| ??? 05/18/08 23:18 Read: times |
#154870 - In this case, it is the antiques ... Responding to: ???'s previous message |
Erik Malund said: I'd be very interested in where and how you got that information! Richard Erlacher said:
For exactly that reason, I substituted Intel, AMD, Signetics, Philips MCU's ... P8032 MCU ... which I believe is a 'dynamic version' which DEFINITELY will show 'something funny' when going into reset. Please forget the antiques and stay with modern parts. Unfortunately, those "modern" parts aren't the ones that interest me. I'm required, once in a while, to support some equipment that I installed back in the '80's, which uses the i8032. But for the fact someone else started that project, I'd never have allowed the use of an MCU with a positive-going RESET. That Philips part is a P80C32 ... 24 Mhz, I think. It had a date code in the '90's somewhere. The first thing the Philips datasheet said:
is, "The Philips 80C31/32 is a high-performance static 80C51 design fabricated with Philips high-density CMOS technology with operation from 2.7 V to 5.5 V." Richard Erlacher said:
Which NXP processor would you recommend? e.g. the P89V51Rx2 which is UART ISP. Also it is a 6 clocker which brings it close to your DS part. Not really very close. The DS89C4x0's are one-clockers. That allows me to leave the clocks alone and yet run much faster simply by replacing the on-board 2764's with CY7C261's (35 ns) and the 120 ns SRAMs with 15 ns parts, for both of which conversions I have long had adapters. Until I add the code to change the external bus timing, the DS89C4x0 is a four-clocker when using the external memory bus. That's still 3x as fast as the "standard" timing allows. I prefer 2-clock speed, though I have considered using fast SRAMs in place of the EPROMs and copying the EPROM content into them. I could, of course, copy internal memory content there as well, and then cause a change in the state of nEA and a RESET, which would change everything. I haven't yet had to do any of that, though. I'm interested in your comment once made about using a 10K-ohm resistor between two sources of RESET. I don't see how this will help, since there's at least one pulldown resistor, namely that within the MCU, and that's going to prevent the RESET signal from going high enough. Can you shed a bit more light on that? I have that worked out somewhere at work, but am currently working from home. Erik From what little "playing" with Vcc I've done, aside from that one long-term test that showed the problem I'm chasing doesn't occur when Vcc rises faster than 10 ms and falls very fast, I've not yet found a set of reproducible conditions that routinely send the MCU into runaway during Vcc decay. Sometime soon, as time allows, I'm going to go back to that original circuit board, which has a broken (meaning broken off) component that I have to replace, and, being a small and very dense wire-wrapped board of ancient vintage, it will take some time to repair. It has only a few square millimeters of space remaining on the board. However, I should be able to recreate that failure there ... it has a couple of 330 uF caps on the VCC rail, which will ensure that the rise time is long and slow, (all the latches and gates are CMOS) and that the Vcc decay time is even longer. I'll just have to trigger on nWR during RESET to capture the event. That shouldn't be hard. I really can't envision any other way in which the BBRAM could become corrupted. I will monitor the behavior of Vcc as well as the control signals, though. RE |
| Topic | Author | Date |
| Problem with NVRAM - DS12887 | 01/01/70 00:00 | |
| Background | 01/01/70 00:00 | |
| So where's the schematic, then? | 01/01/70 00:00 | |
| No schematic | 01/01/70 00:00 | |
| till you show your schematic no answers | 01/01/70 00:00 | |
| Which MCU type? | 01/01/70 00:00 | |
| Which MCU type? | 01/01/70 00:00 | |
| That base has been covered ... sort-of ... | 01/01/70 00:00 | |
| comments, answers | 01/01/70 00:00 | |
| In this case, it is the antiques ... | 01/01/70 00:00 | |
| you forgot the 'C' in your first post | 01/01/70 00:00 | |
Yes I did ... sorry about that .... | 01/01/70 00:00 | |
| Where did you buy it from? | 01/01/70 00:00 | |
| What??? | 01/01/70 00:00 |