Simple block detector - zero loss

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 03.03.2014, 21:13

rjversluis wrote:Hi Richard,

if you force a shortcut what ADC value is reported by your I2C solution?
Or is there a threshold possible for shortcut/overload of a section?
I have no idea. I haven't got to the shortcut stage yet. Sorry.

Rich

rjversluis
Site Admin
Posts: 41184
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Simple block detector - zero loss

Post by rjversluis » 04.03.2014, 13:09

Hi Richard,

in Rocnetnode 6915+ the AOUT pin(15) will be set to 255 in case one or more analog inputs are reporting a value above the threshold.
On the AOUT you can connect a resistor and LED.

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 04.03.2014, 17:20

Yes you can connect a resistor and LED to the DAC output. Having said that, I have seen some leakage between the DAC output and the ADC inputs. I didn't pursue why that was occurring as I considered the analog output as unused. The maximum amount of leakage was about "6-8" when reading the registers.

Ramp the DAC and turn it into a "I am alive" indicator. Could flash the LED to indicate which sensor is activated too. 1 flash = sensor one, etc.

Other than a activity indicator I could not think of a use for that single DAC output.

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 04.03.2014, 17:33

One more item regarding leakage. The amount of leakage I saw is added to the value read.

Isn't playing with hardware fun? :D

rjversluis
Site Admin
Posts: 41184
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Simple block detector - zero loss

Post by rjversluis » 04.03.2014, 18:02

Make sure your reference voltage is stable.
We encountered the same problem with RocMouse, but was solved after using a stable 3V3 from a voltage regular.
http://wiki.rocrail.net/lib/exe/fetch.p ... 04_sch.pdf

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 05.03.2014, 01:30

Thanks for the tip. The next time I run a test on the ADC I will check into it.

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 05.03.2014, 05:28

If you want max sensor capacity, check out the LTC2495.

WIth 16 channels and 27 addresses that yields 432 sensors per I2C bus. Using both buses on a Rpi, that makes for a total of 864 sensors.

rjversluis
Site Admin
Posts: 41184
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Simple block detector - zero loss

Post by rjversluis » 05.03.2014, 08:08

Richard-TX wrote:If you want max sensor capacity, check out the LTC2495.

WIth 16 channels and 27 addresses that yields 432 sensors per I2C bus. Using both buses on a Rpi, that makes for a total of 864 sensors.
For those who need it...

The Rocnetnode concept is distribution not concentrating.
In this case the 32 zero loss sensors per Rocnetnode fits in that concept.
http://wiki.rocrail.net/doku.php?id=gca-pi07-en

Liviu M
Posts: 932
Joined: 03.12.2011, 20:44

Re: Simple block detector - zero loss

Post by Liviu M » 05.03.2014, 08:25

Hello Rob, hello Peter,

are you sure the A0 for the one of the PCFs can remain floating? Looking in the Datasheet I'm almost sure it hasn't any pull resistors internally.

Regards,
Liviu

rjversluis
Site Admin
Posts: 41184
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Simple block detector - zero loss

Post by rjversluis » 05.03.2014, 08:29

This is already corrected in the file; Don't worry.

rjversluis
Site Admin
Posts: 41184
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Simple block detector - zero loss

Post by rjversluis » 14.03.2014, 17:57

Today I tested the Pi07 with 1 MOhm drain resistors...
I must be in a heavy WLAN/DECT/UMTS field because all sensors reported over more than the threshold of 15 without anything connected.
After replacing the drain resistors with 10K it all works OK.
With 10K drain a 10K load is not really detected. A load of 4k7 is OK.

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 14.03.2014, 19:46

If you think that the cause is RFI, then I would try bridging the coil with a .001 uf capacitor. That should kill all of the RFI.

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 15.03.2014, 12:50

Rob,

Peter and I are exchanging emails regarding your RFI (Radio Frequency Interference) from GSM. I gave him a few things to try. Sadly RFI is one of those things that there is no "do this and it will be fixed". Nearly every solution is a custom one that involves a certain amount of trial and error.

One thing to think about is the tracks. Track makes for a darn good antenna for all kinds of RF. GSM included.

We will get it resolved.

Richard

Richard-TX
Posts: 1320
Joined: 10.07.2012, 04:00
Location: Texas, USA
Contact:

Re: Simple block detector - zero loss

Post by Richard-TX » 21.03.2014, 05:12

Rob,

I am looking at the code for turning on the DAC output when any of the sensors become active (rocnetnode.c) , but I don't see any thing that will shut if off if all the sensors are off.

Is that the way it is supposed to work?

rjversluis
Site Admin
Posts: 41184
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Simple block detector - zero loss

Post by rjversluis » 21.03.2014, 06:42

Hi Richard,

with the Pi07 it works if you set the threshold correct in the Rocnetnode option panel.
At line 1968 the ADC sensors are reported off.

Post Reply

Return to “DIY Hardware”