3
\$\begingroup\$

The Altera CycloneIII starter kit that I have provides a jumper (J6) that is connected to VCCINT of the CycloneIII FPGA core for what seems to be the purpose of measuring the core's power consumption.

Here is the reference page for the kit: http://www.altera.com/products/devkits/altera/kit-cyc3-starter.html

What I had intended to do was to take measurements from this jumper with a scope (voltage across the jumper should be proportional to power) - but I'm afraid that the readings from this jumper may not be as straight-forward as I had originally anticipated.

More specifically - after looking at the pin-out for the CycloneIII FPGA, each "bank" in the FPGA has its own VCCINT. I'm curious as to how these relate to the VCCINT that is said to be tied to J6.

How representative will this reading be?

Any insights?

\$\endgroup\$
5
\$\begingroup\$

VCCINT is indeed the power to the logic core of the chip. The "banks" refer to the I/O pads, which are divided into groups; each I/O bank can have a different supply voltage for compatibility with different kinds of external interfaces. Those bank VCC connections supply power only to the I/O pads, and have nothing to do with the core logic supply.

To measure the core supply current, you're going to want to remove the jumper and wire the two pins to a low-value shunt resistor. Use your scope in differential (A - B) mode to measure the voltage drop across this resistor, from which you can infer the current. You might be able to solder an SMT resistor directly to the pins of the jumper (on the solder side of the board); then, reinstalling the jumper will simply short it out when not needed for measurements.

\$\endgroup\$
  • \$\begingroup\$ I forgot to mention that when the jumper is removed, VCCINT of the core is tied to a current-sense resistor. Looking at the definition of current-sense vs. shunt, for these purposes, seem to be relatively close (please correct me if I'm wrong) - so I think that will take care of the first part of this solution. For the differential measurement - I don't explicitly see a "differential mode" setting (I have an Agilent InfiniiVision) but there is a math function to subtract one channel from another - is this sufficient? Thanks! \$\endgroup\$ – kbarber Sep 21 '12 at 0:00
  • \$\begingroup\$ The "A-B" refers to the subtraction (differential) mode. Set it to this and connect one probe to each pin. The scope will display the voltage across the shunt resistor, and the current is V / Rshunt. \$\endgroup\$ – Oli Glaser Sep 21 '12 at 1:56
  • \$\begingroup\$ Okay - I assumed (A-B) was subtraction - I just wanted to be sure that there wasn't an explicit setting to place the scope in this mode and that using the calculation was sufficient. Do you happen to know if there is a difference between current-sense and shunt resistors? \$\endgroup\$ – kbarber Sep 21 '12 at 11:49
3
\$\begingroup\$

I can't access the schematic without downloading the whole board package, so this is not completely certain:

To measure the current you need to use a multimeter/ammeter on current setting and attach a lead to each pin, so the meter is in series with the supply. Or for dynamic readings use the shunt resistor and scope it as Dave mentions. Measuring the voltage to ground, just does exactly that, measure voltage.

It's likely that if they say this measure current for the whole core, then this is the main connection for all the VCCINTs (i.e. the current for all of them runs through this point) This can be confirmed by checking the schematic.

It's quite common to provide this on a dev board - for example the STM32 Discovery board I have here has such a jumper.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.