Topic: Davis VP-2 sensor(s) apparently drawing too much current - ideas?

[kobuki]

Yeah. Sorry, I haven't followed this as closely as you have. Now I remember asking once if Elagache tried to just put back the same battery since they, as you confirmed as well, aren't really depleted and maybe the battery removal in itself helps for a while. IIRC he didn't understand why I asked and I stopped following the soap opera for a while

[dalecoy]

Yes, and I suggested just leaving the battery in until at least the second failure, to see at what time (etc.) that occurs.

...and his Mac is frequently rebooted due to memory leaks...

...and the WeatherCat software is updated with a new version (and restarted) every week or so...

[elagache]

Dear Dale, Kobuki, and WXForum Davis troubleshooters,

I have stated this before, but I guess I have to repeat myself.

Yes, and I suggested just leaving the battery in until at least the second failure, to see at what time (etc.) that occurs.

I have left the battery in and the results are very simple.  The outages increase in duration and eventually occur at other times of the day until the ISS stops transmitting all together.  You'all seem to forget this problem has been going on since - October 2014.

Dale you have no concerns with asking me to lose more data.  If you really want to crack this puzzle, I suggest that you go over to your ISS and measure the voltage drop on your 123 battery - daily.  I think you'll soon realize that even if the voltage loss are relatively small, they are nonetheless far in excess what the ISS board should be causing on a daily basis.  Isn't a 123 battery supposed to last around 4 years before needing replacement?

I have already spent around $500 of new parts to move the ISS transmitter from its present location on the small deck to the present location of my temperature/humidity probe and switch from solar power to the Davis AC adapter.  I have a new temperature/humidity probe with the factory Davis cable - no more extension issues.  The 3 extension data cables for the rain gauge, solar sensor, and anemometer have been finally pulled through the conduits.  I have set up an extension cord to reach the current bracket for the temperature humidity probe.  I have replaced the Davis power connector jack with an 1/8" inline jack so I can power the aspirator fan using a universal AC/DC adapter.  I'm hoping to complete the work and switch the station over to this new configuration on Monday.

I'm going to try as hard as possible to not lose any more data until then.  As far as I'm considered there isn't any more worthwhile troubleshooting to be accomplished here.  Perhaps we don't have a definitive diagnosis, but it is clear to me that Davis has changed the power switching logic in the ISS circuit board and that is the cause of the trouble.  My best guess is that the system isn't switching to the battery quickly enough and there are conditions when the super-capacitor cannot supply enough power to the long cable run for my present temperature/humidity probe.  That's a good enough explanation for me.

Edouard

[dalecoy]

Dear Dale, Kobuki, and WXForum Davis troubleshooters,

I have stated this before, but I guess I have to repeat myself.

Yes, and I suggested just leaving the battery in until at least the second failure, to see at what time (etc.) that occurs.

I have left the battery in and the results are very simple.  The outages increase in duration and eventually occur at other times of the day until the ISS stops transmitting all together.  You'all seem to forget this problem has been going on since - October 2014.

I know you have stated that before.  I've reviewed all three threads that you started, and perhaps I missed where you give details, or when that last happened.  Could you point to that, please?

If that happened in October or November, since then you have made a lot of changes - replaced the ISS board, rewired things, changed versions of WeatherCat several times, etc. 

Dale you have no concerns with asking me to lose more data.  If you really want to crack this puzzle, I suggest that you go over to your ISS and measure the voltage drop on your 123 battery - daily.  I think you'll soon realize that even if the voltage loss are relatively small, they are nonetheless far in excess what the ISS board should be causing on a daily basis. 

I simply cannot do that.  There is no such battery in my Cabled VP2.  Perhaps someone else will. 

However, would you please do the test on your system?  That is, daily, measure the voltage under load on the battery? 

it is clear to me that Davis has changed the power switching logic in the ISS circuit board and that is the cause of the trouble. 

Why do you believe that Davis changed that?  And haven't you used 3 different circuit boards?

  My best guess is that the system isn't switching to the battery quickly enough and there are conditions when the super-capacitor cannot supply enough power to the long cable run for my present temperature/humidity probe.

I'm a bit confused by that.  If the "sensor errors" are caused by a lack of supercap power (and too-late switching to battery power), why don't the errors occur every night? 

But the above are relatively unimportant issues.

I do appreciate your desire to not lose data, of course.  And I really hope your new installation will have no more sensor errors, without the need to change batteries every 4 days.

[dalecoy]

Version of WeatherCat:
   V2.10, Build 960
WeatherCat uptime:
   21 hours, 27 minutes

Good data elements received from weather station:
   32150
Data link timeout errors:
   0
Data link invalid data errors:
   0
Weather station sensor errors:
   2480

Between approx. 5:30 and 5:45 AM

[elagache]

Dear Dale and WXForum Davis troubleshooters,

Version of WeatherCat:
   V2.10, Build 960
WeatherCat uptime:
   21 hours, 27 minutes

Good data elements received from weather station:
   32150
Data link timeout errors:
   0
Data link invalid data errors:
   0
Weather station sensor errors:
   2480

Between approx. 5:30 and 5:45 AM

Afraid so.  Had a similar scenario on January 16th when errors occurred only two days after replacing the battery.  The battery voltage under load this morning was 3.03 Volts.  The battery I had put in 18 hours earlier or so was new and had the normal charge of 3.23 Volts under load.

The circumstances of this morning event are also similar.  We have been having a period of sunny dry weather which allowed the solar panel to provide more power.  Before these two episodes, there was several days of partly cloudy conditions that would have significantly reduced solar power generation.

It appears under these conditions, the Davis power switching circuitry is putting an unreasonable load on the battery during the period when the sensor errors are occurring.  One possible way this could happen is if the super capacitor is relatively discharged, but the switching circuitry attempts to power the station with the super capacitor.  In this situation, the battery would end up trying to recharge the super capacitor and power the station.  That is the sort of load that might drop the battery voltage enough to generate the errors.

The plan remains to switch the station over on Monday.  I need to clean up the paths of the 3 extension cables in between the runs of conduit and do whatever weather-proofing is required.  On Monday, I need remove the ISS circuit board from the ISS enclosure, relocate the board into the 6382 enclosure, splice the data cables from the solar sensor, rain gauge, and anemometer, and complete the installation.  Once all that's done, I'll have to troubleshoot any problems that develop.

Wish me luck!

Edouard

[dalecoy]

Good luck.

And you have several times mentioned how the "Davis power switching circuitry" works.  Can you share the technical details (circuit diagram, etc.) of that circuitry with us, please?

[elagache]

Dear Dale and WXForum Davis troubleshooters,

And you have several times mentioned how the "Davis power switching circuitry" works.  Can you share the technical details (circuit diagram, etc.) of that circuitry with us, please?

I'm afraid I'll I know about it is what moehoward4 expressed in this post early in this thread:

http://www.wxforum.net/index.php?topic=24426.msg234869#msg234869

In that post he indicated that he had two Davis boards which were malfunctioning in the way that they switched from solar panel to super capacitor and finally to the battery.  There has to be some circuitry that monitors the electrical activity of each component and based on that either adds charge to the super capacitor while powering the instruments or discharges the super capacitor and finally the battery itself.

I've long forgotten my understanding of how capacitors work.  However, I do remember that they behave very differently from a battery.  A traditional capacitor can only collect static charge.  So the amount of current that can be supplied by a capacitor has to be limited.

The clear mistake in my setup is that I have a very long cable to my temperature/humidity probe with a splice.  At the very least, that should require more voltage and it might take a little more current.  Since Davis can test their stations with great precision, they clearly can determine precisely the power requirements for each instrument and fine-tune the way that the ISS switches from solar panel, to super capacitor, to battery to get the longest possible life of the battery.  Between 2009 and 2014, they made some adjustment to the ISS circuitry based on their altered recommendations about temperature/humidity probe cable length.  That change presumably includes lowering the available voltage (and possibly current) available to power the temperature/humidity probe.  Since my probe requires more than has been allocated, under some conditions, instead of being powered by the super-capacitor, the battery is forced to power the ISS transmitter and attempts to keep the super-capacitor charged up at the same time.  I believe that is the mechanism that is causing the excessive battery consumption.

That's the best guess I can come up with.  Since the only difference between my setup and apparently every other Davis installation out there, somehow that additional cable length is effectively throwing a monkey wrench into the way the ISS manages is power resources.

Edouard 

[dalecoy]

The technical explanation provided was:

Somewhere on the PCBA is something (circuit(s)) that tells the board where to get power.

Personally, without more details on the circuitry, I would not conclude that " the system isn't switching to the battery quickly enough and there are conditions when the super-capacitor cannot supply enough power to the long cable run". 

Nor would I conclude, if the system isn't switching to the battery quickly enough, that changing the battery every 4 days would help.

  Since the only difference between my setup and apparently every other Davis installation out there, somehow that additional cable length is effectively throwing a monkey wrench into the way the ISS manages is power resources.

Your problem is "sensor errors" reported by WeatherCat software.

In addition to the cable length and wiring, differences between your setup and most other Davis installations include:
1.  Your physical location, RF environment, etc.  And other electronics running in/around your home.
2.  You are running a Mac (with Macintosh Operating system (OS-X:) 10.10.1)
3.  You are running WeatherCat weather software.
4.  (Perhaps other software running on the Mac)
5.  You are using an Envoy for logging (most other installations will be using a console).
6.  Probably some other differences.............

Your upcoming new configuration will eliminate one of those differences, and I really hope that it cures the problem you are experiencing.  I'm looking forward to reports.

[elagache]

Dear WXForum Davis station troubleshooters,

I quick update on my plans to resolve this problem.

It was a terrific struggle, but today I finally completed relocating my station transmitter from the Davis factory position at the end of the rain gauge to a 6382 enclosure next to my temperature/humidity probe under the deck.  This meant running 3 50+ foot data cables from the anemometer, rain gauge, and solar radiation sensor back to under the deck.  It took a lot of prep work, but today I actually spliced the data cables, moved the ISS transmitter board into the new enclosure and replaced the old temperature/humidity probe with a new fan-aspirated assembly (fan is also AC powered via a universal AC adapter.)  I had to buy an extra Davis AC adapter to power the ISS board since solar power was no longer available.

This afternoon, I powered up the ISS board and waited . . . . .   At first the console couldn't find the instruments, but eventually everything worked normally.  I ran some water through the rain gauge and used a blower to get the anemometer moving just to be sure.  I then deleted the data logger in the Weather Envoy using the Davis WeatherLink software so all erroneous data was purged before WeatherCat was launched.  Everything has been working fine since.

I promise to provide a write-up of my changes, but I've had to put some other projects on hold that are now pressing.  So it may be while.

In a word Phew!

Cheers, Edouard

[dalecoy]

Thanks for the report.  That should do it.

[fish-on]

I have been reading this post for quite awhile and there has been a troubleshooting approach that is missing the most basic of test. First off this method will take about a month to get to the root cause. Maybe you have taken this approach, but I have not got that from any post. So here goes.
1) buy new battery and test voltage record value and temp of surrounding air at time of test. make sure battery settles for half an hour.
2) in the iss of concern, remove ALL sensor connections
3) install battery and test voltage and record. If sunlight is low or poor. check every hour till dark and record voltage.
4) Do this for the next week. Yes your console and internet data will not like it, but there is only one to find problem if there is a bad sensor, cable, that attaches to electronics module.
5) after a week, if you have lost 10% or more battery, then the issues are more than likely lack of sunlight, poor setting for the gathering of sunlight, or a bad solar cell or internal circuit leakage.
5A) remove the solar panel if you have a greater than 10% drain in battery and record voltage. run another week and check voltage. If voltage is stable, replace panel, still dropping, the super cap sounds like it is leaking. I see you replaced it. If you did not heat sink the leads and bend leads very carefully with needle nose pliers when installing, there is a good chance you damaged the cap.
6)So after step 4 you did not find a loss of voltage, then its time to find out which sensor it is. 7) Plug in anemometer for a week. check voltage, any loss greater than 10% yes, this could be a bad sensor. if not disconnect and plug in the tipping bucket.
7) keep this up till you find the sensor that is drawing all the current from the battery.
now plug in all the sensors that did not have a large effect on the battery, check battery voltage and run another week. if minimal lose of voltage is measured, then you found your sensor.

Even though we live in an electronic world, all devices use current to function. If ohms law or any calculation you would use to calculated current draw for an electronic sensor, transistor, ic, etc, this is the only way to truly know what the calculated use would be. Another check is the environment. Do you live in high humidity area. This can cause leakage and with dirty air or even salty ocean air to create a resistive circuit to form and now you have a load which will use current and drain battery.

just my 2 bits. take it or leave it. hope this helps.

[dalecoy]

I have been reading this post for quite awhile ..........

That would have been a good procedure.  However,

1.  elagache has (essentially) today replaced the entire Davis apparatus with "new stuff", some of it in different locations, and with different wiring.

2.  And there is no (current) evidence that the new configuration has any problem with "drawing too much current".

In other words - if the new configuration does not have the problem, then no troubleshooting procedure will find the problem.