Topic: Sharing an ISS with 2 or more wired consoles...

[W9LRT]

I have read in numerous places (here and elsewhere) that using multiple consoles on a wired Vantage Pro 2 system is possible.

I have a couple questions that maybe someone experienced with this configuration might be able to answer:

1.  I would like to put steering diodes in the power line so the ISS can be powered by either console (in case one is powered off).  Does anybody know if the .7v drop across the diode will cause abnormal ISS operation?  I could use germanium diodes and get it down to .3v drop.  Does anyone know the current draw for the ISS?

2.  I have an older firmware console that is compatible with VirtualVP and a brand new one with the latest firmware.  Any issues running them together?

When I get this all sorted out, I will post a schematic and photos of the finished product.

Thanks in advance for your input.

[dalecoy]

I don't know the answer to #1 - but a third alternative would be to provide a separate power supply for the ISS (not depending on power from a console).  On the other hand, I'm interested in what you might find out about using isolation diodes.

Re: #2 - I can't think of a reason that configuration would not work - there is no communication from the console to the ISS.  The ISS just sends a data stream that each console interprets.

Note: Technically the communication is RS-422 which means the console data lines should be wired in series.  Therefore, disconnecting either console would interrupt communication to the other.  I am not sure whether powering off a console would interrupt the communications (depends on how the interface is wired internally).  HOWEVER - by years of experience, I know that - at least with two consoles - I wire the data lines in parallel and it works fine on both consoles.

[dalecoy]

On a wired? Interesting didn't know it was possible

I'm not sure what you are commenting on.  Power?  Or parallel connection? 

One footnote:  one "problem" with having two wired consoles, is that you have to manually keep the console times identical.  Otherwise, rain events that last over midnight will show different "daily rain" on the consoles.  [Also, of course, if you clear something on one console it doesn't clear on the other one, etc.]

I think it's RS422, but there's not much difference between that and RS485.  And by the way I suspect that the "required" 120-ohm terminating resistor is present in each console.

http://us.brainboxes.com/faq/items/what-is-the-difference-between-rs422-communication-and-rs485-com

[BCJKiwi]

The VP2 ISS CABLED specifications Document 6152C_6162C_SS indicates a current draw of 5mA (average) at 4 to 6 volts for ISS only. 10mA average for both console and ISS. http://www.davisnet.com/product_documents/weather/spec_sheets/6152C_6162C_SS.pdf.

The VP2 Console manual RevK also states a 4 to 6 volt range.

So, if you are still concerned about the diode volt drop you could use a 6V power supply on both consoles with the .7 v Drop.

However, you should be aware that the volt drop on a diode is dependent on current. The spec sheets typically rate the forward volt drop at a low current so even a schottky may well drop more than .3v if the current is high enough.
An onsemi spec sheet can be found here http://www.onsemi.com/pub_link/Collateral/1N5817-D.PDF Note also that it is preferable to use the higher voltage rated diode as the reverse current is less on the higher voltage device. See figures 7 & 9 on page 5 for details.
My pick would be good quality 1N5819 devices, after all they don't cost much.

[Coolerman]

The main difference between RS-485 comms and RS-422 comms is that the 485 devices are addressable. This means each device is assigned a two digit address  (up to 32 devices) usually via dip switches. ALL devices on the 485 network, see ALL the commands on the network ,but only respond to commands that have their address as the second character in the command string. My company does a lot of 485 comm work.  We use Cat 5 cable for all our 485 comm work. Since most of the time we use two wire 485, we can get 4 individual circuits on one Cat 5 cable. If using 4 wire 485 then only two circuits.

[W9LRT]

Thanks BCJKiwi ---

That is the information I was looking for...  My Google-Fu must have been weak that day.  I will run some tests to determine the actual drop across the different diodes.  Given the minimal current draw, I may just use a 3 terminal regulator (I have 12V backup power everywhere) and power it independently of the consoles.

As for the accuracy issues between the two consoles, the second one is for my 'financial advisor' to use in the kitchen.  She is a lot more willing to accommodate my hobbies when they can benefit her as well.  She will not be concerned about small inaccuracies in rain totals.

Thanks again.  I will be assembling this over the weekend and will post details later...
 

[dalecoy]

The main difference between RS-485 comms and RS-422 comms is that the 485 devices are addressable. This means each device is assigned a two digit address  (up to 32 devices) usually via dip switches.

...so, the Davis setup is RS-422 (as I indicated)?

[Coolerman]

I will have to take your word that Davis uses RS-422 for their comms, as I am a newbie when it comes to Davis communications protocols. 

Main reason I went wireless with my Davis unit was to not have to deal with long comm lines and all the issues that come with them. I spend way too much time in the field figuring out why a customers RS-422/485 comms have stopped working as it is! 

It's usually flooded conduits and non-water proof cabling. I have most of them converted over to ethernet using Vlinx serial to ethernet converters. We replace the flooded cables with Cat 5 underground rated cable. Then their IT department is responsible for the new ethernet long runs.  I only have to deal with the devices and the short runs to the converters.

[W9LRT]

Since my tower climbing days are mostly over, I decided changing batteries on the ISS didn't need to be on my to-do list.  Everything I can find indicates that is RS422 or something closely resembling it (5v balanced serial)... 

Given the amount of RF at my location (wifi, amateur radio repeaters, etc.), wired will be less trouble for me overall.

[dalecoy]

I will have to take your word that Davis uses RS-422 for their comms, as I am a newbie when it comes to Davis communications protocols. 

I don't have any inside knowledge, either.  All I know is that it's differential signaling, and that the wired system uses one transmitter and can accommodate multiple receivers, and there's no (obvious) way to change addresses.  It could, of course, be 485 with a hard-wired address. 

I "claim" it's rs-422 because it acts that way.

[Coolerman]

Since the transmitter can broadcast to multiple receivers over longer distances than RS-232, and they all receive identical data, it's more than likely the transmission protocol is RS-422.

A simple way to find out would be to take a RS-422 to RS-232 converter, connect the RS-422 side to the Davis transmitter and  the RS-232 side to a com port on a PC, set the baud rate, parity and stop bits on the PC to match the Davis and run HyperTerminal or something else that would show ASCII text. If you see the "data" transmitted from the Davis as readable characters then you will know it's RS-422.

[dalecoy]

Since the transmitter can broadcast to multiple receivers over longer distances than RS-232, and they all receive identical data, it's more than likely the transmission protocol is RS-422.

That was my conclusion.  And I had considered doing the test with a converter, but it isn't very close to the top of my "to do" list.

[W9LRT]

I realize that I am reviving an almost 4 year old thread, but figured it was time to update with the results of everyone's input.  My weather station took a lightning strike late last fall and it gave me the downtime over the winter to develop and test this circuit.  I built it in a standard wall outlet box with a 3 port keystone jack plate.

The end result is everything works perfectly.  The interface is RS422 (I found the actual interface chip Davis uses and its spec sheet). The device will work down to 3.3 volts (so the diode isolation doesn't impact operation) and it claims to support up to 32 devices on the interface (so my 2 or 3 were not an issue).  Given the recent lightning experience, I threw a few MOV devices in the circuit.  It won't help it survive another direct hit, but it should suppress minor spikes that come along.



The schematic and description can be found here: [size=78%]http://www.w9lrt.com/weather/DavisSplitterInstructions.pdf[/size]


The spec sheet for the interface chip (ADM3485E) is here: [/size][size=78%]http://www.w9lrt.com/weather/ADM3485ESpec.pdf[/size]


Pics of the construction:













Installed and Operating:

[dalecoy]

I really appreciate all of that analysis, and particularly the design and construction.

I wish I had had that when I was hunting for a solution.  (I suppose that I should point out my previous posting about the "clearly wrong" approach -- wiring two cabled consoles with their data lines in parallel.  Which is still working fine, by the way).