My first project - multi-sensor

[fvhemert]

When I was testing the NodeMCU for Greig, all 4 of my test boards ran for a month with 5V supplied to the V in (5Vs) pin. From the pictures of your boards yours have a V in pin as well.

[ATTACH]60543[/ATTACH]

The bottom pin of the left row - I think it also supplies 5V to the USB conector

So here is what I understand from doing research and looking at the schematics.

Your board has 2 ways to provide power:

- Through USB
- Through the pin marked Vin

Both will end up at the input side of a onboard voltage regulator that steps it down to the required 3.3v.

If you need 5V to for example power a LCD you can take that from the pin marked VU.

I have seen reports where people use the Vin as a 5V supply when using USB but the version I am using does not seem to have a direct connection to the 5V supplied by USB so mileage may vary.

[zwolfpack]

... the version I am using does not seem to have a direct connection to the 5V supplied by USB so mileage may vary.

Generally these designs will have a diode similar to this one. This is so the USB VBUS can power the board's VIN, but not vice-versa. This is required by USB spec and is a good thing; without this, if your USB host (PC) was powered down, your micro-powered USB device might be trying to power a bunch of other attached USB peripherals!
When powered from USB, VIN will be the USB VBUS voltage minus the (small) diode drop.

[petez69]

Generally these designs will have a diode similar to this one. This is so the USB VBUS can power the board's VIN, but not vice-versa. This is required by USB spec and is a good thing; without this, if your USB host (PC) was powered down, your micro-powered USB device might be trying to power a bunch of other attached USB peripherals!
When powered from USB, VIN will be the USB VBUS voltage minus the (small) diode drop.

Just ordered a few samples of this device, they look good. Thanks for sharing

[fvhemert]

Generally these designs will have a diode similar to this one. This is so the USB VBUS can power the board's VIN, but not vice-versa. This is required by USB spec and is a good thing; without this, if your USB host (PC) was powered down, your micro-powered USB device might be trying to power a bunch of other attached USB peripherals!
When powered from USB, VIN will be the USB VBUS voltage minus the (small) diode drop.

Yes, and some boards do not even have a Vin or Vusb pin, they just show Reserved (one zillion versions of boards )

[petez69]

Some of the Chinese clones are just nasty :-) Its fun though to buy things from Aliexpress and experiment with them...

[fvhemert]

Some of the Chinese clones are just nasty :-) Its fun though to buy things from Aliexpress and experiment with them...

Agree EURO 3.59 (shipping included) for a NodeMCU is worth the try, it makes me wonder how they can still make money...

[mrhappy]

How hard would it be to change it to use a 9v battery or AA batteries. I wonder how long the batteries would last?

Not very long at all on battery, you are talking hours rather than weeks. They have a varied current demand through their cycle but if you took it at 100mA on average then at 5V if you used a 1000mAh USB power bank you are talking ten hours and that's saying things are being 100% efficient. You've then got to think of what is being powered by it, whether the battery is 100% charged, how often the device is used etc.

You can access some of the more involved sleep methods to lower the current on the ESP8266 but it can get advanced and I still don't know you would get the mileage out of one, IMHO if you have to replace the batteries more than once every six months then it rapidly becomes a chore and annoyance.

[logbuilder]

IMHO if you have to replace the batteries more than once every six months then it rapidly becomes a chore and annoyance.

I agree with this. The thing that motivated me to explore Arduino was two fold. First, the high cost of Insteon sensors. Second, the need to replace batteries. The straw that broke this camels back was a battery powered motion sensor in a heavily used area such that battery usage was poor and (I can't prove this) it seemed to me that the sensor started to miss triggers when the battery got low.

[logbuilder]

The multi-sensor is now in a live environment. It is in my garage which is 200 feet away from the house. I was concerned about wifi connectivity but that has not been a problem as far as I can tell.

The problem I have found is with the PIR motion sensor. It triggers every minute or so. There is no motion out there. I even covered it up with a cloth and it still triggers. I got these PIR sensors from Radio Shack. They were 60% off due to store closing. Problem is they have no adjustment for sensitivity or trigger delay. I will replace the PIR with a version that has the adjustments.

[Blade]

What is the difference between these 2 nodeMCUs?

http://www.ebay.com/itm/191905456448...%3AMEBIDX%3AIT

http://www.ebay.com/itm/NodeMcu-Lua-...-/192113216526

[logbuilder]

What is the difference between these 2 nodeMCUs?

http://www.ebay.com/itm/191905456448...%3AMEBIDX%3AIT

http://www.ebay.com/itm/NodeMcu-Lua-...-/192113216526

One is V2 and the other is V3. I'm providing a link below with the whole story on the different versions of the nodeMCU. I've have one of each. The V3 board is wider by 2 rows of pins. As such, it does not work well on a breadboard. I'd get the V2 if it were me.

https://frightanic.com/iot/compariso...ent-boards/#v3

[bpm32]

The problem I have found is with the PIR motion sensor. It triggers every minute or so. There is no motion out there. I even covered it up with a cloth and it still triggers. I got these PIR sensors from Radio Shack. They were 60% off due to store closing. Problem is they have no adjustment for sensitivity or trigger delay. I will replace the PIR with a version that has the adjustments.

I have built several units similar to this for around my home using PIR sensors from Amazon. (Not sure of the specific seller) I experience the same issue as you (albeit on a different time variance than you) and attributed it to the wifi chip on the NodeMCE causing interference that is detectable by the PIR.

A couple different solutions:
1. Put the wifi chip to sleep until the unit detects motion. This slows reporting down as wifi reconnection isn't instantaneous.
2. Move the PIR off board by no less than a foot. This seems to get it out of the wifi field.

Aluminum foil wrapped between the PIR and the wifi chip may help, but that wasn't my experience.

Congratulations on your project - I think you'll love it. I really do like how mine came out. I even 3D printed custom boxes for mine. They look slick... if I do say so myself.

HTH,

Brian

[fvhemert]

One is V2 and the other is V3. I'm providing a link below with the whole story on the different versions of the nodeMCU. I've have one of each. The V3 board is wider by 2 rows of pins. As such, it does not work well on a breadboard. I'd get the V2 if it were me.

https://frightanic.com/iot/compariso...ent-boards/#v3

THis is the board I use http://www.ebay.com/itm/172304083455 been running it for 2 weeks now without issues. Might be interesting, it's half the price

[logbuilder]

I see many components that are way low priced. The thing they often have in common is that they are from China. Delivery times are very slow. The product you reference has a delivery time of a month. If you are willing to wait, it might be a good deal. But who knows what it would be like to get satisfaction should you experience a problem.

[logbuilder]

I have built several units similar to this for around my home using PIR sensors from Amazon. (Not sure of the specific seller) I experience the same issue as you (albeit on a different time variance than you) and attributed it to the wifi chip on the NodeMCE causing interference that is detectable by the PIR.

A couple different solutions:
1. Put the wifi chip to sleep until the unit detects motion. This slows reporting down as wifi reconnection isn't instantaneous.
2. Move the PIR off board by no less than a foot. This seems to get it out of the wifi field.

Aluminum foil wrapped between the PIR and the wifi chip may help, but that wasn't my experience.

Congratulations on your project - I think you'll love it. I really do like how mine came out. I even 3D printed custom boxes for mine. They look slick... if I do say so myself.

HTH,

Brian

I think you are onto something there. When I was testing inside on the bench, I kept the cover open and the PIR was pointed at the table it was sitting on. No motions unless I moved the lid to test the sensor. When I moved it to the garage for real life testing, I closed the cover. It has fired that PIR ever since. I didn't make the connection. Just went outside and opened the cover and it has not fired for 10 minutes. Gonna go out and move it to see if I get any motion logged. Just tested it. Yep, motion was logged. Your hunch is looking more and more true. Humm... gotta think of the alternatives.

On a weird note, I kind of miss the false motions since they kind of gave me a heartbeat on the device.