Can I give feedback too?

I won't wait for your approval. :-)

1) I would use a reset generator chip. R/C circuits will bring problems with brownouts and voltage flutter later when you can least affort to muck around with them. An R/C seems easy now but wait till your debugging some strange problem that starts to render you bald (becasue you are pulling it out) and you wish then you had followed what we are saying now. I don't know your advisor but unless he has years of real full scale production product design with modern parts under his belt he is hardly the person to give you the final word on R/C reset circuit applicability. I recall a LONG discussion on this subject in the recent past here on the forum and it is pretty much the agreement that academia tends to promote the continuance of old fashioned ideas that were tried for 8051s from a couple of decades ago.

2) You should bench check your particular LCD module Vee voltage requirements. I have worked with many a module that just could not achieve a suitable display bias with Vee equal to -5V (as referenced to the Vcc pin of the LCD module). I have found time and again that a bias toward -7V (referenced to Vcc of module) is more applicable unless you want a washed out dim display. You can get the supply rail for this, which replaces your current GND connection on your 10K pot, by one of several ways. A) Use a 7660 chip in 8-pin package to invert +5 to -5 volts, B) Tap off the V- pin of an RS232 transceiver chip, or C) Provide a negative supply input. Another option is to tie a timer output port pin, toggling at a 1KHz or so rate, into a capacitor / diode voltage trippler circuit. This scheme does not work well with standard 8051 type I/O pins and would only be applicable if your chip is similar to the Cygnal parts where a pin can be programmed to a push-pull output.

3) If the primary battery has a short lived capability to deliver power to the 7805 relative to the length of time that you want to keep the time in the RTC valid you may want to consider adding a second backup battery in the form of a lithium coin cell to the RTC power connection. If you use a NON rechargeable lithium battery of 3.6 volts this can be added to the RTC with the simplest of circuits using two low forward drop low leakage schottky diodes. The RTC chip also has an Aux battery capability that you may want to use. (I see after I wrote this that the RTC chip has an integral battery so maybe ignore this comment unless you also want to provide for use of the DS17885 chip which may very well be cheaper in a production situation. Of course then you also need to have a crystal too).

4) The RTC chip uses a multiplexed bus for its address and data lines. Address moves along with ALE and data with RD/ and WR/ lines. You will need to provide another port pin connection to the ALE pin so that your bit bang access routine for the RTC can transfer the address to the RTC. In this same discussion, while you are bit banging this connection the port 0 lines will be running open drain unless your processor has push-pull programmable I/O pins. It may be that you will need pullups on these lines.

5) The RTC chip IRQ line is an open drain output. I would recommend a 10K pullup on the corresponding INT0 line going to the processor to prevent possible false re-trigger of the 8051 interrupt at the time of slow rise time of the signal if just the weak pullup of the processor is used. This false retrigger could occur due to a noisey environment on your board.

6) If you do not intend to use the KS/ pin of the RTC chip (Kick Start pin) it should be best be tied to a high level.

Michael Karas