help needed with multiple temp sensor data logging project

[wallyllama]

If you search "1-wire hub", there are premade options, and diy.

[MasterCATZ]

thanks for your guidance
parts arrived yesterday and just got the probes reading

how ever I could not get OWFS to work,
https://www.domoticz.com/wiki/1Wire

sudo i2cdetect -y 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --

but using /sys/bus/w1/devices
they are there

cat w1_slave
a2 01 4b 46 7f ff 0c 10 49 : crc=49 YES
a2 01 4b 46 7f ff 0c 10 49 t=26125

edit

actually quite happy, using 5v and daisy chained a heap of cat5 cables together for testing, over 100m all working good

[wallyllama]

@MasterCATZ :smiley_cat:

[zboblamont]

@MasterCATZ I'm following this with interest as looking at a WAY smaller scale use with the small 3 legged 18B20 to give individual room temperatures in the house, 4 upstairs, 6 downstairs.
Had originally considered individual room nodes, but this seems a slightly simpler solution (other than lifting parts of the upstairs floor).

[MasterCATZ]

@zboblamont I will be able to test the lengths today, making another 20x temperature sockets and I have a heap of 20m cat5 cables to chain from

I am using a trim pot to adjust at the breadboard, so far still 4.7

if I was going to do it again would use a 5 or 10k high precision the 5% pot too dodgy

still yet to use resistors on each drop and I have a 10v 1000uf capacitor on each ground / vcc, no idea if it helps or not

you could also run a 100mbs LAN plus temp probes each using 4x of the 8 cat5 pins

http://www.108relays.ca/dl/1_Wire_Design_Guide_v1.0.pdf

edit

it looks like you have to keep the lengths and probe numbers equal if your splitting them off like a tree

if I had 1 side 6m longer with more probes than the other side a probe read 0 deg

added more probes on the other side and they read fine again
( touch wood )

I have placed a probe at the end with no capacitor, can not get it to connect

when I make more probes tomorrow will try that spot again with a capacitor

for now will try the one with out capacitor at the first split near the pi to see what happens, it's pitch black now time for bed :P

strange thing just tried another stock probe with no capacitor and can not get that to work either and I was running these ones inside all day
for some reason voltmeter reads neg millivolts

now I just want to build another cap probe to see whats going on

edit , either I wired something backwards or maybe i just hit the limits adjusting trim pot now as even the DS18B20 on the bread board stopped responding

edit, somethings not right pulled the main link and its receiving 0 deg from sensors not even plugged in.

new probe and bread board temps working tho

set @ 4k , tried adding another probe at 1st split all died again,
I'd say at my limits without the 100~150 resistors trying to push 6x 20m cables
voltage still 4+

but I am sus as to why a probe felt warm where a cap was done, unsure if it was still from the soldering heat or not so I might have something miss wired as I was running as the day light was dropping fast
( this setup is for monitoring tanks in my polytunnels no lights apart from the flash light in my mouth )

the goal was 3 x 20m on each side split at the breadboard with ~5m long branches running of each join with 2x probes about every 1 m with 1m long lengths approx 60x DS18B20 and about 210m of cable

unfortunately, I was 1 section short the 5th section would not work
so 50 probes 175m cable

I'll try again in a daisy chain formation later on

currently using raspberry pi 2 B and domoticz

I ended up doing rj45 as they worked out cheaper than the rj11/14
I paid $20 for 100 , also why I did not add the capacitor inside like I was going to so they could be used in other areas for LAN, hopefully, no one tries to plug a laptop into them

http://www.ebay.com.au/itm/CS-5X-US-10x-Practical-3-Way-RJ45-LAN-Network-Ethernet-Splitter-Connector-Beige/201963874318

[MasterCATZ]

might be best to NOT buy the ones I got , just discovered 1x port of just about all of them was not working correctly
( pins on slight angle off setting alignment )

causing the probes to be in parasitic mode

[zboblamont]

@MasterCATZ I should not have the same issues as the scale in my case is small, but will set up a test with the devices individually just in case before installing. This is the type on it's way (not from this supplier), I can only hope they are genuine components. link text
I will also have to read the guide again as the devices placement and harmonics were a bit of a puzzle, and on one hand they recommended Cat5e yet went on to say twisted pair was better.

[MasterCATZ]

anyone know the correct way I should be using an external power source

I powered them up directly with + - still running to the pi and blew the external 2a 5v power pack ( it also powered up the pi and I believe back-feeding this way removes surge protection )

then I tried with the GND and DQ connected with the pi doing the pull up , external PSU ( 30A rated 5v rail 450w 3.3v 12v 5v ) looked like it was shorting out dropping .2v per probe and changing sound as tho it was under load how ever I am currently using its 3A port

I then tried with just the DQ connected not much luck

then I moved the bread board over the PSU and had the 4.7k pull up on its end and DQ / GND connected to the pi , I seem to have some connections, but getting a lot of unknown vendor type errors
but PSU does not seem to be under load

http://datasheets.maximintegrated.com/en/ds/DS2482-100.pdf
I might get one of these I2C one wire bridge that can adjust the strength of the pull-up dynamically as the network grows

http://videos.cctvcamerapros.com/voltage-drop-calculator

has got me thinking I might double up on the strands 24 gauge has too much voltage drop @ 5v

should I just treat each pair as a wire or should I have ground on each twist ?

ie
( p1,p2 GND) Green pair
( p3,p4 VCC ) Blue pair
(p5.p6 DQ ) Orange pair
( p6,p7 GND ) Brown pair

or
(p1 ground p2 VCC ) Green pair
(p3,ground p4 VCC ) Blue pair
(p5,ground,p6 DQ ) Orange pair
(p7,ground,p8 DQ ) Brown pair

or any other combo, I would rather use a 2amp PSU than a 30 amp .. but thats so far the only way to keep these things powered up in a daisy chain

https://forum.allaboutcircuits.com/threads/24-gauge-wire-if-you-double-up-strands-what-equivalence-gauge.108649/

[wallyllama]

It looks like the standard for power over ethernet uses a twisted pair as a "single wire". So blue/blue-white is + and brown/brown-white is -

Seems like a reasonable way to do it, even though you arent wiring ethernet.

[zboblamont]

@MasterCATZ Perhaps you are expecting too much of the network you wish to monitor, which to be fair is not small. Splitting the network into smaller more controllable and manageable segments may be counter intuitive, but can be efficient even if you derive a solution beyond your own pre-determinations.

[MasterCATZ]

@wallyllama thanks that is what I feared, and would have made the job a lot easier
how ever I was thinking about doing it the other way to keep the other probes already made backward compatible as they used pins 1-4

just trying to weigh up the pro's and cons
also out of all those 3 way splitters only 15x actually have all pins functional ...

so trying to think of a way to still use the dodgy ones by having redundancy's in place

I think the way the PSU is wired up at the moment is the correct way
things don't seem to be stressed out this time

I am also raising my polling times as I think I read somewhere it takes the sensors 750ms to perform a function and uses 1.5 mA each?

Sensor Poll Period: 1000 to 30000
Switch Poll Period: 100 to 1000

my way of thinking this should still have all 60 sensors read in 1 min?
60 sec / 1000ms for Switch Poll period?

edit kept my original pin out just doubling up instead of having in/out

I will just change the colors to make it easier to put into the jacks
with out doing A / B standard Ethernet wiring

Pin 1 (White/Green) - Power return or ground
Pin 2 (Green) - +5V
Pin 3 (White/Blue) - Power return or ground
Pin 4 (Blue) - DQ (data in)

Pin 5 (White/Orange) - 1-Wire return or ground
Pin 6 (Orange) - +5V
Pin 7 (White/Brown) Power return or ground
Pin 8 (Brown) - DQ (data out)

this way a ground will be on every twisted pair
voltage is higher now

and now that I just soldered everything up, I notice I don't even need the pull-up because I am using external PSU anyhow
I knew I should have put in a bypass switch

[wallyllama]

@MasterCATZ said in help needed with multiple temp sensor data logging project:

@wallyllama thanks that is what I feared, and would have made the job a lot easier
how ever I was thinking about doing it the other way to keep the other probes already made backward compatible as they used pins 1-4

just trying to weigh up the pro's and cons
also out of all those 3 way splitters only 15x actually have all pins functional ...

The big "con" is the risk that someone will come along and try to use it in a standard way.

so trying to think of a way to still use the dodgy ones by having redundancy's in place

I think the way the PSU is wired up at the moment is the correct way
things don't seem to be stressed out this time

I am also raising my polling times as I think I read somewhere it takes the sensors 750ms to perform a function and uses 1.5 mA each?

Sensor Poll Period: 1000 to 30000
Switch Poll Period: 100 to 1000

my way of thinking this should still have all 60 sensors read in 1 min?
60 sec / 1000ms for Switch Poll period?

There is a function that tells all the sensors to start a temperature conversion, you could then read each one individually. So it would be 750ms + (60 * pollingtime) to get them all. I dont know if you trade speed for reliability or not. You may be able to interleave conversion requests and polling.

Convreq1 poll30 convreq2 poll31 8convreq3 poll 32 .... convreq 60 poll29 - loop

Maybe overkill if 1 reading per minute is enough