Help on renewable energy ckt (buck converter)

[gudxors]

Hello,

I am currently working on solar systems and DC-DC converters. Since this is a competely new topic for me, i have lots of questions and need some guidance. I have spent lots of time searching online, but couldnt get much information. Please, any help will appriciated.

First, i am designing a buck DC-DC converter, which takes inconsistant input voltage from the solar panels. The voltages generated by panels are varying from 200V ~ 300V, which is input voltage for the DC-DC converter, and i want to have 110V DC at the output so later i can inverter that to 110V AC by using full bridge inverter. I started with 10V buck converter using two IRFR320 mosfets and it worked well. However, when i increase my input voltage to over 200V, it is not really working well. i am using these values L=80mH, Cout=2.5uF, and duty cycle = 66% with 6us period. Any recommendations on component values or mosfet selection?

Second, since my input voltage varies, my output value varies as well. I want to have 110VDC at output even though the input voltage changes. What should i do for this?
By the way, i am planning on using a DSP based microcontroller to control the gates of mosfets. I figured that Texas Instruments have TMS320C2000 series for solar applications.

Thank you so much in advance.

Tyler

 

[WimRFP]

Hello,

For making 110Vac (sine wave), you need a bus voltage (for your H-bridge inverter) that is above 156Vdc. Mostly, a higher bus voltage is used so you don't get close to zero or 100% duty cycle.

In my opinion, I would drop the DC-DC converter and use the inconsistent 200..300V as supply voltage for a full bridge circuit that runs at 20 kHz or above and make a sine wave inverter.

 

[gudxors]

Hello,

For making 110Vac (sine wave), you need a bus voltage (for your H-bridge inverter) that is above 156Vdc. Mostly, a higher bus voltage is used so you don't get close to zero or 100% duty cycle.

In my opinion, I would drop the DC-DC converter and use the inconsistent 200..300V as supply voltage for a full bridge circuit that runs at 20 kHz or above and make a sine wave inverter.

alright, but i need some voltage stablizer or something, because the voltage generated by PV modules are very choppy and noisy. However, if anybody have different ideas or opinions, they will help too.

 

[WimRFP]

Hello,

The voltage fluctuation of the solar panel will not be that fast. You can stabilize the output of the inverter by modulating the amplitude of the 50/60Hz signal that you feed to your PWM modulator. In the analog domain you need a multiplier for that. Of course you can do that digitally (together with the feed back loop and/or combined with feed forward).

At higher input voltage, the actual swing in duty cycle will be less in that case, when the bus voltage drops to almost 156V, the duty cycle has to vary from 0 to 100% to make the 110Vacrms voltage.

So if you want a sine wave inverter, I think you can skip the DC-DC converter as you mentioned that your input voltage is > 200V.

Of course when you want to make just a 50/60Hz square wave inverter, you need the DC-DC converter

 

[gudxors]

Hello,

The voltage fluctuation of the solar panel will not be that fast. You can stabilize the output of the inverter by modulating the amplitude of the 50/60Hz signal that you feed to your PWM modulator. In the analog domain you need a multiplier for that. Of course you can do that digitally (together with the feed back loop and/or combined with feed forward).

At higher input voltage, the actual swing in duty cycle will be less in that case, when the bus voltage drops to almost 156V, the duty cycle has to vary from 0 to 100% to make the 110Vacrms voltage.

So if you want a sine wave inverter, I think you can skip the DC-DC converter as you mentioned that your input voltage is > 200V.


Of course when you want to make just a 50/60Hz square wave inverter, you need the DC-DC converter

I guess your idea sounds like a good plan. However, since i have an energy storage(Lead Acid Battery), i need some kind of charger controller to prevent the overcharging and stuff. For that reason, i need DC-DC converter or charge controller.

also, what do you mean by "You can stabilize the output of the inverter by modulating the amplitude of the 50/60Hz signal that you feed to your PWM modulator"? i didn't quite get what you are saying there.

 

[WimRFP]

Hello,

Fully clear, in case of feeding batteries you need a DC-DC converter to operate the panels at maximum power output and to avoid wrong charging. I assume that your inverter runs from the stack of Lead-Acid batteries. I think you have to design your DC-DC converter yourself based on IGBT/MOSFET with a high-side driver IC. You have lots of choice for you voltage range.

Regarding the duty cycle of the full bridge circuit. If you have a sine wave that modulates the pulse width, your sine wave output amplitude will be 1*Vbus when the duty cycle swings from 0 to 100% (that is: 1-0 = 1*Vbus).

When your duty cycle swings from 20 to 80% your sine wave will have a amplitude of 0.6*Vbus. in case of a swing from 25 to 75%, it will be 0.75-0.25 = 0.5*Vbus. So by varying the strength of the PWM, you can control the output voltage of your full bridge, hence you can handle variations in Vbus.

You will need this control scheme with a battery back up also as (battery cell voltage during charging)/(battery cell voltage when almost depleted) ratio is large.

Not related with your question directly: both the 0 and live of your inverter output are floating and have a galvanic connection to the DC circuit. If you don't want this you have to built a more complex isolated DC-DC converter.

 

[gudxors]

Hello,

Fully clear, in case of feeding batteries you need a DC-DC converter to operate the panels at maximum power output and to avoid wrong charging. I assume that your inverter runs from the stack of Lead-Acid batteries. I think you have to design your DC-DC converter yourself based on IGBT/MOSFET with a high-side driver IC. You have lots of choice for you voltage range.

Regarding the duty cycle of the full bridge circuit. If you have a sine wave that modulates the pulse width, your sine wave output amplitude will be 1*Vbus when the duty cycle swings from 0 to 100% (that is: 1-0 = 1*Vbus).

When your duty cycle swings from 20 to 80% your sine wave will have a amplitude of 0.6*Vbus. in case of a swing from 25 to 75%, it will be 0.75-0.25 = 0.5*Vbus. So by varying the strength of the PWM, you can control the output voltage of your full bridge, hence you can handle variations in Vbus.

You will need this control scheme with a battery back up also as (battery cell voltage during charging)/(battery cell voltage when almost depleted) ratio is large.

Not related with your question directly: both the 0 and live of your inverter output are floating and have a galvanic connection to the DC circuit. If you don't want this you have to built a more complex isolated DC-DC converter.

I see, thanks for clearing this up. i think i still have a few more questions if thats ok with you.

1. i decided to go with MOSFETs for switches, but it is very hard to find high voltage mosfets. I was able to find a few from International Rectifiers (IRF series), but they are limited to only 300V. For my case, it could be even higher than 300V. So i was curious if you know any companies that make very high voltage mosfets for this application.

2. i just designed my boost converter to test the spice model of IRFR320(upto 300V). I started with 14V input to see if it is working properly, then i was going to increase it about 150V later, however, with 14V input, it is not boosting it up. I am only getting about 11V max at the output. I have been playing around with the duty cycle, but it never goes above 11V. Can you take a look at my .sp file and give me some advice on it?

Thank you so much for your help.



p.s. i could not update either .sp or .txt files due to its limitation, so i am just pasting it to here.
****************************************************

.OPTIONS LIST NODE POST
.options METHOD=GEAR
.Tran 300n 0.3m

******************************

V1 Vin 0 14
V2 Vpulse 0 pulse(0 17 0 1n 1n 4u 6u )
V3 Vpulse2 0 pulse(0 17 4u 1n 1n 2u 6u )

*******************PWM POWER STAGE*************************************
X1 D1 Vpulse 0 irfru320
X2 D1 Vpulse2 Vout irfru320

L1 Vin D1 8.0E-6
Rf1 Vout 0 1k
C Vout 0 50E-6

.print i(L1)

 

[WimRFP]

Hello,

Maybe you can post a circuit diagram in .png format (screen shot), to make it easier for us.

Regarding the mosfets, Check Fairchild, Onsemi, Vishay, ST, etc, sure you will find high voltage mosfets,

When your switching frequency is not that high, you may consider IGBT's The newer ones have reasonable switching speed and low on-state loss. I used IGBTs (HGTG30N60A4) recently in a resonant converter with somewhat above 300V supply voltage at >100 kHz, really good efficiency. So I am sure that for the hard-switched case it is worth to consider IGBT in case of about 20 kHz switching frequency.

Your driver has to provide more voltage with respect to mosfets. Think of about 15V. When selecting your full bridge driver you have to check whether it can supply that or not.

 

[gudxors]

Hello,

Maybe you can post a circuit diagram in .png format (screen shot), to make it easier for us.

Regarding the mosfets, Check Fairchild, Onsemi, Vishay, ST, etc, sure you will find high voltage mosfets,

When your switching frequency is not that high, you may consider IGBT's The newer ones have reasonable switching speed and low on-state loss. I used IGBTs (HGTG30N60A4) recently in a resonant converter with somewhat above 300V supply voltage at >100 kHz, really good efficiency. So I am sure that for the hard-switched case it is worth to consider IGBT in case of about 20 kHz switching frequency.

Your driver has to provide more voltage with respect to mosfets. Think of about 15V. When selecting your full bridge driver you have to check whether it can supply that or not.

thanks a lot. I will test out both MOSFET and IGBT and see which one works better. Regarding o the driver voltage i am using 17V as shown in the code. isnt that high enough? what do you mean by think of about 15V?

Regarding to the pv output fluctuation, do you think MPPT should be good enough for reducing the fluctuation? any other methods or ckts using these days for fluctuation?

Many thanks to you!!