WP34s flashing procedure

[MarkHaysHarris777]

(02-27-2015 08:05 PM)John Galt Wrote:  I really don't understand what Marcus means when describing a "strobe" function; it's probably attributable to my understanding of the terminology or regional English differences.

With all due respect, the paper clip in the back of the unit is doable (I did it even) but its dangerous; words to the wise.

First let me tell you what I don't mean, regarding strobe. Strobe sometimes is used to refer to a signal that is momentarily received (by several gates) in a multiplexing environment to 'select' (or write enable) the particular chips being 'strobed' (because *all* the chips are on the same bus or parallel com lattice and some way is needed to tell the chips which 'one' is being written to now. ( I do not mean strobe in that sense, obviously, I think).

When I refer to strobe I mean a 'clean' leading edge pulse (rising from low to high over time, usually very shot time) that is used to communicate an 'ON' condition or some other input to a MOS or CMOS input (but can also be TTL). Versus a trailing edge pulse (also clean) running from high to low over time (again short time) normally referred to as a ground shot, in TTL logic.

When I speak of strobe I mean a very clean noiseless high signal that is held for some time and then released, as in the 'erase' strobe of the WP34s.

This is as 'apposed to' a noisy (non contact bounce conditioned ) electrical push button switch (finger actuated) which creates for the MOS or CMOS input what amounts to an electrical train wreck on the capacitive input; the worst for the condition is a damaged MOS gate, with the probable resulting condition being erratic behavior or non behavior. To put it simply a bare metal contact switch should (bold here, NEVER) not be used to signal MOS or CMOS gates. The 20b-30b flash cable (the one commercially produced for HP, I'm not knocking anyone present in the DIY community) is an over-simplified electrical possibility which is going to be unreliable and which may do damage, IMHO. But hey, HP was trying to provide to the community a least expensive 'probably-will-work-most-of-the-time' setup (most people are not going to invest in a JTAG development system, nor are they likely to build their own flash-box--probably not.

The preceding discussion does not change the harsh reality that (although the micro-controller is fairly robust) there isn't much protection on the Tx Rx pins, and the commercial flash cable does really nothing to help the user understand what is happening in the process (it either works or it does not work, and if it does not work, oh well---try again).

This article about contact debounce conditioning is an excellent read on debouncing MOS and CMOS inputs, the various hardware options for same, and why its important.

Cheers,
marcus