Announcement

Collapse
No announcement yet.

Arduino Plugin feature requests

Collapse
This is a sticky topic.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • enigmatheatre
    replied
    Doug,

    Ok I think I get what you want now and I may have misunderstood what you were after. The problem is that I receive and handle the data from the onewire bus in the plugin and if it is strange data there is an error. I have no way of polling the bus from the plugin so I would need to try and handle the errors on the board or find a way to poll the device if there is an error.
    I do not have a lot of time at the moment but I will try and set up a test to see if this is possible to add.

    Greig.

    Leave a comment:


  • DMcKnight
    replied
    Originally posted by enigmatheatre View Post
    Doug,

    I can confirm that the "9999999" is the error value of my plugin so this is being handled by my plugin is showing on the device as an error. The problem is that you are graphing the value so it records the "9999999".

    Greig.

    Hi Greig,
    I'm a little confused by your reply. I think that's what I expected. The plug-in is detecting an error and so reports an unambiguously error-style of value. I think that behaviour makes sense.

    What I'm wondering is if the plug-in could be enhanced to try again, if it detects an error condition like this.

    For example, if it could be configured to "retry 3 times" on error, for example, then I think my system would work just fine. The errors are quite rare and are proving quite stubbornly difficult to track down and eliminate in hardware.

    regards
    Doug

    Leave a comment:


  • enigmatheatre
    replied
    Doug,

    I can confirm that the "9999999" is the error value of my plugin so this is being handled by my plugin is showing on the device as an error. The problem is that you are graphing the value so it records the "9999999".

    Greig.

    Leave a comment:


  • mihaium
    replied
    Hello Greig.
    Today is happened the same error in the my system. Is 1 error to 3-4 hours. I have 1 DS18B20 (TO92 capsule ) with 8m CAT5 cable and 1 kohm resistor.


    Inviato dal mio SM-G920F utilizzando Tapatalk

    Leave a comment:


  • enigmatheatre
    replied
    Originally posted by DMcKnight View Post
    Debug log sent by email.
    Regards
    Doug
    Doug,

    I will need to check the code when I get home but looking at the log I think my plugin is reporting an error for your devices. If you look at the rom device that has the temp in it is the status value for error not "9999999" and what you are seeing is Jon00's plugin logging this value. I will confirm this when I have more time and also see if there is a way to stop this.

    Greig.

    Leave a comment:


  • DMcKnight
    replied
    Originally posted by enigmatheatre View Post
    Doug what version are you running as there is code to trap bus errors in the beta. If you are running it then capture and send me a debug log so I can see what is going on.

    Greig.

    Sent from my HTC One using Tapatalk

    Debug log sent by email.
    Regards
    Doug

    Leave a comment:


  • rprade
    replied
    Originally posted by DMcKnight View Post
    I'm using 1.0.0.85. If the 1.0.0.95 Beta is compatible with the 1.0.0.85 Arduino sketches I'll install it right now, otherwise I'll have to wait until I'm next home.

    Doug
    It *should* be just fine. The 1.0.0.95 version uses V1.0.0.84 sketches.

    Leave a comment:


  • DMcKnight
    replied
    Originally posted by enigmatheatre View Post
    Doug what version are you running as there is code to trap bus errors in the beta. If you are running it then capture and send me a debug log so I can see what is going on.

    Greig.

    Sent from my HTC One using Tapatalk

    I'm using 1.0.0.85. If the 1.0.0.95 Beta is compatible with the 1.0.0.85 Arduino sketches I'll install it right now, otherwise I'll have to wait until I'm next home.

    Doug

    Leave a comment:


  • enigmatheatre
    replied
    Doug what version are you running as there is code to trap bus errors in the beta. If you are running it then capture and send me a debug log so I can see what is going on.

    Greig.

    Sent from my HTC One using Tapatalk

    Leave a comment:


  • rprade
    replied
    Originally posted by DMcKnight View Post
    Thanks for the suggestion. I have already tried many different resistance values, including as low as 1k. I've also tried a small resistor in series with the data pin, which I've found helpful with one wire before.

    I think the proper hardware solution is to use a "real" one-wire bus master with an active pull-up, but this is so close to working I think a simple software trap would do the trick.

    regards
    Doug
    Give Greig some time. I'll bet he can figure out how to trap OneWire errors within the code.

    FWIW I have 4 DS18B20 sensors on a Cat 5 cable of about the same length, with a 2.2K pull-up and haven't seen an error. This is on a test board. For my production temperature sensors I use EDS OWServers because they are so darn reliable.

    Leave a comment:


  • DMcKnight
    replied
    Originally posted by mihaium View Post
    Hello.

    Forse the long distance is necessary change the pull-up resistor. Put the 1 kohm resistor.

    Inviato dal mio SM-G920F utilizzando Tapatalk

    Thanks for the suggestion. I have already tried many different resistance values, including as low as 1k. I've also tried a small resistor in series with the data pin, which I've found helpful with one wire before.

    I think the proper hardware solution is to use a "real" one-wire bus master with an active pull-up, but this is so close to working I think a simple software trap would do the trick.

    regards
    Doug

    Leave a comment:


  • mihaium
    replied
    Hello.

    Forse the long distance is necessary change the pull-up resistor. Put the 1 kohm resistor.

    Inviato dal mio SM-G920F utilizzando Tapatalk

    Leave a comment:


  • DMcKnight
    replied
    Error trapping on 1-wire temperature measurments

    Hi Greig,

    I've used Arduinos in the past to measure temperature on multiple one-wire sensors and I'm aware of the limitations of "bit banging" with a single output (as distinct from using a more sophisticated active pull-up).

    As it is, I've got three temperature sensors on a cat 5 cable that's about 15 meters long and I've got it *almost* working perfectly. By that, I mean that I can measure temperatures every 6 minutes and only get about 1 or 2 measurements per day that show up as 9999999.

    So, my question is: can these erroneous readings be trapped and handled sensibly somehow? I expect it could all be done in a script, but it seems more appropriate for it to be handled close to the sensor, such as in the plug-in. Perhaps this kind of reading could cause a warning to be written to the log, and the measurement re-tried?

    Regards
    Doug

    Leave a comment:


  • Daggy67
    replied
    Update help file for pinouts

    There are different configurations for the pinouts depending on USB, ethernet Mega or UNO. It's taken me some time find what is needed selecting inputs or outputs in the board setup to get it working depending on my set up analog or digital. It would be good reference to update the help file reflecting these differences. Great plugin always works as advertised, much to learn but satisfaction is guaranteed.

    Leave a comment:


  • enigmatheatre
    replied
    Originally posted by piever View Post
    Hello Greig,

    I have a chipkit max32 board and would like to connect this board to HS3.
    The latest rev http://www.chipkit.net/started is here.
    I tested this version but its sees not to connect to hs3.

    The newer do not support chipkit 32-bit boards.
    Is it possible to use this board?

    Regards piever. Netherland
    Emailed a reply.

    Leave a comment:

Working...
X