Intel® FPGA University Program
University Program Material, Education Boards, and Laboratory Exercises
Announcements
Intel Support hours are Monday-Fridays, 8am-5pm PST, except Holidays. Thanks to our community members who provide support during our down time or before we get to your questions. We appreciate you!

Search our FPGA Knowledge Articles here.

## Measure Power on DE2

Honored Contributor I
1,267 Views

Hi everybody,

I am using Altera DE2 board for the implementation of some VHDL designs.

I wonder is it possible to find the way to MEASURE the power consumed by the design?

Could I, hypothetically, insert the resistor between the FPGA power supply (VCCINT) and FPGA device and by measuring the current through the resistor to caclulate the power ?

http://www.mypicx.com/01312012//

This means to unsolder the TC11 capacitor and switch it with the serial connection of the same capacitor and one jumper. Connecting the resistor on the jumper and measuring the voltage on it I will be able to calculate the current my design will consume (see the attachment).

Please discuss with me and share your point of view. I would be very grateful to you.

Best regards,

Bojan

7 Replies
Honored Contributor I
69 Views

--- Quote Start ---

This means to unsolder the TC11 capacitor and switch it with the serial connection of the same capacitor and one jumper. Connecting the resistor on the jumper and measuring the voltage on it I will be able to calculate the current my design will consume (see the attachment).

Please discuss with me and share your point of view. I would be very grateful to you.

--- Quote End ---

The way you have drawn will not work.

The voltage drop across R53 is what you need to measure. You need to replace that resistor with something low-valued, so that you measure a voltage across it. How do you know the value?

1) Use the PowerPlay analyzer to get an estimate of your maximum design current, Imax.

2) Determine the maximum allowable drop in the core supply, eg., 1.2V core probably has a maximum drop of dV = 5mV or so (read the handbook to get the actual value).

3) The resistor value is something smaller than R = dV/Imax.

4) Measure the voltage difference using; two channels of an oscilloscope, a differential probe, or an instrumentation amplifier, or a current-sense amplifier.

Cheers,

Dave
Honored Contributor I
69 Views

Dear Dave,

Thank you so much ! Thank you from the bottom of my heart. Your advices were so precious to me.

I understand you very well. It is now up to me to switch R53 with the jumper and to find out the allowed drop of the actual 1.2V core supply.

Dividing the measured differential voltage on the new R53 resistor with its nominal value I obtain the current value (Idesign).

Idesign*1.2V=Ptotal_measured ! Am I right ? :D

Cheers mate,

Bojan
Honored Contributor I
69 Views

--- Quote Start ---

Dividing the measured differential voltage on the new R53 resistor with its nominal value I obtain the current value (Idesign).

Idesign*1.2V=Ptotal_measured ! Am I right ? :D

--- Quote End ---

Yes.

Here's what I have done in the past;

1) Create a design with a block of control registers with control bits.

2) Route a control bit to a block of registers that can be enabled to toggle (50-percent duty cycle).

3) Write a Modelsim simulation to generate a .vcd file for various numbers of toggling registers enabled, and use Quartus to provide power estimates.

4) Download the design to the board, and enable various numbers of toggling registers, and measure the power.

5) Compare (3) and (4) to see how good the estimates are.

I think I saw an agreement within the 10% to 20% range.

Cheers,

Dave
Honored Contributor I
69 Views

OK, thanks for the suggestions. Just one little thing... did you use differential probe, instrumentation amplifier or current-sense amplifier ? :rolleyes:

Honored Contributor I
69 Views

--- Quote Start ---

OK, thanks for the suggestions. Just one little thing... did you use differential probe, instrumentation amplifier or current-sense amplifier ? :rolleyes:

--- Quote End ---

I had Texas Instruments INA209 I2C current sense devices wired on the board.

If I wanted to do your measurement on the DE2, I'd use a pair of scope probes.

Cheers,

Dave
Honored Contributor I
69 Views

My idea is to use some oscilloscope with the differential probe.

1) To apply some 1000 input values to the design, to measure differential voltage and to save it in .txt format.

2) Change the probe ends and repeat the procedure.

3) From the saved measured values (in .txt files) calculate MEAN voltage value (Vr_mean)

4) Idesign=Vr_mean/R

5) Pmeasured_total=Idesign*1.2V

6) :cool:
Honored Contributor I
69 Views

--- Quote Start ---

3) From the saved measured values (in .txt files) calculate MEAN voltage value (Vr_mean)

--- Quote End ---

My scope has a math function that allows me do perform the averaged difference directly :) ... Your method will work too.

Cheers,

Dave