re; any help mppt

Thread Starter

olusola

Joined May 21, 2007
74
please can you throw more light on this?

The mosfet has a Vgs MAX value of 5V. I hesitate to recommend powering the DAC with 5 volts as I think that might cause problems if the output of the DAC opamp can't swing close to +5V rail.
 

hgmjr

Joined Jan 28, 2005
9,027
hgmjr said:
The mosfet has a Vgs MAX value of 5V. I hesitate to recommend powering the DAC with 5 volts as I think that might cause problems if the output of the DAC opamp can't swing close to +5V rail.
please can you throw more light on this?
This parameter Vgs is specified in the manufacturer's data sheet for the FDA79N15 mosfet.

Vgs is the voltage difference that must be developed between the mosfet's gate terminal and the mosfet's source terminal to fully turn on the mosfet so that its Rds(on) resistance is at its minimum. That means that when the active load represented by your mosfet needs to be at 10 ohms, the voltage measure between the mosfet's source terminal and its gate terminal could approach 5 volts.

In an N-channel enhancement-type mosfet the gate voltage is more positive than the mosfet's source terminal voltage to turn the mosfet on.

Remember, I mentioned earlier that you will need to ground the negative side of the solarcell and the bottom of the 1.25 ohm resistor in order for the ATMEGA16 to measure the voltages correctly.

Does that help you any?

hgmjr
 

Thread Starter

olusola

Joined May 21, 2007
74
yes it helps thanks.
and i'll ground the negative side.

how do you suggest i get a voltage difference of 5V to turn on the mosfet?
does this mean that to get more resistance i would need more voltage at the gate?

oh also so if i use a 10v reference for my DAC, will that help?

thanks
 

hgmjr

Joined Jan 28, 2005
9,027
yes it helps thanks.
and i'll ground the negative side.

how do you suggest i get a voltage difference of 5V to turn on the mosfet?

That is where the DAC comes into the picture. It is the output of the DAC opamp that will connect to the mosfet gate and provide the control that you need to adjust the mosfet's on resistance to suite your active load requirement.

does this mean that to get more resistance i would need more voltage at the gate?
Yes.

oh also so if i use a 10v reference for my DAC, will that help?
If you have a 10 volt source I see no harm in that. If you power the DAC with 10 volts you might as well use +-10 volts on the opamps also.

hgmjr
 

Thread Starter

olusola

Joined May 21, 2007
74
wow, cool then . ok , do i need the 1k resistor just before the mosfet?
so am i guessing the rest will now be software?

please what do you have to say about my connections of the DAC:
i have the LSB to MSB connected to the microcontroller , using a +15supply and now a 10V reference, then the Rfb (pin16) is connected to the 2nd op-amp (did you say i shluld put it back to the 1st op-amp?), then the out1 (pin1) to the inverting input (-) of the op-amp and out2(pin2) to the non-inverting (+) input of the op-amp? and also (not shown in diagram) a 0.1uC capacitor between Rfb (pin16) and the inverting input?

also for the microcontroller?

thanks

olu
 

hgmjr

Joined Jan 28, 2005
9,027
wow, cool then . ok , do i need the 1k resistor just before the mosfet?
You can use a 1K or a 10K in series with the gate. It should not be too critical. The gate signal is not high frequency so pretty much anything between 1K and 10K should be good.

so am i guessing the rest will now be software?
We are almost to that point. The schematic needs to be cleaned up and solidified.

please what do you have to say about my connections of the DAC:
i have the LSB to MSB connected to the microcontroller , using a +15supply and now a 10V reference, then the Rfb (pin16) is connected to the 2nd op-amp (did you say i shluld put it back to the 1st op-amp?), then the out1 (pin1) to the inverting input (-) of the op-amp and out2(pin2) to the non-inverting (+) input of the op-amp? and also (not shown in diagram) a 0.1uC capacitor between Rfb (pin16) and the inverting input?
I would ditch the second opamp and just connect up the DAC as shown in the manufacturer's datasheet for the unipolar configuration as a starting point. The output of the DAC opamp will then be routed to the gate of the mosfet through a 1K or 10K resistor.

also for the microcontroller?
I don't have any problems at the moment in the way you connected up the microcontroller to the DAC or the measurement points in the circuit.

That may change but for now leave it as it is.

hgmjr
 

JoeJester

Joined Apr 26, 2005
4,390
With all the changes, I would recommend to olusola to post a new diagram of the circuit to ensure everyone is on the "same page".
 

Thread Starter

olusola

Joined May 21, 2007
74
hi,
well in the flow chart, though i was not sure how to put it, i was going to have there to be a first reading of voltage and current to the the first power P0 ;
then a second reading of voltage and current is taken again to get the second power P1 ;
then rest of the flow chart.
after that, current P1 becomes P0 and then another reading is taken . and so on ..

thanks

olu
 

hgmjr

Joined Jan 28, 2005
9,027
I guess I was expecting that you would read the voltage and current and then calculate the near realtime power and then compare the measured value to a reference power (10W) and then adjust the active load up or down as needed
until the difference between the measured power and the 10W reference was close to zero.

hgmjr
 

Thread Starter

olusola

Joined May 21, 2007
74
oh well , em, what can i say ,
because i want the system to be flexible, but really because of light intensities you might not always have the power at 10W, and so one will really have to let it do the adjusting by itself. - it is called the hill climbing algorithm for mppt.

the major problem with it , is fast changing light intensities - for example where a cloud goes over the pv cell rapidly, the system could detect a small power and then suddenly a large power.

do you think this system can handle this?

sola
 

hgmjr

Joined Jan 28, 2005
9,027
The software should be able to do pretty much whatever the hardware will allow it to do. I believe the hardware is up to the task.

hgmjr
 

hgmjr

Joined Jan 28, 2005
9,027
Once you have had a chance to update the schematic to reflect some of the changes I have suggested then I would post it as joejester has suggested?

I would take the time to add reference designators to the components as it will make it a lot easier to refer to the various components when discussing the circuitry.

hgmjr
 
Top