×

With the recent surge in Omicron cases, shipping may be slower than stated times. We are working to build, ship and respond to everything as quickly as possible. Please see all COVID-19 updates here. Thank you for your continued support.

AdamMil

Member Since: October 30, 2009

Country: United States

  • If I could change the hardware, I'd also make these changes:
    * 5V for AVR, 3.3V for flash chip. This way, it'd work with all AVR programmers, and could record input from normal 5v circuits without external voltage dividers.
    * Onboard voltage divider network (6 resistors and a 3-DIP switch?), for recording inputs above the ADC voltage.
    * A more compatible power plug!
    * I don't care much about this, but an external 12.8mhz oscillator would allow the ADC to run at its recommended maximum rate (200khz), for the highest possible sampling rate.

  • There are improvements that could be made to the code. :-)
    * The current code wastes a lot of space. It could get 60% more recording time from the same amount of memory.
    * The hardware should be able to record seamlessly at up to ~14000 samples/second, so a variable recording rate would be nice. At low rates, it could use the ADC sleep mode and averaged sampling to increase precision.
    * The current code uses only one flash buffer. It could instead fill one buffer while the other is being committed, for a higher maximum recording rate.
    * By turning off unused channels, you can record at a higher rate or for a longer time on the remaining channels, and reduce time spent waiting for ADC stabilization. In the common case of only one channel, you could get ~9000 samples/sec at the current CPU speed, or 13-14k samples/sec at a higher speed.
    * The current code constantly uses 100% CPU time, even between samples and after the memory is full. It could be using low-power sleep modes.
    I could implement those pretty easily if I could get it to work at all. :-)

  • Running on 5 volts would also be good because then I could hook it up to typical 5V circuits without needing to hook up 3 voltage divider circuits too.

  • The power plug is frustratingly incompatible with my existing battery packs and connectors (mostly from Pololu). They're almost exactly the same, but the plug on the uLog is a bit too narrow...
    I believe Pololu connectors would accept Sparkfun batteries, but Sparkfun connectors don't accept Pololu batteries. It'd be nice if those were compatible.
    Not quite sure what to do now... I have a few spare connectors, but they're through-hole and not surface mount. And I don't have a 3.3V regulator to apply power directly to the pins. :-(
    Also, not many programmers work on 3.3V. Most require 5V.
    Would the device work on 5 volts?

No public wish lists :(