Hi,
I've run into a very interesting problem, and I've spent the past few days debugging. Still no solution. May be you guys could help?
I have two rigs based on the ADS1292R - one is a prototype rig based off a Shmartboard with wires all over, the other is a PCB version of it. They are identical in terms of schematic, but one works and the other's MISO never replies correctly. I've swapped chips from both rigs, and still same behavior. It's most definitely a problem with hardware on the PCB, but I can't find any difference between the two boards!
Here's my schematic :
Specifically here is the one that doesn't work. Notice that the MISO reacts to CS correctly ( same way as my other rig) : it goes high when CS is disabled, then low when its enabled. This is zoomed out a bit, if I zoom in I can see the individual data transmission and it looks good. I'm running SPI at 1MHz, and like I said it works perfectly and I can read registers and use the chip just fine on the other rig.
I've verified that all the pins are getting the right signals, and that power connection is all correct. PWDN is High as it should be.
I noticed in a schematic online that the CLKSEL pin was pulled up with a 10k resistor. I have a direct connection to 3.3V, no pull-up resistor. I've verified the voltage on CLKSEL with a multimeter. My far fetched theory now is that I need a pullup on the CLKSEL, and the only reason it worked on the prototype board was that I used wires ( which have a bit of internal resistance). Think this would hold? I don't want to try it ( It involves messing up the PCB) unless it holds some water.
What sort of hardware issue makes MISO react to CS but not send data?
Thanks for your time.
Hi Eric,
This sounds strange - can you send the 'non-zoomed' version of your data transfer? There really should be no difference in using a pull-up versus taking CLKSEL directly to the rail in the ADS1292R.
Regards,
Tom
Thanks for the quick reply. I uploaded code to write to the CLK register to have it ClockOut on the CLK pin. Identical code on both rigs, same setup on both rigs, but one works the other doesn't.
The broken one:
The correct one, notice the clk output and the MISO output
NOTE: I know SPI Enable appears High, I forgot to connect my logic analyzer lead to CS , but I checked before and with a multimeter, I can assure you it is actually Low, not high as the logic analyzer suggests.
Any idea what could be going on? Its a hardware issue, but I keep checking and everything seems identical.
Should I remove all capacitors and then try it? Would the chip work without any caps attached? ( I'm thinking maybe caps might be giving problems?)
Thanks
I removed all caps besides the caps on the Vrefp and Vrefn. I still get this same weird behavior.
Any idea what else to check or try?
Can you verify what voltages you see at the reference pins and VCAP pins? Can you send along the layout details? Perhaps there is something there that was over looked.
I feel very silly. The test pads I had on the PCB all showed the right signal. But, I just found out, I had a typo on the connection to the actual pin on the ADS1292R - SCK instead of SCLK. Silly me :P Wired the clock by hand, and now its fine and dandy.
Thanks for your time ! Hope I wasn't too much trouble.
No trouble at all! I'm happy to hear you found the problem!