[werk_ag]

An interesting study, but I think the consumption is overstated.

150mA consuption for a small RS fan is excessive. The one I'm using, and it can be smaller, consumes only 80mA (just measure).

The SHT11 consumption is only 3.2uW (microWatt) making measurements every second, and we just do measurements every 59 seconds or 19 seconds in the case of a SHT21.

As soon as weather permits, at night, I will go up to the roof and make real measurements of current draw from the battery during an hour period. One hour with a fan on, another hour with fan off.

A few questions, so I can further improve the needs estimation:

- How often are readings taken then? Every 3 seconds?
- Can you please share the settings you are using (temperature and wind speed set points) to turn the fan on? This would allow me to simulate behavior by using one year of past meteo data.
- I do not seem to see power being supplied to the Wind Cups and RainMeter. Am I correct to assume these will produce an electrical impulse on their own using motion?

-Instruments are read at very diferent intervals, and sent at diferent time intervals. On average, we have a data packet sent every 3 seconds. Wind data packets are sent every 5 seconds.

- Each one may use diferent settings, my RS fan is on just when outside temp is greater than 30ºC AND wind is below 4Km/h.

- Wind and Rain instruments gets power directly from the Arduino pin where they are connected. It's consumption is negligible, around 0.25mA in the brief time that the read switch closes.

We can also lower the overall consumption by using 3mm LEDs that can fully light with just 4mA. In this case we must raise the value of the correspondent limiting current resistor.

[Brett]

Can I also suggest that if you're looking at increasing efficiency, that you err away from the "PWM" variety of solar controllers, and move to the MPPT version? The PWM version basically throws away any voltage above the required charge voltage by using PWM, whereas the MPPT will use a buck-boost approach to keep both the charge voltage at the desired level and the maximum draw current from the panel that it can sustain.

A few years ago I did an uncontrolled test, I had that *exact* PWM controller and a 10w panel, charging an SLA battery. I had a micro hooked up to it, logging the time and battery voltage, and having it burn some power (I had it flashing LEDs). I would monitor the time that the voltage dropped to the point that the solar controller switched it off. The tests were done on days where there was ample sun available - summer in Australia - and I ran it over about 3 weeks to get trends. I then ran the same tests with a similar MPPT controller, over similar length of time, and it was astounding the difference - instead of hitting the low-voltage cutoff at say 3am, it was making it to ~6am which in some cases was all night as the sunrise happened and kicked off recharge.

The controllers I have opted for now are similarly priced (I think). I've been buying the 5A model - on ebay the item number is 261392830593, http://www.ebay.com.au/itm/261392830593?...1439.l2649

[ And as an aside when comparing this stuff to the FO approach - they are extremely frugal with transmits and receives. At startup or when the RX and TX units sync, they get into a known TX/RX pattern. The receiver turns on when it expects to hear from the TX, and then the TX transmits and shuts down, and the RX shuts down even if it doesn't get a signal. By keeping a low transmit rate, and this closely coupled cycle, they keep the power usage extremely low. It would also seem that they have purpose-selected a processor that allows them to run interrupt counting frugally and in low power state. All of these things are possible to replicate but more complex, and usually beefing up the power capacity is an easier thing ]

---

Oh and another question for Werk_AG, while you're on the roof, can you measure the dimensions of that box ? It looks awesome, and I'm struggling to find one that I think matches - particularly that's UV safe as here in AU that is a real concern. I am almost expecting to pay 10x the price for that than the solar controller !

[werk_ag]

Hi Brett

Thanks for the hint about those MPPT controllers. I didn't know that.
Despite I never had any problem with my system and it has never cut the power to my TX unit, I ordered one of those units just after read your post.
I'm planning to make my second TX unit, be solar powered too

Oh and another question for Werk_AG, while you're on the roof, can you measure the dimensions of that box ? It looks awesome, and I'm struggling to find one that I think matches - particularly that's UV safe as here in AU that is a real concern. I am almost expecting to pay 10x the price for that than the solar controller !

I will do that!
That box cost me around 38€, can't remember exactly.

[uncle_bob]

That box cost me around 38€, can't remember exactly.

I used to work in electrical wholesale and I can tell you those types of enclosure were never cheap (and still ain't by the looks of ebay etc).

[Brett]

I used to work in electrical wholesale and I can tell you those types of enclosure were never cheap (and still ain't by the looks of ebay etc).

Tell me if you find a good + cheap source ok ? Surely with "contacts" from the previous life you'll have a channel?

[uncle_bob]

Tell me if you find a good + cheap source ok ? Surely with "contacts" from the previous life you'll have a channel?

I've been out of the loop for awhile (15 years) and the freight from the East Island may be the killer

[tyntop]

Hi,
I have another query with regards, to TX power supply, on the TX pcb the power supply connection has the following pins marked :

+Bat
12v
GND

As far as I am aware these connections also have something to do with showing the battery voltage in the software, So would I be correct when wiring this to the battery that you would have :

GND to negative/black terminal on battery
+Bat/12v to positive/red terminal on battery

I am hoping that I have understood this properly If not please let me know.

regards

tyntop

[hvalentim]

150mA consuption for a small RS fan is excessive. The one I'm using, and it can be smaller, consumes only 80mA (just measure).

Agreed. I am planning to use a 8x8cm model as this will more or less fit the circular hole I drilled in the radiation shield to house the FineOffset thing. Just found one rated at 66MA.

The SHT11 consumption is only 3.2uW (microWatt) making measurements every second, and we just do measurements every 59 seconds or 19 seconds in the case of a SHT21.

The SHT21 also seems to draw much less power.

As soon as weather permits, at night, I will go up to the roof and make real measurements of current draw from the battery during an hour period. One hour with a fan on, another hour with fan off.

That would be awesome. And a more reliable way to estimate power consumption.

On average, we have a data packet sent every 3 seconds. Wind data packets are sent every 5 seconds.

So all the arithmetic (wind speed average and so on) is done by the RX unit?

Each one may use diferent settings, my RS fan is on just when outside temp is greater than 30ºC AND wind is below 4Km/h.

That should be rare and happen in Summer only as I expected. Actually in most cases that means the power will be supplied by the Solar Cell directly.

Overall I am pretty happy with my RS, not sure I actually need the fan that much. But will apply it, just because...

Thanks for the remaining info. On the light of it I think we can drop the bar to a little under 2Ah per day. Perhaps even 1.5Ah per day.

---

Can I also suggest that if you're looking at increasing efficiency, that you err away from the "PWM" variety of solar controllers, and move to the MPPT version?

Yep. A good MPPT regulator can increase the production of an array as much as 30%. I was unaware of such an affordable model, though.

comparing this stuff to the FO approach

I think the sole remote virtue the FineOffset approach might have is annoying and aggravating users to the point they swear they will build their next weather station themselves and actually understand what the hell is going on...

[uncle_bob]

Hi,
I have another query with regards, to TX power supply, on the TX pcb the power supply connection has the following pins marked :

+Bat
12v
GND

As far as I am aware these connections also have something to do with showing the battery voltage in the software, So would I be correct when wiring this to the battery that you would have :

GND to negative/black terminal on battery
+Bat/12v to positive/red terminal on battery

I am hoping that I have understood this properly If not please let me know.

regards

tyntop

Here's a snippet from an upcoming magazine article on this matter

When powered by a solar system, the TX unit has the ability to allow remote monitoring of the battery voltage. Just connect the +Bat terminal to the +12v battery terminal, and the 12V terminal to the output of your solar charger controller. If the TX unit is powered by a power pack supply, just connect +Bat and 12V terminal together.

[werk_ag]

So all the arithmetic (wind speed average and so on) is done by the RX unit?

RX unit only does some unit conversion to meet Davis protocol (mph) and to display data in the user selected unit (Km/h, mph or m/s).
Each Wind data packet sent by the TX, includes Wind Gust and Wind Speed which is a rolling average of the last 2.5 minutes. Values are sent in m/s (more precisely m/s * 1000, to be sent as an integer value).

... Actually in most cases that means the power will be supplied by the Solar Cell directly.

Yes, I think that during daylight, the 10W solar panel is enough to feed it and still charge the battery. The TX unit only draws power from the battery at night or on very dark days. I say this because even in days without bright sun, battery voltage stays almost always, between 12.8V to 13.8V