Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
08/11/06 14:54
Read: times


 		 #122056 - I don't think so
Responding to: ???'s previous message
Erik Malund said:
I think you are wrong. You can check if the cells are "charged enough" to read correctly TODAY, I see no way of checking if they are "charged enough" to read TOMORROW.

This is provided by design of the FLASH cell and the programming circuitry.

In fact, the "parallel programmer" which used to determine the voltages and the timing, is simply built into the chip. When you parallel program the FLASH, you simply read out the signal of this circuit, indicating if it finished or not. You even don't provide the programming high voltage (it IS needed for FLASH, even if it says "5V programming" or so), it is generated internally. If the chip has a high voltage input for programming, it is only a "enable" logic signal and only a few uA is drawn from it (check the datasheets).

Believe me or not, this is the way how it is.

Erik said:
Actually, according to a realiable source, the reason Philips stopped providing parallel programming info was a lawsuit as to chips losing the code which was eventually traced to the programming being too 'soft'


I'd acknowledge this, but I'd believe this was in the times of EPROM - and they continue in this practice simply because it is convenient.

As I said, in those times (and maybe for early FLASH, too) the timing and voltages were actually determined by the parallel programmer. Underprogramming DOES result in problems with the declared retention, really, but the same applies for overprogramming; and also overerasing (I experienced that, but it was really a brutal UV dose, using an improper "homemade" eraser).


Jan Waclawek

List of 53 messages in thread
TopicAuthorDate
AT89s52 error with programmers            01/01/70 00:00      
   the 'weekly' question get the weekly answer            01/01/70 00:00      
   Add some pull up resistors (4.7K\'s) on LS device.            01/01/70 00:00      
      risque business            01/01/70 00:00      
   Home made JTAG            01/01/70 00:00      
      It is so refreshing to see thing like this instead            01/01/70 00:00      
         Tested!            01/01/70 00:00      
            have you 'tested' that the flash caps are fully            01/01/70 00:00      
               Atmel does that            01/01/70 00:00      
                  I think you are wrong            01/01/70 00:00      
                     I don't think so            01/01/70 00:00      
                        yes, and no            01/01/70 00:00      
                           Bah, shoot that designer :-)            01/01/70 00:00      
                              not the chip designer            01/01/70 00:00      
                                 JTAG x Flash Data Retention            01/01/70 00:00      
                                    nope            01/01/70 00:00      
   Tnx for answers...!            01/01/70 00:00      
   89s52 programmer            01/01/70 00:00      
      Only LS chips            01/01/70 00:00      
         there is the mail            01/01/70 00:00      
            here is world´s end            01/01/70 00:00      
               However far it from nowhere, I am sure you have ma            01/01/70 00:00      
                  You can buy here!            01/01/70 00:00      
                     he can do it himself...            01/01/70 00:00      
            What more fast than 9 ns?!            01/01/70 00:00      
               I do not understand. What I said is not to find s            01/01/70 00:00      
         Then stop doing in electronics!            01/01/70 00:00      
   re: AT89S52 error            01/01/70 00:00      
   Somebody test buffer CD4010?            01/01/70 00:00      
      the CD40 is sloooooooow compared to HC            01/01/70 00:00      
         What more fast than 9 ns?!            01/01/70 00:00      
            Not the speed is reason only            01/01/70 00:00      
   But and the pull ups in buffer´s out?            01/01/70 00:00      
      Does that suggest anything to you?            01/01/70 00:00      
      Type "74VHC244" into the search window            01/01/70 00:00      
         show the calculations            01/01/70 00:00      
      Schematic of programmer using 74LS244            01/01/70 00:00      
         Circuit that i am using            01/01/70 00:00      
            the very reason            01/01/70 00:00      
               Right...!            01/01/70 00:00      
                  have you conmsidered a 'radical' approach            01/01/70 00:00      
                     RX2...?!            01/01/70 00:00      
                        Rx2            01/01/70 00:00      
            Try this one            01/01/70 00:00      
               a full circle            01/01/70 00:00      
                  Time for a decision...            01/01/70 00:00      
               I did it...!            01/01/70 00:00      
                  Can\'t you order the parts by mail?            01/01/70 00:00      
      the old STUPID "It is tested"            01/01/70 00:00      
         Gee, another BS answer from Erik.            01/01/70 00:00      
            I can BE QUITE specific about what I post            01/01/70 00:00      
         "tested" does not mean tested            01/01/70 00:00      
            and even then ...            01/01/70 00:00      

Back to Subject List