Not really any changes I can think of. When I first tried to install the Atom chip, I beant some of the leads. But I carefully realigned all of the leads and inserted into the socket. I am pretty sure all of them are making good contact as I have checked for low resistence to the bottom of the board. I have tried using the 6v battery, a fresh 9v battery, with and without the PS2 cable installed, but still no luck.
I have also tried using a logic probe to make sure that there is some signal changes. When I push the Program button in the software, I do notice that the ATN line alternates high and Low several times. There is also shorter pulses in the SIN pin, but not much activity on the SOUT pin. I may try later to breadboard the chip directly to a serial cable to see if I have any luck that way.
First, I’d like both of you having problems to double check your Advanced comport settings. Hit the Defaults button and sete Reset Hold time to 25. Try reprogramming your modules. If that doesn’t work try changing the reset hodl time to 10 and 50 testing at each change. USB to Serial adapters can be the cause of problems. Some just can’t handle the quick changes in direction(send/receive) programming the Atom/AtomPro modules require. I know for a fact that some don’t work. Keyspan adapters are hit or miss. Some models work and some don’t so I’d stay away from them. Anything based on a Prolific chipset should work(assuming they use prolifics drivers and didn’t do something custom). Anything based on a SiLabs CP210X should also work. I’m not sure about FTDI based stuff but I haven’t heard of any problems either.
The last thing to try is simply uninstall all the softwaare(make sure it’s all gone). Also go into the registry and delete the BasicMicro keys under HKEY_CURRENT_USER/Software. Then reinstall the software and start the reset hold test all over again. Check that your modules are getting overly hot(they will be warm at most ussually).
The USB to serial drivers on my Portable are FTDI based. They have worked well for me on several other things such as AVR programming, JTAG, SSC-32… I am currently trying to run the BasiMicro IDE on my Home machine which has a hardware serial port. Still no luck on all of the different Reset hold times.
If I get a chance I will try setting up a cable to program the chip directly without the AAB board to make sure I have all of the signals.
I will probably take Jim up on his offer and ship him my Card and ship and see if he has better luck.
I did take Jim up on his offer and I shipped him the AAB with the Atom chip. Yesterday I also received my additional AAB and Atom Pro chip that I ordered and last night I soldered in the appropriate pull-up resistor, replaced the 1" stand offs with 3/4" stand offs, such that my battery can now fit under the SSC-32 card and hooked it up the new AAB+Pro with a PS2 on I/O pins P12-15, moved the TTL serial interface from P15 to P11…
I compiled my updated code with the appropriate signal and timing changes and then I downloaded it to the new board! The board appears to be responding to my PS/2. The legs are not moving properly yet. Maybe I screwed up on my offsets, or maybe timing of Serial communications, … But that is where the fun begins!
Yes, it’s looking like you will have it going very soon. When you press the “O” button the legs should go to all = 1500. Does it look just as it did after you did the servo offset procedure? If so you still may have installed the servos in the wrong pins on the SSC-32. This will definetely cause it not to walk correctly. We just added some new programs to the tutorial for this bot yesterday. Lots of additional functionality. Funny thing… I was doing the offsets for my test bot with the new (stylish) legs installed. I did the math wrong. I subtracted the adjusted value from 1500 instead of subtracting 1500 from the adjusted value. Doing this will cause the pulse offset to double the error instead of correct for it. LOL So after I figured out my mistake it walked perfectly. It IS fun!
Oh one other thing, the PO command must only be done one time. It was implemented with a relative value, not an absolute one. So long story short, power cycle the SSC-32 if you reprogram the atom with new pulse offsets.
I believe that the pins are correct as it was walking when I had it connected to the ATOM. My guess is that the math functions are not working the same on the Pro as it did on the Atom. Maybe my updates to the acos are not working properly.
I tried running it under the debug mode, but when I try this the code never gets out of the PS2 init code. Most probably timing…
I have now converted the program to output to the debug terminal instead of the SSC32 to see if I can figure out which values are wrong. I wish I had a working atom to compare the outputs…
When the code is run and we are at the initial stance it loos like the vertical hips are all off. The three on the left hand side are all lifted up at about 45 degrees and the hips on the right hand side are all down at about 45 degrees.
I may not be able to work on this for the next week, so I included the outputs from a reset, until is in it’s static mode incase some one sees something obvious.
it to the new board! The board appears to be responding to my PS/2. The legs are not moving properly yet. Maybe I screwed up on my offsets, or maybe timing of Serial communications, … But that is where the fun begins!
I subtracted the adjusted value from 1500 instead of subtracting 1500 from the adjusted value. Doing this will cause the pulse offset to double the error instead of correct for it. LOL So after I figured out my mistake it walked perfectly. 
It IS fun! 