Author Topic: Atlas issue  (Read 1307 times)

0 Members and 1 Guest are viewing this topic.

Offline DoctorKnow

  • Forecaster
  • *****
  • Posts: 1982
Atlas issue
« on: November 14, 2019, 08:20:49 AM »
My Atlas dropped out last night when it got dark, and did not report all night. Now that the sun has come up, it started reporting again. Is this a battery issue or something else? The batteries are not dead... This is the first time I have had this issue with this particular Atlas, but not the first time I have seen this. I have another Atlas that did this no matter how good the batteries were, and it happened every day.

Offline nincehelser

  • Forecaster
  • *****
  • Posts: 3337
Re: Atlas issue
« Reply #1 on: November 14, 2019, 08:36:16 AM »
If the batteries are near a year old, I'd suspect them.

Some others have noticed a night/day issue.  The solar cells are running through the main board (unlike the 5n1) so they might be having some effect.

So far, replacing the batteries seems to clear things up.

Offline John Z

  • Senior Member
  • **
  • Posts: 89
Re: Atlas issue
« Reply #2 on: November 14, 2019, 09:31:33 AM »
Doctor Know,

My Atlas sun-cycled as yours did when the batteries got weak. The battery low message was never sent, though. When testing the pulled batteries I found that one in each series string was thoroughly depleted, while the other remained strong. I'm very puzzled by that. They were all new when they went in, same batch.

My unit went down again at 1am the night the Arctic blast came through. It has not come up since. No solar cycling this time. I suspect it's batteries again, and again, no warning message. It will be a few days before I can get to it.

Offline DoctorKnow

  • Forecaster
  • *****
  • Posts: 1982
Re: Atlas issue
« Reply #3 on: November 14, 2019, 09:40:14 AM »
I wonder if those caps under the big solar panel are failing when it gets cold. It just recently got cold here, with yesterdays high in the low 40s.

Offline DoctorKnow

  • Forecaster
  • *****
  • Posts: 1982
Re: Atlas issue
« Reply #4 on: November 14, 2019, 10:12:38 AM »
Doctor Know,

 When testing the pulled batteries I found that one in each series string was thoroughly depleted, while the other remained strong. I'm very puzzled by that. They were all new when they went in, same batch.

Seems to indicate two banks not making contact at a connection somewhere. It could be the wiring or it could be the board? I have not seen this on my Atlas, and I have checked the batteries several times.

I will try to look at my Atlas later, but I am not sure if I can find some batteries for it... I am not even sure the issue is the batteries.

Offline John Z

  • Senior Member
  • **
  • Posts: 89
Re: Atlas issue
« Reply #5 on: November 14, 2019, 10:28:10 AM »
The batteries that showed imbalanced depletion were Energizer Lithiums.
I have ordinary alkalines in Atlas right now. It will be interesting to see if they ran down in a more balanced way.

I have seen this imbalance issue when using Lithiums in series strings in other products: a portable shortwave radio, some motorized window shades. The motorized shades were especially maddening. They held 8 cells, but would fail if one was depleted. After too many runs up the ladder, those shades are gone.

Update, after some reading. Seems that cells have capacity variation due to manufacturing tolerances, and other factors. In series strings, strong cells will beat down weak ones, driving them to early failure. So, no mystery it seems.
« Last Edit: November 15, 2019, 06:38:08 AM by John Z »

Offline DoctorKnow

  • Forecaster
  • *****
  • Posts: 1982
Re: Atlas issue
« Reply #6 on: November 15, 2019, 02:49:47 PM »
I changed batteries. This seems to have corrected my issue. The batteries were not run down very much at all... I am wondering if the solar panels could actually drain the batteries?

Offline John Z

  • Senior Member
  • **
  • Posts: 89
Re: Atlas issue
« Reply #7 on: November 15, 2019, 03:41:20 PM »
Replacing the batteries got my Atlas on line again, too. It's troubling to not get a batt low warning. The pulled alkaline cells all tested out at 1.3 volts, and are able to send 3.6 mA to the load in my Harbor Freight meter. Four mA is typical of a fresh cell. I would consider these cells to be still good. No sign of imbalanced discharge this time.

Thinking about this, maybe the cold weather caused a logic lockup that only a power reset could clear. That might explain why the unit was not functioning, though the batteries seem good, and no batt low alert.
« Last Edit: November 15, 2019, 04:16:06 PM by John Z »

Offline radioman61

  • Senior Member
  • **
  • Posts: 93
Re: Atlas issue
« Reply #8 on: November 17, 2019, 08:48:13 PM »
I had the same thing happen last month. First the temperature and humidity went off line, followed by the wind measurements. The next morning everything came back up.  I replaced the batteries and it’s been solid ever since. I too had year old Energizer lithium batteries and never received a low battery alert.

Offline John Z

  • Senior Member
  • **
  • Posts: 89
Re: Atlas issue
« Reply #9 on: November 18, 2019, 09:00:42 AM »
DoctorKnow,

Earlier you had asked if solar panels could run down the batteries. Your batteries did not appear to be run down, but you asked a good technical question. Solar panels exhibit a reverse leakage current in darkness that is proportional to their area. Bigger array, more leakage. The solution is to put a diode in line to the batteries. The diode will effectively block the reverse flow. Now the leakage is determined by the reverse current of the diode, which is much, much smaller than that of the panels.

I haven't traced the circuitry in Atlas, but there appears to be at least one diode in about the right place on the main board to do this job.