So after trying several things and failing, I did some more reading. A
couple of things that I tried and was disappointed in that I could not get
them to work were the accumulators without deltas and the weewx accumulator
time page (both that I posted in the previous notice.
In particular, the accumulator page where they define mintime, maxtime,
lastime as extractors within the accumulator looked so promising but
everything I tried created errors within the accum.py program.
I also kept running into issues where it seemed that the accumulators were
reseting at the end of each archive (REC) loop!
So here is my present solution (not yet tested cause I am waiting for
thunderstorms to come through!!!) .
There are three sections to this.
The first section is the [STDCalibrate] in weewx.conf
the lightning_distance and lightning_strike_count come from the Deltas
defined in the SDR.py
[StdCalibrate]
[[Corrections]]
# For each type, an arbitrary calibration expression can be given.
# It should be in the units defined in the StdConvert section.
# Example:
foo = foo + 0.2
# lightning_strike_count is a internal variable that is used by
weewx in reports. It shows the numer of lightning strikes that occured
within a loop/archive interval
# obtain information about lightning strikes for reporting to
hubitat
# lightning_count_sum is the sum of lightning_strike_count for the
day
# lightning_strike_last_distance = distance of the last lightning
strike
# lightning_strike_min_distance = distance of the closest lightning
strike
#
# still have work to do to obtain the time of the two lightning
strikes
# still have testing to do on what is presently done
lightning_count_sum = lightning_strike_count
lightning_strike_min_distance = lightning_distance
lightning_strike_last_distance = lightning_distance
lightning_strike_last_time = lightning_distance
lightning_strike_min_time = lightning_distance
The second is the [Accumulator} section
[Accumulator]
#
# Set up accumulators for monitoring the lightning strikes
#
# Set them all for daily accumulation (period = daily)
#
[[lightning_count_sum]]
period = daily
extractor = sum
[[lightning_strike_last_distance]]
period = daily
extractor = last
[[lightning_strike_min_distance]]
period = daily
extractor = min
[[lightning_strike_last_time]]
period = daily
lastime = user.accumulators.LastTimeAccumulator
[[lightning_strike_min_time]]
period = daily
mintime = user.accumulatos.MinTimeAccumulator
##############################################################################
you will see to custom accumulators in the Accumulators... they are
stored in a file /etc/weewx/bin/user/accumulators.py
# user/accumulators.py
import weewx.accum
class MinTimeAccumulator(weewx.accum.Accumulator):
"""Custom accumulator to track the timestamp of the minimum value."""
def __init__(self, obs_type):
super(MinTimeAccumulator, self).__init__(obs_type)
self.min_val = None
self.min_time = None
def accumulate(self, record):
"""Process each record from the archive."""
value = record.get(self.obs_type)
if value is None:
return
ts = record.get('dateTime')
if self.min_val is None or value < self.min_val:
self.min_val = value
self.min_time = ts
def get_result(self):
"""Return the timestamp of the minimum value."""
return self.min_time
class LastTimeAccumulator(weewx.accum.Accumulator):
"""Custom accumulator to track the timestamp of the last value."""
def __init__(self, obs_type):
super(LastTimeAccumulator, self).__init__(obs_type)
self.last_val = None
self.last_time = None
def accumulate(self, record):
"""Process each record from the archive."""
value = record.get(self.obs_type)
if value is None:
return
ts = record.get('dateTime')
if self.last_val is None or value < self.min_val:
self.last_val = value
self.last_time = ts
def get_result(self):
"""Return the timestamp of the minimum value."""
return self.last_time
Again... these have not been tested yet (waiting on a thunderstorm) and I
am expecting some level of disappointment when it gets here!!!! However,
watching the records in the database they are being written and all the
values are 0/Null at this point (which is what they should be). I am
thinking that I might need to put a qualifier on the STDCalibration section
about only doing anything unless lightning_strike_count is none 0 (a
lightning strike occured within the loop interval. I am also expecting
that the mintime and lasttime accumulators that i wrote may not be 100%
correct or may be able to be done more eloquently.
If you have any thoughts let me know.
THanks
Patrick
On Friday, August 7, 2026 at 8:07:49 PM UTC-4 Patrick Sifford wrote:
> Found a couple of articles that may be of interest....
>
> weewx using accumulators without deltas - Search
> <https://www.bing.com/search?q=weewx+using+accumulators+without+deltas&cvid=bfe795017db14f25ae926b91adc02554&gs_lcrp=EgRlZGdlKgYIARBFGDsyCQgAEEUYOxj5BzIGCAEQRRg7MgYIAhBFGDkyBggDEAAYQDIGCAQQABhAMgYIBRAAGEAyBggGEEUYPDIGCAcQRRg8MgYICBBFGDzSAQg2NTU2ajBqOagCCLACAQ&FORM=ANAB01&adppc=EDGEINJP&PC=DCTS>
>
> weewx accumulator time - Search
> <https://www.bing.com/search?pglt=299&q=weewx+accumulator+time&cvid=997f44499e394bb88d7ac38baca61342&gs_lcrp=EgRlZGdlKgkIABBFGDsY-QcyCQgAEEUYOxj5BzIGCAEQRRg5MgYIAhBFGDsyBggDEEUYOzIGCAQQABhAMgYIBRAAGEAyBggGEEUYQTIGCAcQRRg8MgYICBBFGDzSAQgyOTEwajBqN6gCALACAA&FORM=ANNTA1&adppc=EDGEINJP&PC=DCTS&ntref=1&dayref=1&ajf=10>
>
> I think the accumulator time is going to be more useful... I was hoping
> that the accumulators without deltas would be nice but I think it only
> works with extractor last or firstlast... I have not been able to get it
> to run with min max or sum (but that maybe me).
>
>
> On Wednesday, August 5, 2026 at 12:15:26 PM UTC-4 Patrick Sifford wrote:
>
>> Here is the code that I am working on...
>>
>> [StdCalibrate]
>>
>> [[Corrections]]
>> # For each type, an arbitrary calibration expression can be given.
>> # It should be in the units defined in the StdConvert section.
>> # Example:
>> foo = foo + 0.2
>>
>> # lightning_strike_count is a internal variable that is used by
>> weewx in reports. It shows the numer of lightning strikes that occured
>> within a loop/archive interval
>> # obtain information about lightning strikes for reporting to
>> hubitat
>> # lightning_strike_count_start is the number of strikes that the
>> day started with (strike_count)
>> # lightning_strike_count_last is the present number of strikes
>> today
>> # lightning_count is the total count of lightning strikes that
>> have occurred today
>> #
>> # lightning_strike_last_distance = distance of the last lightning
>> strike
>> # lightning_strike_min_distance = distance of the closest
>> lightning strike
>> #
>> # still have work to do to obtain the time of the two lightning
>> strikes
>> # still have testing to do on what is presently done
>> #
>> lightning_strike_count_start = strike_count
>> lightning_strike_count_last = strike_count
>> lightning_count = lightning_count + lightning_strike_count if
>> lightning_strike_count > 0 else None
>> lightning_distance = strike_distance if lightning_strike_count >
>> 0 else None
>> lightning_strike_last_distance = strike_distance if
>> lightning_strike_count > 0 else None
>> lighting_strike_min_distance = strike_distance if
>> lightning_strike_count > 0 else None
>>
>> and
>>
>> [[Delta]]
>>
>> [[[lightning_strike_count]]]
>> input = strike_count
>>
>> [[[rain]]]
>> input = rain_total
>>
>> [[[lightning_strike_distance]]]
>> input = strike_distance
>>
>> [Accumulator]
>> [[lightning_count]]
>> extractor = last
>>
>> [[lightning_strike_last_distance]]
>> extractor = last
>> # merger = minmax
>>
>> [[lighting_strike_min_distance]]
>> extractor = min
>> # merger = minmax
>>
>> [[lightning_strike_count_start]]
>> extractor = min
>> [[lightning_strike_count_last]]
>> extractor = last
>>
>> If these work then I will start pulling the datetime information. I
>> believe that the SDR has a datatime available that I plan to then:
>>
>> lightning_strike_last_time = datetime if lightning_strike_count>0 else
>> None
>>
>> if lightning_strike_last_distance is <= lightning_strike_min_dstance then
>> lightning_strike_min_time = lighting_strike_last_time
>>
>> and accumulate with
>> [[lightning_strike_min_time]]
>> extractor = last
>>
>> [[lightning_strike_last_time]]
>> extractor = last
>>
>> random thought of an amature
>>
>> So far on my testing i have been able to show that the data flows over to
>> hubitat..
>>
>> My strike counter this morning was (and still is) 78. that value is
>> going over to hubitat in both the lightning_strike_count_start and
>> lighning_strike_count_last records!!!!
>>
>> what I now need is some bad weather so that I can see if the count_start
>> does not increment untill the new day (or acurite atlas reset) and that
>> strike_count_last value increments correctly.
>> then the next day I expect count_start to be set to the new value which
>> was the last count of the previous day. Or if atlas reset the value to
>> zero (roll over of power reset) that it is set to zero and count_last is
>> operating as expected.
>>
>> After that I will be testing if the distance values are working
>> correctly...
>>
>> then I will be testing the time values to see if they are working
>> correctly.
>>
>> All will be compared to the data being sent to MyAcurite by the acurite
>> hubs.
>>
>> Any thoughts would be very much appreciated!
>>
>> THanks
>>
>> On Wednesday, August 5, 2026 at 10:51:29 AM UTC-4 Patrick Sifford wrote:
>>
>>> Thanks... very interesting read!!!
>>>
>>> we are going two different directions on how to make this work. You
>>> have a much more elegant process that I want to get into to and play with
>>> sometime. Because of my lack of knowledge I was looking at a more brute
>>> force (which normally means wrong) way of doing this.
>>>
>>> My thought on time was to set the nearest time = present time when
>>> present strike distance is equal to or closer to nearest strike...
>>> and setting present time = present time when there is a strike.
>>>
>>> not really databasing it but getting the value available for hubitat...
>>>
>>>
>>> However, looking at your solution it looks really tasty... If I get
>>> mine working... or run into a brick wall that is there I will look at
>>> what you got... tempting as it means there are somethings I can learn
>>> from it!
>>>
>>> Thanks.
>>>
>>> P.S.
>>>
>>> it was suppose to rain this morning so I was hoping that we would get
>>> some lightning.... neither of rain or lightning. So I am going to make
>>> it look good and then will post for you to look at. Would love to have a
>>> test that runs with lightning and switch over at mid night.
>>>
>>> THanks
>>>
>>> On Wednesday, August 5, 2026 at 9:56:06 AM UTC-4 [email protected]
>>> wrote:
>>>
>>>> Patrick,
>>>> I'm looking forward to what you come up with. I did some work and you
>>>> can find it here,
>>>> https://github.com/bellrichm/WeeWX-Extras/blob/master/bin/user/observationtime.py
>>>> Basically I store the first/last/min/max of an observation for the
>>>> archive period. A bit more information is here,
>>>> https://groups.google.com/g/weewx-user/c/oQPj0cuYMZw/m/EJvMn6ggAQAJ
>>>> It is totally an experiment, so it no install and real documentation;
>>>> but maybe something in it will be useful to you.
>>>> rich
>>>>
>>>> On Wednesday, 5 August 2026 at 08:00:14 UTC-4 Patrick Sifford wrote:
>>>>
>>>>> Tom,
>>>>>
>>>>> Your information was very useful. I think I have a handle on most
>>>>> things that I am trying to get through. What I am trying to do is to
>>>>> have
>>>>> values for daily lightning strikes, nearest strike (and its time), and
>>>>> latest strike (and its time) available to send to Hubitat (That is the
>>>>> easy
>>>>> part for me).
>>>>>
>>>>> One thing that i did not see (may of skipped over it) in the wiki page
>>>>> is the time period of a accumulator. It was looking like it was a
>>>>> loop/archive interval, but i found out that happenned to be my data.
>>>>> From
>>>>> reading around it seems that accumulators reset to 0 at midnight. Once I
>>>>> got over that I was able (I think... still waiting on a thunderstorm to
>>>>> check it out) to get a lightning count for the day:
>>>>>
>>>>> lightning_count_start = the min value of strike_count (choose the min
>>>>> versus first incase acurite atlas resets the value during the day for any
>>>>> reason)
>>>>> lightning_count_now = the last value of strike_count
>>>>> lightning_count = lightning_count_now - lightning_count_start
>>>>>
>>>>> then put lightning_count into an accumulator for the last.
>>>>>
>>>>> finding the min and last lightning distance is easy also.... What
>>>>> I can not find any reference on is :
>>>>>
>>>>> Once I have the min lightning distance (or the last) how do I
>>>>> determine the time of that strike?
>>>>>
>>>>> Thanks
>>>>> Patrick
>>>>>
>>>>> On Sunday, August 2, 2026 at 12:33:43 PM UTC-4 Patrick Sifford wrote:
>>>>>
>>>>>> Tom
>>>>>>
>>>>>> Thanks... I think that will help
>>>>>>
>>>>>> I have seen other structures to the Delta functions... Ill give
>>>>>> this a try
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Saturday, August 1, 2026 at 5:45:37 PM UTC-4 Tom Keffer wrote:
>>>>>>
>>>>>>> Accumulators are documented here
>>>>>>> <https://github.com/weewx/weewx/wiki/Accumulators>.
>>>>>>>
>>>>>>> Unfortunately, there is no documentation for deltas, but they are
>>>>>>> pretty simple. If you have some extensive quantity, say "tide", and
>>>>>>> you want to calculate "delta_tide" as the difference in tide
>>>>>>> between archive records, it would look like this:
>>>>>>>
>>>>>>> [StdWXCalculate]
>>>>>>> [[Delta]]
>>>>>>> [[[delta_tide]]]
>>>>>>> input = tide
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Sat, Aug 1, 2026 at 2:13 PM Patrick Sifford <[email protected]>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Is there any where a definitive and correct set of documents that
>>>>>>>> discuss how to use deltas and accumulators and their formats?
>>>>>>>>
>>>>>>>> I am trying to work on Acurite Atlas Rain and Lightning sensors and
>>>>>>>> get confused when I read different discussions on the topics. I think
>>>>>>>> the
>>>>>>>> rain is taken care of (somewhere various rain items are calculated
>>>>>>>> (rainToday, rain24, ... ) that I am able to take advantage of.
>>>>>>>> However, I
>>>>>>>> do not seem to find similar items for Lightning.
>>>>>>>>
>>>>>>>> I have also found multiple solutions which do not seem to be
>>>>>>>> solutions... and they all seem to be using different formats of how
>>>>>>>> deltas and accumulator's work.
>>>>>>>>
>>>>>>>> Not looking for the solution ... looking for good reasonable
>>>>>>>> documentation of how these things work to include extractors...
>>>>>>>>
>>>>>>>> Thanks
>>>>>>>> Patrick
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> 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 [email protected].
>>>>>>>> To view this discussion visit
>>>>>>>> https://groups.google.com/d/msgid/weewx-user/de622fde-796f-43ad-b087-5cc28edaf347n%40googlegroups.com
>>>>>>>>
>>>>>>>> <https://groups.google.com/d/msgid/weewx-user/de622fde-796f-43ad-b087-5cc28edaf347n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>>>>> .
>>>>>>>>
>>>>>>>
--
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 [email protected].
To view this discussion visit
https://groups.google.com/d/msgid/weewx-user/cb5fd68d-b095-40cd-8a16-baa10cc34432n%40googlegroups.com.
