Cumulus Support

Wind sensors unplugged just after 14;30 and plugged back in just after 16:00.So I deduce that this confirms having the wind sensors plugged in causes the problem. It may be to do with the anemometer or it could still be interference. Next test will be to disconnect the anemometer leaving the vane connected. That will be tomorrow now - it's getting dark as well as very cold.

Gina, this may be a wild shot. Have you checked to see if your dielectric grease is contaminated?

Randy

wd40 wrote:Gina, this may be a wild shot. Have you checked to see if your dielectric grease is contaminated?

I'm not using any on my present system - I'm using all the new bits and haven't put any on as yet.

Curiously the "waggling" on the plot seems to have stopped. The spikes continue and still seem to have some relationship to the wind.. There was a period earlier when we had a flat calm and there were no spikes.

One anemometer wire disconnected at 13:10 and reconnected at 15:20. No spikes or noise appeared on the temperature plot during this interval.

I'm now making up a test rig for the opto-coupler to see how much current is required to feed the LED in turn to switch the anemometer logic in the transmitter microchip. That will determine the value of series resistor required. I shall be using an AA size battery as I already have a battery box for that size.

Component layout.

Test rig running now. Website testing page and is now running from the second console locked on to the test rig transmitter.

Now to the details... Digital multimeters measuring the voltage and current required to feed the IR LED that will drop the photo transistor output to logic 0. LED volt drop is about 1 volt and current 120uA. So a 1.5v battery would be barely sufficient so I shall use 2 cells rather than one. The system is currently running with 2.2v from my bench supply and 10K resistor. Two alkaline cells will be nearing the end of their life at this voltage (1.1v per cell). I estimate that alkaline cells should last a year or more. Lithiums quite a lot longer.

Here's the test circuit.

I posted results, then realised I'd still got multimeters and bench supply connected. I've now disconnected the multimeters and I'm running the opto-coupler from a pair of AA alkaline cells. All other clutter has also been removed from the vicinity. So I'm now testing with the circuit as above, a 25ft coil of CAT 5 UTP cable and desk fan blowing the anemometer round. The transmitter is also running off 2 AA alkaline cells. This is now on a clear part of the table.

I'll posts proper results shortly.

that didn't help much.
a spike is normal,thats a hedgehog
isn't there a faulty (smd)part on the pcb
power regulation stable enough
or the thermistor itself may be deteriorated,damaged inside.
it's leads to long(doubt that).

honestly,i haven't got a clue at the moment,there isnt much you can do anymore.

edit"
didn't see your last post

Sorry Hans, our posts crossed. Please see edited post above. I got ahead of myself

Looking better - hedgehog sent packing

My fault, sorry. I edited the post while you were writing your reply - I shouldn't do that! Here's the last couple of hours - the first hour with instrumentation attached and the 2nd without. The real test.And this is the wind speed plot for the same period. The dip is where I disconnected the bench supply and connected the battery.

that looks much better the last hour.
that outdoor unit is a very sensitive little thing.

hope this last over a longer period.but looks promissing

hans wrote:that looks much better the last hour.
that outdoor unit is a very sensitive little thing.

Yes, I was about to say much the same.

hope this last over a longer period.but looks promissing

Yes indeed! But yes, it does look promising so far. I'll leave the test rig running overnight and if that's OK I'll set up the same thing outdoors.

So far so good