Bpod v2.3 ports half working

ckopec

Josh (or anyone who may know),

Just tried building the new v2.3 Bpod (hold up in a my garage trying to keep busy as Princeton is now shut down). Everything seemed good except when I connect a nose port to jacks 1 and 4. Poke in events are not detected (other port functions, LED, water valve work fine in all jacks) and poke in events are detected if I connect the same nose port to jacks 2 and 3. I've been trying to hunt around on the board for a short (seeedstudio did once send us a circuit board that was wired incorrectly with a hidden short in one of the layers) but so far can't seem to find anything. The voltage input to IC12 seems off for those 2 jacks so I think IC12 is functioning fine, it's just getting a bad input and passing that along. When I test the pins on the ethernet port on the Bpod board with a nose port connected everything looks identical between a functioning jack (2 or 3) and the non-functioning jacks (1 and 4). Finally I measured the voltage across the phototransistor on the nose port itself. When connected to a functioning jack I see a voltage of 0.15V when poked out and 4.55V when poked in. However, when I connect the same nose port to a non-functioning jack I see a voltage of 0.15V when poked out but only 0.65V when poked in. I get the same result after removing the new 22k pull-up resistors. If I measure the voltage across the pads for the new pull-up resistors I get a similar behavior. Functioning jack, poke out 4.85V, poke in 0.15V. Non-functioning jack, poke out 4.85V, poke in 4.15V. Any thoughts on what could be causing this? Any other tests I should run to try to find the source of the problem (teensy?, IC12?, PCB?)

Thanks
Chuck

ckopec

After hunting around on the board I discovered that some flux residue lying between the feet of IC12 connected them electrically with very high resistance (in the 10s to 100s of kOhms). After carefully cleaning this the resistance between neighboring pins on IC12 went to Infinity and the behavior of all the nose ports returned to normal. I will note this was a board I made myself (never had any such problems with boards ordered from Sanworks).

Josh

Hi Chuck,

Thanks for sharing the solution!

-Josh

Josh

Max wrote[quote='Max' pid='1441' dateline='1587492141']
Hi I am having troubles my selve with a solve soldered bpod 2.3

am I correct that the rj45 connectors on the left side of the bnc connectors (bnc connectors down) are the ports where the breakout boards should be connected? (see attached drawing)?

and that the ports are numbered 1 to 5 from top to bottom (again bnc connectors on bottom)?

because I cannot address the valve board in pybpod 🙂

Hi Max,

There are 4 behavior ports and 5 module channels - so the side with 4 RJ45 connectors is the behavior port side - in your drawing, the right-hand side. As viewed from the side, channels are numbered from left to right. The laser-cut enclosure contains labels for the channels (e.g. here).

-Josh

ckopec

Max wrote[quote='Max' pid='1441' dateline='1587492141']
Hi I am having troubles my selve with a solve soldered bpod 2.3

am I correct that the rj45 connectors on the left side of the bnc connectors (bnc connectors down) are the ports where the breakout boards should be connected? (see attached drawing)?

and that the ports are numbered 1 to 5 from top to bottom (again bnc connectors on bottom)?

because I cannot address the valve board in pybpod 🙂

The row of 5 RJ45s are for module connections, the row of 4 RJ45s are for port connections.