Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

How to determine Dry-run current in Solar powered AC Pump

Status
Not open for further replies.

abicash

Member
Hello

I am in the final stage of my Solar AC pump project.

I am implementing a dry run current shutdown, but unsure how to do this in a solar powered (VFD) setup.

The motor current is varying based on applied frequency based on solar power.

I implemented the simple current threshold algorithm and wait to see if it stays there.Once it is confirmed, i shut down pump.
Seems this may not work since in late evenings or a cloud cover, the healthy current is anyway going to be low (due to lower solar power) .In some cases , it will be lower than the set threshold.

How can this be achieved?
Can someone throw some light??

Thanks and regards
 
Not exactly sure what you're trying to achieve.

Once a low voltage (or power) set point threshold is hit, why would you want to "wait" to shut down the pump? Is there no supplimental battery power to cover those periods when solar power is momentarily interrupted?

Could you provide a block diagram (as detailed as possible) of your system?
 
Hello cowboybob

Thanks for replying.
I apologize for replying late since i was involved in tuning my controller these days.Now getting back to this problem.

I gathered a lot of information on the web regarding Dry running motor, and found that dry running has to be confirmed for a few seconds before taking any action.
If i were to have a fixed AC voltage at the motor terminals, and a fixed frequency, this task would've been mundane.

The implementation i have is of a solar powered MPPT AC VFD

The circuit determines a Max power point of the Solar Panels with a defined interval. Then a 3-Ph inverter converts this DC volts to 3-Ph Ac volts. The frequency is also varied as per the AC volts as per ratio 400vac/50Hz
This is a basic V/F speed control of an AC Induction motor implemented, so that a constant ratio is maintained to make the Torque constant over solar insolation throughout the day.

Now when i run the pump dry, i dont have confirmed data as to current parameter.
I read that current reduces, but here i observe that current is increasing.

I am really confused. Can you help?
Thanks and regards
 
Is this something like your system block diagram?:
Solar Pump 3P Inverter.JPG

Are there contol set points that can be adjusted (since you said you've been "tuning" the system, I assume there are)?

If so, what are they?

I'm not sure why the current is increasing (free running the motor exceeds some rev limit?) but I'm thinking that monitoring the solar panel array's VDC level might provide an anomaly that can be exploited.
 
Last edited:
Hello

It's similar to what you have shown.
I don't have the intermediate converter.Panel goes direct to 3 phase bridge.
I measure solar voltage, current and motor current.
When motor stops demanding current, solar voltage increases and my algorithm reduces the duty of the inverter and vice versa.which reduces motor ac voltage.
I run the entire show on a Microcontroller.
My question is
How is a sensorless dry run current measured in a solar vfd?
is this a right question?
 
Just noticed your other post on this issue ("Help: Solar pump").

I would say that Northguy has way more knowledge and experience on this subject than I do.

And I might add that, as a former precision measurement equipment specialist, there is a point where "fine tuning" follows the Law of Diminishing Returns: you may achieve your goal but at what cost in time and frustration?
 
As to your most recent post:

Presumably, as the load (pump loses prime) suddenly drops, the VfD will spike momentarily.

Although you say the the pump current increases. Does it stay high? That might also affect the VfD in a manner that can be detected.
 
I will check it again and get back

EDIT:

The current increases from 8A to 10A and stays there.
Input power reduces.
 
Last edited:
Hi Abi,
I followed your post with North Guy and found the subject very informative.
You did ask about the pump/pumping.

Can I ask what type of pump are you using;
Is it a centrifugal pump or a positive displacement pump.
What is the head and volume capacity of the pump. Do you have type number for the pump maybe?
what arrangement do you have for priming the pump or do you have a foot valve.
I presume the pump motor is a 3 phase 50 Hz machine.
On our farm we have several pumps the biggest being a two stage centrifugal driven by a 6HP DC machine. Then there are many single phase small centrifugal pumps from 0.3 to 1.5 kW and we also use some helical rotor pumps for wine transfer operations. So we have a fair bit of experience with pump selection and pump/pipeline system design. I'm sure we can give you some advice.
 
Hello rumpfy
Thanks for replying.

My Pump is a Submersible Pump.
Specs are
Positive displacement pump
325 feet head (25 stages, 4meter/stage)
Voltage 440V/50Hz
capacity :5HP
 
Abi,
The angle from which I was looking was the bit about the pump speed changing as your solar radiation changed. Your post says "positive displacement", but I am certain it is centrifugal. Otherwise there is no need for 25 stages.
Clearly, the pump is a centrifugal pump and most likely the motor speed is 2800 RPM; I have a bit of data for Grundfoss, Lowara and Ritz pumps.
With these things, the head pressure is a function of Speed squared, the flow is a function of speed, and the horse power is a function of speed cubed.
There is a Grundfoss type SP8A, and for this pump range, the 'no flow' head is 500 feet and the rated delivery is 8 cubic metre per hour. At 8 cubic metre per hour this pump has a head pressure of 330 feet. The motor on this pump is 3.7 kW (5HP). This sounds like your pump duty but what is not clear is the actual head/pipeline conditions.
There was lots of stuff in your conversation with North Guy; I was not clear as to whether you both were aware that with changing solar radiation the load on the solar converter was changing rapidly.
With a positive displacement pump (helical rotor), you will have a lower delivery rate as you slow the speed. The head will be developed regardless of the speed. There is a slight reduction of delivery due to increasing head pressure due to internal leakage in the pump. However, because your pump is, I am sure, a centrifugal pump, then the operating characteristics vary all over the place due to the different running speed. It's not clear what the actual load (head/pipeline characteristic) is like, but for your system, it seems likely that at some operating speed, you will have insufficient pressure to get any flow at all. This all goes back to what NG was saying that at low solar radiation, you might just as well switch the pump off.
I'm happy to do more evaluation for you, but it would be necessary to get more info on the pump outlet configuration; depth to water(or total static head, delivery pipe diameter and length, actual pump model. It looks like you have the pump submerged in the water supply, so there is no problem with maintaining pump prime. The real question is what is the delivery rate (of water) as the pump speed slows. From your other post you say you maintain a fixed V/F for the pump power supply, so it shouldn't be too hard to work out a frequency at which you can stop the pump.
hope this helps.
 
I am so glad you replied..since i am not moving ahead..:(

I am based in India. The pump is locally manufactured and the data i shared was received from the manufacturer (rather , the seller)
Currently i do not have an intricate pipeline installed above the pump.I pull water from a tank and throw it back in again.To create pressure, i have a valve which stops the water-flow to show a reading on a gauge (optimum pressure was 12kg/cm2)
I too am not an expert in pumps/motors. so please bear with me.

The actual utilisation will be in a submerged tube-well below ground (300ft).
To maintain a constant v/f ratio, i calculate motor voltage as Vsol/root2 x duty
and based on this parameter i define a frequency 0- 50Hz.
Does this all make any sense, or something wrong?

EDIT : A little search shows that you are right.The pump is a centrifugal pump:facepalm:
When i pull the pump out of water (with a chain-pulley arrangement) to make it seem like a dry run situation, what happens is ..
Since there is no load on the drive, input solar current reduces, but o/p ac voltage is increased
(you said you read about the algorithm i am following with Northguy) and the motor starts demanding more current.
I thought that actually the current should have reduced, but it did noto_O
 
Last edited:
OK.
So you are in the 'development phase'. This is fine and NG has given some very good stuff on tuning the power supply.
Now you have to fix up the load. This is the hydraulic bit attached to the electrical bit.
I imagine that in service, the pump will be submerged in the water table. If this is the case, then you will NOT have the situation where the pump will run dry. If the level of ground water drops then I guess the pump may run dry.
It is important that the pump NOT run dry because there will be the seals in the pump which will NOT be lubricated if this happens. However what will happen, unless you design for it, is that the pump will run, BUT, there will be no flow of water. In this condition, the pump will agitate the water in the pump and cause it to heat. Of course, it should be obvious that if the pump is running, but the pressure developed is not enough to lift the water to the surface, then there is no point in running the pump.
So, you might need to include a flow switch as part of the electrical design. If there is NO FLOW, then the power supply is shut down.
We still haven't got to the place where we can develop a hydraulic design.
I understand what you have done with the flow limiting valve, but I would like you to give some thought to the hydraulic part.
You will have a static head above the pump, and you will have a dynamic head due to the water flow; the dynamic head will be due to the cross sectional area of the casing (or pipe) PLUS the effect of the pipeline system above the ground. The static head is the distance below ground level. Have you got your bore yet?
Get on to your mates at the pump shop and make absolutely certain that they get the Head/Flow characteristic of the pump for you. When you get it, post it. I'm guessing here but I reckon your pump is a copy of something. Find out!. Dont let them fob you off. The head/flow characteristic is critical but there is other stuff too. Often the motor size is selected to guarantee that if the pump loses its hydraulic load, the motor wont overheat. In your case this is not relevant because the static load wont get disconnected so you motor size can probably be reduced. If you need a bit more from me on this, then ask. BUT, get the pump data before you get back to me.
I think we are getting there but for now, its back to you.
hope this helps.
 
Apology for late reply.

I appreciate your patience with me :)
I am checking these parameters and will get back shortly
 
Hello rumpfy

I managed to receive some data from the seller.

6" (150mm) Radial Flow Submersible Pumpsets
Max Pumpset Dia : 142mm , Outlet size : 2"
AC Supply : 3Ph , 415V, 50Hz, 2880RPM
Pump Model : B63 : 3.7KW/5HP , 25 Stage
Discharge in LPM at 100 meters : 90LPM
 
Last edited:
Abi,
attached is a simple example of where we have to go. To study your system, you need to get enough info to get a good picture of the operating characteristic of what you want to build.
For the pump we must have the maximum pump pressure. This will always be at the 'no flow' point. That is, the pump pressure when the outlet is blocked. Your data shows only one operating point (100 metre head at 90 LPM flow). Then we need to have an idea of the maximum flow. This is when the pump is operating with no restrictions on the outlet. In the attachment I have shown 3 'typical;' pump characteristics. These are the dotted lines at the LHS of the picture. They have been 'massaged' from some real pump data. The CR2 and 4 are Grunfoss multistage types; the Onga is an Aussie single stage pump. You can see the steepness of the multistage pump curves compared to the Onga pump shape. You can see how the CR2 would pump more water than the CR4 because one has 110 metre head and the other has an 80 metre head.
Because you want to run your pump over a speed range, we need the head/ flow characteristic curve, to be able to modify the typical characteristics at the 2880 RPM to other speeds.
Next on the graph is the pipeline (or load) characteristic. It shows how the load interacts with the pump. In this case, the load is a pipeline and the pump is located some 55 meter below the pipe outlet. If you extend the pipeline flow back to the vertical axis, you will find the pump pressure required to be developed BEFORE you get any flow.
In your trial to date you say you used a restriction on the pump to replicate the working pressure of the pump. However in that case your load is not exactly doing what you want. Your trial load will allow flow at almost no pump output pressure, whereas your system will require some significant pump speed to get water to flow. Your trial load will be a sloping line like the pipe line characteristic but it will extend down to the horizontal axis.
The depth to the water is needed. It is assumed that the bore hole will have a diameter at least that of the pump. BUT will there be an outlet pipe attached to the pump, and if so will this be a 2 inch pipe.
When the water gets to the surface, what piping is attached. Does the water get put into an above ground tank or dam or a water channel, or is is used to irrigate directly.
I looked on google and found a story about the Indian Government wanting to change 26 million diesel pumpsets for solar. I guess this is part of that program.
 

Attachments

  • pipe-pump_example.jpg
    pipe-pump_example.jpg
    42.9 KB · Views: 237
Last edited:
Hello rumpfy

Thank you for such a comprehensive and enlightening post.Most things are clear.

I think the setup would need to be installed on an actual site now.

You mention about the Govt here wanting to change 26 million diesel pumpsets to solar.This is not directly linked to this program but although it might be in some way.There is abundance of solar energy and it only remains to be harnessed to help farmers (irrigation) in a way to counteract unavailability of power in most areas.

Coming back to the problem mentioned in the heading, I see in most solar pumps here just measure Motor current and install a pot to align the current with a dry run point, which till date i have been unable to determine!
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top