Feeding sensor outputs to multiple receivers

[enigmabg]

I have two "WH1080" systems, one very old 433Mhz system whose sensors have, through age, given up the ghost and an 868Mhz system bought from Clas Ohlson. I should like to have all sensor data displayed on both receivers in different parts of the house - is this possible?
Since the receivers don`t obviously seem to be modifiable, the only way I could think of doing it would be to split the sensor signals to feed both external transmitters which would then transmit the data to the relevant receivers on the two frequencies. Would this work? Or would there be some problem as it will result in effectively linking the inputs to both transmitters together.
Any advice and direction will be gratefully received.
Thanks in anticipation
enigmabg

[uncle_bob]

I'm not sure if it possible with the Fine Offset's and I suspect it isn't.
But it's totally possible if you switched the Fine Offset's electronics out and build the WeatherDuino Pro 2 system in their place.
What is the WeatherDuino Pro2

[AllyCat]

Hi,

the only way I could think of doing it would be to split the sensor signals to feed both external transmitters which would then transmit the data to the relevant receivers on the two frequencies. Would this work?

Welcome to the forum. The external temperature and humidity sensors are inside the transmitter boxes so they will still be sent to their associated Consoles. The Rain and Wind Speed sensors are magnetic reed switches which the transmitter(s) try to "pull up" to the battery voltage with a small current. In principle, these can be connected in parallel (i.e. one sensor feeding both) provided that the supply voltages (batteries) in the transmitters are similar (perhaps no more than half a volt difference).

However, the Wind Vane signal (the outer two pins on the wind connector) is a variable resistance and AFAIK the method that the transmitter uses to read the resistance CANNOT be shared across two transmitters. So you are likely to need two Vanes, or display the wind direction on only one Console.

If you're an experienced electronics hobbyist, it might be possible to "open up" the Clas Ohlson transmitter (the box is glued) and attach a 434 MHz transmitter module (very cheap from ebay) to transmit the same data in parallel. But that CANNOT work with a 434 MHz Console which is more than a few years old, because the wireless protocol (data format) was changed a few years ago.

Cheers, Alan.

[enigmabg]

Wow! Thanks for those rapid, expert responses.
I need now to re-read and get my head around the suggestions.
I`m no electronics expert (my background is mechanical), but I have a VERY small understanding of the principles...though probably insufficient to follow any of the suggestions without much more guidance.
I could probably forgo wind direction from going to both, so perhaps paralleling the wind speed and rain might suffice. Is there likely to be any "interaction" between the two transmitters back through the reed switch inputs which might damage either, or cause a malfunction/error? Would giving it a try be a bad idea?
I shall be back with further queries, I`m sure.
Thanks again
enigmabg

[AllyCat]

Hi,

Actually a number of the practical issues are "mechanical". The "RJ11" connectors are "insulation displacement" type (a spike through the insulation into the conductor) so a crimp tool is recommended (fairly cheap from some ebay "shops"). Therefore, it may be better (and cheaper) to buy ready-made cables. But beware that many cables no longer use "solid" (multi-core) copper but (copper-coated) aluminium or sometimes very fine copper tape wound in a helix around an insulating core. So, if you're planning to solder joints, look for specified "solid copper" cables from e.g. Kenable or CPC.

It would be wise to put a resistor (perhaps 1k ohm) between any signal connections that are being shared. All microcontroller pins have "electrostatic protection" diodes to the supply and ground (battery postive and negative), so if the rail voltages are different by more than about half a volt (or of course if signal and earth connections are accidentally swapped) then undesirable currents might flow.

If you don't have a Multimeter, I suggest getting one, even for this, and certainly for the WeatherDuino project. My recommendation is to get one with a "continuity" function (buzzer), they still can be obtained for less than £5, e.g. from CPC. BTW, note CPC's icon to "sort in ascending price" can save a lot of time.

Cheers, Alan.

[enigmabg]

Thanks for the extra info. I have not yet explored the cabling side of things, but was planning to chop up an old router modem cable (RJ11 at both ends) and try to solder the wires onto a stripped down microfilter board (which has a 2 wire RJ11 socket already fixed).
I did wonder about the current flowing where it shouldn`t really go - hence the query. presumably the 1K resistor suggested doesn`t need to be anything special
Looking at the kit today, without fiddling, I noticed that the wind direction sensor is very "rough" and difficult to rotate in certain positions - looks like that is dying the death.
Hence, I wondered if I should buy a new kit complete (433Mhz from Maplin @£49.99 or 864.3 from Clas Ohlson for £69.99), but the reviews don`t seem too good for either with many complaints of transmitters eating batteries or not getting data to the receivers. Such problems would just replace my current "problem" of how to send data to the two different frequency systems with one of crappy reliability! What are the general opinions from those in the know about these cheapo systems - has quality been paired back since I last bought one?
Are there any reasonably priced units on the market that are reliable and robust? Pictures on the web seem to show many different makes with apparently the same sensing elements as the 1080, so I would be dubious about buying one of those.
Thanks
enigmabg

[AllyCat]

Hi,

Yes, "gritty" bearings is a well-known issue with both wind sensors. A squirt of WD40 may be sufficient, but the bearings are a "standard" size (10mm x 5mm), available quite cheaply, so some members have changed them as a matter of course. Yes, any resistor type should be OK.

"Eating" batteries is not normally an issue, the claimed 2 years is not unrealistic for the transmitter. The Conole will be less, unless powered via the USB cable from a running PC most of the time. But I did have a very early "freak" 3080 (Solar Pod) which did eat transmitter batteries (every 3 months) and was replaced by the supplier.

Yes,all the "badged" stations with similar sensors are likely to be manufacturedf by Fine Offset who have always had an issue more of poor Quality Control rather than poor design (but there are some BAD features). The "cost reduction" a few years ago probably did have an adverse effect on (typical) wireless range and introduced the infamous "USB lockups". But "crosstalk" between the wind and external temperature sensor has been removed and the wireless protocol is more "hackable" and documented for those interested.

Sadly, there doesn't appear to be anything better at anything like the same price (pay 10 times more for a Davis). The Clas Ohlson is arguaby better than the Maplin, with its Radio Controlled Clock, 868MHz wireless, blue backlight and 2-year warranty, but perhaps not £20 better? Personally, I much prefer the non-touchscreen 2xxx and 3xxx Consoles which have a much clearer display and a "reconnect" wireless function, but generally cost nearer to £100.

Cheers, Alan.

[enigmabg]

More information - wonderful/thanks.
I have almost got a "test" configuration together - just finished getting the bits and will hopefully have a go this weekend. During my rumaging, I found a wind direction and rain meter in with all my old stuff (unused - came with the Clas Ohlson kit), so, if the splitting of the of the sensor signals to the two transmitters works, then perhaps a new installation will not be necessary.
When you comment about the 1k resistor location, do you mean soldering it across the two signal pins of each sensor feed BEFORE its splits, or across the two transmitter feeds AFTER it splits (I imagine the latter otherwise stray currents from either transmitter would not be aware of the of it and so the resistor would be a waste - I did tell you I wasn`t electrically skilled, didn`t I!).
Interestingly my original 433MHz unit (from maplin) has an RC time signal sync, as well as the newer 868MHz Clas Ohlson one.
If you confirm the resistor location, I shall update you of my success (or lack of it!)
Thanks
Briang

[AllyCat]

Hi,

No don't put the resistor between signal and earth (which will stop the signal being detected) but between the two transmitters (it's primarily to stop a higher voltage battery in one trying to "charge up" the other). The sensor could go to either end of the resistor (so one transmitter sees the sensor as usual, the other via 1k), or use a "symmetrical" arrangement with two resistors (the sensor feeding each transmitter through about 1k). Note that it is important that the signal and earth lines don't get swapped (i.e. connected differently to the two transmitters), which is why a multimeter (or at least a LED) would be useful to test.

Maybe it was a very early Maplin station, I've only ever seen their "1081" derivatives without the RCC (the RCC receiver is actually located in the external transmitter box which sends "time" data packets just after each hour).

Cheers, Alan.

[enigmabg]

Finally got to spend some time testing the principle of sensor splitting, but whilst it did seem to work, there were significant differences in the values reported for all parameters. The resistive wind direction element when connected to both did indicate on both receivers. Disconnecting this from 1 receiver didn`t seem to affect the value displayed on the other. Does that indicate that it can be split, in principle? I think more playing around is necessary. Either that or just rely on the data on checkwx for RAF Shawbury which is 2 miles up the road for the second data set!
Thanks for all your guidance.
enigmabg

[cancunia]

Interesting stuff in this thread. I also have an old FO console that I'm reluctant to part with via the bin as it's a neat & clear display. The transmitter unit failed sometime last year and I bought a Clas Ohlson branded Fine Offset to replace it. Unfortunately one very windy day, the old sensors had an argument with my wind turbine and came off worse so I'm using most of the Clas O. hardware.
It would be good to get the old console working again so I'll watch with interest in case there's a good solution that evolves.

[AllyCat]

Hi,

there were significant differences in the values reported for all parameters.

Moderate differences or ridiculous differences?

There are separate humidity sensors in each transmitter and they may well differ by 5% or more. The rain and wind speed should be basically the same, but each transmitter is counting over a (different) 48 second "window", so there may well be some differences over the short term. Barometric pressure is measured in the Console so will very probably be different. And Fine Offset "specifications" often seem to be as much wishful thinking as the result of any quantitative measurement.

The "old" FO station transmitters used the same Analogue-Digital Converter for external temperature and wind direction. This meant that even a minor issue with the wind cable (e.g. ingress of water into a connector) could produce crazy external temperatures to be reported. It was well explained by Gina in this thread exactly 5 years ago. That's why I said you should not common up the vane cables.

The resistive wind direction element when connected to both did indicate on both receivers. Disconnecting this from 1 receiver didn`t seem to affect the value displayed on the other. Does that indicate that it can be split, in principle?

More recent stations use the ADC only for the wind direction and I believe the external temperature is obtained (digitally) from the (new) humidity sensor. So maybe the two transmitters can independently "read" the vane (without upsetting the temperature measurement?), but does either actually read the "correct" direction? Some directions use low resistance values (< 1k) so I would expect my suggested 1k "safety" resistor alone to upset at least some directions.

As for a Console continuing to display (changing?) values; bear in mind that the Console is designed to accept up to about 7 "missing" radio transmissions, so it can be up to 6 or 7 minutes before the LCD shows any updated or missing data.

Cheers, Alan.