My aim is to build a simple solar powered pump with a rechargeable battery to water plants.
The idea is to use a 6V 1W Solar Panel connected to a TP4056 (protected) to charge a 18650 Lithium Ion Battery. On the output is an MT3608 2A Boost Converter to step up the voltage to a 3V/6V DC Pump.
I would like to know if the design work and any correction/ improvements on different components or adding in a timing function for the motor.
I have attached a diagram for more information. I am a total beginner and would really appreciate any help or advice.
This may seem like a simple project, but it is not.
The boost converter may not work at all, because the startup current for motors is 5 to 10 times the running current, and there should be no need for it anyway. A 3/6V pump will work fine on 3.7V supplied by a Li battery.
The charger won't work properly if the motor is running, because it is designed to charge only batteries that are disconnected from any loads. Furthermore, if the battery is discharged too deeply, you will ruin the battery and/or the charger won't even start.
For beginners working with solar power, lead acid or NiMH batteries are the way to go. Both can withstand a great deal of abuse. Li-based batteries cannot, are easily destroyed and in the worst cases, start fires.
You need to do the maths to figure out how many amp-hrs the pump motor will need over (say) 7 days and how many amp-hrs the solar panel can produce over the same period. If you have cloudy days where you live you must take that into account.
I suspect you should have a solar panel that can produce about twice or three times what the pump motor needs (averaged over a week) to allow for all the inefficiencies in the system.
And I agree 100% with the suggestion to use a lead-acid or NiMh battery. If you are using a lead-acid battery make sure it gets a very full charge at least once per week. I don't have the same experience of NiMh batteries but the same care can't do any harm.
jremington:
This may seem like a simple project, but it is not.
The boost converter may not work at all, because the startup current for motors is 5 to 10 times the running current, and there should be no need for it anyway. A 3/6V pump will work fine on 3.7V supplied by a LiPo.
The charger won't work properly if the motor is running, because it is designed to charge only batteries that are disconnected from any loads. Furthermore, if the battery is discharged too deeply, you will ruin the battery and/or the charger won't even start.
For beginners working with solar power, lead acid or NiMH batteries are the way to go. Both can withstand a great deal of abuse. Li-based batteries cannot, are easily destroyed and in the worst cases, start fires.
That's a really good point regarding the boost converter. Since the power from the solar panel can fluctuate, I wanted a component to stabilise the output. After reviewing the diagram I realised that it was not serving its purpose!
This model of the TP4056 is the updated version which I believe has over discharge protection - this should prevent discharge damage to the battery. However, I agree that a lead acid battery would work better for this function. I think the TP4056 is only compatible with Lithium Ion and is designed for the 18650 battery.
I am confused why the TP4056 will not work properly while charging the battery as there is the option to connect a load on the board.
Do you have any suggestions on how I could adapt this idea to work?
Robin2:
You need to do the maths to figure out how many amp-hrs the pump motor will need over (say) 7 days and how many amp-hrs the solar panel can produce over the same period. If you have cloudy days where you live you must take that into account.
I suspect you should have a solar panel that can produce about twice or three times what the pump motor needs (averaged over a week) to allow for all the inefficiencies in the system.
And I agree 100% with the suggestion to use a lead-acid or NiMh battery. If you are using a lead-acid battery make sure it gets a very full charge at least once per week. I don't have the same experience of NiMh batteries but the same care can't do any harm.
...R
The reason I chose the 6V 1W solar panel is because its small (110x60). I want to create a design which is compact and portable. I'm not aiming to have the pump running 24/7 so it would be great to incorporate a timer. For reference I have been looking at the company "Irrigatia" who manufacture solar powered pumps which achieve the same function using small solar panels. Is there a simple way to add a timer/ relay module into the circuit to conserve battery life?
jremington:
The TP4056 is not designed for load sharing. The charge cycle assumes that the battery is unloaded, and does not work correctly if a load is attached.
What about an ESP that spends most of the time in deep sleep? Is that too much load?
electricjack:
The reason I chose the 6V 1W solar panel is because its small (110x60). I want to create a design which is compact and portable. I'm not aiming to have the pump running 24/7
I always assumed the pump would only run intermittently.
You cannot choose a solar panel based on size only. You must calculate the energy requirement and then choose a suitable solar panel.
jremington:
The TP4056 is not designed for load sharing. The charge cycle assumes that the battery is unloaded, and does not work correctly if a load is attached.
You are right, I never realised this was an issue! I've done some research and there are very limited charging components which also offer load sharing. However, I was able to find a charger developed by Adafruit with load sharing but it is too pricey! Do you know about any different chargers with load sharing?
SteveMann:
What about an ESP that spends most of the time in deep sleep? Is that too much load?
I had researched the ESP8266 to use the deep sleep cycle for basic automation on the pump but I was also concerned about the load. I found a simple diagram online showing a basic TP4056 connected to an ESP2866 and powered by a lithium ion battery.
Robin2:
I always assumed the pump would only run intermittently.
You cannot choose a solar panel based on size only. You must calculate the energy requirement and then choose a suitable solar panel.
...R
I agree, it would be unrealistic to try and power the entire device using a small solar panel. I definitely need to do the math and work out what size I will need to power it long term. I know it is possible using a small solar panel due to the design of the "Irrigatia" solar pump system.
On their website they have a dedicated page to parts for spares or repair. It's interesting since the system only has one circuit board which ticks all the boxes:
I believe it is custom made to suit the application and is unable to be sourced else where.
Where's the water coming from? If it's mains water, you won't need a pump because of the mains pressure, so you only need to charge a small battery to power the controller.
If it's collected rainwater, is it possible to collect it in a reservoir at a higher level? Again, the pressure provided by the high reservoir would mean no pump required and you are harvesting energy naturally.
If you can't collect it higher, can you store a modest amount higher? You could use solar panel to pump water up to the high reservoir, storing the energy that way instead of in a battery.
SteveMann:
What about an ESP that spends most of the time in deep sleep? Is that too much load?
I have an ESP32 sitting out in the back yard, using solar, and a TP4056. The TP4056 is connected to a battery for night operations. The ESP32 deep sleeps for 6 minutes, powers up, powers up the sensors, does the thing and goes back to sleep. The ESP32 uses WiFI. Charging, typically, starts around 0630 and lasts till 1100ish.
I use a DROK switching regulator to bring the solar cell voltage down to 5V. The 5V is sent to the TP4056. The TP4056 output feeds a 3.3V low drop out regulator that feeds the ESP32 and the sensor. During charging the output of the TP4056 is shared with powering the ESP and charging the battery. I use a 10 watt solar cell and a 26000mAh battery.
Basically, works for me, since Feb., till it doesn't.
Idahowalker:
I have an ESP32 sitting out in the back yard, using solar, and a TP4056. The TP4056 is connected to a battery for night operations. The ESP32 deep sleeps for 6 minutes, powers up, powers up the sensors, does the thing and goes back to sleep. The ESP32 uses WiFI. Charging, typically, starts around 0630 and lasts till 1100ish.
I use a DROK switching regulator to bring the solar cell voltage down to 5V. The 5V is sent to the TP4056. The TP4056 output feeds a 3.3V low drop out regulator that feeds the ESP32 and the sensor. During charging the output of the TP4056 is shared with powering the ESP and charging the battery. I use a 10 watt solar cell and a 26000mAh battery.
Basically, works for me, since Feb., till it doesn't.
I would like to see a schematic of your setup. (And, hopefully the OP?)
SteveMann:
I would like to see a schematic of your setup. (And, hopefully the OP?)
Idahowalker:
I have an ESP32 sitting out in the back yard, using solar, and a TP4056. The TP4056 is connected to a battery for night operations. The ESP32 deep sleeps for 6 minutes, powers up, powers up the sensors, does the thing and goes back to sleep. The ESP32 uses WiFI. Charging, typically, starts around 0630 and lasts till 1100ish.
I use a DROK switching regulator to bring the solar cell voltage down to 5V. The 5V is sent to the TP4056. The TP4056 output feeds a 3.3V low drop out regulator that feeds the ESP32 and the sensor. During charging the output of the TP4056 is shared with powering the ESP and charging the battery. I use a 10 watt solar cell and a 26000mAh battery.
Basically, works for me, since Feb., till it doesn't.
SteveMann:
I would like to see a schematic of your setup. (And, hopefully the OP?)
Me too! This would be a great help with my project as its a similar device. If you have any photos of the setup then it would be greatly appreciated.
PaulRB:
Where's the water coming from? If it's mains water, you won't need a pump because of the mains pressure, so you only need to charge a small battery to power the controller.
If it's collected rainwater, is it possible to collect it in a reservoir at a higher level? Again, the pressure provided by the high reservoir would mean no pump required and you are harvesting energy naturally.
If you can't collect it higher, can you store a modest amount higher? You could use solar panel to pump water up to the high reservoir, storing the energy that way instead of in a battery.
Just some thoughts.
Really interesting ideas, thank you for sharing them. I had intended for the water to be pumped from a small reservoir beneath the plant. What I am aiming for is similar to a basic hydroponics system except it's solar powered.
A gravity fed system would be brilliant since its natural energy. However, the water would still need to be returned back to the elevated tank for the system to be automated. I think your idea about storing the energy in an elevated tank instead of in the battery is brilliant.
