Five beehive sensor ICs (I2C and 1.8V max)

On a sunny day (11 May 2019 15:30:54 -0700) it happened Winfield Hill wrote in :

Lots of sensors This runs here:

date and time CO level alarm, sensor is also sensitive to alcohol (from cleaning work duh :-) ). temperature relative humidity roll pitch air pressure GPS location magnetic heading some more things...

all sensors are either directly in a raspberry as plug-in board, the rest via power over ethernet remote from that computer. Thing has own IP and this shot is from the PC app elsewhere, When planes is active it shows those too, heading, altitude, speed, angle from observer,

and when AIS is active it shows the other boats and data: All RJ45 for the remote sensors. RTL_sdr DVB sticks for receiving. Only GPS is directly serial to raspi /dev/ttyAMA0

The circuit diagrams are on the here well known pencil drawn pieces of paper the C sources for the raspberry and asm sources for the PIC based sensors are on my site (the old ones anyways)

iam obviously showing of .. Now how to make a bee position and flight direction and speed and altitude sensor? This is you next task!

observer,

It may be possible to acquire bee flight data after the fact. When bees return from foraging they use "dance" steps to communicate target type (water, nectar,...), direction, and distance to the other bees already present in the hive. [1] There's an electric field component to the dance and a hive provides a handy stationary spot for EM observation.

Note.

1. formatting link

Thank you, 73,

--
Don Kuenz KB7RPU 
There was a young lady named Bright Whose speed was far faster than light; 
She set out one day In a relative way And returned on the previous night.

On a sunny day (Sun, 12 May 2019 16:27:17 -0000 (UTC)) it happened Don Kuenz wrote in :

Yes, I know that, I was thinking radar, but that needs very high frequency short wavelength so light. So scanning laser, or simpler:

2 video cameras looking at the sky at different angles to get 3D. Bees may go far away, so maybe 2 cameras from a drone, or some high masts, looking down. The very difficult [1] way would be to put a microchip on each bee with GPS and a transmitter, that is for the IC designers here :-) [1] bees may not like that ... But that is foretell

Have you considerd putting transient protection on SCL and SDA, Transzorbs or diodes to supply and ground?

I2C is initially intended for connection between IC's on the same board only. It is pretty exposed to outside influences.

--

-TV

I use both on external i2c sensors: BAV99W on each of SCL and SDA with top and bottom to Vcc and ground. Then a transzorb or zener between Vcc and ground plus of course a decoupling capacitor. Also, some series resistance between the BAV99Ws and the pins of the sensor. I need to check the exact values I used, but I think it was around 22 ohms - calculated to ensure that all the voltage margins were sensible with lowest permitted pullup resistors. A few hundred k units like this have been shipped and they passed all the regulatory testing (with a specified maximum unshielded four core flat cable length of 3m).

John

For sensor systems in remote locations it makes sense to make the i2c power switchable in order to have an "if all else fails" way of recovering from bus lockups.

John

Good points, both of you.

OK, I've added a Bourns CD143A TVS clamp to the two I2C lines, where they enter the sensor board, plus two 22-ohm resistors, plus a 2nd TVS clamp on the I2C chip lines. The CD143A-SR3.3 is spec'd to drop 8.2V drop at 10A, and the 22 ohms limits the current into the 2nd clamp to 200mA. In a previous project, where I used CD143A, their SOT-143 package seemed small, but compared to sensor packages on my new board, they're gigantic.

Hah, while at it, I added another sensor, a TLS2591, for light. It's dark in a hive, even darker with constrictions from our 24 black narrow bee-counting channels. Last summer I observed some unusual goings on at night, and would like to know the light level in there. Hell, we really need a few cameras!

--
 Thanks, 
    - Win

Yes, it makes total sense. All our sensor power goes through a handy FPF2003 current-limiting switch. But the switch's control line was grabbed for some other purposes in the instrument. Our Feather controller only has 21 pins for IO, and we use them mercilessly. Many perform two functions, some three, and one pin does four (!) different things, depending on states of other pins. We already had individual enables, and an Aen line, for power-hungry sensors, etc., so we grabbed the switch pin. Now I wish it was back.

--
 Thanks, 
    - Win

Winfield Hill wrote in news: snipped-for-privacy@drn.newsguy.com:

Have the guys over at Cambridge apply a VantaBlack coating and you will get zero light back from any surface. It actually looks like you are looking at a hole in the universe.

Photos of it appear to have been edited when they have not.