Components aren't made for such a lifespan. There isn't _one_ ting wrong, but several. If i warmed them, the card would crash right there. They were replaced, because if i cooled them, the card would run longer uninterrupted. The replacing of components has made the card a little more stable. That's a very strong argument that the intermittency is on Intermittent fault is internal to a chip, since your recent physicalĭisturbance of the card (replacing components) restored operation for Would also be good to have a very quick way to tell if the card is Localization-both cooling spray and a small but sturdy plastic rod (toĬause local flexing of the card and its solder joints) will be useful. When you're feeling more in control, return to methodical One thing wrong, and you'll never find it if you change more than one thing ![]() Of the card led to a "remission"-that's a big mistake. You changed so many things that you have no idea which physical disturbance That you're currently running a little short on that. This will require a lot of patience to get good localization, and I sense Steps with cooling spray to localize the problem. And don't go overboard and drown in data before you repeat your earlier This is the perfect sell and run opportunity -) - no really, im trying to fix this f****r for good. It looks like if i fix one thing another one seems to break down. The last straw is replacing the slow LS and HC Logic against ALS and AHC(T) ones. I replaced the capacitors for tantal ones as well. Something in this card is fundamentally wrong! I just can't find it. But it only made the card seemingly only a _little_ more stable (faster "recover" from crashes as it seems). I found some flunky (U63, U60, U31) SMD-ICs which i replaced. From then on every restart shortens the time to the next crash until i get 401 and 302 then only 301 errors. Now it has the exact same symptoms as before. The Card worked for about 8 hours in cache-test and a good 2 hours playing Mods afterwards! I played some OotW today with 56Mhz Oscillaror (14mhz just to be safe) and guess what, the card crashed about 30 minutes in (+10-20 minutes GSOS). But counting the hours of work i put into the cards I could have as well payed $300 for a working one -) Luckily i only payed less than $20 for the card together with a ZIP GSX (defective too). I hate myself right now, because the card worked until i couldn't get enough speed. Only thing that could be left is the cachboard, one of the SMD-ICs on the Transwarp or the CPU-cable. No matter what oscillator i tried (tested 24mhz, 25mhz, 28mhz, 36mhz,48mhz,54mhz,64mhz and 66.6mhz). The crashes happen after about 20sec.-5min. I think the Cacheboardmay have died at one time or another. ![]() But neither replacing them with the original ones nor the GALs, CPU etc. At that Time the Transwarp would sporadically freeze and show either error 301 or 302. I realized, that the SRAMs were the limiting factor. I the replaced the GALs against faster ones, again the card would work nicely, but now at up to 12.5mhz. ![]() After replacing the SRAMs, the CPU and the oscillator the card would run nicely at 9mhz. There had been modifications on the cache-board, but one of the SRAMs had the wrong size. Hi, about a Month ago i got a defective Transwarp GS.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |