LVCMOSS33 I/O sink current

I am using spartan3 fpga with lvcmos33 output pin set to 16 mA drive strength and fast slew rate. I am assuming that 16 mA is the source current at 3.3V as my VCCO voltage is 3.3V. I would like to know what is the sink current capability for LVCMOS33 OUTPUT pin is. Or is there any way I can increase the sink current capability?

I am using this output pin as open drain type with 150 Ohm pulllup to 1.8V Plane. With this I am seeing low swing of 400mV. If the sink current was truely

16 mA, then the low swing should have been 0v. I raised the pullup value to 330 Ohms and the low swing was 200mV.

In my opinion the 16 mA is source current capability. I need to make the output go down to 0v.

Any help will be great.

Reply to
Test01
Loading thread data ...

and fast slew rate. I am assuming that 16 mA is the source current at 3.3V as my VCCO voltage is 3.3V. I would like to know what is the sink current capability for LVCMOS33 OUTPUT pin is. Or is there any way I can increase the sink current capability?

Plane. With this I am seeing low swing of 400mV. If the sink current was truely

16 mA, then the low swing should have been 0v. I raised the pullup value to 330 Ohms and the low swing was 200mV.

output go down to 0v.

Reply to
Peter Alfke

and fast slew rate. I am assuming that 16 mA is the source current at 3.3V as my VCCO voltage is 3.3V. I would like to know what is the sink current capability for LVCMOS33 OUTPUT pin is. Or is there any way I can increase the sink current capability?

Plane.

With this I am seeing low swing of 400mV.

If the sink current was truely 16 mA, then the low swing should have been 0v.

I raised the pullup value to 330 Ohms and the low swing was 200mV.

output go down to 0v.

CMOS drive, Sink current is not a pure current source, it is spec'd at some Vol level, and for a given PinDrive selection it will sink MORE than that level. Real CMOS drivers are MOSFETS that look resistors, in the sub-volt saturation region. In your example, the 400mV is the expected drop across that resistor, and that is why it drops to 200mV when you halve the load current, by 150 => 330 Ohms. What is this driving ? - 400mV maybe fine as a Vol, or you could go to a higher Drive option and lower that 400mV.

-jg

Reply to
Jim Granville

As per my understanding, you are suggesting an external transistor in parallel with the n-channel transistor to ground. The external transistor will turn on when the FPGA is driving low and will provide low resistance path to ground.

Also you are suggesitng to raise the sink current value to 24 mA. Currently I am using 16 mA driver. I am assuming that 24 mA sink current will some how lower the resistance of the n-channel mosfet in the FPGA when it is on and help reduce the voltage drop across it. It will be great if you can expand on this. I am assuming that this solution will not require any external transistor to ground.

Thanks for your feedback.

Reply to
Test01

with the n-channel transistor to ground. The external transistor will turn on when the FPGA is driving low and will provide low resistance path to ground. No, no, no. I am not suggesting an external transistor. I just wanted to give you an understanding of the CMOS output structure. A very basic understanding. Peter Alfke

am using 16 mA driver. I am assuming that 24 mA sink current will some how lower the resistance of the n-channel mosfet in the FPGA when it is on and help reduce the voltage drop across it. It will be great if you can expand on this. I am assuming that this solution will not require any external transistor to ground.

Reply to
Peter Alfke

Test01,

The voltage can not be made to go to 0 volts.

That would be a 0 ohm impedance for the driver.

There is no such thing as a FET with 0 ohms ON resistance.

So, realizing this, what voltage do you actually need when driving low?

Austin

Reply to
austin

am using 16 mA driver.

n-channel mosfet in the FPGA

Correct. 24mA has lower resistance MOSFET than 16mA

ground.

No external transistor is needed, it will decrease the 400mV loss to 2/3 of that. If that is still too much, you could parallel pins on the FPGA.

-jg

Reply to
Jim Granville

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.