Cloud Meter

I don't like clouds anymore. After relying too much on NOAA forecast (and as a result getting beaten here on forum, disqualified from NAR's best video contest, etc), I decided not to repeat that mistake ever again.

Here's new rocketry gizmo called "Cloud Meter". It measures temperature, relative humidity, and estimates cloud base altitude above ground level. Priceless for high-altitude flights!



Inside: Arduino Pro Mini, SHT15 temperature and humidity sensor, and a bit of math.

-Alex

Do you have a link to more info, Alex? Is that a packaged version of this https://www.elektor.com/magazines/2010/may/cloud-altitude-meter.1321986.lynkx?tab=1 board?
What we really want is a ceilometer in kit form...

Thanks for compliment but no, it's not from a kit. I built it from standard parts from SparkFun - I can send wiring diagram if interested.

-Alex

How exactly can you estimate cloud altitude by measuring temperature and humidity? If I were to risk flying on a cloudy day I would want to know a lot more about the prediction algorithms...

First, I calculate dew point from temperature and RH; from temperature and dew point I estimate cloud base. Here is short descriptionand here is longer one. Datasheet from sensor I'm using has a ton of info on this subject as well.

We'll see about its accuracy. So far, it's definitely better than relying on NOAA or local ICAO reports.

-Alex

It's amazing to me that that works, but the logic is valid-and if it's valid, and the assumptions are sound, I guess it works... lol

Do you have on-board sensors for temperature and humidity?

azzie said:

First, I calculate dew point from temperature and RH; from temperature and dew point I estimate cloud base. Here is short descriptionand here is longer one. Datasheet from sensor I'm using has a ton of info on this subject as well.

We'll see about its accuracy. So far, it's definitely better than relying on NOAA or local ICAO reports.

-Alex

Click to expand...

This is very cool. How are you determining its accuracy? If you are sending sounding rockets into clouds, don't tell us, just blink twice. :wink:

I know nothing about meteorology, but I assume that the calculations based on ground conditions can predict at what altitude clouds are likely to form, but is there a rate of false positives, so that even though the conditions are amenable to a 3200ft cloud deck there may not be one formed at any given moment (I suppose you could always just look up...)? And are there conditions where "rogue clouds" can drift through under a layer?

I'm at work and haven't had time to read through your links, so if these answers are there, no need to repeat them...

Hey Sully, you are right in a way that it is impossible to know cloud altitude without shooting rocket, balloon, or laser into it. We can do none of it! Still, it is possible to make an estimate under a certain set of assumptions. Here's how:

It is well known how much moisture air can hold at a given temperature and pressure. Once humid air cools down, the moisture will condense into clouds, fog or other "obscuring phenomena" as NAR puts it Also well known that temperature drops with altitude (5.5F per 1,000ft).

Now if we make an assumption (which will affect accuracy the most) that air has uniform distribution of moisture, from ground level and up, we can estimate at which altitude the moisture will condense. Cloud-Meter can not tell if clouds actually formed; it only ballparks at which altitude they will condense under current conditions.

So if I'm standing on the field trying to figure out whether to stick H or J into the rocket, and a cloud passes by, this device will give a clue.

Or so I think. We'll see


-Alex

azzie said:

I don't like clouds anymore. After relying too much on NOAA forecast (and as a result getting beaten here on forum, disqualified from NAR's best video contest, etc), I decided not to repeat that mistake ever again.

Here's new rocketry gizmo called "Cloud Meter". It measures temperature, relative humidity, and estimates cloud base altitude above ground level. Priceless for high-altitude flights!



Inside: Arduino Pro Mini, SHT15 temperature and humidity sensor, and a bit of math.

-Alex

Click to expand...

This project is very interesting. Thank your for sharing. The dew point is a very important peace of data derivate from the temperature and humidity. The dew point formula is kind of ugly. When or where up in the sky the dew point and temperature meet, we see fog, clouds, or rain. This project is showing that replacing somehow a 27K Dollars ceilometer.

Here is an online calculator to verify results:
https://www.csgnetwork.com/cloudaltcalc.html

We have a guy who has what he calls his "beater" rockets.
They have been loved a little too well. But he uses them as sounders.
A device like this would be pretty handy though!

The physics assumption in this calculation is that the adiabatic lapse rate applies all the way from ground level to cloudbase; i.e. you have a single airmass with no stratification. The assumption fails in the presence of any kind of inversion or vertical instability, both of which are pretty common. Good for an estimate but I'd not risk firing a large rocket into clouds without a real measurement.

caveduck said:

The physics assumption in this calculation is that the adiabatic lapse rate applies all the way from ground level to cloudbase; i.e. you have a single airmass with no stratification. The assumption fails in the presence of any kind of inversion or vertical instability, both of which are pretty common. Good for an estimate but I'd not risk firing a large rocket into clouds without a real measurement.

Click to expand...

I could be wrong, but I think if you have an inversion, this method predicts a lower cloud base than would actually exist. If so, the prediction is conservative and therefore safe.

azzie said:

Hey Sully, you are right in a way that it is impossible to know cloud altitude without shooting rocket, balloon, or laser into it. We can do none of it! Still, it is possible to make an estimate under a certain set of assumptions. Here's how:

It is well known how much moisture air can hold at a given temperature and pressure. Once humid air cools down, the moisture will condense into clouds, fog or other "obscuring phenomena" as NAR puts it Also well known that temperature drops with altitude (5.5F per 1,000ft).

Now if we make an assumption (which will affect accuracy the most) that air has uniform distribution of moisture, from ground level and up, we can estimate at which altitude the moisture will condense. Cloud-Meter can not tell if clouds actually formed; it only ballparks at which altitude they will condense under current conditions.

So if I'm standing on the field trying to figure out whether to stick H or J into the rocket, and a cloud passes by, this device will give a clue.

Or so I think. We'll see


-Alex

Click to expand...

It's a nice unit Alex.

The prediction accuracy depends on the temperature lapse rate you use to perform the calculation, which can vary from 5.5 F/1 kft to 3.0 F/1 kft depending on the RH of the air. Do you assume a fixed lapse rate or do you calculate and estimated lapse rate based on the actual ground level temperature and RH measurements? The latter method should give more accurate prediuctions.

bobkrech said:

I could be wrong, but I think if you have an inversion, this method predicts a lower cloud base than would actually exist. If so, the prediction is conservative and therefore safe.

Click to expand...

Complications abound. Stratifications are usually cold over warm, so the temperature gaps down at the boundary. If relative humidity levels were comparable (definitely not guaranteed here!) then cloudbase will be lower than that predicted by looking at RH/temp at the ground. I think the upper colder airmass is often drier, which would have opposite influence. Seems like no clear winner; we need data!

This is all interesting enough I think I could have an R&D project for NARAM. Last year they were relying on WX observations from a station 25 miles away to try to determine if the ceiling was adequate for G SD...during a day with convective turbulence and at least 3 visible cloud decks.

Edit: Somebody who has one of the Nikon, Bushnell or similar shooters/golf laser rangefinders should try it out on low cloud decks and post the results. Clouds are pretty reflective in the near IR used by these devices. It's possible that they'll work beyond their rated range. IR rangefinders exist that are rated out to 1-1.6 km. Prices range from $100 for Chinese ones to $300 for top-end Nikon 1km to $600 for Bushnell 1-mile rated.

I did find this web-based cumulus cloud base calculator ...

https://www.csgnetwork.com/estcloudbasecalc.html

Greg

The lapse rate might not be what you think it is. A while back I stumbled (more or less at random) on this set of sounding data:
Not only is the temperature nearly constant over this interval but the dew point is decreasing with altitude.

Interesting sounding...looks like about 3 layers. Boundaries are where the temperature gaps upward, followed by (lagging, sensor response related?) rises in moisture content and RH. Though it never gets close to condensing, it's a pretty good example of why a ground conditions based lapse rate estimate is not going to be too reliable.

What I want to see but have never seen sold at anything other than huge expense as professional equipment is a remote weather station for remote launch sites which would, just on weekends and holidays or during scheduled launch days, broadcast via formatted text messages (so a more expensive data plan isn't needed) data such as temp, average wind speed, wind gusts as the minimum data set to allow people who want to launch that day to determine if it's worth the long drive to the launch site. I've seen many Arduino projects that do this via short range wireless to the home or via a wifi network to the Internet, but never one for remote use via cell phone texting. I'd work on modding one of those projects to use a cheap eBay Chinese cellular module or a more expensive one from Adafruit, but I have other projects I'd prefer to work on.

Of course, this could also be done via a text message from a club person responsible for opening the site, but it's always amazing to me how clubs can be so resistant to good ideas.

Winston said:

What I want to see but have never seen sold at anything other than huge expense as professional equipment is a remote weather station for remote launch sites which would, just on weekends and holidays or during scheduled launch days, broadcast via formatted text messages (so a more expensive data plan isn't needed) data such as temp, average wind speed, wind gusts as the minimum data set to allow people who want to launch that day to determine if it's worth the long drive to the launch site. I've seen many Arduino projects that do this via short range wireless to the home or via a wifi network to the Internet, but never one for remote use via cell phone texting. I'd work on modding one of those projects to use a cheap eBay Chinese cellular module or a more expensive one from Adafruit, but I have other projects I'd prefer to work on.

Of course, this could also be done via a text message from a club person responsible for opening the site, but it's always amazing to me how clubs can be so resistant to good ideas.

Click to expand...

I saw an ~$70 one at Costco a while back...
Wind speed, direction, humidity, temperature, maybe some other stuff, and sends to phone. I though it would be nice if the info could be embedded onto the club website.

Winston said:

What I want to see but have never seen sold at anything other than huge expense as professional equipment is a remote weather station for remote launch sites which would, just on weekends and holidays or during scheduled launch days, broadcast via formatted text messages (so a more expensive data plan isn't needed) data such as temp, average wind speed, wind gusts as the minimum data set to allow people who want to launch that day to determine if it's worth the long drive to the launch site. I've seen many Arduino projects that do this via short range wireless to the home or via a wifi network to the Internet, but never one for remote use via cell phone texting. I'd work on modding one of those projects to use a cheap eBay Chinese cellular module or a more expensive one from Adafruit, but I have other projects I'd prefer to work on.

Of course, this could also be done via a text message from a club person responsible for opening the site, but it's always amazing to me how clubs can be so resistant to good ideas.

Click to expand...

It's not rocket science, nor is it complicated or expensive. It's a solved problem that has evolved over the past several decades.

https://www.wunderground.com/weatherstation/overview.asp

Personal weather stations have been around for several decades. They typically cost several hundred dollars compared to several tens of thousands of dollars for commercial quality weather stations. Many are connected to the Personal Weather Station Network which makes the information available to all on the internet through outlets like Weather Underground or any number of other similar internet sources.

To operate a personal weather station remotely, you need a power source, a way to access the internet, a small computer or microprocessor and the personal weather station. If you are really remote, you need a 200 watt solar panel, a charge controller and a lead acid storage battery. That's probably the most expensive item costing $500 to $1000. You should be able to pick up a cheap notebook or pad of some type for $100-$200 to operate the software. The weather station will cost $200-$500.

If your field is located near a cell tower, a cheap cell phone with a hot spot is all you need to get to the internet and log into the Personal Weather Station Network. If not you will need some type of ratio transmitter which is more expensive.

caveduck said:

Complications abound. Stratifications are usually cold over warm, so the temperature gaps down at the boundary. If relative humidity levels were comparable (definitely not guaranteed here!) then cloudbase will be lower than that predicted by looking at RH/temp at the ground. I think the upper colder airmass is often drier, which would have opposite influence. Seems like no clear winner; we need data!

This is all interesting enough I think I could have an R&D project for NARAM. Last year they were relying on WX observations from a station 25 miles away to try to determine if the ceiling was adequate for G SD...during a day with convective turbulence and at least 3 visible cloud decks.

Edit: Somebody who has one of the Nikon, Bushnell or similar shooters/golf laser rangefinders should try it out on low cloud decks and post the results. Clouds are pretty reflective in the near IR used by these devices. It's possible that they'll work beyond their rated range. IR rangefinders exist that are rated out to 1-1.6 km. Prices range from $100 for Chinese ones to $300 for top-end Nikon 1km to $600 for Bushnell 1-mile rated.

Click to expand...

A commercial laser range finder does not have the sensitivity or the electronics to make the measurement. A laser rangefinder is designed to look for the return from a solid object, not the backscattered light from a volume scattering source. Light travels at ~1' per nanosecond. If you are ranging a target at 1000 yards (3000') it take the laser light pulse 3000 ns (3 us) to hit the target and another 3 us (6000 ns) to return to the laser range finder. A typical laser pulse width is 5 ns so the laser rangefinder gate width is set for ~5 ns or 5' long pulse of light. The brightness of the laser beam fall off inversely proportional to the distance. If you were to measure the brightness at 1 yard and compare it to the brightness at 1000 yards, the distance ratio is 1/1000 so the brightness is 1/1,000,000! As the round trip is 2000 yards, if you could collect the entire illumination spot from a perfect reflector, the intensity would be down another factor or 4, however in reality, it is much lower as you are usually ranging on a diffuse reflector that absorbs some of the laser radiation, sometimes by several orders of magnitude. That's why in the specifications for laser rangefinders they are careful to define the type of surface that can be detected at various ranges.

The backscatter off rain droplets or other small particles are weak, orders of magnitude lower than a solid surface, and since it is generated in depth, the return pulse is much longer. You need a much longer range gate and a much higher optical gain to see the return. In layman's terms: big expensive optics and far more complicated electronics and computing power.

https://www.osram-os.com/Graphics/XPic2/00054201_0.pdf/Range Finding using Pulsed Laser Diodes.pdf

https://jultika.oulu.fi/files/isbn9514272625.pdf

https://en.wikipedia.org/wiki/Laser_rangefinder

UhClem said:

The lapse rate might not be what you think it is. A while back I stumbled (more or less at random) on this set of sounding data:

Code:

    PRES   HGHT   TEMP   DWPT   RELH   MIXR   DRCT   SKNT   THTA   THTE   THTV
    hPa     m      C      C      %    g/kg    deg   knot     K      K      K 
-----------------------------------------------------------------------------
 1000.0     93                                                               
  988.0    171   10.2   -3.8     37   2.93    310     12  284.3  292.9  284.8
  980.0    242    9.8   -6.2     32   2.47    312     15  284.6  291.9  285.0
  944.0    565    7.2   -7.8     34   2.26    319     30  285.0  291.8  285.4
  939.0    610    7.9   -8.3     31   2.19    320     32  286.1  292.7  286.5
  931.0    684    9.0   -9.0     27   2.09    314     34  288.0  294.3  288.3
  925.0    740    9.4  -11.6     21   1.71    310     35  288.9  294.2  289.2
  918.0    803    9.6  -14.4     17   1.37    306     35  289.8  294.1  290.0
  905.7    914    9.2  -13.8     18   1.46    300     36  290.5  295.0  290.7
  872.9   1219    8.1  -12.2     22   1.73    285     27  292.4  297.8  292.7
  857.0   1370    7.6  -11.4     25   1.88    275     26  293.4  299.3  293.7
  851.0   1428    8.4  -14.6     18   1.46    271     25  294.8  299.5  295.1
  850.0   1438    8.2  -15.8     17   1.32    270     25  294.7  299.0  295.0
  841.2   1524    8.1  -21.3     10   0.84    270     25  295.5  298.3  295.6
  832.0   1614    8.0  -27.0      6   0.51    261     28  296.3  298.1  296.4
  823.0   1704    9.2  -19.8     11   0.97    252     31  298.5  301.7  298.7
  816.0   1775    9.0  -13.0     20   1.73    245     33  299.0  304.6  299.3
  810.6   1829    8.5  -12.6     21   1.79    240     35  299.1  304.8  299.4

Not only is the temperature nearly constant over this interval but the dew point is decreasing with altitude.

Click to expand...

The good news with that data (if accurate) is, if you look up, you won't see a cloud, so there is not need for an estimate or a measurement!

-----------------------------------------------------------------------------------------------------------------------------

The physics of clouds requires the air temperature is below the dew point of the air at altitude. If the ground level air temperature equals the dew point, you will have either fog or rain or both. If the dew point is lower than the ground level temperature, and there is no temperature inversion, a cloud can form at the altitude where the air temperature equals the dew point temperature.

If the air is dry, the standard lapse rate is 8 C/km or 4.4 F/kft. If you look at the chart below you can determine the dew point if you measure the ground level temperature and the relative humidity.

For example on a nice summer day the temperature could be 30 C (86F) with a 50% RH. The dew point for this condition is 18.4 C (65.1F). The difference in temperatures is 11.6 C (20.9 F). The cloud height would be 11.6 C / 8 C/km = 1.45 km or 20.9 F / 4.4 F/kft = 4.75 kft.

https://en.wikipedia.org/wiki/Lapse_rate

I keep it simple and have no need for any device that measures cloud ceilings. I just look to the East at the highest mountains and if their peaks are visible, then I know I have clouds over 2000 feet in height. The lower mountains are 500 feet in height and if the Columbia River has white caps, it is blowing too hard to fly.

Winston said:

What I want to see but have never seen sold at anything other than huge expense as professional equipment is a remote weather station for remote launch sites which would, just on weekends and holidays or during scheduled launch days, broadcast via formatted text messages (so a more expensive data plan isn't needed) data such as temp, average wind speed, wind gusts as the minimum data set to allow people who want to launch that day to determine if it's worth the long drive to the launch site. I've seen many Arduino projects that do this via short range wireless to the home or via a wifi network to the Internet, but never one for remote use via cell phone texting. I'd work on modding one of those projects to use a cheap eBay Chinese cellular module or a more expensive one from Adafruit, but I have other projects I'd prefer to work on.

Of course, this could also be done via a text message from a club person responsible for opening the site, but it's always amazing to me how clubs can be so resistant to good ideas.

Click to expand...

It might be already out there: www.aprs.fi. Find your launch area. Click on "show all" and see if there might already be some WX stations near your launch site. Click on them for the latest information.

Yeah, one might not have any WX stations nearby but it's worth a shot. My local site has four nearby stations to peruse.

The problem with leaving any kind of equipment on a field is it could get stolen unless it's in a permanent setting with a power source and a high fence with barbed wire on the top. I have seen a remote station out in the country that is exactly that.
Likely professional grade but appears on the Wunderground site. Try the APRS-IS thing and see what you get. You got nothing to lose. Kurt