own house? Check out http://ubasics.com/home/ -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist