Topic: Acurite, Ambient, Davis - Model Comparison

[johnd]

Have you installed one of the mainstream Linux distro lately? Dont confuse UNIX with Linux.

Please, this isn't my perception - I'm quite happy using a Pi for instance. (In fact you've just prompted me to check my CumulusMX/Pi0W dashboard across my LAN.) But I'm just trying to reflect what in my experience the typical newbie customer is likely to think. And yes I agree that radio hams are of course pretty computer literate and open-minded (at least in an IT sense  ). But, again IME, they're only 5-10% of the general-interest/hobbyist market.

Where I do agree more is that if a Linux system can be sold as a turnkey solution then it's likely to access a larger slice of the general-interest market. But it does need to be backed up by good instructions/documentation and support, both of which add to the cost of course. The Pi is quite an attractive platform for this because the manufacturer/dealer could in principle just sell a suitably configured SD card just leaving the user a minimal amount of initial set-up to do, though I don't see many examples of this actually happening.

[pfletch101]

Being a little bit nitpicky but the Atlas display is a TFT LCD display. TFT = touch screen

The Atlas display may well have a touch screen, but TFT == Thin Film Transistor technology. See, for example https://en.wikipedia.org/wiki/Thin-film-transistor_liquid-crystal_display.

[worachj]

Being a little bit nitpicky but the Atlas display is a TFT LCD display. TFT = touch screen

The Atlas display may well have a touch screen, but TFT == Thin Film Transistor technology. See, for example https://en.wikipedia.org/wiki/Thin-film-transistor_liquid-crystal_display.

I know that, it was clumsy writing. I shorten what I was trying to say.

A TFT touch screen is a combination device that includes a TFT LCD display and a touch technology overlay on the screen.

[Dennis Rogers]

I have the vantage vue and still runs as good as when they first came out is when I got mine.

The only issue I had once was the rain gauge, but a clean soon fixed that. The only time I look at it to maintain it is when its battery change time and still happy with it.

I did recently get an ambient ws 2029 and love the screen.

Miles better than the vue screen which looks something out of the 80's techology.

My ambient clone is as accurate of my vantage vue and feel its projected forecasting is better as the barometer seems to react quicker to air pressure differinces.

I calibrated my ambient to my Davis air pressure readings as I know that's pretty much correct.

On weather underground my ambient clone accuracy is as good as any.

I use it to go online as was cheaper than getting my Davis online.

I have the USB data logger which works fine but sick of having to tether it up to PC to get online or run weather link,? Etc.

The ambient is a much better solution and as has been said they dont sell the ambient stations here and the clone is pretty much the same as the ws 2029 and was cheaper.

I believe ambient are made in the same factory anyway with slight twinks.

I am running it though ecowitt.net also and love how I dont have to have it tethered to my computer like Davis unless I get their wifi addon, Which I'd extra cost and more expensive to get than my ambient clone.

I think when my Davis finally craps out I wont bother replacing it as my ambient clone gives as good weather stats and pretty much neck and neck with my Davis.

[galfert]

I did recently get an ambient ws 2029 and love the screen.

Again with the bogus model number. Did you not read my other post?

Miles better than the vue screen which looks something out of the 80's techology.

If you think your new screen looks good you should check out the Ecowitt HP2551 / Ambient WS-2000. It does not have the limited viewing angles. You can get just the display as a worthy upgrade.

My ambient clone is as accurate of my vantage vue and feel its projected forecasting is better as the barometer seems to react quicker to air pressure differinces.

I calibrated my ambient to my Davis air pressure readings as I know that's pretty much correct.

On weather underground my ambient clone accuracy is as good as any.

I'm a very happy Ambient customer too but some of your claims are not based on reality.  Your perceived barometric precision is a by product of how the Fine Offset sensor used works with actually less accuracy. What you perceive as greater accuracy is actually the sensor being unsure of itself and vacilating between readings. The Fine Offset sensor also may be reporting with a shorter reporting period during these less accurate readings. Davis barometric accuracy is ±0.03 inHg and Fine Offset clones are ±0.08 inHg. The Fine Offset clone is still great but it's no Davis.

I use it to go online as was cheaper than getting my Davis online.

I have the USB data logger which works fine but sick of having to tether it up to PC to get online or run weather link,? Etc.

You don't have to stick the Davis USB logger in your PC. You could use a Raspberry Pi. Much more convenient and way cheaper than having gotten a Fine Offset clone. Then you could even send data with that Pi to way more places than the Fine Offset clone, to places like CWOP, AWEKAS, PWSweather, Windy, and many more including to your own website. The Fine Offset station can do this too but only by adding hardware/software.

Don't get me wrong I'm not saying you shouldn't be happy with your purchase, but rather just know that there are Pros and Cons no matter what you choose. Which is the purpose of why I started this thread. I didn't start it to say any one is better than the other. The purpose was to identify and inform of the shortcomings of all the current options. This way people can be informed and decide what they value more all the while realizing that there are things that they are giving up.

[ukwoody]

I Have an old "F" series vantage vue, which is dying on its windy bits   The anemometer broke in half and is glued back together with a nail and PU glue.  The cups (despite lubricating them) wont register over 34.5mph, and the supercap fails halfway through the night now.
I did have to replace the Hum sensor a few years back and I'm fairly certain it is starting to fail again.  I run an early type of logger that Belfrey made that has been superb - and still going strong.

Davis build a really good, robust fairly reliable unit.  It is (was) very accurate compared to two local professional stations.
However, I am reluctant to buy another davis.
Why?

The console would have looked old in 1976 and hasnt changed with modern times.  I VERY strongly object to having to pay a fortune for a proper logger/internet access. (yes I know I could keep my existing one and current console), but at 9 years old how much longer does it have??)
I feel Davis charge over the top for goods that lets be honest are years and years old and have barely changed and must have paid for them selves many times over).
Cost of spares very expensive especially here in UK.
They do not seem to have any interaction or communication with their userbase at all, and indeed almost seem to take them for granted.  They seem to have taken their eye off the ball. (weather the new ownship will make a difference remains to be seen).

Then I look at Ecowitt.  A new company.  Yes they are dealing with Fine offset, which quality wise may not be the best. BUT they are listening to users, working with them on forums, happy to make improvements and additions, will ship anywhere in the world and are creeping up fast behind Davis especially with their new seperates unit.  MUCH cheaper than Davis.  If it lasts 5 years I would still have saved money if I had to buy another one!  Plus, they are interent enabled, loggers etc etc.

Woody

[Dennis Rogers]

For me, one of the most important comparisons is stated accuracy.

What are you measuring accuracy to though? Your station may not reflect your countries offical temps; rain, etc and a station 1 mile away can report differently than yours.

I know where I live at 6km away from me the temps and rain very by quite a bit.

The only real way to gauge accuracy if your station is manually. Manual rain gauge and a standard liquid or mercury thermometer.

So how do you gauge accuracy? As your going to be different to someone 5 miles away..

[galfert]

For me, one of the most important comparisons is stated accuracy.

What are you measuring accuracy to though?

I'm sure the reference was to comparing the spec sheet stated accuracy of one brand/model to another. Knowing to what degree of accuracy your equipment will perform elevates the level of trust in the data after calibration. That said a broken sensor is a broken sensor. But usually when a sensor fails it is often obvious.

[johnd]

I'm sure the reference was to comparing the spec sheet stated accuracy of one brand/model to another.

The problem is that the spec sheets typically show only intrinsic sensor accuracy, ie how the sensor element would perform in a lab environment. Just to take one example, what's important for temperature accuracy is how well the sensor element performs in its radiation shield under different insolation values and different wind speeds. If a cheaper station has a poorly performing shield then the values it measure are also going to be less accurate even if the two are fitted with identical T/H sensor with identical intrinsic accuracy.

[galfert]

I still don't see how any of this is relevant.

I'm talking about comparing spec sheet to spec sheet in order to choose a better performing system. Spec sheet to spec sheet were all tests done in likely equal lab environments.

[CW2274]

I still don't see how any of this is relevant.

It is when one buys a PWS with a "spec" that's acceptable to the buyer only to get it properly sat and then realize the shield it's inside of ain't worth a damn, then it's relevant.

[galfert]

I still don't see how any of this is relevant.

It is when one buys a PWS with a "spec" that's acceptable to the buyer only to get it properly sat and then realize the shield it's inside of ain't worth a damn, then it's relevant.

Still I'm talking about station manufacturer's spec sheet, not sensor manufacturer's spec sheet.

[CW2274]

I still don't see how any of this is relevant.

It is when one buys a PWS with a "spec" that's acceptable to the buyer only to get it properly sat and then realize the shield it's inside of ain't worth a damn, then it's relevant.

Still I'm talking about station manufacturer's spec sheet, not sensor manufacturer's spec sheet.

What the hell good is the best sensor in the world if it's not properly protected...you're not buying a sensor, you're buying a PWS.  I do know that Davis does differentiate between sensor spec and inside the radiation shield spec with the VP2, aspirated and passive. How about the others? Is it merely a sensor spec or in the field spec? I'm betting the former because a lot of shields would make the sensor spec meaningless and far less attractive to the potential buyer.

[galfert]

From reviewing Ambient and Davis the specs from the manufacturer are different than from the sensor manufacturers (this is good). Not sure about Acurite but I would suspect that they do their own specifications too.

[CW2274]

Sensirion states the SHT31 at 0.2C temp accuracy, Davis states 0.3C. This quote is straight from Davis about the added error from the shield:

Outside temperature accuracy is based on the temperature sensor itself and not on the sensor and the passive shielding together. The solar radiation induced error for standard radiation shield: +4F (2C) at solar noon; for fan aspirated radiation shield: +0.6F (0.3C) at solar noon (insolation = 1040 W/m2, avg. wind speed 2 mph (1 m/s), reference: RM Young Model 43408 Fan-Aspirated Radiation Shield)

Not saying there isn't, but I know of no one else who'll state this, at least at this level.

[galfert]

Not saying there isn't, but I know of no one else who'll state this, at least at this level.

Other station manufacturers do it too, but I agree with you, definitely not at the level of detail that Davis does.
If I pull up Fine Offset specs (given in metric) you can see that they are different than what Sensirion publishes for the SHT30. For example Sensirion says SHT30 humidity is accurate to +-2% and Fine Offset says it is +-5%. Granted the range is different for those stated accuracies. But still Sensirion says worst case is +-4% for the full range and FineOffset says its +-5%. My point is there is clearly different specs than from the sensor manufacturer.

[CW2274]

Not saying there isn't, but I know of no one else who'll state this, at least at this level.

For example Sensirion says SHT30 humidity is accurate to +-2% and Fine Offset says it is +-5%.

Then that would be a direct relation to the error of the shield temperature wise, because a shield doesn't provide insolation protection from the humidity. 

[johnd]

From reviewing Ambient and Davis the specs from the manufacturer are different than from the sensor manufacturers (this is good).

From what I recall, the Davis T/H specs are identical to the Sensirion ones - this is exactly what I meant by the specs relating only to the intrinsic sensor specs. Of course there have been at least 3 different T/H sensors used by Davis in the recent past, so important to try (not always easy) to check that the Davis specs relate to a particular T/H sensor type. And that's for Davis which in general has pretty good documentation - I suspect that other cheaper makes don't even go that far.

AFAIK, Davis don't release operational specs, but trial results are available from eg the Stephen Burt reviews or eg the WMO (I think it was) report on shield performance. But TBH the specs are really a minefield IMO. For instance, T/H specs are quoted ±a value for accuracy but without saying what that ± value represents. In statistical terms it really ought to represent SEM of mean readings (or some such statistic) from a set of stations selected at random from production, but I bet it doesn't. But without some definition of the nature of the accuracy spec it starts to become less meaningful.

Without doubt, what we really need is a definitive long-term test site that is well-instrumented with high-end sensors to provide genuinely accurate reference readings against which test stations can be evaluated over eg 12-month periods.

[CW2274]

From reviewing Ambient and Davis the specs from the manufacturer are different than from the sensor manufacturers (this is good).

Without doubt, what we really need is a definitive long-term test site that is well-instrumented with high-end sensors to provide genuinely accurate reference readings against which test stations can be evaluated over eg 12-month periods.

Just as soon as you pay for it....

[johnd]

Just as soon as you pay for it....

Not out of the question, but it would need to be somewhere near me - maybe you could put in a good word with your guys at Lakenheath (or Mildenhall, though that seems to be closing unless there's a change of mind) and let us use a bit of their land. It's the practicalities of organising it that are the main problem.

[CW7491]

Just as soon as you pay for it....

Not out of the question, but it would need to be somewhere near me - maybe you could put in a good word with your guys at Lakenheath (or Mildenhall, though that seems to be closing unless there's a change of mind) and let us use a bit of their land. It's the practicalities of organising it that are the main problem.

At least as far as the VP2, Blue Hill Observatory outside of Boston, MA has pretty much taken care of organizing it and providing the data for probably at least 10 years with a co-located VP2 and ASOS. The data should all be on CWOP and Mesowest ... The Blue Hill, MA (East Milton, MA) ASOS is KMQE and the VP2 is CW2236.

https://weather.gladstonefamily.net/site/search?site=kmqe&Get+information=Get+information

https://weather.gladstonefamily.net/site/C2236

 [ You are not allowed to view attachments ]

How well the VP2 has been maintained and which T/H sensor version they were using at various times would be potential issues. But for someone with the time, desire and know how, there should be a wealth of data available at that particular site.

[galfert]

I've updated the OP with a note that Ecowitt will soon release a new sensor with the SHT35 chip sensor. There will be an indoor and outdoor version.

WH32-EP outdoor
WH31-EP indoor