Starting w RFM69 & OpenHAB (Mail Box Node)

[sunbloquer]

A question. Is it normal for the voltage regulator to heat up? To the touch almost burning ... use the option of 4 AAA batteries of 1.5 v. that would be 6 v and with the multimeter at 3.65 - 4.00 v. between 5v and GND of various parts of the Arduino Standalone.

If the RFM should go to 3.3v, it will not burn with 4v.?

A greeting.

[papa]

Yes, sunbloquer, I believe the regulator can get pretty warm & how hot depends on how much the input is above the specified output.

You wrote "3.65 - 4.00 v" If I understand you correctly that you are sometimes reading 4 volts on the DIY Arduino node, I wonder if the regulator is defective.

Safe voltage for the RFM69? See page 11 of this document. www.hoperf.com/upload/rf/RFM69HW-V1.3.pdf
Safe operating range is specified as 1.8-3.6 volts. Above that risks long term reliability.

As an option to supply power for a DIY Arduino, computourist once suggested omitting the voltage regular. Instead use a 3.8 volt lithium ion battery in series with a germanium diode to somewhat reduce the voltage to a safer range for an RFM69.

[sunbloquer]

Thanks papa. The defective one I think is me. I had placed the trainers backward ¡omg!. Now I have a measurement of 2.44v stable, that if, the regulator is heated enough to lower 6 v. at 2.44v and the regulator is supposed to be 3.3v probe several and of different types and all give me the same measure. At that temperature I do not know how long it will last, I will leave it overnight in a controlled place, depending on what happens I will try to use it at 3.8 volt lithium ion battery in series with a germanium diode. What I do not know is if giving 2.44v and being the ATMEGA set to work at 3.3v will work ........

With a digital thermometer it gives me a temperature of 200 Fahrenheit .... I have it put as shown in the picture the capacitor on the left of 100uF 25V and the next one of 100uF 16V. All other components are of the same value that appears in the scheme.


cdn.instructables.com/FQ2/AP5C/HIQ4NZLM/FQ2AP5CHIQ4NZLM.SMALL.jpg

Something I must have more, they heat up to the batteries.

I will start from the beginning in a larger format. Now it only gives me a measure of 1.04v

[papa]

sunbloquer, if I understand you correctly, you somehow did not connect the voltage regulator correctly. I believe I did that once & the regulator did not work correctly after that. We all make mistakes. The important thing is to learn from them & do better next time.

[sunbloquer]

Well, until I find the fault and to be able to continue advancing with the project I have seen that maybe I could use

www.sparkfun.com/products/11114

Afterwards I continue with the standalone, which I have material now to make them in series

[papa]

sunbloquer, instead of an Arduino pro mini, you might try a Mini-WirelessHW at your chosen RFM69 frequency. See them on this page.

These compact devices with installed RFM69 radios can be programmed & used much like the RFM69 nodes our forum documents. My node choices sketch offers a Miniwireless definition. I've often thought that the Miniwireless could be powered by battery, but have not tried it. If so, that could be good for something like a mail box node.

Our documentation & programming includes being able to program the Miniwireless wirelessly (OTA). See this thread & this thread.  Although this function would be tricky if a Miniwireless node was using the SLEEPY mode to make the battery last.

A caution about getting 3.8 volt lithium batteries: Don't go too cheap as I did. As I found out, cheap ones are often fakes. They may provide 3.8 volts, but the current is so low, one can get nothing done with them.

[sunbloquer]

Every time I have more material to work, I do not know what I'm going to do with so much. I got the Pro mini and discovered the failure of the Arduino Standalone, rather failures, many and serious ........ how hard is to be ignorant.

Not all voltage regulators have the same pins, I was trying something impossible with 2 different types. The most serious is that I took the voltage at the entrance and not at the exit of the regulator, I had it placed backwards ......... today I just did the assembly I think that correctly, in the protoboard I had a correct output of 3.30v without fluctuations.

From mistakes you learn, the important thing is to keep trying.

[papa]

sunbloquer: "From mistakes you learn, the important thing is to keep trying."

papa: That has been very true for me in electronics & life. I learn as much or more from my mistakes as from my successes. Hang in there. I admire your perseverance.

Yes, on many things, study the pins many times before connecting.

[sunbloquer]

Hello again, we continue with ours. I have a mini arduino connected to the RFM, I upload the skecht and the serial monitor looks

_{You can try the mailbox node on a 3.3 volt manufactured Arduino compatible: Make sure lline 193 is UNcommented (reads #define DEBUG ...) so the Arduino serial monitor will use 115200 baud (line 213) to display node debugging output. With Arduino IDE running on a computer & the Mailbox Node connected via USB, upload the sketch program to the node. Then with the node still on USB, the Gateway successfully running connected to your Ethernet network & the IDE serial monitor running at 115200 baud, the serial monitor will display connection & data info about the Mailbox node.}



This is my sensor

www.aliexpress.com/item/Tilt-Sensor-Module-Switch-Microcontroller-Electronic-Building-Blocks-for-arduino-robot/32820195378.html

Change of the sensor can sense the object angle

Working voltage 3.3V-5V

Digital output form output switch (0 and 1)

A fixed bolt holes, easy to install

Small PCB size: 3cm * 1.6cm

aeProduct.getSubject()

The tilt sensor using MEC original tilt switch SW-460D, high sensitivity, the change is usually used to detect angle.

Module without tilt or slope angle is not up to the preset threshold, DO port output high level, when the sensor tilt angle exceeds the preset threshold, module D0 output low level;

Small digital output D0 can be directly connected with the MCU, the MCU to detect changes of level, thereby to detect objects angle;

Small digital output DO can directly drive the relay module, which can form a tilt switch;



I do not understand how to connect it and where and what it refers to ...

_{With the built, programmed Mail Box End Node connected to your computer's USB, the Arduino IDE Serial Monitor Output should look something like the following (before & after D3 connects to ground through some switch or jumper.}

In my Openhab

[sunbloquer]

OMG seems to have one to write it to ask for help and realize the mistake made. I need the cable DIO0 to D2. Nothing for the moment. I will continue informing. Short and close

papa:  Yes, it is very important that the RFM69's DIO0 pin is connected to Arduino D2.  Without that, you will get no radio communication.

Also the mail box node spends most of its time sleeping & saving battery power.  It only does radio communication when Arduino D3 gets a signal.

[papa]

sunbloquer, compared to the one I used for the project, your tilt sensor is more complex & bigger, perhaps more than is needed. With mine, you only connect to two pins (Arduino GND & D3) & it is small.

It looks like to wire your sensor, connect sensor's VCC pin to Arduino 3.3 volts, sensor's GND pin to Arduino GND, & connect sensor's DO pin to Arduino D3.

        Here is the English version of the product listing:

"Module without tilt or slope angle is not up to the preset threshold, DO port output high level, when the sensor tilt angle exceeds the preset threshold, module DO output low level."

That sounds like your sensor's DO pin outputs 1 (HIGH) when it is NOT tilted & outputs 0 (LOW) when it is tilted. For now, we'll assume that when sensor's DO connects to Arduino D3, that Arduino D3 will have the same state as sensor's DO.

It also looks like you adjust the sensor's sensitivity (how much tilt is required to output HIGH?) by adjusting the potentiometer.

I wondered "Where is the sensor when it's NOT tilted or From where does the sensor tilt? even though it should make no difference to the way I coded the mail box node & its OpenHAB configurations.

    From the sensor documentation that sunbloquer found:
"[This tilt sensor] only gives two values, either it is vertical or it is inclined (more than 15º). It is a switch, or makes contact [when vertical] or does not make contact [not vertical]."

So the sensor is vertical when it is not tilted & the switch is closed & DO outputs 1 (HIGH).  When the sensor inclines more than 15 degrees from vertical, the switch opens & DO outputs 0 (LOW)

The following posts look at the sketch & OpenHAB configs to see if the logic needs to change for this sensor. (I believe they do not need to change.)

[papa]

The choices sketch's code sections that might be affected by sunbloquer's different tilt sensor:

    #ifdef MAIL
#define SWCH4 3 // pin 3 for switch, usually HIGH via internal resistor
// pinMode(SWCH4, INPUT_PULLUP); // pull high w internal resistor, disabled for sunbloquer tilt sensor
    #endif // ifdef MAIL

    #ifdef MAIL // use LOW for d3 on pullup, use HIGH for d3 on pull down
if (send46) { // papa added Binary switch input read [not Button pressed]
    mes.devID = 46;
    if (curSWCH4 == LOW) mes.intVal = 1; // state (integer) !! D3 = LOW = 0 = tilted, not vertical, on sunbloquer sensor
    if (curSWCH4 == HIGH) mes.intVal = 0; // state (integer) // !! D3 = HIGH = 1 = no tilt, vertical, on sunbloquer sensor
    send46 = false;
    txRadio();

I believe the logic of the 2nd code section is OK since the sensor's DO states function the same way as the code, but I edited the comments.

In the first section of code above, I disabled a pullup to HIGH state, since sunbloquer's tilt sensor already functions this way.  Edit to match the total line in blue text.

[papa]

    The OpenHAB configurations that might be affected by sunbloquer's different tilt sensor:

Contact Mail82 "Mail82 [%s]" <lock>
{mqtt="<[mosquitto:home/rfm_gw/nb/node07/dev46:state:OPEN:OFF],<[mosquitto:home/rfm_gw/nb/node07/dev46:state:CLOSED:ON]"}

rule "update mailbox"
    when
        Item Mail82 received update
    then
        postUpdate(mailbox_time, new DateTimeType())
end

Note: When the mailbox door is opened & closed, the node sends the change of mailbox state. When OpenHAB receives that update of D3's state via device 46, the rule above also updates the date & time to be posted on the UI.  The node will also send signal strength & voltage unless that part of the sketch is disabled.

Text item=mailbox_time
    ==============================

In the Contact item above, I  believe OPEN:OFF should be OPEN:ON & CLOSED:ON should be CLOSED:OFF.

However, I believe the OpenHAB configs will work as they are because the mailbox door will both open & close.

If an approach different from mine is used to signal movement of the mail box door, the Contact item may need to be changed.

[papa]

    Using the Mailbox Node with Sunbloquer's Mailbox & Tilt Sensor

      Sunbloquer pictured his mail box like this: 

I assume most of the node will be in the lower chamber.  I assume the tilt sensor will be mounted flat against the Top Door's lower side (the side facing us in the image above) with the sensor cylinder pointing toward the top door's top lip.  IF the top door & sensor move to less than 15 degrees of vertical, the sensor's D0 changes output to HIGH. IF from vertical, the top door then swings at least 15 degrees toward closing, the sensor's DO changes output to LOW. Both state changes should cause state updates on the D3 pin & generate interrupts to wake the node & send the update to OpenHAB & the OpenHAB rule should time stamp the last state change.

The mail box node should sleep on low power most of the time (stretching battery charge).  It should wake up only when the Top Door opens vertical enough & closes & only at that time, send the data enabled in the sketch: tilt switch state plus perhaps signal strength & voltage).  In that case, via the MAIL code in the choices sketch & the OpenHAB configs, the
OpenHAB User Interface should show the latest time the Top Door was
opened & closed, a signal that perhaps new mail was received.

For this to work, the top door & sensor may need to open near enough to vertical.  To achieve this, sunbloquer may need to adjust the sensitivity of the tilt sensor OR somehow position the sensor so that it will usually get near enough to vertical when the top door is open.

[sunbloquer]

Well, I have achieved communication between the gateway and the node. In openhab I do not see the data that voltage and signal should show. 1 hour connected to usb and takes about 30 minutes connected to external source, at least that data should be updated truth?



I have tried several ways to connect the sensor but I do not see any that work, it has two leds

Pw Led - It works when I give it tension

DO Led - Off. It turns on when I change position

Question - For the sensor

Maybe for the sensor it would be better to use the analog input instead of the digital one, I do not need to know how much it is inclined, we are only interested in knowing the connection / disconnection.

D0 to A0

GND to GND

VCC to VCC

I think I should make a copy and paste it by changing the node82 data in My.rules. Its correct?

// query every 1 min. Example of "pulling data" from a node

rule "refresh Node02 Voltage, RSSI"

when

Time cron "0 0/1 * * * ?"

then

sendCommand(getNode2Voltage, "READ")

sendCommand(getNode2Rssi, "READ")

end

Do not bother to check code, change sensor or remove the sensor to that module and that's it. Thank you very much

I found this documentation in Spanish

translate.google.com/translate?sl=es&tl=en&js=y&prev=_t&hl=es&ie=UTF-8&u=http%3A%2F%2Fkio4.com%2Farduino%2F40interruptorinclinacion.htm&edit-text=

[sunbloquer]

You were faster than me. Thank you very much papa, tomorrow we will try.

papa:  Yes, that documentation of the tilt sensor is helpful.  Your tilt sensor is the first type documented: 3 pins, DO sends a digital signal of 1 or zero.  With our coding, we are using an interrupt to wake the sleeping mail box node.  The documentation says to use Arduino D2 for interrupts.  We already use D2 (interrupts possible) for the RFM69 radio.  So we use Arduino D3, the other pin where interrupts can be attached.

If the sensor's DO is connected to Arduino A0, opening & closing the mail box door canNOT wake the sleeping node because an interrupt cannot be attached to A0.

For the mail box node, DO must connect to Arduino D3 (a pin to which one can attach an interrupt).

[papa]

sunbloquer: In openhab, I do not see the data that voltage and signal should show

 I think I should make a copy and paste it by changing the node82 data in My.rules. Its correct?

// query every 1 min. Example of "pulling data" from a node

rule "refresh Node02 Voltage, RSSI" ...

papa:  The rule like that will probably not help. That rule commands an AWAKE NODE to send voltage & RSSI data to OpenHAB. A mail box node that is SLEEPING most of the time will not receive or respond to that command.

The sketch code already sends OpenHAB voltage & RSSI, but ONLY when tilting the sensor wakes up the node.

Your Serial Monitor image shows your node is already sending data for voltage & RSSI, but maybe you mean voltage & RSSI are not showing on OpenHAB's User Interface (as I can see from that image).

    Check that you have something like these items:
Number Node82Rssi "Mail82_RSSI [%.3f db]" <network> (Nodes) {mqtt="<[mosquitto:home/rfm_gw/nb/node82/dev02:state:default]"}
Number Node82Voltage "Mail82Voltage [%.3fVdd]" <battery> (Nodes) {mqtt="<[mosquitto:home/rfm_gw/nb/node82/dev04:state:default]"}

     Check that you have something like these sitemap entries:
Text item=Node82Rssi
Text item=Node82Voltage

If your items & sitemap have correct RSSI & Voltage entries, sometimes things get out of sync & data updates do not show on the User Interface.  Try refreshing the browser page that holds the User Interface. To get things back into sync, try rebooting OpenHAB (maybe easiest to reboot your Pi that runs OpenHAB).

[sunbloquer]

ok I think I understood it. Tomorrow we will try again and I will present the results.

I already have to invite you to more than one beer 😉

papa:  You are welcome, sunbloquer. A beer would be nice, but one of us might have to travel a long ways for that to happen.  Then again, I've never been to Spain, but who knows.

Just remember that when you are able to contribute new documentation later, please do so.

In the meantime, others are probably learning from your questions & statements & our conversation.

[sunbloquer]

Feb 14, 2018 18:55:03 GMT papa said:

The choices sketch's code sections that might be affected by sunbloquer's different tilt sensor:

    txRadio();
}

You forgot this I think 

In the sitemaps was missing this

Text item=Node82Voltage

homeautomation.proboards.com/post/1517/thread

We are moving forward but I think that today I will not be able to do more tests

2018-02-15 15:00:08.726 [vent.ItemStateChangedEvent] - Mail82 changed from NULL to CLOSED



2018-02-15 15:48:23.992 [vent.ItemStateChangedEvent] - Mail82 changed from NULL to CLOSED

I have a pending visit to the State of Texas, specifically to Houston. Sites I've visited ..... Washington, Chicago, Los Angeles, Las Vegas, San Francisco ......

Thanks again. When I have this working properly I review all the information I have, the steps to do it and encourage others to do it. Above all it is entertaining. 

Between my English and the google translator, it's horrible to read me....

[papa]

papa: Yes, sunbloquer, I also noticed that the sitemap entry for voltage was missing & I added it.

When you look at this thread today, be sure you look at all my posts from yesterday.

In this post (based on my understanding of the tilt sensor tutorial), I said, "For this to work, the top door & sensor may need to open near enough to vertical. To achieve this, sunbloquer may need to adjust the sensitivity of the tilt sensor OR somehow position the sensor so that it will usually get near enough to vertical when the top door is open."

I am wondering about the 15 degrees mentioned in the tilt sensor tutorial & rethinking what I said: I would think that when the sensor is near level, the balls could roll either way to OFF = zero = LOW or to ON = 1 = HIGH. By the time the sensor cylinder end of the whole unit tips up 15 degrees from level & is 75 degrees from vertical, I would think the ball would roll & close the switch. Then by the time the pins' end of the whole unit tips up 15 degrees from level & is 75 degrees from vertical, I would think the ball would roll & open the switch. << If this is correct enough, then how I suggested to mount the tilt sensor unit in that post should work nicely.  Once the mail carrier opens the Top Door at least 15 degrees above level (likely needed to be able to put mail through the top slot), I believe the switch would close & change the state of sensor DO & Arduino D3 (to HIGH) & wake the node to send data.

You will find out for sure by experimenting.

By the way, for your type of mail box, I thought of another approach that would keep everything inside of the lower chamber:  Instead of connecting a tilt sensor to Arduino D3, one might position a micro switch with a long lever just under the top slot in the box.  Mail dropped through the slot could hit the lever to close the switch & change the state of Arduino D3, & wake up the mail box node to send data.  If this works, it would truly signal that mail was received, not just that someone opened the Top Door.

With all the places you visit in the United States, we might get that beer together yet.  : - )  You're doing OK in English, much better than I'd do in Spanish.

[sunbloquer]

Feb 15, 2018 14:56:50 GMT papa said:

By the way, for your type of mail box, I thought of another approach that would keep everything inside of the lower chamber:&nbsp; Instead of connecting a tilt sensor to Arduino D3, one might position a micro switch with a long lever just under the top slot in the box.&nbsp; Mail dropped through the slot could hit the lever to close the switch &amp; change the state of Arduino D3, &amp; wake up the mail box node to send data.&nbsp; If this works, it would truly signal that mail was received, not just that someone opened the Top Door.

I had thought something similar to what you describe. I do not intend to put the sensor in the lid of the mailbox, I would have to mechanize the box so that it would go in, I had thought to do it in some way that a wire or wire moves to the sensor but being inside the mailbox, in this way it would be totally invisible. However, I already told the postman about my project and almost that he is more interested in seeing it finished than me, he is curious to see it working, I think to try to deceive him .

The final placement form will not know until you experiment with it.

With 2-year-old twins I think I'll be a season without traveling, we went with them to Venice (Italy) and it was a really traumatic experience

[sunbloquer]

I have it a bit stopped a step from finishing but I can not take time ....... I will continue ;-)

papa: Good to hear from you, sunbloquer.  I had missed you.  Enjoy children while you have them.  They grow up so fast.

Report back when you have opportunity.