PWM Buck booster for under 13V?

SolarPowered · August 2013

Hello All,

I've been working on a 440uF and 660uF 16Volt (micro farad) buck booster for PWM during low voltage/amp during early PVwatts for panels that are perfect south angle of azimuth fixed not tracked.

The proto type system I have right now uses (2) Sun Solar USA 60watt panels plugged in parrallel, 1 panel STC (VMP 18.2) (IMP 3.3), PWM controller used for this mock up system is a sunforce 12volt/10amp digital PWM controller, which uses a chipset and mosfets, the mosfet's actually allow the controller to peak surge at 12amps, however this is not recommended. As we know it is not true MPPT as the range this PWM works in charge mode is between 13.5V (+/-0.4) and peak at 14.4V (+/-0.4). So there is a lot from the voltage spectrum not being used through out the course of the day.

During early morning hours until after 9:30am to 11:30am 90*F panel VMP is measured on average @ 12.5 volts highest VMP during hot summer days is about 14.9volts, and the voltage gets better during the spring and fall months at about 15.2VMP.

I've been considering engineering a small buck booster. To keep it simple its just a circuit board that runs on the inline load side between panel and PWM controller. The circuit board uses a 18Volt diode on the (in) side of load of the circuit board (this is to prevent reverse polarity during evening which would cause polarization to the panels via capacitor).

The math used for the capacitor was simple. Its so I don't blow the PWM controller.

Peak 3.1 IMP recorded per panel during my Physical Test Conditions (regardless of STC), in parallel I will just say that it is roughly 6 AMPS PTC recognized, the controller has the potential to surge at 12amps.

A 440uF 16V capacitor in 1 second emits a 7.04 coulombs at 16V with a energy capacity (joules) 56.32watts. The charge and discharge is Low Resistance (+/- 40%).

The 660uF 16V capacitor seems as if it may be over kill.

in 1 second it would discharge 10.56 Coulombs at 16V with a energy capacity (joules) 84.84watts, (which would mean I would need smaller panels to offset peak watts). The charge and discharge of this capacitor is Low Resistance (+/-20%).

So in theory would it work to just use the 440uF. I'm just looking for a little added performance without going all out affording or building a MPPT controller.

Kindest Regards

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

So I ran the test on the capacitor today.
After running the numbers on leakage and with Low Resistance losses I will be implementing a few more capacitors with different capacities. I will also be installing a 550uF, and a 330uF. Based on the charge rates and times of each capacitor I don't believe that I will need to add any additional switching but I will figure that out as the tests continue.
These tests were ran at absolutely no load, tilt was at 0 degress from 180* azimuth, weather at 72*F, full sun, @6:58pm voltage from panel was 18.59V amperage from panel recorded at .26AMPs
Charge accepted by capacitor 18.12Volts amperage from capacitor recorded @ .25amps

Under no load (60watt panels) produced 4.883 watts
Under no load the capacitor charged 4.53watts.
Charge loss is @ approximately 4.7065% will round up and call it 5%.

Then I performed the leakage test.
Test performed per minute showed the capacitor rapidly leaks voltage roughly 1.28 volts per minute, roughly 2.0175 amp per every 1 minute. (This is why it is bad to install Car Stereo FARAD caps to off grid 12V pure sin wave inverters! It will drain the battery when not in use unless you install a cut switch between battery and stereo cap!)

These pics are the test results from my sample bread board fully disconnected 15 minutes after charge.

Attachment not found.Attachment not found.

mike95490 · August 2013

Re: PWM Buck booster for under 13V?

Just to throw a wrench into the mix - what's the ESR of the caps, at the frequency you are pulsing at? That ESR, along with the inductor's reactance, can have a big effect on the circuit.

A 440uF 16V capacitor in 1 second emits a 7.04 coulombs at 16V with a energy capacity (joules) 56.32watts.

Huh ? You can't be running a 1 Hz buck/boost, so I think something is not really right.

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

mike95490 wrote: »

Just to throw a wrench into the mix - what's the ESR of the caps, at the frequency you are pulsing at? That ESR, along with the inductor's reactance, can have a big effect on the circuit.

Huh ? You can't be running a 1 Hz buck/boost, so I think something is not really right.

Hello Mike, its not a wrench in the mix, its a work in progress anylizing each part of each circuit to see what the end result will have to be. Trial and error from a newbie in tech.

Maybe I am fooling myself and its really 7 seconds, not 1 second according to 1 amp, (or 1 amp equally to 1 coulomb), and I have my math backwards? In most cases though wouldn't a 20-50hz copper wrapped magnetic coil at the output end of the circuit board correct the problem from inhibiting foiling and frequency to the PWM? Just another part I have to buy, but like I said trial and error, and something is telling me through this due process I will also need a few 10volt 220uF capacitors as well, which are not becoming easy to find. An RC is in the works but to what impedance and watt at this time I do not know, need more time to work it out.

This capacitor emits a really high ESR, that of which I don't know, as I do not have a capanalyzer to detect exactly. The spec on the capacitor indicates 16V 440uF (+/- 40%).

Good news is the leakage on the capacitor stops at 15.01 volts, .01 amps which means all I need is amps to fill the gap. Which this was just a leakage, not a full drain test.

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

O.K so I believe I am on the right path now.
Worked out a few erroneous calculations on the math, added circuits to RC, and this is the quick drawing mock up. Any suggestions or feedback would be good, as i am new to boosters.
I found a good deal on a Toroidial Inductor which is much better on EMI and less on losses. How ever at that point I might as well start looking into chip sets and going mppt. I have a pic of the Toroidial that I am looking at. Good price for under $30.
Attachment not found.Attachment not found.

NorthGuy · August 2013

Re: PWM Buck booster for under 13V?

I don't quite understand what happens between 9:30 and 11:30 (should be a lot of light unless you're close to South Pole).

Does the panel voltage drops below battery voltage and the controller disconnects?

or

Controller goes 100% duty cycle, but the batteries are so discharged/loaded that they drag the voltage down to 12.5V?

or

Something else?

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

That is where the majority or most of the I-V curve shifting power is occurring, it happens again roughly between 6 pm until dark. Early morning to evening times voltage runs great until there is no sun at all, amps are low in morning and evening. This is because it is a "fixed tilt" not a tracking system. There is no tracking , no tracking of pv watts, and no means of buffering or "boosting".

I live in the valley cold nights-hot days. Where I live there is up to a 15*~ 30* F differential in temperature within a 24 hour time period, that significantly impacts voltage.A few weeks ago during the late june and through mid july tempratures were exceeding 100*F as early as 10am. When I ran my leakage test for the capacitor after 6 pm at 0*tilt the panel showed more voltage than STC (under load), yet amperage was under .26 way below STC.

My battery for 125aH recommends a 12AMP charge, regardless panels produce a 6 amp charge even when the controller is fully functioning. So regardless of the charge controller and its function to operate within the voltage spectrum the amps during early morning and later evening are at a significant loss that doesn't allow for a substantial "float charge". The PWM activates at within 10mA however that doesn't mean that the battery due to internal resistance will accept that charge or accept it well. According to what I've shown in my tests using the PWM and taking a recording or measure of sun hours "PV watts", the potential state of charging and efficiency is at 45%. Even if I get only 15 or 30 minutes off the booster, I would hope for would increase efficiency to 55% from 45%, and I feel that to be pretty conservative and modest figures, and still no where near the range of MPPT which has a much better efficiency of 95~97%. But if you can make a booster with $15 in parts and get 10% increase from an investment that initially cost a grand, I would say the expectation is modest and fair.

I'm just looking to make a booster that captures volts/amps prior to that I-V change, get an additional boost in current hopefully before leakage from the capacitors becomes to significant.

I've also considered adding concentrators to the 60watt panels, and adding additional silicon sealant to the back of the panels for water proofing and installing a mist miser to cool the back of the panels which I've seen tests that it can drop temperature by 10~15*F at the panel.

Attachment not found.

NorthGuy · August 2013

Re: PWM Buck booster for under 13V?

I don't understand your answer.

Your 18 Vmp panels should not have any problems producing voltage enough to charge 12V batteries.

You say they don't. If so, something wrong with either panels or the controller. Do you know what it is?

Can you measure voltage on PV side and on battery side at the same time?

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

On hot days of over 100*F, yes voltage is under 12V. 2 weeks ago I measured a Vmp of 13.2 on a 95*F day at 2pm.
In early morning to later evening amps is the problem.

So for example if I can take a total hours lets say 5 hours of full sun through out the course of the day my panels would yield only 2.4 hours of "efficient" charging to the battery.

When I have time, some time this week I will draw a graph to explain the I-V curve scenario.

NorthGuy · August 2013

Re: PWM Buck booster for under 13V?

SolarPowered wrote: »

2 weeks ago I measured a Vmp of 13.2 on a 95*F day at 2pm.

PWM controller doesn't track the MPPT point. Vmp is the voltage at the MPPT pont. Unless you have an MPPT controller capable of moving the voltage to the MPPT point, you cannot measure Vmp.

I guess the voltages you're measuring are on the PV side. You need to measure the voltage on the battery side too at the same time.

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

Bingo! Measuring from the PV side under load.
Then when applied at the battery terminal side it's less.

I've been improving the system as I progress adding shunts and metering before and after PWM.

I'm looking more, to stay consistent with a 13v~14.5v point of operation during scorching hot weather. Which I believe can be stored from early morning "cooler"operation when it's not as hot, then when sunlight reaches a point of yielding enough amps then the discharge from the Resistor Circuit would happen.

I could explain better showing the I-V graph where the potential "boost" occurs.

NorthGuy · August 2013

Re: PWM Buck booster for under 13V?

SolarPowered wrote: »

Bingo! Measuring from the PV side under load.
Then when applied at the battery terminal side it's less.

How much less? Can you post both numbers?

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

Don't know how that will help you.
13.1 PV from my multi meter
12.8 Battery from the PWM digital display

Float mode is off as the PWM controller states that it will float charge at 13.5V (+/-0.4)
PWM digital accuracy is within (+/-0.4)

PWM is limited between 13.5~14.4V, give or take (+/-.4). Sure if panels make more than 13.15VMP GREAT! After 14.4, will call it 14.8 for accuracy and purpose of load, all that over voltage is lost and not being converted, and anything less than the float is a waste of time and energy. Therefore accuracy for its potential to displace power and charge battery on a hot day is moot.
I NEED 13.5V FOR CHARGE

NorthGuy · August 2013

Re: PWM Buck booster for under 13V?

SolarPowered wrote: »

Don't know how that will help you.
13.1 PV from my multi meter
12.8 Battery from the PWM digital display

That tells me that the PWM controller is doing 100% duty cycle. 0.3V drop is because of the switch (MOSFET or alike).

In simple words, at the moment when you did the measurements, the controller connected the battery directly to the panels without switching. All the current that the panels are producing is flowing into the batteries. This current (minus what is taken by loads, if any) rises battery voltage to 12.8V. There's absolutely no way in the world you can rise the voltage higher without increasing the current. If you built a boost controller that increased voltage to 100V, the battery would still drag it down to 12.8V.

Even if you could increase voltage without increasing current, this would not do you any good. State of the charge of the batteries depends on how much current flows in and does not depend on the voltage that you used to push the current in.

The MPPT controller would keep panel voltage at Vmp, which is 18V and 3A current, and convert it to battery voltage and higher current, say 4.5A. This would rise the battery voltage and charge it faster.

Using more panels would increase current too.

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

NorthGuy wrote: »

The MPPT controller would keep panel voltage at Vmp, which is 18V and 3A current, and convert it to battery voltage and higher current, say 4.5A. This would rise the battery voltage and charge it faster.

Using more panels would increase current too.

Here's my problem.

I can't increase the number of panels as this is a solar generator.
Generator weighs 148LB.s, anything more would make it that much more difficult for transport. The panels are fold out for fixed tilt, and rack mounted to the box via unistrut and rod assembly, it can fit easily into any half ton truck or SUV, Panels are already 17.36 conversion efficient AR coated.

I've been running loads of tests prior for perfecting the unit for extreme wear and tear conditions.

So I am left with only the option of
-Fold out concentrators at 25* angle which will eventually hinder voltage due to iradiant heat intensity per meter square.
-Or have a modified I-V curve booster

Only 2 options I have.

I-V curve boosting seems reasonable since its in almost every AC 110 volt surge protection equipment, AC and DC radio's, and amplifiers, not to mention even MPPT's use proportionate I-V curve boosting via either a 25v or 30V capacitor in conjunction with torodial inductors, that is in part to what makes the tracking possible with microprocessor, not to mention loads of mosfets.

It was suppose to be the bang for your buck under $1250.00 generator and MPPT was not in the budget for this build. Its engineered to be efficient for as low as the costs can get.

The lowest cost MPPT I could find that was almost within reason of my budget was the MPP Solar MPPT made in Tawain which a distributor was going to offer me in bulk at $110 per MPPT for a quantity of 25. That doesn't meet my credit expectations as of yet.

NorthGuy · August 2013

Re: PWM Buck booster for under 13V?

$1250 seems like a lot for a generator based on 100W panel (wgich with PWM controller translates to 70W). Also 150 pounds is too heavy.

If you can solder, you can build your own controller. All you need is a capacitor followed by a buck SMPS (not boost because battery voltage is lower than Vmp). You can put it at fixed 16V without microcontroller. If you can also prohram a microcontroller, you can track MPPT. Lot of work, but will be under $100 for your size.

People here posted limks to MPPT-on-a-chip which tracks MPPT (made by TI, can't find the link), so you can use one of them instead of microcontroller. They're very small and hard to solder though.

SolarPowered · August 2013

Re: PWM Buck booster for under 13V?

NorthGuy wrote: »

$1250 seems like a lot for a generator based on 100W panel (wgich with PWM controller translates to 70W). Also 150 pounds is too heavy.

Correction they are (2) 60 watt panels (120watt). I could of bought 2 pallets at .89 per watt, but instead didn't have enough info on the panels. So I started with 2 @ $1.20 per watt. will buy more later after I conclude the testing on them. They are pretty durable. Here is the spec sheet that came included with them.

Got a great deal on a SunForce 1000watt Pure Sine Wave inverter for $179.00 n clearance. The VMAX tanks 125Ah cost $260.00. The bulk of the cost came with all the 4AWG conductor, inline fusing/ameters/shunts/usb port/depth of discharge gauges, and the most expensive believe it or not was the unistrut racking which cost more per watt than all the components which was 25% of the total cost.

Most of the Bulk weight of the generator comes from the battery which is 78.9LB.s

Fry's electronics here in the bay area sells the MPPT chip set it cost $22, the scratch board I need is another $14, and miscellaneous such as mosfets/capacitor/transformer is another $45 in parts.

Either way my combination comes with panels and is 30% lower in cost than Goal Zero's yetti after you buy their panels sold seperately, and outperforms the yetti by 20%.

I also cheated a bit with a 12 AMP AC/DC 110V battery wall charger so when I'm done with the unit for camping I load it in the garage and let it charge. I'm also a general contractor so I use it at jobsites when I have absolutely no means of power, or I have a service panel upgrade and have to supply my clients with power.

This whole concept started because when we are camping and boating some friends need a jump for their boat, my wife enjoys pina coladas so I needed something to power a blender, and the kids are little tribal maniacs unless I can pop on a movie with the bluray player LCD. Prior to the generator I just hooked up some panels to the SUV with a cheap 7amp controller (2) 20watt panels, and a modified sin wave inverter that was just horrible for everything to power our small camp site. when it came to camping or for work I was tired of paying for fuel just the time and energy plus what gasoline costs was killing me financially.

Attachment not found.

PWM Buck booster for under 13V?

Comments

Categories