Hi, I suggest to focus on the 1V2 supply first, since only the FPGA core (VCCINT) is using it. Disconnect all other supply voltages to reduce the possibility of complex interactions between several components. The FPGA should be fine with having only the 1V2 voltage applied and should remain in power-on-reset (POR) mode, where it draws minimal current (UG394 p.32).
As Uwe suggested, this might also be simply a problem with the current limited supply and the regulators. So here's a simple fix: remove L12 and then inject directly the 1V2 into the board with your current-limited supply. Tell me if there is still an over-current condition then - if there is not, do the same thing with 3V3, then put back all the ferrite beads into place and it should be safe to power up everything for real :) By the way, the Spartan 6 documentations strongly advise against a non- monotonic (i.e. with dips) rise of the supply voltages, so you may also want to check with an oscilloscope what the power-up ramps look like. Sébastien On Wednesday 16 June 2010 08:09:23 Adam Wang wrote: > Hi, > > The static current is too high (I first very carefully limited the > power source to be only 0.1A, then step by step set to 0.5A) when I > plugeg +5V in, but it still turns into CC mode on my power supply. > It's not normal. > > So I did steps on #1 pcba are as: > > 1, Still set the limited current to as 0.1 Amp only, take apart > L7(3V3), L8(2V5), L9(1V8), L12(1V2), L2(AUDIO_AVDD) > then meausre all outputs of U10, U11, U12, U13 without any loads, > then all voltage results are good. > 2, so step by step, shorting the FBs back on L8, L9, L2; I got: > > only L9 (1V8), the static current = 0.07A, reasonable/acceptable > added L8(2V5), the total static current increases up into 0.08A, > ok, means only 0.01A 2V5 on roughly fpga and etc., good. > > 3, but when I added L12 or L7, high current over 0.15A(I clamped at > power supply), so I am still checking why 1V2 and 3V3 got high > current... > > 4, #4 are also the same status, so at least they have the same phenomenon. > I checked all 1.2V connections/traces of U22 fpga on the bare pcb > are all good. > > So you can give me some comments, I am still debugging.. > > P.S.: the results of X-Ray inspection is uploaded: still under editing... > http://en.qi-hardware.com/wiki/Milkymist_One_SMT/DIP_Process_Flow#X-Ray_Ins > pection.5B9.5D tks, > Adam > _______________________________________________ > http://lists.milkymist.org/listinfo.cgi/devel-milkymist.org > IRC: #milkym...@freenode > Webchat: www.milkymist.org/irc.html > Wiki: www.milkymist.org/wiki > _______________________________________________ http://lists.milkymist.org/listinfo.cgi/devel-milkymist.org IRC: #milkym...@freenode Webchat: www.milkymist.org/irc.html Wiki: www.milkymist.org/wiki
