Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter

[hochopeper]

That doesn't explain why it didn't work with the isol board in the circuit though does it? I was reading this part of the manual today and the way I read it you shouldn't need to remove L11 when using the isolator board, only if powering the two boards separately, which is when the issue came up evidently. This is because you've got 5V supply from the FIFO regulators and 6.6V supply from the battery both connected to the same clock supply rail ...

[bigpandahk]

Originally Posted by qusp

yep, you didnt prepare the clock board

it is indeed an inductor, its L11,..the one you are supposed to remove before applying independent power....

whoops

from the clock board manual

I thought if I connect the clock board directly with FIFO and want to use individual supply for the clock , then I need to remove it. With the isolator board in between, I don't think I have to.

[qusp]

erm, I would reassess my understanding given the evidence at hand =) given its the one part that was vaporised, i'm thinking there was a problem with it being there...

why would it matter if the iso board is there? you are still connecting 2 parts of the circuit that are assumed unconnected. without the board here or a schematic its hard to say. but i'm guessing its because L11 provides a path from the clock power supply, which would normally be coming from the fifo board, but since this line isnt used or transmitted with the isolator due to separate power being assumed, it now may connect to ground

[qusp]

there may be another problem, maybe i'm wrong here (its been known to happen) it just seems a bit coincidental that L11 and only L11 (as far as we know at this point... we hope) said goodbye. I went back and read that post by ian and it does read that way in post 1160. but hes talking about removing the clock LDOs, shorting across them and then using J5 to directly power the clocks with a 3v3 battery. you have a different situation here, your LDOs are still in place yes?

[bigpandahk]

Originally Posted by qusp

there may be another problem, maybe i'm wrong here (its been known to happen) it just seems a bit coincidental that L11 and only L11 (as far as we know at this point... we hope) said goodbye. I went back and read that post by ian and it does read that way in post 1160. but hes talking about removing the clock LDOs, shorting across them and then using J5 to directly power the clocks with a 3v3 battery. you have a different situation here, your LDOs are still in place yes?

Yes, the LDO still in place. The clock board LED still on when it burnt.

[qusp]

Actually looking closer L11 may be powering the clock board side of the isolator via pin 9 on J9, in that case it could have been a short anywhere on the clock side of the isolator that caused that. can you disconnect the isolator and test resistance from pin 9 to ground (pin 10 or 6) on J9 (isolator board output FFC header) and tell me what you get? it should be open circuit

[qusp]

ah actually it wont be open as the regulator input bypass cap and the reg itself will interfere, but if you have a short or a low resistance then we have our answer

thank god, Ians here to the rescue (name down bottom, hello Ian!) since I dont have the fifo here, or the schematic i'm flying blind, but still trying to help

[hochopeper]

Originally Posted by qusp

Actually looking closer L11 may be powering the clock board side of the isolator via pin 9 on J9, in that case it could have been a short anywhere on the clock side of the isolator that caused that. can you disconnect the isolator and test resistance from pin 9 to ground (pin 10 or 6) on J9 (isolator board output FFC header) and tell me what you get? it should be open circuit

That's what I just came back to write ...


I'd be checking around U3, C4 and L1, just a hunch.

Do the same test on J4 of the clock board, before hooking anything up again, just to be sure.

[bigpandahk]

Originally Posted by qusp

Actually looking closer L11 may be powering the clock board side of the isolator via pin 9 on J9, in that case it could have been a short anywhere on the clock side of the isolator that caused that. can you disconnect the isolator and test resistance from pin 9 to ground (pin 10 or 6) on J9 (isolator board output FFC header) and tell me what you get? it should be open circuit

It's 4.37K

[qusp]

nothing connected to the board? is that a steady value? that seems pretty low given you are measuring across an LC filter with an inductor in series and a 10uf cap to ground. other than checking the parts mentioned above by hochopeper L1, C4, L2 and U3 i'm out of ideas given I dont have schema, or board here. hopefully Ian pipes up shortly, sorry I couldnt be of more help