Cumulus Support

robynfali wrote:

AllyCat wrote: Maplin kit

It's "napkins" not maplins get a grip lol (do a search for napkins to understand )

Rob is correct, no such thing as Maplin, it's Napkin.

trust you to stick your head up from the sand!!!

Errrm, sand? I'm in Brighton, just pebbles!

*bitchslaps paul for spitting hairs!

Nice scientific observations alleycat!
Since reading your report, I unplugged my display, removed the batteries, then re plugged the USB cable and inserted the batteries.

No evidence of the TX battery indicator lighting up! Tried it several times but no display.

However, sometime over the next 10 hours, during the night, the TX battery icon appeared -flashing-. During half an hour of watching it this morning, it is still flashing.

The interesting point here is that the transmitter is still in Costa Rica, minus batteries, while the display is in Canada. So, does the TX low battery indicator really work????

AllyCat wrote:But the Transmitter unit does appear to "monitor" its battery voltage and transmit data (probably a single "battery status" flag bit) in each transmitted message

If you could provoke the 'low Tx battery' condition with Cumulus connected and the debug log turned on, we should be able to see if one of the status bits is set in the status area in the data which Cumulus reads (which has the 'sensor contact lost' flag). If so, I could do something with it.

You might be able to provoke the low battery by using one battery and one battery shaped rolled piece of tinfoil or something similar maybe?

Hi Steve,

I have plans to "improve" some (or many) aspects of the FO sensor hardware before commissioning the system. So I'm not running Cumulus yet and am not clear what the debug log may contain. However, I will try to do some more tests before finally fitting the transmitter onto the mast (which needs to be rather large as I'm in an urban environment).

It's easy enough to trigger the transmitted low battery status (flag), by lowering an external variable-voltage power supply through the threshold. But rather more difficult to ascertain at what point between the RF receiver firmware and the LCD drivers that the flag gets "lost" (or rather fails to get updated).

Cheers, Alan.

Hi,

sanramonrover wrote: So, does the TX low battery indicator really work????

For anyone following this thread, the answer now appears to be Yes.

However, it only changes at midnight.

But this is actually good news. It means that it's quite easy to create a debug log whilst the icon changes. Personally I don't think that the status/flag is recorded in memory (so can't be accessed by Cumulus) but I've posted debug logs in this thread.

Cheers, Alan.

A bit late, because the battery indicators have been discovered (sometimes using a torch) a long time ago:
As a recent owner of the WH1080 I noticed that the Dutch / French manual offers much more information about the base station display than the English version(s):


26 Low battery indication base station (lege batterij indicatie binnenunit)
28 Low battery indication transmitter (lege batterij indicatie buitenunit)

Full manual: http://www.alecto.nl/Images/Documents/W ... -FR2.0.pdf

Descriptions, translated NL -> EN with Google translate:
1: Time Reception indicator
2: indicator DST (DST) na. combined DCF receiver
3: DCF reception indication signal
4: Time Display
5: WWVB reception indicator signal (WWVB clock signal is American)
6: Alarm indication on / off (alarm)
7: display alarm time display
8: alarm time and date display
9: display date display
10: indication of the volume memory
11: Pressure history from 0 to -12 Hours or from 0 to -24 hours.
12: pressure indicator history graphic form ek
13: display current pressure
14: measurement of HPA (european) inHg (USA) mmHg (mercury column)
15: relative or absolute indication air pressure
16: indicate rainfall in mm or inches
17: display rainfall
18: rain amount per indication unit (1 hour - 24 hours - week - Months)
19: max indicate rainfall alarm
20: wind direction indicator
21: Wind Display Unit (m / s - km / h - knots - mph - bft)
22: windspeed
23: max wind speed alarm
24: display during contact with outdoor unit
25: humidity value
26: low battery indicator indoor unit
27: High / low alarm value humidity
28: Outdoor unit low battery indicator
29: Temperature display in
30: high / low alarms inside temperature
31: display display outside (temperature - wind chill temperature and dew)
32: outdoor display
33: High / low alarm outside
34: Control + key
35: trend indicator
36: On / Off button
37: Operation - key

Hi folks,

My TX icon started flashing last night, and after an hour or so it went solid, and it's been solid ever since.

I turned my pc off over night as usual, and noted this morning that the icon is still solid. I've also tried shutting down cumulus and unplugging the unit, but to no avail.

I'm now a bit concerned, as, if this is genuinely telling of low TX battery then the unit is consuming a pair of brand new batteries a month. That's clearly no good at all. The batteries in use have 03/2118 as the use by date, about as fresh as one could reasonably expect I'd guess. Energizer ultra plus.

Also between the 13th and now I have had just one lost sensor contact, and that one recovered right away on the next read.

This set up keeps coming up with little mysteries that don't feature in the manual.

Hi Ian,

Did the Low TX icon come on at midnight and is it still on 24 hours later? In my experience it only updates at midnight. The threshold on my unit is about 2.5 volts (1.25v per cell) so there is still some life in the cells, but they certainly shouldn't be that low in only a month (I've tested mine and it definitely consumes a current consistent with at least a two-year battery life). I don't know if there is any significance in a "flashing" versus "solid" icon.

It seems rather too soon for corrosion on the battery contacts but it's worth cleaning them anyway. If you don't have a battery tester then try the cells in a high current device like a camera, shaver or krypton torch, etc. to ensure that they really are are "weak".

However, I did post more than 3 weeks ago that I thought your transmitter is probably faulty.

Cheers, Alan.

AllyCat wrote:Hi Ian,

Hi Alan

AllyCat wrote:Did the Low TX icon come on at midnight and is it still on 24 hours later? In my experience it only updates at midnight. The threshold on my unit is about 2.5 volts (1.25v per cell) so there is still some life in the cells, but they certainly shouldn't be that low in only a month (I've tested mine and it definitely consumes a current consistent with at least a two-year battery life). I don't know if there is any significance in a "flashing" versus "solid" icon.

It started flashing at maybe 19:00 to 21:00 or so, and then went solid later, but I did not think to check if it was exactly midnight, but it could well have been. It's been on solidly every day since.

I agree about the rapid rate of consumption if the warning is proven credible. I do have a plan for it, and I'll know more in a day or five!

AllyCat wrote: It seems rather too soon for corrosion on the battery contacts but it's worth cleaning them anyway. If you don't have a battery tester then try the cells in a high current device like a camera, shaver or krypton torch, etc. to ensure that they really are are "weak".

I've got a meter and intend to test, I've also gone a little belt and braces, and ordered a really basic (EasyOne by TFA Dostman) wireless temperature gizmo, @ 868Mhz, I had planned to do this later anyway, but moved it up the "to do" list a little in view of how things were presently going. I'm intending to see if they will power the new unit mostly for the sake of comparison.

AllyCat wrote: However, I did post more than 3 weeks ago that I thought your transmitter is probably faulty.
Cheers, Alan.

You did, and I do agree with what you said at the time. In fact I have been on to Napkins, and they have issued a return number and instructions along with a request for as much detailed information for their engineers to test it on arrival with them. Reading between the lines I take this to be related to the fact that they actually have no stock to send me a replacement! In the meantime I have been doing more tests so they get a very clear indication of what the problem is and how to reproduce it. I've also been looking to make sure all the other issues that have so far arisen get a fair mention too! I'm making a stevenson (maybe even two or three) at present, and sending the Fine Offset unit back could interrupt that, hence moving the "planned for later" temperature probe forward in time. I can carry on tinkering with the new probe while they get their act/stock together. I know it won't log to cumulus, but it will give me min max to mess around with, and will act as a comparison unit when the replacement Fine Offset is at hand. I chose that "EasyOne" for being on another frequency so as not to clash, and not being too spendy into the bargain, although it's bigger sibling rain gauge looked equally tempting, for only a £10.00 difference it was a tough call until the frequencies of both became known!

I'm trying hard not to shoot myself in the foot here, but there's been quite a lot to arrange to get it all to work out timing wise!

Flying Eye wrote:they have issued a return number and instructions along with a request for as much detailed information for their engineers to test it on arrival with them.

Hi Ian,

Poor transmitter range is probably the easiest fault to demonstrate, but lots of space is needed - just a kitchen table (or even laboratory bench) isn't large enough.

Measuring the (average) current consumption isn't easy because the transmitter draws (relatively) large current pulses. The quiescent current should be less than a few tens of microamps (mainly through the wind sensor resistors). However the wireless transmitter uses at least 20 mA, but only for 50ms every 48 seconds (so an average of about 20 uA). Four seconds before transmission (and the red LED flashing) the A/D converter runs for up to 1 second and consumes about 1mA, again averaging about 20uA. If your transmitter doesn't behave like this then a the fault is obvious. But there is also the possibility of the software "crashing" and causing a "high-current" stage to be switched on for too long (thus flattening the battery), which is hard to prove.

The method I used to monitor the transmitter current was to tape wires onto one end of each AA cell and insert the cells such that they were isolated (by the tape) from the "bridge" conductor at the lower end. Then connect a multimeter (or other measuring device) between the two wires.

Cheers, Alan.

Alan,

Thanks very much for the advice, I can see it coming in handy.

Due to my abilities (physical limitations) I deliberately chose a small place to live last time I moved, as maintaining anything more inside and outside would be problematic for my reach. On the upside, this means the units can't really ever be more than say 10 metres apart, and to do that the units would be in places where a) I would be crazy to site a station, and b) could not get to the panel to read it or connect to it!

Even if I was being fully daft I'd not realistically be able to see more than 6 metres between them. Mostly they will only ever be 3.5 metres apart on average. Now these numbers will change a little as I get my proper mounting pole sorted out next year, but we could safely predict max ever required at less than 10 metres and more probably it'll be no more than 6. More likely is that it'll need to manage 4-5 metres at most. It's a bungalow, so it's not got multiple storeys to encounter either. Presently it's got two favoured spots and they see the two units 3.2 and 3.5 metres apart. It's been a long time since I saw lost sensor contacts, the singular one that did happen was immediately recovered without need for my intervention. Apart from that it's been stable for 14 days.

The setup has got double skinned brick, and double glazing to deal with but, that is it. Mostly it has just the glass to contend with, as I generally try avoid going too near to window frames with this kit as I feel it's a safe bet that all uPVC is very probably metal cored. If trying to test for short range issues, I have sometimes arranged for there to be the odd window frame upright in the way to make things tougher, so as to challenge it a little. They don't appear to be the culprit so far.

Sometimes small has distinct benefits, in this case, if it can't work at my place then it's clearly not up to all that much!

As far as I can reasonably know, this only leaves the battery thing to figure out really. Mostly I'm happy to let "Napkins" sort it out.