Using several EEPROMNs to have more than 8 programs

[bcnx]

Hi,

I would like to have more than 8 programs. I noticed that https://serwis.avt.pl/manuals/AVT5215.pdf uses two 24LC128 EEPROMS but I'm not sure how this works (and I also don't speak Polish), How should I go about this? What do J5 and J7 do in the PDF?

thanks,

BC

[bcnx]

OK, I put some of the PDF in Google Translate and apparently the author says he uses just one of the EEPROMs at the same time. I suppose he uses DIL sockets. J5 seems to be for his EEPROM programmer.
So at this point it does not seem to be possible to have more than 8 programs, is this correct?

Cheers,

BC

[frank]

You can have more tan 8 but you need to disable one EEPROM so that one of the address pins is at +3.3V as the FV-1 reads the device where A2=A1=A0=0V

After changing the A pin you need to make the FV-1 read in the program from the enabled EEPROM by changing the S pins to the new program. Of course if the same program number is desired (i.e. program 0) and you only changed an address pin the FV-1 will not detect the address pin change so you need to load any other program then go back to the desired program.

[bcnx]

Hi Frank,

makes sense, thanks.
The problem of the same program might be fixed by a rotary switch to select programs. To get to a different program you will switch up or down to another program and it will always be different from the original program. But do correct me if I'm wrong.

Cheers,

BC

[bcnx]

You can have more tan 8 but you need to disable one EEPROM so that one of the address pins is at +3.3V as the FV-1 reads the device where A2=A1=A0=0V

After changing the A pin you need to make the FV-1 read in the program from the enabled EEPROM by changing the S pins to the new program. Of course if the same program number is desired (i.e. program 0) and you only changed an address pin the FV-1 will not detect the address pin change so you need to load any other program then go back to the desired program.

Currently doing this, but how do I avoid timing issues? I need to have this order I suppose:
- disable EEPROM 1
- enable EEPROM 2
- change the FV-1 program number

I was planning to use a rotary switch to select programs. Will the time needed for the FV-1 to change programs be sufficiently high to be able to switch EEPROMS?

Cheers,
BC

[frank]

Currently doing this, but how do I avoid timing issues? I need to have this order I suppose:
- disable EEPROM 1
- enable EEPROM 2
- change the FV-1 program number

I was planning to use a rotary switch to select programs. Will the time needed for the FV-1 to change programs be sufficiently high to be able to switch EEPROMS?

That is really something you will need to look at the datasheet for the EEPROM to answer. The FV-1 will start reading as fast as 1/2 of a sample period after a program select pin changes. You also need to consider bounce on the signals from the switch as you need to make sure the disable/enable happens and is stable prior to changing the program select number. This could be done with simple RC making the RC for the program select slower than the enable/disable.

[bcnx]

Hi Frank,

I was indeed thinking about adding an RC network to avoid bouncing. Now you mention it, it dawns on me the RC can also be used to slow down the program select enough to allow the EEPROMs for being switched. I think I need to use some sort of flip-flop that addresses one EEPROM while the other one is "un-adressed".

BC

[frank]

I think I need to use some sort of flip-flop that addresses one EEPROM while the other one is "un-adressed".

A simple inverter should do, one EEPROM is addressed by the debounced switch and the other by the debounced switch through the inverter. As long as A0 (or which ever pin is convenient) is stable prior to SDA/SCL then it should be good.

[bcnx]

I think I need to use some sort of flip-flop that addresses one EEPROM while the other one is "un-adressed".

A simple inverter should do, one EEPROM is addressed by the debounced switch and the other by the debounced switch through the inverter. As long as A0 (or which ever pin is convenient) is stable prior to SDA/SCL then it should be good.

I currently have a rotary switch connected to a 74HC148, which connects to the select lines of the FV-1. The common ping of the rotary switch is connected to ground. Does bouncing still occur in this case? The 74HC148 does have resistors connected to the power supply at the inputs.

UPDATE: I've been told in another forum that contact bouncing would not be an issue, so I guess I only need to add some sort of delay.

BC

[bcnx]

Small update, I did not find any way to combine the outputs of 2 74HC148 ICs, so I decided to switch to a Arduino Nano so I can deal with the timing in software.

BC