This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hi,
so this is a sort of a continuation of my previous post:
but as I have now discovered that it is probably not the C2000 problem and because I more or less just had a monolog with my self, I'm opening another post here.
So I have a board with multiple SD modulators. I have multiple AMC1303E2520 (Manchester coded internal clock) and 3 AMC1306E25 (Manchester coded external clock, the 3 are clocked together). The AMC1306E25 gave me sporadic erroneous resultants and with some more experimentation I have found that when I'm using AMC1306E25 the sample frequency changes a lot (see my previous post). I had nothing connected to the board just started to test the hardware.
I have now ruled out board layout or the MCU with this text: I have cut the connection to one AMC1303E2520 and patched a neighboring AMC1306E25 into the line that was previously AMC1303E2520. With this now both filters have changing sample frequency. Then I reversed my patch so that the one AMC1303E2520 was patched the lines for AMC1306E25. Now both filters have a stable sample rate.
And this is true for all 3 AMC1306E25. Now the only thing that the AMC1306E25 share is their clock line. I have looked at the clock with a scope (500 MHz scope, short ground clip, 15pF x10 probe) and it is perfect, nice and square, no overshoot, no undershot, no jitter, no missing clock pulses, ... (I have set up a pass fail mask on the scope with tight tolerances - picked up no problems). Then for just good measure I have also AC terminated the clock signals but still I still have the same problems with my changing sample frequency.
The only thing that I can now think off if is somehow I received a whole pack of defective AMC1306E25. I bought them at Mauser, so they should be legit.
Any ideas what I could try? I have tried a lot of stuff in the MCU, but now with this test I think I have ruled out any possibility that MCU has anything to do with my problem.
Best regards,
Marko
Hi,
yes that would be my question, because this seems to be the source of my problems.
This problem is about 6 data converters, measuring 3 phase voltage and current. They are arranged in pairs. 2 for each phase. These pairs are on a common insulated "island" and are powered together.
The power supply for the insulated side is 5V. The power is generated by SN6501 and 750313626 transformer - so 7V. Each data converter has its own zenner diod (to make 5.1V) and a ferite bead in front and then good bypassing. Both converters (in a pair) are powered exactly the same so if there would be a problem there I would have problems with the second converter (in the pair) also. The Dvdd is 3.3V form the rest of the board.
The data lines are directly connected with max 130 mm length of pcb trace (terminated correctly with AC termination). Again all the converters are located closely so if the pcb connection trace would be the problem I would have problems with both converters in a pari. But I don't.
If only 1 converter would be acting up, I would say I soldered it wrong, or ESD killed it, or I had some transient power, or the power supply was wrong. But in the 3 pairs, the AMC1303E2520 seems to be ok (the conversion results) and the AMC1306E25 seems to have big problems.
The momentary drop of sample rate for AMC1303E2520 could be attributed to MCU or to the debugger or anything like that. But the totally jittery rate of the AMC1306E25 is defiantly a problem (with the converter).
Both the AMC1306E25 and AMC1303E2520 are connected exactly as the data sheet suggested. I also took out the analog front end buy soldering the AinP, AinN and Agind pin together.
Best regards,
Marko
P.S. I clicked something wrong and it marked your post as this resolved my problem. And I don't know how to remove it. Can you remove the resolved my problem status?