I have added in the MCLK but I have a strange issue and I'm not sure how to proceed.
The input for MCLK seems to be dragging down the line. I've checked for shorts with USB microscope but cannot see anything. The programming seemed to go ok and was the same project that I used with my previous PCM2DSD board, which works fine. The FPGA came from the same source and appears to look identical to the previous one.
Is it possible the FPGA pin is configured as output and driving low? (Possible failure to program properly?)
I can try a reprogram again this evening but I'm running out of ideas at that point...
I'm not familiar with impact software so I may have made some error in configuring. However, when it asked if I wanted to use the previous project I just accepted that and went with it and did a program, which seemed to program the flash memory and the FPGA but perhaps not? Maybe it only did the flash?
The input for MCLK seems to be dragging down the line. I've checked for shorts with USB microscope but cannot see anything. The programming seemed to go ok and was the same project that I used with my previous PCM2DSD board, which works fine. The FPGA came from the same source and appears to look identical to the previous one.
Is it possible the FPGA pin is configured as output and driving low? (Possible failure to program properly?)
I can try a reprogram again this evening but I'm running out of ideas at that point...
I'm not familiar with impact software so I may have made some error in configuring. However, when it asked if I wanted to use the previous project I just accepted that and went with it and did a program, which seemed to program the flash memory and the FPGA but perhaps not? Maybe it only did the flash?
Well, further to that last post, I have it working now. I reprogrammed it but still no joy. So I went between the pins with a very fine needle and now it is working, so all I can think is there was a very fine trace of solder that I couldn't see on the microscope but was there in reality. Or else I am losing my mind, which is also possible. 
Well, I spoke too soon. It was working fine and then suddenly started pulling way too much current and hitting the (deliberately low) current limit I'd set on my bench supplies.
I've been trying to figure out what went wrong but as yet not got to the conclusion. I am suspecting one of the tantalum caps may have failed but it could also be the ADM7150 which is getting very hot after just a few seconds.
I guess I will just have to remove capacitors until either the problem goes away or I have to conclude it might be the FPGA? Does the FPGA use the 3.3v in addition to the 1.2v? Is it likely to be a failure of the FPGA or a faulty cap or ADM7150?
I'd set FB2 instead of FB3 so the 1.2v regulator was fed directly from the 5v input rather than cascaded from the ADM7150.
Any advice in diagnosing please?
I've been trying to figure out what went wrong but as yet not got to the conclusion. I am suspecting one of the tantalum caps may have failed but it could also be the ADM7150 which is getting very hot after just a few seconds.
I guess I will just have to remove capacitors until either the problem goes away or I have to conclude it might be the FPGA? Does the FPGA use the 3.3v in addition to the 1.2v? Is it likely to be a failure of the FPGA or a faulty cap or ADM7150?
I'd set FB2 instead of FB3 so the 1.2v regulator was fed directly from the 5v input rather than cascaded from the ADM7150.
Any advice in diagnosing please?
Hi PJotr,
2nd photo was from before it was finished of course but only the headers fitted after.
I went over the FPGA pins again with the USB microscope but all look to be good. I am getting around 3.2v on pin 1 of the 7510 but nothing on pin 2, the output. So it looks like something is shorted on the 3.3v line. Perhaps I should remove C30 first?
Many thanks,
Tony
2nd photo was from before it was finished of course but only the headers fitted after.
I went over the FPGA pins again with the USB microscope but all look to be good. I am getting around 3.2v on pin 1 of the 7510 but nothing on pin 2, the output. So it looks like something is shorted on the 3.3v line. Perhaps I should remove C30 first?
Many thanks,
Tony
I don't see the polarity of C33 and C40. That's where I'd start.
C40 is connected to +3.3V. C33 is connected to the input AMS1117.
Can't see the FPGA well, but try Desoldering Wire to remove some tin from the pins, for all pins.
Maybe there's a short circuit somewhere...
C40 is connected to +3.3V. C33 is connected to the input AMS1117.
Can't see the FPGA well, but try Desoldering Wire to remove some tin from the pins, for all pins.
Maybe there's a short circuit somewhere...
Last edited:
How much current is way too much? How much current @1.2v is the FPGA supposed to draw?
Why? Sometimes LDOs can go into self-protection if the voltage drop across them combined with the output current causes sufficient self-heating. Running the 7150 from 3.3v input should help keep it cool. Also make sure the bottom solder pad connection is good, as that is needed for cooling too. Does the bottom of the PCB under the 7150 get hot too?
EDIT: Another thing to maybe try would be to turn off the power and measure the resistance to from the 7150 output pin to ground with a DVM. This assumes you have a DVM that uses a very low voltage to measure resistance. Typically that voltage is too low to bias on a semiconductor junction (maybe best to check the manual first). A shorted cap might still show up as shorted. Otherwise a normal cap of large value will slowly start charging up, so the resistance will keep going up until the meter is more or less displaying infinite or OL.
Last edited:
I meant that the 7150 (3.3v reg) didn't also have the AMS1117-1.2 downstream of it so ruling out that possibility. (I also removed the AMS1117 to rule it out completely and the problem still exists. Yes, the 7150 is also in good contact with the board, the reverse side gets hot too. I will try removing C33 and C40 later today when I get chance.
They're a different make to the ones I got from Mouser, I think I got them from Farnell in the UK. They do have a line for the +ve side, like the Vishay ones but very indistinct. I'm sure they are mounted the right way around though. Having said that I also have Vishay now so can replace,
Well, I had chance to do some diagnostics and so far it is not looking good. I think either the FPGA or the memory chip has failed.
I initially removed the capacitors as suggested but no luck with that. So I did as Mark suggested and cut the track from the 3.3v regulator to the group of vias. That removed the short on the regulator and I got 3.3 volts on its output. So at least the expensive 7150 is ok and will live to fight another day.
That was the good news. The bad news is I have 0.5 ohms on the (now disconnected) 3.3v line and after removing some more caps and hunting around I think I will admit defeat and build a new one.
Unless anyone can give me any more inspiration? It's not looking good though, since it did work fine for a couple of hours prior to failing. So some component must've just given up. Luckily not the expensive (and almost impossible to obtain) 3.3 v reg.
I initially removed the capacitors as suggested but no luck with that. So I did as Mark suggested and cut the track from the 3.3v regulator to the group of vias. That removed the short on the regulator and I got 3.3 volts on its output. So at least the expensive 7150 is ok and will live to fight another day.
That was the good news. The bad news is I have 0.5 ohms on the (now disconnected) 3.3v line and after removing some more caps and hunting around I think I will admit defeat and build a new one.
Unless anyone can give me any more inspiration? It's not looking good though, since it did work fine for a couple of hours prior to failing. So some component must've just given up. Luckily not the expensive (and almost impossible to obtain) 3.3 v reg.
Yes, it is this one: https://www.mouser.co.uk/ProductDetail/913-M25P80-VMN3TPB
Yes, worth it I guess to at least rule it out. Only a 80p part vs the Spartan. I will try that in the morning.