RC Circuit

[Balakumaran Selvaraji]

Hello again  ,
      I now moved onto making the output of my computer by using some EEPROM's. So, the datasheet says that the write time to the EEPROM must be between 100 and 1000 nanoseconds. To accomplish that, I need to built an RC Circuit with a tactile switch so that with the push of a button, the value would be stored in the EEPROM. So, I could easily achieve that, lets say I use a 1nf(nano farad) capacitor and 680 ohm resistor would give me about 680 nanosecond. But unfortunately, I don't have 1nF capacitor and would like to know that I have various kinds of resistors but only 0.1microfarad and 1microfarad capacitors.... So, can you please tell me a circuit using the limited supplies to make an RC Circuit which give me a Write Pulse Width between 100 to 1000 ns.

Thank You.

[danadak]

Are you using PC to write to EEPROM ?

What type of EEPROM are you using ? I2C or Parallel or SPI interface ?

What is controlling the process, a PC or a UP ?

What is source of data being written to EEPROM ?



Regards, Dana.

[Balakumaran Selvaraji]

I am not using a PC to write the EEPROM. I am using the ATMEL 28C16 Parallel EEPROM. I connect some led's to the eight I/O Lines of the EEPROM and modify data using some jumper wires and hit the button to store that value in it.

[danadak]

Ignore the name of this website.

Shows using a 555 time with a P-B to generate the pulse.

Using just a RC may cause problems because of Trise of the
pulse, thats why using a circuit designed to be a logic pulse
generator is best way.


Regards, Dana.

[Balakumaran Selvaraji]

Yeah, but I am just using this circuit just to see what is in my eeprom and I dont want to build a logic generator circuit again. Because, I am going to be programming with Arduino Nano and I will insert the programmed eeprom in the circuit to see whether I get the necessary output.

[danadak]

Then use the Arduino to detect the push button and generate the
pulse.

That's an even better approach than the 555, no discrete or external
components needed except for push button.

Regards, Dana.

[Balakumaran Selvaraji]

So, I have a doubt : the time constant ? is given as:   T = R x C :
So If I need to generate a pulse width of about 100ns to 1000ns. What R would I need so that C =0.1uf (or) 1uf.

[Zero999]

So, I have a doubt : the time constant ? is given as:   T = R x C :
So If I need to generate a pulse width of about 100ns to 1000ns. What R would I need so that C =0.1uf (or) 1uf.

Forget the RC circuit. Use the microcontroller to generate the pulse.

[rstofer]

Here's an Arduino programmer:
http://forum.6502.org/viewtopic.php?f=4&t=3156

What I don't see it the V_H (12V + 0.5V) for erasing the chip.  I guess that must be done elsewhere.  It too has a timing requirement but it only takes one shot to erase the entire chip.

As to how many Tau you need, well, you only get to 63% output in 1 Tau and you get to 99.99% in 6 Tau.
Equation is:

Vout = Vin * (1 - e^-t/T) where T = RC and t is time, both in seconds.

http://www.electronics-tutorials.ws/rc/rc_1.html

This is a particularly poor way to program a device.  In fact, I have no idea what you're going to do with this exponential rising edge.  You still need a defined period and a fallling edge.

What you need is a one-shot or monostable multivibrator perhaps using a 555 timer.

http://www.ohmslawcalculator.com/555-monostable-calculator

[David Hess]

... lets say I use a 1nf(nano farad) capacitor and 680 ohm resistor would give me about 680 nanosecond. But unfortunately, I don't have 1nF capacitor and would like to know that I have various kinds of resistors but only 0.1microfarad and 1microfarad capacitors.... So, can you please tell me a circuit using the limited supplies to make an RC Circuit which give me a Write Pulse Width between 100 to 1000 ns.

So time = k x R x C.  In your case, 680ns = k x 680 ohms x 1nF so k = 1.  Note that typically k = 0.7 or 2.2 depending on the logic circuit.  For the same time constant , we can simplify to R1C1 = R2C2.  680 x 1nF = R2 x 100nF so R = 6.8 ohms when C = 0.1uF (100nF).  That is too low of a value to be practically driven by a normal gate.  You could place 10 0.1uF capacitors in series to get 0.01uF so a 68 ohm resistor could be used.  Or use a different circuit.

[rstofer]

Maybe the 555 isn't such a good idea as the minimum pulse width is 10 us.

http://www.ti.com/lit/ds/symlink/lm555.pdf

See figure 13.

I would try to see if it is possible to get a 1us pulse from an Arduino.

The 74ls123 can generate a narrow pulse.  Look at line 5 of the truth table.  A rising edge on B1 causes a pulse with the other pins as shown

http://www.ti.com/lit/ds/sdls043/sdls043.pdf

I doubt that you can get away with scratching an input for something that puts out such a narrow pulse.  You are likely to get 100s of triggers.  I would use a ring tailed NAND gate on a SPDT switch to trigger the one-shot.

[danadak]

My aplogies on 555 timer recommendation, its min PW in one shot mode does exceed your max desired
PW.

Same seems to be true for Arduino, unless you want to code that part in machine language. The
Arduino language PWM for stuff like LED fading, not fast precise pulse generation.


Possibly a schmidt based approach - https://www.fairchildsemi.com/application-notes/AN/AN-140.pdf
use tiny gate logic for the schmidt element.


Or PSOC 4, see attached. This generates a ~ 1 uS pulse, can be shorter if you wish, with a resolution
of ~42 nS.



Regards, Dana.

[David Hess]

There are all kinds of ways to make gate based monostable multivibrator and pulse circuits.

http://www.electronics-tutorials.ws/sequential/seq_3.html

[Balakumaran Selvaraji]

I have made the RC Circuit by using 0.01uF Capacitor and I am able to write to the EEPROM properly.

Thank You.

[Zero999]

Maybe the 555 isn't such a good idea as the minimum pulse width is 10 us.

http://www.ti.com/lit/ds/symlink/lm555.pdf

See figure 13.

I would try to see if it is possible to get a 1us pulse from an Arduino.

The 74ls123 can generate a narrow pulse.  Look at line 5 of the truth table.  A rising edge on B1 causes a pulse with the other pins as shown

http://www.ti.com/lit/ds/sdls043/sdls043.pdf

I doubt that you can get away with scratching an input for something that puts out such a narrow pulse.  You are likely to get 100s of triggers.  I would use a ring tailed NAND gate on a SPDT switch to trigger the one-shot.

That's sensible, but I'd recommend the 74HC(T)123 for new designs.

Here's an interesting app-note about the 74LVC1G123.
http://www.ti.com/lit/an/slva720/slva720.pdf