I/O breakout hardware

I need to get wiring access to Input 1 for counter/frequency data massaging. This will be used for an anemometer that puts out dry contact pulses. While this seems fairly easy I need some input on the hardware to use.
 
Looking at the board I see I probably need
- 16 pin ribbon cable with 16 pin DIL header male connector on both ends,
- 16 pin breakout board for above with wire terminals.
 
Some questions.
- CAI board pictures seem to show non-standard pinout labelling to others 87654321 vs. 13579...
   am I reading this wrong?
   are they hardware compatible with label translation?
 
- Can somebody direct me to a good source for these. (I really only want terminals on the end of a cable that plugs into CAI WC)
 
Thanks

Ross makes some IO board with terminals for sales.  You can contact him directly through private message on this board.
Ross is also working with us on the WC32 IO board.  We have been beta testing that also.

LarrylLix said:

- Can somebody direct me to a good source for these. (I really only want terminals on the end of a cable that plugs into CAI WC)

Click to expand...

 
http://webcontrol.rossw.net/
Just get the bare board, or bare board plus the connector if you wish.

Very nice! I dug through my junk box and found some old DIL headers with cable into D-style shell connectors. That is working fine but want to get one of your boards. Nice design and a board with the just the terminals and DIL header connector would work really well.
 
State  price with  shipping to Canada (N4L1W5) and I'll order one.
 
I suppose you won't mind if I end up advertising this a bit on the ISY site. Many people are starting to prefer the RPi boards but for a small dedicated job the CAI is better due to built in 1Wire sensor input. CAI needs to get some inertia going there or they will lose this one to RPi. The RPi needs too many extras to get basic things working for the ISY HA guys.  It supports Insteon and ZWave directly but basic analogue inputs are a PITA.
 
Insteon is a dual band (powerline and RF) mess network where most devices repeat signal for a limited number of hops. Confirmations and retries are included making it very secure.
Programmes can be written to turn things on and off at set temperatures and conditions with complex logic.
 
Here is the problem. Suppose you want to set temperature in a room based on a setpoint temperature with an offset for humidity
 
eg. Setpoint = Temp = Humidity/40. The analogue values are not available to store in variables. Along comes 1 wire sensors and Rpi or CAI. The URLs can set the temperature directly into the ISY variables via the REST interface over Ethernet. ISY can auto trigger based on an update even and cause thing to happen. 
 
The cheapest Insteon wireless thermostat 2441ZTH costs about $80. Temperature and humidity are available for logical switching of loads in the home or industry but hard to get into variable for math.
,
 
Thanks!

LarrylLix said:

Very nice! I dug through my junk box and found some old DIL headers with cable into D-style shell connectors. That is working fine but want to get one of your boards. Nice design and a board with the just the terminals and DIL header connector would work really well.

Click to expand...

 
The amplifier and gain pots can be left off, or if you have a use for them later, it's easier to put them on first up.
 
 

LarrylLix said:

State  price with  shipping to Canada (N4L1W5) and I'll order one.

Click to expand...

 
I'll confirm the postage PM you.
 
 

LarrylLix said:

I suppose you won't mind if I end up advertising this a bit on the ISY site.

Click to expand...

 
Not at all... although see following, you may prefer to hold off!
 

LarrylLix said:

Here is the problem. Suppose you want to set temperature in a room based on a setpoint temperature with an offset for humidity

Click to expand...

 
I will be making a new version of this board. Since CAI brought out I2C support (and later SPI), there's even more need for another board that brings out those connectors.
I'm thinking of dropping the (not insignificant) space that I'd left for pullup/pulldown resistors on the inputs too. It seemed like a good idea at the time, but I don't know anyone (including me!) who actually used them!
 
Also, on humidity.... the AN2321 temperature and humidity sensor is cheaper, more precise and wider operating range than the HIH4000 sensor, and gives you both temperature and humidity in one small device. For HVAC type applications, I think it's a clear winner over using a DS18B20 + HIH4000.
 
I'm also seriously considering adding a DS1307 for RTC (in stand-alone applications) and 24C32 for a little more storage (for those using the WC as a logger), but am open to thoughts and requests at this "early" stage.



 

Thanks.
 
I assume you are referring to a device like this? I can't find any mention of an AN2321 device.

http://www.silabs.com/Support%20Documents/TechnicalDocs/Si7005.pdf

 
The HA crowd people are stuffing probes into fish tanks and furnace ducts to measure inside/outside/intake/output differences. Some  dependability of these measurements is expected and 1-wire seems to be the way of choice offering distance, environmental toughness, availability, plug'n play, and price.
 
I only see the I2C stuff as never meant to leave the PCB and it doesn't seem to be designed with robust signal levels to be passed through a 50m cable through electrical wiring. I know very little of these technologies. I have done some RS-485 work in 14kv environments.

LarrylLix said:

I assume you are referring to a device like this? I can't find any mention of an AN2321 device.

Click to expand...

 
Sorry, typo.  AM2321.
 

LarrylLix said:

I only see the I2C stuff as never meant to leave the PCB and it doesn't seem to be designed with robust signal levels to be passed through a 50m cable through electrical wiring.

Click to expand...

 
There are I2C bus extenders that specifically claim cable lengths of 50m, and others that show optocouplers for galvanic isolation of the I2C bus, also over significant length.
 
I am (mostly) using DS18B20 sensors in epoxy-filled, stainless steel capped waterproof probes, measuring quite literally hundreds of air and liquid temperatures.

I still find no spec sheets on that exact part (AM2321) but  I do find information on the related series.
 
I had a few of those (DHT22) on order and had to cancel as they are not Dallas/Maxim 1-Wire compatible. It will be interesting to see how you make out with interfacing these devices. Many spec sheets on these conflict on this but 0.5C is not enough resolution for thermostat users. For outside temperature usage it would be good.

http://dlnmh9ip6v2uc.cloudfront.net/datasheets/Sensors/Weather/RHT03.pdf

The only reason I chose the CAI board over RPi was I could get my desired inputs without a stack of  interface cards (and connectors) due to the built in 1-wire I/f.  If I knew I needed as many peripheral boards I would have went to the Pi where the price may have been lower, with much more power, code and ram space, and more capability.
 
I hope the old connector issue has been resolved. Years ago ss-50 and S100 bus computers suffered massive connector problems due to resistance built up. I believe gold plating has solved much of this.

AM2321 datasheet can be download from their web site, but somehow they only has Chinese version.  \
http://aosong.com/asp_bin/Products/AM2321.pdf
 
Although it is in Chinese, its chart with timing diagram helps a lot for understand how to control it through I2C.  The problem with this chip is its I2C address is fixed, 0xB8, so that one I2C bus can only have one AM2321.
 
Si7005 is another I2C temperature and humidity sensor.
 
If you like Honeywell sensor, look digikey part number  480-3651-2-ND

LarrylLix said:

I had a few of those (DHT22) on order and had to cancel as they are not Dallas/Maxim 1-Wire compatible. It will be interesting to see how you make out with interfacing these devices. Many spec sheets on these conflict on this but 0.5C is not enough resolution for thermostat users. For outside temperature usage it would be good.

Click to expand...

 
I believe the DHT series use a resistive sensor which is not as linear, sensitive or long-term stable as the capacitive sensor used in the AM2321 (or SHT series)
 
The 0.5 deg C is its ACCURACY, not it's RESOLUTION.
The AM2321 is perfectly capable of 0.1 degrees resolution, and 0.1% RH
 
Interfacing the device was a non-event. Seriously, the grand total of parts was:  ZERO.  A length of 4-conductor wire. Period. Ende.
I also use a BMP180 precision barometric-pressure sensor on the same bus.
 
I published the code I used in another thread here, on the Weatherstation.

The game changer will be a visual programming environment allowing networked master and slave µCs, using local and server based data, local and server based logging etc... Of course the communication will be transparent and also including WiFi
The target platform will be very likely based on Linux but CAI could introduce a programming environment on its own, easing access to its PLC hardware, reducing hassle setting up more complicated distributed systems. 

LarrylLix said:

Many people are starting to prefer the RPi boards but for a small dedicated job the CAI is better due to built in 1Wire sensor input. CAI needs to get some inertia going there or they will lose this one to RPi. The RPi needs too many extras to get basic things working for the ISY HA guys.  It supports Insteon and ZWave directly but basic analogue inputs are a PITA.

Click to expand...

rossw said:

I believe the DHT series use a resistive sensor which is not as linear, sensitive or long-term stable as the capacitive sensor used in the AM2321 (or SHT series)
 
The 0.5 deg C is its ACCURACY, not it's RESOLUTION.
The AM2321 is perfectly capable of 0.1 degrees resolution, and 0.1% RH
 
Interfacing the device was a non-event. Seriously, the grand total of parts was:  ZERO.  A length of 4-conductor wire. Period. Ende.
I also use a BMP180 precision barometric-pressure sensor on the same bus.
 
I published the code I used in another thread here, on the Weatherstation.

Click to expand...

Could you please point us towards that thread, I only found c-Code so far.

Efried said:

Could you please point us towards that thread, I only found c-Code so far.

Click to expand...

 
The thread is http://cocoontech.com/forums/topic/26708-weatherstation-based-on-webcontrol8/?hl=+weatherstation
 
I've made lots of changes since then, but there's a chunk of useful code still there.