Welcome to the Parallax Discussion Forums, sign-up to participate.

# Is PING sensor temperature compensated?

Posts: 5
edited January 2006
Does anyone know if the PING sensor is temperature compensated?
The spec. sheet for other ultrasonic sensors (eg. Senscomp·http://www.senscomp.com/specs/Mini-A%20PB%20spec.pdf)
state that they are.

• Posts: 6,491
edited October 2005
No, there is no temperature compensation on the Ping -- it is not designed for temperature extremes that would require compensation. And given the Ping's short range and ~1/2" accuracy, temperature compensation would just add cost and complication without benefit to the user.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax
• Posts: 6,241
edited October 2005
The speed of sound depends on temperature. About 1120 feet per second at room temperature.

Here is a calculator:
hyperphysics.phy-astr.gsu.edu/hbase/sound/souspe.html

If you measure temperature in your environment, you could do the temperature compensation yourself.

There is also a complicated dependence of the speed of sound on altitude. The URL above has links to breaking the sound barrier, with fascinating photos of clouds that form around an aircraft.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
• Posts: 5
edited October 2005
Thanks for the feedback.

I was trying to make an inexpensive snow depth sensor that would automatically post the data to the web (I thought it would be fun to add to the weather stations and lightning detectors that I've been building for the local schools).

I did a quick experiment with a rabbit processor board from Imaginetools (boo hiss... I know this is a Stamp forum :-). I used a thermistor and a R/C to measure the temperature and the PING to measure the distance to the ground.

I used the formula· 331.4 + 0.6 * (T -273.0) to calculate the speed of sound.

Over a period of 24 hours I took data and I see a 0.5" drift without compensation and almost 0.8" drift with compensation. That's what made me think that the PING might have onboard compensation.

There are other sensors... but the 5/\$100 was in the right range for the school projects. I'm hoping to build something with 1/10th " accuracy.

Last year I built some IR snow depth sensors (http://www.howmuchsnow.com/snow/), but thought I'd try out the PING because of it's price and increased range.
• Posts: 6,241
edited October 2005
Hi Chris,

The sensor was solidly mounted and pointed at a solid flat ground surface, right? The echo signal from the PING))) should be longer duration at low temperatures, right?
20 deg C, 1126.7 feet per second, 74.0 microseconds for 1" difference in go and return
0 deg C, 1087.3 feet per second, 76.6 microseconds for 1" difference in go and return.
That is not much difference, marginally detectable by the original BASIC Stamp, but easily detectable by the faster Stamps.
I'm a little suspicious of the R/C measurement of temperature. I'd be more confident with something like an LM34 or a DS1620.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
• Posts: 5
edited November 2005
Here is an example of using the PING sensor (http://www.howmuchsnow.com/ping/) for snow depth measurement. Temperature compensation makes a big difference, but the sensor drift seems to be around +/- 0.3". I'm still waiting for some snow to test the accuracy on :-)
• Posts: 1,563
edited January 2006
Hi nafis,

I notice you have the hotplates listed on your website. Do you have anything to do with these?
We have a set running at the moment at one of our field sites (I work for Environment Canada)....and they seem to run quite well.
We also have a POSS (Precipitation Occurance Sensing System) that is basically a police radar (bistatic) pointing straight up. It measures the reflected energy (for drop size/flake size) and also the rate of fall. When used with a temperature sensor, it does quite well to give us info on what kind of precip is falling.
We've also got your a-typical 'weighing' gauge. Many exist...ours uses a vibrating wire and a bucket. As the bucket fills...more tension goes to the wire, which increases its vibrating frequency (like a guitar string). With a dose of anti-freeze (to melt snow) and transformer oil (to prevent evaporation), we do quite well for accumulation values.
We've also got a tipping bucket gauge (calibrated see-saw with cups on the ends) that will tip for every Xmm of rain (only good in rain).
We've also got a Capacitance plate that works well for moderate rain/snow...but seems to get swamped with heavy precips.

I enjoyed reading your webpage and was also looking to use the PING as a snow depth sensor. We do have some ultrasonic snow sensors in teh field, but they cost WAY too much (IMHO). Haven't tested it just yet....too busy with the other gear.

Another sensor we will be getting is a GPS rain sensor. I'm not entirely sure how it operates...but its something along the lines of measuring the GPS signal delay from each satellite and how it worsens in precip.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve

"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
• Posts: 5
edited January 2006
Hi Steve,

Sounds like you have a lot of nice·tools to measure precipitation. I only have unheated rain gauges and several prototype snow sensors. I think the IR sensor ([url=http://www.howmuchsnow.com/snow/) is]http://www.howmuchsnow.com/snow/) is[/url] more accurate for small snowfalls. I think I need to use a better temperature sensor·with the PING sensor.· The thermister is not tracking my other 1-wire temperature sensors. I may be reading it to often and heating it up?

Take a look at Steinar's sensor (http://voksenlia.net/met/snow.php)

····························· Chris
• Posts: 1,563
edited January 2006
Have you thought about going with a larger onsite 'stevenson' screen with a "psychrometer motor"?
Keeps a relative flow.
Mostly used for Dry and Wet bulbs for Rh calculations....
I have to try and find the calculation table....

We've got a couple of the \$30 thermistors and they seem fairly good. (part # YSI55006: glass thermistor).
I'm just a tech here...not really in to the data side of it, so couldn't say for sure which ones work the best....I think our better temp sensors are the CAmpbell Scientific ones. The HMP35C (with Rh sensing) are popular with us on their dataloggers.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve

"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
• Posts: 6,241
edited January 2006
Hi Chris, I'm interested in followup information too, about how well the PING)))-- How did it work when the snow started falling?

Something like the DS1620 or SHT15 would probably better for the temperature compensation? I think the thermistor is plenty accurate, but RCtime has quite a drift.
www.emesystems.com/BS2rct.htm#drift:

A group from the geology dept. at UC had a data logger next to a standing snowfall gage with glycerin (I think it was glycerin) inside to melt the snow. That is, until a bear came along and thought it smelled pretty sweet. He took out the rain gage and knocked around all the other equipment too. The logger kept running, so at least we know when it happened!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
• Posts: 1,563
edited January 2006
I think we use ethyl glycol to melt our snow.
I see some other precip sensors use heaters....which isn't the way to go if you're a remote station on solar power.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·

Steve

"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."