Announcement

Collapse
No announcement yet.

DIY battery operated tilting blinds

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    DIY battery operated tilting blinds

    Anyone done it?
    Last edited by NeverDie; June 18, 2015, 08:52 AM.

    #2
    i have heard good things about the nordic chip:
    http://www.nordicsemi.com/eng/Produc...z-RF/nRF24L01P

    you might also contact the guy who owns rollertrol, their motors are super low standby current with RF.

    http://rollertrol.com/store/en/solar...ind-motor.html
    HS3 Pro on Windows 8 64bit
    53 Z-wave nodes(46 devices, 7 remotes), 15 DS10a's, 10 ms16a's, 9 Oregon Sensors, W800, RFXCOMtrx433, Way2Call, 3 HSTouch Clients, 2xRussound CAS44, Global Cache GC100-12,10 Rollertrol blinds(+ zwave) ,3 Squeezebox Radios and 1 Squeezebox Boom,DMX Arduino via ethernet,Rain8Net,3x Echo Dot's


    Check out my electronics blog here:
    https://www.facebook.com/RaptorsIrrationalInventions

    Comment


      #3
      The Nordic NRF24 is a good option and it what's used by the mysensors eco-system. Here's a forum post on the mysensors forum for a blind project: http://forum.mysensors.org/topic/833...trol-project/1

      Cheers
      Al
      HS 4.2.8.0: 2134 Devices 1252 Events
      Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

      Comment


        #4
        Originally posted by NeverDie
        Looks like he ran wire to the window to power it. Though there is an argument to be made for doing that, if you can get 3-5 years per set of 4x D-cell batteries (as Lutron does), I think it may just as good to do it that way. If you can get to 10+ years per set of batteries, the ease of installation can make it pretty compelling.
        Yes in that particular posting. You can run the whole thing on batteries as well. There are lots of websites on how to reduce Arduino power use to a bare minimum. Not sure if 10+ years will be achievable with a home-brew solution.

        Cheers
        Al
        HS 4.2.8.0: 2134 Devices 1252 Events
        Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

        Comment


          #5
          Originally posted by NeverDie
          Thanks for the link. "standby current consumption: 0.03 mA". Is the radio still listening for commands while in standby? If so, then it's in the ball plark. How long does it take to start reacting to a command? Is it seemingly instantaneous, or does it seem to take a while?

          It also says "compatible with Vera and RFXCOM USB transceiver." Correct me if I'm wrong, but isn't that a 433Mhz OOK transceiver?

          Edit: For comparison, a specialized "ultra low power" OOK wake-up receiver might consume 69 microamps at 1.2v (http://liu.diva-portal.org/smash/get...FULLTEXT01.pdf). The datasheet description is vague on details, but it appears rollertrol is claiming it's whole enchilada is consuming 30 microamps at 12v? If so, that sounds pretty good. Maybe too good. Does anyone here own one?
          I own one, but have no way of verifying the claims. it reacts instantly.
          HS3 Pro on Windows 8 64bit
          53 Z-wave nodes(46 devices, 7 remotes), 15 DS10a's, 10 ms16a's, 9 Oregon Sensors, W800, RFXCOMtrx433, Way2Call, 3 HSTouch Clients, 2xRussound CAS44, Global Cache GC100-12,10 Rollertrol blinds(+ zwave) ,3 Squeezebox Radios and 1 Squeezebox Boom,DMX Arduino via ethernet,Rain8Net,3x Echo Dot's


          Check out my electronics blog here:
          https://www.facebook.com/RaptorsIrrationalInventions

          Comment


            #6
            The specs for the NRF24L01 are:
            power consumption as low as 11.3mA transmitting,
            13.5mA receiving at 2Mbps,
            900nA in power down, and
            26uA in standby.
            Nominal voltage is about 3v.

            Therefore, standby means "not receiving." Hence, the rollertrol is going to be worse. So, for now, I'm scratching the rollertrol from consideration.
            Last edited by NeverDie; June 18, 2015, 08:50 AM.

            Comment


              #7
              Take a look at the ESP8266 as well. Looks like you could control a servo directly from it: http://www.roboremo.com/esp8266-servo.html as an example. Here's a site with a good overview of the ESP8266: https://nurdspace.nl/ESP8266.

              The Electric Imp could be an option too: https://electricimp.com/docs/examples/pwm-servo/. More details here: https://electricimp.com/docs/hardware/power_story/. It's likely not a cost-effective option in the quantity you need.

              I have used the NRF24L01+ and electric imp modules, and have some ESP8266 modules on their way from China. However, I've typcially used them in mains & usb powered applications.

              Cheers
              Al
              Last edited by sparkman; May 14, 2015, 10:18 PM.
              HS 4.2.8.0: 2134 Devices 1252 Events
              Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

              Comment


                #8
                Originally posted by NeverDie View Post
                Apparently the NRF24L01 can change modes very rapidly, so the idea is to keep it powered down or in standby. If it were to check for messages every half-second, the question would be: what percentage of that time can it be kept powered down? As a very rough calculation, if I assume power-up, check for messages, and power-down again all in under 1ms, then the average power consumption would be less than 28 microamps. Close enough.
                One thing to note is that there are reports of some of the NRF24L01+ clones that are coming from China don't have as good a power performance. However, it seems to be hit or miss. There are people on the mysensors board that have reported battery life in excess of a year using a couple of AAs powering a sensor, Arduino and the NRF24L01+ module.

                Cheers
                Al
                HS 4.2.8.0: 2134 Devices 1252 Events
                Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

                Comment


                  #9
                  I lost a lot of time trying to get my blinds automated but in the end I figured mine were bordering on the impossible because of they appeared to have some gearing that prevented them being turned by that central rod. I must of bought 10 stepper motors of varying shapes and sizes but never really thought of a servo.

                  I think you will struggle with the ESP8266 because it has some quite high current draw (potentially an amp) so running it off batteries could last a short time indeed.

                  Comment


                    #10
                    Originally posted by mrhappy View Post
                    I think you will struggle with the ESP8266 because it has some quite high current draw (potentially an amp) so running it off batteries could last a short time indeed.
                    Thinking about this more, the ESP8266 would need to stay powered on longer than the NRF24 module as well as it needs to associate with an AP, etc. before sending data whereas the NRF24 can just power up and send. The extra power draw of the microcontroller you need with the NRF24 is likely less than what's used by the extra on time of the ESP8266.

                    I have just started playing with servos (one of my sons is building an RC car) so don't have much experience with them yet. The only ones I have right now are clones from China: http://www.ebay.com/itm/351185473583 and http://www.ebay.com/itm/281402653743. They are not particularly quiet, but not that noisy either, but I don't have any high-quality ones to compare against.

                    Cheers
                    Al
                    HS 4.2.8.0: 2134 Devices 1252 Events
                    Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

                    Comment


                      #11
                      Originally posted by NeverDie
                      Thanks for your post. If it's possible to drive a stepper smoothly (?), I suspect it may be the quietest option. In looking just now at the stepper motors on Sparkfun, I notice that none of them have gearboxes. In contrast, the servos I've been looking at have all had gearboxes. Consequently I'm concerned servos are likely to produce, to one degree or another, a toylike noise when they operate. We could probably live with it, but if possible I'd prefer we avoid using a noisy actuator.

                      Out of the 10 stepper motors you tried, which were the quietest and which did you like the best overall?
                      They were all much of a muchness really, I salvaged various out of old printers, lighting equipment and others off places like DealExtreme (like this https://www.youtube.com/watch?v=M1PS-oMF350) so don't have makes/models.

                      The smaller ones are pretty much silent in use, some are probably higher current than some servos and a good proportion are circular which has some mounting issues inside a blind mechanism. In terms of driving them smoothly you will see in the specifications that you get a degree per step, the higher this is the less smooth each step will be. I anticipated that I would need higher torque than I could get from a servo which was why I tried them but never really got them to drive the rod in the blinds because of the mounting issues (and the fact the rod in my blinds was square rather than round).

                      I'm not sure after all the headache whether it may be easier to buy something, I think that is the direction I am going to go in this time...

                      Comment


                        #12
                        Originally posted by NeverDie
                        Meanwhile, back to the NRF24L01. Because the blind's header channel is metal, and because the NRF24L01 appears to have only a trace antenna, it looks as though I'll have to mount the entire NRF24L01 outside the channel. I suppose it could be hid behind the valence, but I'd rather keep it inside the channel and just run a wire whip antenna behind the valence. Are there any NRF24L01 breakout boards with a through-hole that allows for adding a wire whip antenna?
                        I haven't seen any with just a through-hole, but you can get ones with a connector for an external antenna. Most of those that you see include an PA/LNA and get better range, but also draw much more power when transmitting. Here's a couple of examples of ones with an antenna connector: http://www.aliexpress.com/item/Free-...968512470.html and http://www.ebay.com/itm/200932672566. And here's an example of ones with a connector and PA/LNA: http://www.ebay.com/itm/310651702557. I've used the ones with a PA/LNA and the ones with the trace antenna, but not the ones with the antenna connector but no PA/LNA.

                        Cheers
                        Al
                        HS 4.2.8.0: 2134 Devices 1252 Events
                        Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

                        Comment


                          #13
                          Originally posted by NeverDie
                          Sheesh! It turns out IPX connectors come in different sizes:

                          but the above ebay listing doesn't say which size it uses.
                          Given that the pin spacing on the module should be .1" or 2.54mm, I'm guessing it's a 2.5mm connector. No guarantees though . Good price on those modules too. There are reports on the ones with PA/LNA drawing significantly more in sleep mode than what is advertised. One way around that is to power the whole module from a digital pin on the Arduino. Given the max current draw of the module, you can't power it directly so would need an intermediate transistor.

                          Cheers
                          Al
                          HS 4.2.8.0: 2134 Devices 1252 Events
                          Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

                          Comment


                            #14
                            Originally posted by NeverDie
                            Without adding some kind of encoder to a stepper motor, to make it closed loop, manual adjustments to a blind seem like they would be problematic: without manual adjustments, all adjustments would have to be done through the motor, which may not be convenient, especially for other family members or guests.

                            By definition, a servo is closed loop, so manual adjustments shouldn't cause any problems.
                            If you're using a stepper, I would remove the rod that gives manual control and allow others to adjust using a remote control (talking to the central controller, rather than each blind direct) or provide local buttons.

                            Cheers
                            Al
                            HS 4.2.8.0: 2134 Devices 1252 Events
                            Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

                            Comment


                              #15
                              Originally posted by NeverDie
                              Anyhow, it occurs to me that most of the blinds only need to open in the morning and close in the evening, and for that RF isn't even needed, except to maybe to occasionally report battery capacity. By allowing manual adjustments, it becomes a simple 80-20 solution for maximizing battery life. For that matter, an LED which flashes very slowly would be all that's needed to know when it's time to change batteries, and then RF wouldn't be needed at all. That said, for convenience, I'd probably still prefer to find a way that would allow OTA re-programming of the micro-controller.

                              So,by relaxing the above constraints, the challenge is mainly reduced to finding a quiet servo with enough torque to tilt the blinds.
                              For me there are large changes between sunrise and sunset and then twice a year adjustments for DST and I want to control some blinds depending on how much sun there is and how hot it is, so for me RF is needed. We still need to select new blinds that allow for motorized operation, so I'm still a ways off on implementing. Keep us posted on what you find for servos.

                              Cheers
                              Al
                              HS 4.2.8.0: 2134 Devices 1252 Events
                              Z-Wave 3.0.10.0: 133 Nodes on one Z-Net

                              Comment

                              Working...
                              X