| ??? 07/26/07 15:06 Read: times |
#142352 - the lpc9xx is the glaring BAD example... Responding to: ???'s previous message |
... of how this relatively simple task can be #$%&^$% up.
P89LPC932A1 ERRATA SHEET Version 1.1, 2005 Aug 04 said:
To generate a proper Power-On-Reset (POR), VDD must have dropped below 0.2V before being powered back up.
Power-cycling without VDD having dropped below 0.2V may result in incorrect Program Counter values. Please also see the VPOR specification in LPC932A1 Datasheet, DC electrical characteristics. Section 8.15 (Reset) states that during a power cycle, VDD must fall below VPOR . Be also sure that in the LPC932A1 Datasheet (nor in the User Manual - an another example of NXP's datasheet guy's idiocy), there no "DC electrical characteristics" section, no section 8.15, and no single mention of VPOR. But the datasheet says the following - and they were scared to death by some of the importand high-volume customer so that they put it there several times: "P89LPC932A1 Datasheet Rev. 03 — 12 March 2007 said:
When using an oscillator frequency above 12MHz, the reset input function of P1.5 must be enabled. An external circuit is required to hold the device in reset at power-up until VDD has reached its specified level. When system power is removed VDD will fall below the minimum specified operating voltage. When using an oscillator frequency above 12MHz, in some applications, an external brownout detect circuit may be required to hold the device in reset when VDD falls below the minimum specified operating voltage. (I even was so brave and dig into nxp.com to check for the new version of the datasheet, the link is easy and straightforward, http://www.nxp.com/#/pip/cb=[type=product,path=/50809/45992/45995,final=P89LPC932A1_3]|pip=[pip=P89LPC932A1_3]|[5] - when will they finally fire that stupid webmaking group/company? - sure they have there an older errata than I have from the very same website... OK but that's not the point. The new version is there but the only change is that they find/replace-d Philips for NXP... :-( When will they finally fire the guy who writes the '51 datasheets?) This, combined with the weird requirement of not holding down the pin shared between reset and an extra input upon powerup, renders the whole scheme unusable. But what #%^&%^ me more than this fact is, that Philips/NXP pushes its head into sand, and instead of saying out loudly: "look, people, this is bad, so please use a reset IC if it's not a toy", they insist on the designers making provisions for VDD to fall below 0.2V! Isn't this ridiculous - the same argument - the designer should have full control over VDD - just in a different setup... A related post http://www.8052.com/forum/read.phtml?id=99079 and the whole that thread. JW |
