Current to sink PROG_B low?

I have desingned a board with an Atmega128 and Spartan3, such that the Atmega128 configures the FPGA via the slave serial method.

In my current rev, the Atmega128 is powered by a 5V supply. The FPGA power supply is correct. The HSWAP_EN is tied low to ground. An general I/O on the microcontroller is tied to the FPGA pins.

Anyways, the problem I am having is that I am not able to pull PROG_B low to reset the FPGA. The miroprocessor pin goes low but it is not able to pull the PROG_B low at all?

The way I test this, is I have a JTAG port which I used to program the FPGA. After configruation the DONE signal goes high. So my understanding was that if I pull the PROG_B low from the microprocessor, I should be able to manually reset the FPGA and the DONE signal should go low.

I need to solve this problem before I move on to the actual programming of the board using the CCLK and DIN signals, to which I will have attach series resistors because of the Atmega128 (5V) and FPGA(3.3V).

Any suggestions or ideas appreciated.

Thanks

Reply to
Yaju Nagaonkar
Loading thread data ...

What do you really mean? The output pin goes Low, but the PROG pin does not? What is between these pins, more than a strip of copper? Pewter Alfke, Xilinx

Reply to
Peter Alfke

There was a 242 ohm resistor to limit the current into PROG_B since, the microcontroller is at 5V and PROG_B has to be at VCCAUX(2.5V).

I tried to remove the resistor and connected the PROG_B to the microncontroller I/O. The FPGA still would not reset when the Microcontroller I/O was held low. The FPGA resets only when I connect a wire from the PROG_B pin to GND directly.

Reply to
Yaju Nagaonkar

Measure the current in that wire with a milliampere-meter. That gives you a feel for the strength of the pull-up transistor or resistor that you must overcome. Also, there is no need for 242 Ohm. You want to limit the current between 5 V and 2.5V + one diode drop. 100 Ohm would limit it to max 18 mA, which is fine. Peter Alfke

Reply to
Peter Alfke

Measure the current in that wire with a milliampere-meter. That gives you a feel for the strength of the pull-up transistor or resistor that you must overcome. Also, there is no need for 242 Ohm. You want to limit the current between 5 V and 2.5V + one diode drop. 100 Ohm would limit it to max 18 mA, which is fine. Peter Alfke

Reply to
Peter Alfke

Measure the current in that wire with a milliampere-meter. That gives you a feel for the strength of the pull-up transistor or resistor that you must overcome. Also, there is no need for 242 Ohm. You want to limit the current between 5 V and 2.5V + one diode drop. 100 Ohm would limit it to max 18 mA, which is fine. Peter Alfke

Reply to
Peter Alfke

"Yaju Nagaonkar" schrieb im Newsbeitrag news: snipped-for-privacy@g43g2000cwa.googlegroups.com...

PROG_B is input only on FPGA with virtually no current, so if you nee strong pulldown to get it to low level there must be something else pulling it up, not FPGA

Antti

Reply to
Antti Lukats

Xilinx Website is a good source for information too...

formatting link

Reply to
jpdullius

Ok. It works now. I cleaned up all the solder from the pin connections and I disconnected a pull-up on the PROG_B pin. The pull up was tied to

2.5V(VCCAUX) vias 4.7K. It was a simple mistake which I should have fixed on my own. Basically I need to learn to solder things neatly.

I have already looked at the website which earlier JPdull has mentioned. I will move on to the other signals, names CCLK and DIN. Again, since my micro-controller (5v) is driving these signals, I will be using series resistors (300ohm) to limit the current to these pins.

Thanks everyone.

Reply to
Yaju Nagaonkar

Pewter? Time to lay off the beer, Peter.

Bob

Reply to
Bob

why not powering the atmel at 3.3V ? speed ? Aurash

Reply to
Aurelian Lazarut

That was an error in my design. I hope to run it at 3.3V in the next revision. That would have made things much easier.

Yaju

Reply to
Yaju Nagaonkar

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.