|
Post by lbendlin on Jan 28, 2019 4:33:21 GMT -8
You would need to make the Peltier part of the box's wall somehow (with radiators attached to each side?), and then reverse the current flow in summer so the inside is cooled rather than warmed. The Peltier I am playing with is tiny (maybe a square inch) - not sure if they come in more usable sizes.
|
|
|
Post by SDL on Jan 29, 2019 8:55:46 GMT -8
Lutz,
Exactly. I plan to play with these in the near future. I bought two of them them the you suggested. I'll use a QPBM board to do the switching.
BP
|
|
|
Post by lbendlin on Feb 2, 2019 9:08:20 GMT -8
What would happen if I would put a 5.1V Zener diode across VDDM and GND on the SunControl board? My load is pulling 500mW so a single 1N4733A should do. My hope is that this would reign in the runaway output voltage on the Suncontrol, and also help to stabilize the I2C bus voltage.
|
|
|
Post by SDL on Feb 3, 2019 13:58:07 GMT -8
If I remember right, you are seeing what voltage on the output of the SunControl board? It should not be higher than 5.25V. Adding the diode would fix the problem you are describing.
BP
|
|
|
Post by lbendlin on Feb 4, 2019 4:51:36 GMT -8
I've seen as high as 5.9V, but that might also have been caused by a mistake I made (adding thermistor without removing the 10K SMD). I'll see how the board behaves now that I remedied that.
|
|
|
Post by lbendlin on Feb 9, 2019 10:14:59 GMT -8
Adding the diode didn't really fix the issue. It definitely increased the consumption a lot, most likely because of the voltage mismatch (5.2V output circuit fighting with 5.1V Zener ?)
The board still shows runaway output voltage, even with the thermistor situation corrected and with only one solar panel attached.
{"measurements":{
"battery":{"load":3.92,"current":-970.80}
,"solarcell":{"load":5.97,"current":-1161.60}
,"output":{"load":5.39,"current":112.40}
,"battery_raw":{"load":3.74,"current":158.40}
,"servo":{"load":4.17,"current":0.00}
},"timestamp":"2019-02-09 13:05:14","temperature":37.8,"humidity":37.5,"pressure":1021.2,"switches":"00101"}
Looks like the issue might manifest itself when the current coming from the solarcell(s) exceeds 1A. Here's a dataset with a little more shade on the cell(s).
{"measurements":{
"battery":{"load":3.90,"current":-810.00}
,"solarcell":{"load":5.04,"current":-1000.40}
,"output":{"load":5.21,"current":110.80}
,"battery_raw":{"load":3.75,"current":157.60}
,"servo":{"load":4.17,"current":0.00}
},"timestamp":"2019-02-09 13:10:15","temperature":37.2,"humidity":43.2,"pressure":1021.3,"switches":"00111"}
|
|
|
Post by SDL on Feb 10, 2019 10:39:19 GMT -8
The zener will dramatically increase the current on the output. You can put a current limiting resistor inline, but that will change the voltage. I'm charging a battery right now to do the test on the board you returned and another board.
John
|
|
|
Post by lbendlin on Feb 28, 2019 7:07:46 GMT -8
Update: replacing the 6.2V Zener over the solar panels with a 5.6V Zener resolved that issue. Just make sure that the Zener has a useful heatsink attached. Beyond that it will require active solar panel management (ie selectively switching them on and off , for example via the QPM) to control the "Full sun/Full battery" situation.
|
|