Rain gauge errors

[TheChoctlab]

My VP2 rain gauge, not unlike other tipping bucket gauges, is often inaccurate. I have a Stratus RG202 manual gauge to compare it to.

I understand that I will need to manipulate weewx.sdb to reflect the readings from the manual gauge.  But, I am not sure what exactly to manipulate.


Guidance please.

[Andrew Milner]

Depends what you mean by 'manipulate'!!

If you have a formula that would convert VP2 reading to match manual reading then the formula could be put into the calibration section of weewx.

If you want to change existing readings already in the database then you can use SQLite to issue sql statements direcly

eg to double all rain readings before epochdate xxxxxxxxx would be something like:

UPDATE archive SET rain = rain * 2 WHERE dateTime < xxxxxxxxx;

[Andrew Milner]

I forgot to add

1. Take a copy of the database before manipulating/updating values

2. If you 'correct' the readings MAKE A NOTE OF THE DATE TO WHICH YOU DID THE UPDATE and then on subsequent manipulations it would be

UPDATE archive SET rain = rain * 2 WHERE dateTime > xxxxxxxxx;

[TheChoctlab]

Andrew,

In this case, "manipulate" = "correct."

Unfortunately, the difference between what the tipping bucket and the manual gauge reads is not linear.  So, multiplying readings by a constant number will not work.  I likely will have to make the correction(s) after every day that it rains.

[TheChoctlab]

After re-reading your post, the following occurred to me:

UPDATE archive SET rain = rain + yy WHERE dateTime > xxxxxxxxx;

yy being the difference between the tipping bucket and manual gauges.

On Monday, September 18, 2017 at 8:47:22 PM UTC-7, Andrew Milner wrote:

[Andrew Milner]

It should not be that incorrect if the tipping gauge is mounted in the correct manner. They are not that bad – seriously.

 

Sent from my Windows 10 phone

--
You received this message because you are subscribed to a topic in the Google Groups "weewx-user" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/weewx-user/ygSQ1OzWV88/unsubscribe.
To unsubscribe from this group and all its topics, send an email to weewx-user+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

 

[Andrew Milner]

A single addition would work for the daily total in the daily rain table, but not fir the individual archive records in the archive table. 

 

Sent from my Windows 10 phone

 

From: TheChoctlab
Sent: 19 September 2017 08:44
To: weewx-user
Subject: [weewx-user] Re: Rain gauge errors

 

After re-reading your post, the following occurred to me:

--

[TheChoctlab]

Got it.  Thank you.

[Andrew Milner]

The tipping bucket must be ut in a linear fashion as the only thing which can be wrong is the bucket size itself, assuming it is lubricated and moving freely. Have you ecperimented with dripping water into the gauge and measuring the number if tips for a fixed quantity of water? It must be a linear relationship as near as dammit.

[Thomas Keffer]

If you truly believe that your tipping gauge is inaccurate, why not just use the StdCalibrate facility? That's what it's there for.

How much of an error are we talking about?

-tk

To unsubscribe from this group and all its topics, send an email to weewx-user+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

 

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+unsubscribe@googlegroups.com.