LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

peak detection width problem (PV panel testing)

Hi,

 

I working on a project 'An electronic load for testing PV panel'. Here I am used to draw I-V characteristics of a panel. I am able to read  voltage and current samples, here comes the problem as the samples will contain ripple more like alternating DC. So decided to track the peak values of the ripple DC waveform.

 

I used peak detection fucntion palette and included the parameters, It is giving me the error "the width must meet 0<width<samples. I have attched my Vi could anyone help me out.

0 Kudos
Message 1 of 10
(3,015 Views)

Hello,

 

Your VI includes the data acquisition part, so it does not work for others.    Please save the waveform data as the default value which you pass to the peak detection VI.  

 

Thanks,

Michael

0 Kudos
Message 2 of 10
(2,985 Views)

HI,

 

thanks for your reply.

 

I sorted the problem, for me I just need to find the peak values of the waveform, so I used default value in width section.

 

Here I am stcuk with sample rate and samples to read in DAQmx. I am totally wrong where I am using 200 rate and 300 samples. But this specification giving me the I-V but thats too weird ones. Please find the attached front panel with I-V curve.

 

So decided to take the 20 samples and make an average. But this is not working for me.

 

Any suggestions

Download All
0 Kudos
Message 3 of 10
(2,970 Views)

It looks like your problem is more with the data than with the software. How did you get the data which you plotted in SolarFront.jpg?  I am not talking about the software but the solar panel, loads, measurement system, irradation, ...

 

Lynn

0 Kudos
Message 4 of 10
(2,960 Views)

Yes, you are true. The problem is more with data. Here I am using low voltage supply (30V,10A) instead of solar panel. All the way I am driving the cirucit using Electronic load (MOSFET) from 0-5V Vg to have I-V curve from 5V open circuit to 6A short circuit. For the measurement of current using LEM kit4 current transducer. Presently not considering the calculation of irradiance.

0 Kudos
Message 5 of 10
(2,947 Views)

Is your power supply going into current limiting where some of the transients in your data appear?   A typical power supply does not act much like a solar panel, so id does not surprise me that you do not get the kind of curve you expect.

 

What settings do you use on the power supply while taking that data?

 

Lynn

0 Kudos
Message 6 of 10
(2,941 Views)

It has 2 current limiting modes with 8A and 10 A. Here I used 8A. The DC power supply I have used is having signal more like full wave rectifier, not a pure DC. As solar panel do not give pure DC, that is why I am trying with this suply.

 

while taking the data I set the voltage to 5V (~5.4). As you see in front panel it has limited to 0.7 V from 5V. As far as I have tested today,the MOSFET has a voltage drop of 0.2 Vand I think remaining 0.5V is getting drop over conductor as I used long cables for the circuit with a resistance of 0.14. 

 

I am not sure that I could succeed in this task. As you said using "A typical power supply does not act much like a solar panel". Please find the attached setup.

0 Kudos
Message 7 of 10
(2,930 Views)

From trying to read between the lines to guess what you are doing, I suspect that you are connecting your Analog Output voltage directly to the gate of  the MOSFET.  If that is the case, then you likely go from very little current in your load to a lot of current with a very small change in gate voltage.  The image below shows the transfer characteristic of a typical 10 A, 60 V transistor such as you might be using.  All of the current range of your power supply is covered with a gate voltage change from about 3.5 to 4.5 V.  The position of the transfer characterstic curve will change as the transistor heats up.

 

STP20NF06.png

 

Once the power supply goes into current limiting, it is difficlut to predict how the output will behave.

 

My suggestion: Configure the electronic load to operate in a current regulated mode so that you can specify the current drawn by the load with your AO voltage, prefereably scaled to 1 A/V.

 

Then ramp the AO voltage (and the load current) up slowly.

 

Lynn

Message 8 of 10
(2,921 Views)

Hi,

 

If I am not wrong I figured the problem, here I am taking average of the samples for one run in while loop as it was not pure DC as you said.

 

And then without doing any manual operation of gate,I tried to run the VI from 0 to 5 (Vg). Mostly with 2.4 to 3.6 in a current regualated mode

 

Please find the attached result I got. If I am doing wrong give me suggestions and one thing is I am getting errro 200279 from the DAQmx read. The values are getting over written but not reading the current data values with given 1Ks/S and 200 samples to read. But it is giving me the data I required. Will it there be any problem in future like I am doing with dummy supply not considering solar panel.

 

Thanks for your previous suggestions. It got me think to apply the necessary procedure for testing.

 

 

Download All
0 Kudos
Message 9 of 10
(2,896 Views)

That data looks much more reasonable.

 

You have a 0.5 second delay in the loop but need to read 5 times per second to get all the data.  The Reads will wait until the data is available so you may not need the delay.  I cannot tell how much time the DAQ Assistants take.  You may want to learn how to use a Producer/Consumer Architecture to separate the data acquistion from the the analysis and display.

 

Lynn

0 Kudos
Message 10 of 10
(2,886 Views)