Maximum wind speed

[wxtech]

Attached image of plot; simulation of anemometer cup rotation in revolutions per second on the X axis bottom scale with the wind speed along the Y axis left side scale.
The green plot is the theoritical value of wind speed.
The red plot is the console indicated wind speed.

[MikeInTX]

Regarding the repetitive numbers for max speed... I see them all the time - 11.5mph max 28 days since 10/09, 19.6mph max 19 days since 10/09, 20.7mph max 13 days since 10/09, etc. The higher the speed, the less repetitive.

I have hit the same all time max wind speed of 35mph several times since 02/09 and have attributed this reading to the anemometer sitting of 15ft (thanks to home owner association restrictions ). Based on its surroundings, houses, trees, etc. I don't expect it to read anything much greater than that.

My anemometer is sitting at 27 ft. unobstructed north and south with trees in the neighborhood east & west.  My max has been 35 mph in WL and observed on the VP2 console as well.  I do feel like there is something wrong with the readings.  As I stated earlier with a 140 mph leaf blower at as close as I can get about 3 ft. below the cups my max was only 56 mph observed on the console.

Your 56 mph indicated may be the effect that I noticed when the speed maxed and then started over at zero.

I just tried the leaf blower again.  This time I slowly increased the speed over about a minute and the max shows on WL Weather Bulletin to have hit 67 mph.  Also this time I got to within about 1.5 feet from the cups.   I guess this could be right but I'm not sure yet.

I'd like to hear someone else report their findings with a blower.

[dalecoy]

Attached image of plot; simulation of anemometer cup rotation in revolutions per second on the X axis bottom scale with the wind speed along the Y axis left side scale.
The green plot is the theoritical value of wind speed.
The red plot is the console indicated wind speed.

If you haven't dismantled the test equipment yet, it would be interesting to insert (say 100 ft of) cable, to see what effect that has.

[C5250]

157 mph was the highest wind speed that I saw on the console.  As the frequency increased, the wind speed started over at zero and increased with frequency.

I'll assume that was with the large wind cups configured. As the large wind cup lookup table ends at 0x9C. The small wind cup lookup table ends at 0xB4.

[Cienega32]

Dalecoy - I did the re-seating routine....

I was suggesting that to jeffgod.   Sorry I didn't make that clear.

It was very clear. I was just adding additional info/results to a thread that seems to have a few of us in the same boat...

[Cienega32]

Thanks for the time and effort, Al - that's great stuff!

[Cienega32]

pat,
I know my "high" is not huge.  But we really did not have any crazy winter storms (or any for that matter) this year and I only have data back to October.  Hopefully catch a bigger gust come first spring t-storm.

Andrew - your high gust is still only 42.56MPH. If you ever catch a 77.n KPH gust, my eyebrows will jump off my face.

Not so much a matter of "size" but a matter of KMH = MPH. If you matched my 48MPH, that would say something to me on the limiting issue. I can't wait for my wind storm to stop to hit mine with the leaf blower.

[wxtech]

If you haven't dismantled the test equipment yet, it would be interesting to insert (say 100 ft of) cable, to see what effect that has.

I don't have an O'scope with bandwidth to measure the loss. 

[dalecoy]

Davis specs:
reed switch closes/opens once per revolution with 50% symmetry

__________my__actual__tests___________________________________
A Davis VP2 wired on my workbench with a signal generator producing squarewaves;
...

Did you verify the Davis spec of 50% symmetry? 

Given the geometry of the device, I'm not sure why it would be symmetric (square wave), nor am I really certain that it would be the same on/off ratio at different speeds.  I would expect a shorter "closed" time than "open" time.

I don't suppose it's likely to matter (unless the pulse got really short).

If I get a chance in the next few days (and if you didn't already do it), I'll take a look at that aspect.

[wxtech]

Did you verify the Davis spec of 50% symmetry?  
Given the geometry of the device, I'm not sure why it would be symmetric (square wave), nor am I really certain that it would be the same on/off ratio at different speeds.  I would expect a shorter "closed" time than "open" time.
I don't suppose it's likely to matter (unless the pulse got really short).
If I get a chance in the next few days (and if you didn't already do it), I'll take a look at that aspect.

I'm eager to see what you learn.  I also expected a short switch closure, thinking it would be a brief pulse.  I also wish to learn more about how the SIM counts the wind speed data.  I have my own opinions on these but let's see yours.  
I have the set up to change the symmetry to various off/on ratios and see what effect it has on console indicated wind speed.  
Thanks, Al

[W3DRM]

Al,

When I was having VP2 anemometer problems I did some measuring of resistance values for the leads coming from the anny assembly. I noticed that the reed switch was closed for only about 10 degrees during a 360 rotation. If you look at the magnet and reed switch orientation with both the magnet and reed switch being horizontal, the switch is closed only while the magnet sweeps past it. Thus, there would not be a 50% symmetry (on half the time and off half the time). It would be much less than that.

For an on for 10 degrees and off for 350 degrees, the pulse duration on your signal generator should only be 2.777%. Since I didn't measure the exact duration of the switch contact closure, this is only a guess and it may be different in reality.

[dalecoy]

I'm eager to see what you learn.  I also expected a short switch closure, thinking it would be a brief pulse.  I also wish to learn more about how the SIM counts the wind speed data.  I have my own opinions on these but let's see yours.  
I have the set up to change the symmetry to various off/on ratios and see what effect it has on console indicated wind speed.  
Thanks, Al

OK.  I had a new Davis anemometer assembly on the shelf.  Here's what I measured:

Resistance measurement (between red and black) - approx 100 ohms when the reed switch is closed (actually 97.5).  As expected, infinity when the reed switch is open.

Angle where the reed switch is closed - 70 degrees, as near as I could measure it several times.  Call that plus/minus 5 degrees.  This was done simply by manually moving the assembly through its rotation (the proper direction), and marking the position of the edge of one bucket on a piece of paper under the assembly.

That would be a pulse duration of 20%. 

I approximately confirmed that on an oscilloscope.  As you can imagine, it's difficult to achieve a steady speed of rotation so that a synchronized display can be read.  I actually managed to capture one view on my storage scope that seemed to confirm the 20% pulse width.

As expected, this was a purely on/off waveform, as best I could observe.

An even tougher question - any change in pulse width with different speeds of rotation.  I did not observe any change, but that's not conclusive. 

[wxtech]

My conclusion is that the symmetry isn't important.
A substitue pulse source was connected instead of the anemometer.  Using an analog pulse generator, pulses were put into the SIM and the wind speed was read on the console.  The pulse frequency and symmetry was read on the o'scope.
I did these tests at different frequencies, trying for a long positive duration; equal positive & negative durations; and a long negative duration.

HZ/RPS    + DURATION                - DURATION                   WIND SPEED INDICATED ON CONSOLE
10.86 Hz          46ms, 50%                   46 ms, 50%           25 mph       square wave
10.86 Hz          2.4 ms, 2%                   90 ms, 98%           24 mph       brief pos pulse
20.00 Hz          24.8 ms, 49%               25.2 ms, 51%         45 mph       square wave
20.00 Hz          4 ms, 8%                      46 ms, 92%           44 mph       brief pos pulse
50 Hz              1.2 ms, 6%                    18.8 ms, 94%        98 mph       brief pos pulse
50 Hz              15.6 ms, 78%                 4.4 ms, 22%         97 mph        brief neg pulse
50 Hz              10 ms, 50%                    10 ms, 50%           97 mph      square wave

[dalecoy]

[Posted simultaneously - and edited, to say that I agree]

Thorough tests are important, even if the results prove the negative.

I dug deeper in the junk box, and found an older, well-used Davis anemometer assembly.  This one has an original short cable (about 6 ft. long or so) with a "wind" label at the end.  I think it may have come from a Weather Monitor II - but I'm not sure.

Measuring the "switch closed" resistance, it was about 1.6 ohms - in other words, this one doesn't have a current-limiting resistor in it.

With the cup assembly pushed fully on to the shaft (and no drip shield was on this one), I got a "switch closed" angle of very close to 180 degrees (perhaps 175).  

Moving the cup assembly to positions slightly farther out on the shaft, I could get "switch closed" angles ranging down to about 60 degrees, before the reed switch didn't close at all.  

[mackbig]

I should have tonight.  Big cold front came through. Wind Warning, it was forecast to get 60k winds, with 70-90k gusts. 
I only got a 64.3....   they canceled the warning early, so I dont think it came thruogh how they originally thought.

Andrew

Andrew - your high gust is still only 42.56MPH. If you ever catch a 77.n KPH gust, my eyebrows will jump off my face.

[LFWX]

Keep two things in mind

1) Davis measured winds will always be in full numbers of MPH... 1.0, 2.0, 3.0... (or the metric equals)

2) Airports are normally located in wide open areas, the environment around you will add friction even if you measure at 33 feet (10 meters). This could result in a point at which much higher winds aloft are needed to get surface winds above a certain point.

The same cold front (the south end) passed through my area and the high wind gusts varied for the nearby airports (Butler County)
http://mesonet.agron.iastate.edu/p.php?pid=201004032044-KILN-NOUS41-PNSILN

[Wx4U]

I think what most of us are wondering; is there a peak for the sensor. It appears most of the comments were about repeatedly hitting the same max speed time after time.
I have had speeds of around 50 mph, but the last significant gusts repeatedly maxed out 37 mph. Odd?

[wxtech]

It appears that the theoretical max is 180 mph when set up for small cups and 157 when set up for large cups.
Quoting C5250; "I'll assume that was with the large wind cups configured. As the large wind cup look up table ends at 0x9C. The small wind cup look up table ends at 0xB4."
I removed the mechanical sensor and used an oscillator to get a reading of maximum of 157 mph on the console.
Worn or dirty bearings/shaft may cause a limit on the rotation speed, causing a familiar max speed seen.

[dalecoy]

I think what most of us are wondering; is there a peak for the sensor. It appears most of the comments were about repeatedly hitting the same max speed time after time.
I have had speeds of around 50 mph, but the last significant gusts repeatedly maxed out 37 mph. Odd?

I suggest that you look at the WeatherLink strip chart function, with either/both wind speed and "hi wind speed" graphs, and scan over relatively long periods. 

Think about this graph.  If you see "flat tops" on the graph, then (for your individual setup) you have a peak limitation.  If you don't have "flat tops", but you see wind speed readings that match how you think they should vary for your location, then what you are seeing is "real".

Let us know what you see, and what you conclude from thinking about the graphs.  Or, post a graph or two.

[jeffgod]

I have Cumulus telling me about the .3 of 48.3, I hope that clears up some confusion. I am also using the large cups. I am going to go into WeatherLink and do some research. Once again wxtech, thank you for all the hard work.

[MikeInTX]

I have Cumulus telling me about the .3 of 48.3, I hope that clears up some confusion. I am also using the large cups. I am going to go into WeatherLink and do some research. Once again wxtech, thank you for all the hard work.

The question about the .3 of 48.3 was only if using MPH.  Another poster (Cienega32) had said MPH but then responded that he was using Weather Display.  I use Cumulus and WeatherLink and the tenths don't show from my VP2.

[ocala]

I have Cumulus telling me about the .3 of 48.3, I hope that clears up some confusion. I am also using the large cups. I am going to go into WeatherLink and do some research. Once again wxtech, thank you for all the hard work.

The question about the .3 of 48.3 was only if using MPH.  Another poster (Cienega32) had said MPH but then responded that he was using Weather Display.  I use Cumulus and WeatherLink and the tenths don't show from my VP2.

I am using Cumulus Version 1.8.9 Build 927 and it does show tenths of MPH for the wind with my VP2. Strange that yours doesn't.

[MikeInTX]

I have Cumulus telling me about the .3 of 48.3, I hope that clears up some confusion. I am also using the large cups. I am going to go into WeatherLink and do some research. Once again wxtech, thank you for all the hard work.

The question about the .3 of 48.3 was only if using MPH.  Another poster (Cienega32) had said MPH but then responded that he was using Weather Display.  I use Cumulus and WeatherLink and the tenths don't show from my VP2.

I am using Cumulus Version 1.8.9 Build 927 and it does show tenths of MPH for the wind with my VP2. Strange that yours doesn't.

OH, well I'll have to talk to Steve at Cumulus about this.  I was using Version 1.8.9 until three days ago and am now trying out v1.9.0 while looking for another problem for Steve.  Both versions only show .0 for the tenths.  My VP2 wired console firmware is Rev 1.90.

[Weather Display]

its only accurate to the whole mph though, so really not much point in showing the decimal place for mph
software should have choice to show the decimal place if wanted to though to please

rounding changes from changing units will show up a decimal place within the software

[MikeInTX]

I have Cumulus telling me about the .3 of 48.3, I hope that clears up some confusion. I am also using the large cups. I am going to go into WeatherLink and do some research. Once again wxtech, thank you for all the hard work.


The question about the .3 of 48.3 was only if using MPH.  Another poster (Cienega32) had said MPH but then responded that he was using Weather Display.  I use Cumulus and WeatherLink and the tenths don't show from my VP2.


I am using Cumulus Version 1.8.9 Build 927 and it does show tenths of MPH for the wind with my VP2. Strange that yours doesn't.


OH, well I'll have to talk to Steve at Cumulus about this.  I was using Version 1.8.9 until three days ago and am now trying out v1.9.0 while looking for another problem for Steve.  Both versions only show .0 for the tenths.  My VP2 wired console firmware is Rev 1.90.

Steve at Cumulus responded "Cumulus displays VP2 wind speeds with a spurious decimal place. Normally I would expect the decimal to be zero, but in cases where Cumulus has to do a calculation, e.g. if you have set a "calibration" multiplier, or have asked it to calculate the average, or you don't use MPH, the decimal will contain a non-zero value. This is an error which I will fix in 1.9 at some point."