Not sure if anyone's still interested, but after some investigation, I think the reed switch design detecting the bucket tips for this class of rain gauge is flawed. I too noticed that on the bench the bucket tipping mechanism and counting works fine, but after sitting idle for a few hours, manually tipping it does nothing. Only by tapping it or using a very strong external magnet will it come to life. So when mounted outdoors, it eventually locks up until I get on the ladder to thump it. Reed circuit applications are usually associated with entry detection, e.g. burglar alarm systems. A door or window opens, and the switch is activated as the magnet moves away. But in such applications the magnet moves a substantial distance from the switch, not the case for this rain gauge. The unit's stationary reed switch is mounted vertically on a small circuit board very close to the bucket movement. There's a channel on the side of the bucket that has a small magnet embedded opposite the reed switch. The intent is that when the magnet swings by the reed switch, it briefly completes the detection loop. On paper that ought to work fine, but my speculation is that when the bucket's at rest, the magnet really isn't too far from the switch; far enough that the switch releases, but it's still in the near field of the magnet. Since the magnet is tilted relative to the switch, the static field is dominated by one polarity. And my sense is that over time, the reed switch's two elements retain enough of that magnetic bias to actually repel and overcome the magnet's passage for subsequent tips of the the bucket. This also could be aggravated by weakening of the magnet over time, or by the unit being subjected to summer temperatures. I think the bucket movement is great, and gives nicely calibrated rainfall increments. So I set out to make an optical detector. There's plenty of room inside the housing, and with an LED and phototransistor mounted side by side, a simple foil reflector on the arm of the bucket seems to detect each tip reliably. We'll see how well it does this winter! On Wednesday, June 8, 2016 at 7:08:33 AM UTC-7 Luc Heijst wrote:
> I forgot to mention: leave the wires of the solar panel in place! > (I had bad experiences with cutting them, so I reconnected them again). > > Luc > > > On Wednesday, 8 June 2016 11:01:24 UTC-3, Luc Heijst wrote: >> >> Hi Dik, >> >> I have experiences with three types of batteries, all with more or less >> good results. >> >> 1. Two AAA rechargeable NI-MH batteries (1.2 V). It works, but only >> for a few weeks, the the batteries have to be recharged in a separate >> vNI-MH charger. >> 2. Two AAA alkaline batteries (1.5 V). The lasted longer, but after a >> few months they were exausted. >> 3. Two C-type Alkaline batteries in an external battery compartment >> which is connected to the anemometer via a loudspeaker cable (one lead is >> marked black, easier to keep the polatity right when installing), see >> photo's. >> >> I'm currently using option 3 and just two days ago the battery status of >> the wind meter went to 'low battery', so I changed my two C-type batteries >> today. They lasted 8.5 months (outside temperatures here varied from 20-34 >> degrees Celsius). >> >> PS 1. On top of the wind mast you will notice the arm of the wind meter >> of my Davis Vantage Pro2 weather station, And yes, for those who will >> notice the smallest details, the arm is facing south !!! ;-) >> PS 2. The 'battery holder' is in fact a broken temp/humidity sensor of my >> TFA Primus weater station. Now only the battery compartment is used! >> >> Luc >> >> On Wednesday, 8 June 2016 10:01:08 UTC-3, Dik wrote: >>> >>> It appears my solar Wind sensor is not working anymore. I check the >>> original rechargeable batteries in the wind gauge and the Voltage is below >>> 1.2V. I suppose they can’t be recharged anymore after a one and a half year >>> of duty. Now I have bought two AAA Lithium 1.5V batteries (ENERGIZER >>> Ultimate Lithium). Can I safely replace my original rechargeable Alkaline >>> batteries with these lithium batteries? Or do I have to change something in >>> the circuit, so that these non rechargeable Lithium batteries won’t be >>> charged? I can cut the wires to the solar panel… >>> >>> >>> Dik >>> >>> >>> Op woensdag 23 september 2015 19:10:02 UTC+2 schreef Luc Heijst: >>>> >>>> No, my console had two times low batteries reported, the rain meter >>>> once and the wind meter several times. Only the batteries of the >>>> thermo/hygro sensor were after 2,5 years of operation still going strong. >>>> Nevertheless I replaced them recentlybecause this device reveives the data >>>> of the wind and rain sensor and send these together with its own data to >>>> the console. In 2,5 years the original voltage of 3.1 V dropped only 0.4 >>>> V, >>>> so theoretically the battreries were 'half way'. Attached my battery year >>>> plot of cmion. >>>> >>>> Luc >>>> >>>> On Wednesday, 23 September 2015 12:08:17 UTC-3, Ξ wrote: >>>>> >>>>> Incidentally, back in August I had 'outage' because as it turned out >>>>> the console batteries were flat but that wasn't reported in the log. >>>>> >>>>> Do you have similar experience? >>>>> >>>> -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/033081e7-fa63-4805-91bb-440ba61b100an%40googlegroups.com.
