+ Reply to Thread
Page 5 of 9 FirstFirst 1 2 3 4 5 6 7 8 9 LastLast
Results 41 to 50 of 86

Thread: The ACME computer/rebreather controller

  1. #41
    RBW Member Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse's Avatar
    Join Date
    Jan 2008
    Location
    Auckland New Zealand
    Posts
    1,647
    rEvoIII mini hCCR

    RG-UMF,IDA 71,rEvo mCCR

    Re: The ACME computer/rebreather controller

    How does the netduino differ from standard Arduino?

    Im using Sanguino for my DIY HUD.
    s.

  2. #42
    Dive different. cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys's Avatar
    Join Date
    Feb 2005
    Location
    S. California
    Posts
    733
    MK15.5, MK15, LARV

    KISS Classic

    Re: The ACME computer/rebreather controller

    Quote Originally Posted by Packhorse  View Original Post
    How does the netduino differ from standard Arduino?

    Im using Sanguino for my DIY HUD.
    Sanguino. What a cool name.

    Anyway Netduino software is built on .NET micro framework. It uses C# and Visual Studio 2010 to debug and deploy things. Netduino's are based on the AT91SAM7X512 microcontroller, running at 48 MHz with 152KB code storage and 64 KB RAM.

  3. #43
    Dive different. cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys's Avatar
    Join Date
    Feb 2005
    Location
    S. California
    Posts
    733
    MK15.5, MK15, LARV

    KISS Classic

    Re: The ACME computer/rebreather controller

    Things are moving along. I was pleasantly surprised to find when I actually wired the ACME electronics up to the rebreather today (as opposed to modeling what would happen when I did) that it looks like I won't be needing the pre-amp after all -- got very stable readings without the need for deep sampling of the ADC. I will continue to study the situation, but the pre-amp is a complication I am happy to be avoiding for the time being. What kept me from wiring them up earlier? Well, getting deco and NDL debugged and tackling the eeprom... I wrote a driver for the flash memory chip this past week. Now that I can write calibration factors to memory I can actually think about taking the controller out for a spin before long.

    Chris

    .
    Attached Images

  4. #44
    RBW Member Apeturbo is an unknown quantity at this point Apeturbo's Avatar
    Join Date
    Jun 2010
    Location
    Italy - Toscana - Versilia
    Posts
    30
    Custom Dolphin mCCR

    Re: The ACME computer/rebreather controller

    Hey Chris,
    how many sample do you take each reading? have you found the noise frequency and than fixed the number of sample?
    Actually i'm not expert in oversampling...
    With my electronic i've found a very very stable reading with something like 14000 sample, but it require long time, so i fixed it to 1400 sample and i've had to use also a op-amp.
    Adc noise is a microcontroller and circuit problem, so i'm curios to know if netduino is a better choice for this problem.

    Daniele

  5. #45
    RBW Member Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse's Avatar
    Join Date
    Jan 2008
    Location
    Auckland New Zealand
    Posts
    1,647
    rEvoIII mini hCCR

    RG-UMF,IDA 71,rEvo mCCR

    Re: The ACME computer/rebreather controller

    For my DIY HUD I use a 12 x multiplier on the ADV and simply read the input once. No averaging or any thing. Never had an issue.
    Has anyone else?

  6. #46
    Dive different. cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys's Avatar
    Join Date
    Feb 2005
    Location
    S. California
    Posts
    733
    MK15.5, MK15, LARV

    KISS Classic

    Re: The ACME computer/rebreather controller

    Quote Originally Posted by Apeturbo  View Original Post
    Hey Chris,
    how many sample do you take each reading? have you found the noise frequency and than fixed the number of sample?
    Actually i'm not expert in oversampling...
    With my electronic i've found a very very stable reading with something like 14000 sample, but it require long time, so i fixed it to 1400 sample and i've had to use also a op-amp.
    Adc noise is a microcontroller and circuit problem, so i'm curios to know if netduino is a better choice for this problem.

    Daniele
    Hi Daniele,

    Right now I'm averaging only 50 samples. Before I was using 500 - 1000. I know what you are saying about ADC noise being a controller and circuit problem. I tossed out one lcd display option because of ADC noise. I keep in mind the ideal circumstance that the signal improves with the square root of the number measurements averaged. In any case I haven't tried going below 50 as doing the latter for each sensor is not much of a burden.

    Chris

  7. #47
    Dive different. cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys's Avatar
    Join Date
    Feb 2005
    Location
    S. California
    Posts
    733
    MK15.5, MK15, LARV

    KISS Classic

    Re: The ACME computer/rebreather controller

    Quote Originally Posted by Packhorse  View Original Post
    For my DIY HUD I use a 12 x multiplier on the ADV and simply read the input once. No averaging or any thing. Never had an issue.
    Has anyone else?
    It was an issue when the development board was powered by USB which apparently provided noisy power. I'm shaving down the number of measurements as much as possible as things progress.

  8. #48
    RBW Member Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse is a glorious beacon of light Packhorse's Avatar
    Join Date
    Jan 2008
    Location
    Auckland New Zealand
    Posts
    1,647
    rEvoIII mini hCCR

    RG-UMF,IDA 71,rEvo mCCR

    Re: The ACME computer/rebreather controller

    Ah, dont think I had an issue when I ran mine via USB.
    Although I did when I tried powering the OLED via a digital pin ( or two). The readings were lower than expected and erratic. May have been a sampling issue. Although I put it down to voltage drop at the time.

  9. #49
    RBW Member Bottom Dollar Diver is an unknown quantity at this point Bottom Dollar Diver's Avatar
    Join Date
    Sep 2009
    Location
    Southern California
    Posts
    51
    Classic Inspiration

    Re: The ACME computer/rebreather controller

    Hi,
    Noise and unstable readings of O2 cell voltage might be partly caused by the nature of the cells. They are plastic and have no "ground" or shielding and the output is a high impedence. This makes the system very prone to noise pickup from any signal coupled to the input lines. A few ideas:

    Be sure the load on each cell is 10 K ohms.

    Put some capacitance across each cell. This will help the sampling process by lowering the impedence. Since the cell output is a slow moving signal, it won't hurt to "load it down" with some capacitance.

    Use "single point grounding" techniques to avoid ground loops. Don't combine or cross connect ground lines from the cells or any other part of the circuit.

    Shielded wire and circuit layout changes might help.


    I don't know anything about the micro-controllers that you are using but problems with noise and instability in low level circuits are common to all kinds of equipment. Unfortunately the typical O2 cell is unstable enough without additional uncertainty in the measuring circuit! Good luck.

    Mark Thompson

  10. #50
    Dive different. cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys has much to be proud of cys's Avatar
    Join Date
    Feb 2005
    Location
    S. California
    Posts
    733
    MK15.5, MK15, LARV

    KISS Classic

    Re: The ACME computer/rebreather controller

    Now the complete ACME controller *is* ready to dive. I think I have been visiting the 90/90 rule -- the first 90% of the project takes 90% of the time and the last 10% takes 90% of the time ...

    Implemented a revised pcb with all surface mount components (except for a couple things like the mini board and some large caps).

    Hacked a dedicated rs232 output pin on the mini to serve as a 0-5V i/o pin as I needed just one more pin than the mini provided.

    Got the flash memory chip fully functioning for calibration factor storage.

    Now that the controller and deco functions are merged, the next issue could be power, but I'll see before long.

    Chris


    .
    Last edited by cys; 23rd September 2012 at 07:59.

+ Reply to Thread
Page 5 of 9 FirstFirst 1 2 3 4 5 6 7 8 9 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts