Updated: Wind direction indicator - tail mods plus some Q's

[Charlie]

I'll look forward to the results One question, why the lead on the top - if it's for balancing then surely it should be on the nose?

It isn't particularly for balance - more a thought I had regarding extra mass, centrally placed to act as a damper (there's around 2grm of blutac too). It's far from centrally placed now that I've stolen some of the blutac! I may beat the lead to form a larger cap at a later time.

First off, I really like what you've done with the tail - excellent job! I'm not sure about adding the mass to the top, however. It may or may not help, but I've always found that when I change 2 things at once I can never tell the impact of either. It might make sense to remove it until you've assessed the impact of the tail mod.
Also, my experiment with adding mass resulted in more stable performance at some wind speeds, but more eratic performance at others. As luck would have it, the decrease in stability happened at the most common speeds in my area, so the weights went in the bin.
Finally, I also found that if you don't carefully balance front and rear, then if your pole is evenly slightly off vertical, at low wind speeds the vane wants to park with the heavier part "downhill". If that's not the wind direction, it will wander with speed variations, turning them into direction variations.
I'm also really keen to see the results!

[EmmBeeGee]

First off, I really like what you've done with the tail - excellent job! I'm not sure about adding the mass to the top, however. It may or may not help, but I've always found that when I change 2 things at once I can never tell the impact of either. It might make sense to remove it until you've assessed the impact of the tail mod.
Also, my experiment with adding mass resulted in more stable performance at some wind speeds, but more eratic performance at others. As luck would have it, the decrease in stability happened at the most common speeds in my area, so the weights went in the bin.
Finally, I also found that if you don't carefully balance front and rear, then if your pole is evenly slightly off vertical, at low wind speeds the vane wants to park with the heavier part "downhill". If that's not the wind direction, it will wander with speed variations, turning them into direction variations.
I'm also really keen to see the results!

Thank you Charlie for the encouraging words.

I discover that overnight the station has been "interfered with" - from the Rain Data and breaks in wireless contact I suspect inquisitive deer. I know that the slightest disturbance to the rain gauge when making mods to the sensors can result in glitches.

The most consclusive results thus far obtained, indicate that the verdict on the mods is... well... inconclusive!

windd 25 April.png

windsm.png

Certainly the straight line on the graph looks encouraging but much of that I suspect is due to the many breaks in contact with the sensors. Couple that with the almost total lack of air movements and I am afraid that I must await conditions more condusive to gaining reliable data.

Edit: The sudden spike in the rain data - no appreciable rain has fallen here for a few days, most "indications" are as a result of my "adjustments".

rainsm.png

[EmmBeeGee]

Without raising the sensors to a much greater height I dont think I shall "fettle" any more.

Fettling!


This iteration of the fins has recently produced the following results:-

windd 30 April.png

wind speed 30 April.png

Version 1a.JPG

Wooden dowel has been replaced by plastic wall plugs so that I can use screws to fix the spacers; two rubber washers on the rear spacer hold the fins laterally.

Yes I know the lump of lead atop the WDI is not only questionable in its efficacy but also inelegant, however some inexact testing is persuading me that this mass is of some value.

Fettling.

[EmmBeeGee]

Wind direction hasn't changed an awful lot but speed is more variable.

wind speed 02 May.png

wind direction 02 May.png

What is the general concensus of opinion as to the efficacy of the fins?

[Charlie]

You need a wider range of wind speed to be completely sure, but the results look good to me. I'm completely convinced this modification improves things, and I've made my version of the modification permanent. If I need to take it down at some future date, I may change to your design as it is more elegant. Case closed for me.
Another thing you may wish to try... I added a small (6 mm) super magnet on top of the existing directional magnet to increase it's strength. As you can read elsewhere, there are only reed switches for 8 points, and it relies on closing 2 switches at once to expand to 16 points. If you look at your data, the frequency of this happening is rare. When I added the magnet last January, this improved considerably. It still does not happen as often as a single switch closure, but the improvement was enough for me given our typically balmy January weather (-10 C to -20 C) is not conducive to outdoor science experiments. It just occured to me that putting it on the underside might have been better, but it's way up a pole now, so this experiment will wait for a bit too.

[EmmBeeGee]

You need a wider range of wind speed to be completely sure, but the results look good to me. I'm completely convinced this modification improves things, and I've made my version of the modification permanent. If I need to take it down at some future date, I may change to your design as it is more elegant. Case closed for me.
Another thing you may wish to try... I added a small (6 mm) super magnet on top of the existing directional magnet to increase it's strength. As you can read elsewhere, there are only reed switches for 8 points, and it relies on closing 2 switches at once to expand to 16 points. If you look at your data, the frequency of this happening is rare. When I added the magnet last January, this improved considerably. It still does not happen as often as a single switch closure, but the improvement was enough for me given our typically balmy January weather (-10 C to -20 C) is not conducive to outdoor science experiments. It just occured to me that putting it on the underside might have been better, but it's way up a pole now, so this experiment will wait for a bit too.

Due to being absent for a couple of days the wind direction plot is lacking some essential detail - wind speed picked up in my absence.

Charlie, other than a few cosmetic updates I shall leave the WDI more or less as shown in the most recent photo - including the asymmetrically positioned mass atop the unit.
I don't dare attempt any modifications that entail intervention into the sensors, as any breakages will render the station inoperable due to lack of funds to replace units.

Before finally mounting the station aloft on the gable-end of the house I may mount the whole kit & kaboodle on the intended TV type poles but then raise them in the current position in the field.

Here are the latest data:-

wind speed 5_6 May.png

wind direction 5_6 May.png

[mcrossley]

One observation I would make is that all the graphs you have provided have shown the wind direction pretty much from the NW. With your mods if there is an imbalance in the sensor it will be very susceptible to having a strong bias to a particular direction if is not perfectly level. Your photos do seem to show it being tail heavy (it is tipped down towards the tail).

May or may not be relevant?

[EmmBeeGee]

One observation I would make is that all the graphs you have provided have shown the wind direction pretty much from the NW. With your mods if there is an imbalance in the sensor it will be very susceptible to having a strong bias to a particular direction if is not perfectly level. Your photos do seem to show it being tail heavy (it is tipped down towards the tail).

May or may not be relevant?

Anything is possible!

In positioning the mass eccentrically on top of the WDI I hope to have avoided the effects that you mention.
Previously I have endeavoured to balance the mass of the added fins by applying lead to the "pointer", varying the "moment"! Unfortunately in the low wind strengths at the time prevailing, all I achieved was a WDI that spun in circles at each little zephyr!

Notwithstanding the above, flakes of snow mixed with the heavy recent rain gives witness, added to my two days hospitalised-forced observation of a church weather vane, to a wind persistantly in a generally Northerly direction.

The behaviour of this girouette took my attention when it pointed due north in light winds; there was little movement. This situation changed as wind speed increased and I noticed that the swing was in the order of 5° either side of the mean wind direction. As the speed of the wind increased the oscillations increased to around 10 - 15°.

Significantly (to me) was the variation in the indicated wind direction at higher wind speed; the girouette swung from N to WSW (followed by a return to the north) in various stages; the maximum single observed change was estimated as NW to WSW.

These observations give me a little more confidence in the value of those data recorded by my WDI.

Mark, I have in mind to attempt at a later time to use model aircraft techniques to establish the best achievable balance of the WDI.

[Orion]

Here is my wind vane mod
Fin cut from a piece of 2mm acrylic and glued to bottom of wind vane weight 5grams
and to balance it i used 2 x 15mm brass olives and cut them down to 13mm to fit over the nose and glued into position with araldite rapid
and a bit of red touch-up car paint to finish off
so far seems to work well but not much wind ATM

[Gina]

I'll be very interested to see if your mod improves things I extended my vane backwards and upwards (as shown elsewhere here) and it only made things worse! I do have it in mind to do more research on wind vanes later.

[MickinMoulden]

The wind vane mods are very interesting to read and I think have really tidied up the results.
At the start of these threads they were mentioning mods to the rain collector. I too made adjustments as I found my WS-1081 was 9% under reading (would read 91mm when it should read 100mm). I tested this by calculating the area of the collector (mine was 50mm x 110mm), and worked out that 550mL of water through the meter should read 100mm. Being aware that the highest rainfall rate I have had to date (only 4 days!) is 60mm/hr which equates to the meter only tipping over approx. 3.3 times per minute (every time mine tips it registers 0.3mm), you can not pour water in it to get accurate readings, instead you have to use a dropper (like for giving medicine). So I wouldn't be there all night dripping 1/2 litre of water, I downsized the experiment. 9 clicks of the meter should represent 2.7mm using 14.85mL (550mL/(100/2.7)). When I tested it, it took 16.2mL. That's a 9% under reading! (At this stage you'll be saying to use Cumulus Calibration mode to fix this, but I wanted it correct for my station too).
Don't bother adding mass to the tipping cups to take up volume so less water is required as the extra weight requires more water to overcome the extra weight.
What I did was to reduce the tipping angle so that less water would be required to overcome the weight of the empty side. I did this by hot gluing a match stick to both sides of the base (inside the unit), reducing the distance (thus angle) of the tipping by about 1mm each side. BINGO! After testing this, it was perfect!
Another mod I did was increase the wall height of the collector as I would presume there would be "splash out", further reducing the readings. They need to be the exact area of collection of the original (50x110 as mentioned). I made them out of laminating sheets (put through the laminator to stiffen them up) and hot glued together, then sealed it onto the rain gauge. An extra bonus of it being clear (and knowing that it is reducing splash out), is you can see rain droplets being caught and running back into the unit from the comfort of ground level (looking up, my unit is 5.1 meters up).

[Ned]

All good stuff. However, the height will cause under reading in windy conditions, when rain is blown across the collector, resulting in losses. Official rain gauges are mounted very close to the ground where wind effects are minimised, so you may find calibration is still necessary - it's a good idea to install a manual rain gauge for comparative purposes. My gauge is also mounted high (no good alternative) and the differences between my low level manual gauge cannot be compensated for with a single correction for all conditions. Eventually one stops worrying about it and adopts a mean value (hopefully)

[Charlie]

Since this thread is becoming active again...

Last March I posted here about my tail mod (about page 9, I think), and I've been happy with it since then. It's still providing the same improvement it did 9 months ago, and it is still the original mylar/tape solution up the pole on top of the roof. I did not have a reason to go up on the roof all summer, so it has not been made sturdier in any fashion. But despite it's (mostly invisible, but highly kludgy) looks, it has held up well. The wedge shape damps nicely.

Here's hoping it continues - the roof will not be fun to visit over the next few months!

[Gina]

Thanks for that Charlie - very interesting Not that I'm doing anything with what's up the pole until the weather permits! I haven't even ventured to the base of the mast for a couple of days! There's been enough wind to keep the wind sensors working.

[Krumlov]

Any updates on what people are doing with the WDI? Anything that has worked without side effects?