20 minutes ago, Erastus said:

Iron core transformer has an efficiency of 90 - 95%. In other words it need 5 - 15% to step up or step down.  Thus if I use a panel of 8 amps then according to you the loss is between 5 & 10%

DC doesnt go through a transformer. It gets converted by using components like Plonkster described above. AC uses a Iron core transformer to Step up or down in basic circuits. There is electronic ways of doing voltage control on AC, but that is not related to this. 

The efficiency of DC-DC converters is in general much higher than the efficiency of  stepping AC down or up using an Iron Core transformer.  

Plonkster,

I see where I am missing the point. Let me try again.

There are two sets of efficiencies:

1. Max power harvested per array.
2. Kw harvested per day

If I understand you correctly you have 1800 W array therefore you should harvest 19800 Watts per day. Do you mind telling me what you are harvesting per day. Lets allow 2% for Mppt then you should harvest 19000 Watts per day.

Thus if you harvest 19Kw per day you are running @ 98%
Not sure what your figures are.

Regards

Edited September 18, 20187 yr by Erastus

57 minutes ago, Erastus said:

A Iron core transformer has an efficiency of 90 - 95%. In other words it need 5 - 15% to step up or step down.  Thus if I use a panel of 8 amps then according to you the loss is between 5 & 10%.

MPPTs don't use iron-core. They typically use ferrite cores, usually toroidal, and the topology is usually not a transformer, in other words there's just one winding. Look for Buck Converter on Wikipedia, it essentially has three components. It's a kind of "charge pump" utilising a magnetic core (made of ferrite in this instance) as temporary storage. It also runs at higher frequencies, which means the efficiency is better and you can use smaller magnetics.

It's a bit hard to explain, but lets take the simple example of a 36-cell panel that makes 22V open circuit, makes the maximum power at 18V, but someone connected it to a 12V battery. The marketing dept labelled that panel at the maximum it will make, so if it is a 20W panel, it's going to make about 1 amp (close enough, it makes the math easy) at 18V. Because a PV panel is a constant-current device, if you pull it down to 12V, it is going to make 12V at 1A or 12W.

But laws of physics maintain that energy cannot be lost, it has to go somewhere. What happened to the lost 8W? The panel got a tiny bit hotter, that is what happened. It turned into heat.

If only I had an 18V battery you say! Or if only I had a 12V solar panel! But now you have a new problem, because the battery will charge up to 14.5V... so how do I get a variable voltage solar panel? Easy, I use a DC/DC converter. The property of a DC/DC converter is that power-in = power-out (minus a bit lost through inefficiency). And that is all there is to that.

If I  use a DC/DC converter on my 20W example panel, it will take that 18V at 1A and convert it into 12V at 1.5A (assuming no losses). Note how 18 * 1 = 12 * 1.5. Pin=Pout, or more accurately, Pout=Pin*0.98... which is a great deal better than just making heat.

26 minutes ago, Erastus said:

19800 Watts per day

19800 Watt-hours, or 19.8KWh (killowatt hour). A killowatt hour is 3.6 million joules. A joule is the SI unit of energy.

3 hours ago, Erastus said:

nnecessary evil to have 
If you are running at 26V and you are chargin

Tell me the watts you are harvesting per day. As per the system you have. That would be very interesting

5 minutes ago, plonkster said:

MPPTs don't use iron-core. They typically use ferrite cores, usually toroidal, and the topology is usually not a transformer, in other words there's just one winding. Look for Buck Converter on Wikipedia, it essentially has three components. It's a kind of "charge pump" utilising a magnetic core (made of ferrite in this instance) as temporary storage. It also runs at higher frequencies, which means the efficiency is better and you can use smaller magnetics.

It's a bit hard to explain, but lets take the simple example of a 36-cell panel that makes 22V open circuit, makes the maximum power at 18V, but someone connected it to a 12V battery. The marketing dept labelled that panel at the maximum it will make, so if it is a 20W panel, it's going to make about 1 amp (close enough, it makes the math easy) at 18V. Because a PV panel is a constant-current device, if you pull it down to 12V, it is going to make 12V at 1A or 12W.

But laws of physics maintain that energy cannot be lost, it has to go somewhere. What happened to the lost 8W? The panel got a tiny bit hotter, that is what happened. It turned into heat.

If only I had an 18V battery you say! Or if only I had a 12V solar panel! But now you have a new problem, because the battery will charge up to 14.5V... so how do I get a variable voltage solar panel? Easy, I use a DC/DC converter. The property of a DC/DC converter is that power-in = power-out (minus a bit lost through inefficiency). And that is all there is to that.

If I  use a DC/DC converter on my 20W example panel, it will take that 18V at 1A and convert it into 12V at 1.5A (assuming no losses). Note how 18 * 1 = 12 * 1.5. Pin=Pout, or more accurately, Pout=Pin*0.98... which is a great deal better than just making heat.

Nope not saying this at all. I would l;ike to understand the amount of watts you are getting from the sun per day

8 minutes ago, Erastus said:

Tell me the watts you are harvesting per day

you can not harvest watts (power). You can only harvest watt hours (energy).

16 minutes ago, Erastus said:

Tell me the watts you are harvesting per day. As per the system you have. That would be very interesting

I have about 1.8KWp of panels on the roof. That means that at peak insolation, they make 1.8kw, which is only an hour or so per day and usually only in summer. I make up to 11Kwh a day in summer, in winter it depends on rain etc. Today I hardly made around 3kwh.

This is my site on pvoutput.org.

Power vs Torque vs energy:

1. Suppose I have a 20meter high cliff, with a pulley on the edge, a rope going across it, and a 100kg load at the bottom.

2. I can use a horse or an ox to do it. If I use a horse to pull the load up, I can do it twice as fast as with an ox.

3. The horse can't pull a 200kg load up. The ox can do it, but only 10 meters up before he's too tired.

From this I can say:

1. The horse has double the power of the ox.

2. The ox has more "torque".

3. In all three cases exactly the same amount of work was done, since the potential energy is mgh (mass times gravity times height) and it is identical in all three cases.

Hence knowing how many horses were involved tells me nothing about how much work was done :-)

O dear me.  What shall I then ask for as a novice. I am trying to understand the amount of electricity that one does not have to pay Eskom or the councils in a normal day with proper sunshine.
Let me try and put a formula together:

with a sample rate of 1 reading per milli second:

This is my source code to determine the current harvested per day @ 1mSec intervals:

        SolarCurrInstGF = SolarCurrTGF / 1000;

        SolarCurrInstGF = SolarCurrInstGF * SolarCurrMulGF;
        Temp1GD = (SolarCurrInstGF/360);  // 60 sec * 60 mint = 1 hour ( 3600) 10 000 samples per sec
        SolarCurrDayTotalGF += Temp1GD;
        SolarCurrTotalGF  += Temp1GD;
        SolarInstCurrRmsGD = sqrt( Temp1GD * Temp1GD );
        SolarCurrGF += SolarInstCurrRmsGD;
        SolarCurrRmsTotalGD    += SolarInstCurrRmsGD;

 

Many thanks Plonkster. Your best day was on the 911 with 10.6Kw. That day I harvested more than 16.9Kw with 1880 array.
That is the difference with tracking and no tracking.
Many thanks  for the information. SO to be on par you need another 900Watts at least 4 panels or R 4800 - the MPPT 1800 = R 3000 difference.
My tracking is doing very well tnx for sharing information.
I was worried but now I am very happy>

My calculations worked well but I was worried but not anymore.
Many Many thanks
Erastus

 

Edited September 18, 20187 yr by Erastus

12 minutes ago, Erastus said:

That day I harvested more than 16.9Kw with 1880 array.
That is the difference with tracking and no tracking.

I have to keep this in mind, I am currently busy with the biggest design I have ever attempted, and had tracking in the back of my mind. I should certainly look into it if this is the results. 

 

 

2 minutes ago, Erastus said:

That is the difference with tracking and no tracking.

No argument there. I'm saying that if you add an MPPT as well you will do even better.

My best day was actually last year when I made almost 12kwh on a particular day. But there is no point in splitting hairs. With a stationary array you can make 5kwh-6kwh for every 1kwp of PV panels you have. With a tracker it's over 7. So none of this is really that surprising.

Sure, it will likely cost another 10k (frames, PV panels, etc) to get to the same level of production... but no moving parts. Less potential for wind damage. I've seen the damage a bad wind does to a windmill. I've seen a 12m tower on a Namibian farm twisted up like liquorice, heads ripped clean out of the tower where the hold-down bracket was not installed. Not unassailable challenges, of course, but buck converters is more my expertise than steel structures :-)

Jaco it is worth it in my opinion. At least 40 - 55% more power. I had to do certain frames to fight our strong winds and then I thought lets go tracker. Looked at the costs and had a heart attack. The I designed my own. Now that I can compare with others I am not sorry.

It is my first design in steel and I am very happy with the results

sOLARcOMPLETE.pdf

bearing10.pdf

7 minutes ago, plonkster said:

No argument there. I'm saying that if you add an MPPT as well you will do even better.

My best day was actually last year when I made almost 12kwh on a particular day. But there is no point in splitting hairs. With a stationary array you can make 5kwh-6kwh for every 1kwp of PV panels you have. With a tracker it's over 7. So none of this is really that surprising.

Sure, it will likely cost another 10k (frames, PV panels, etc) to get to the same level of production... but no moving parts. Less potential for wind damage. I've seen the damage a bad wind does to a windmill. I've seen a 12m tower on a Namibian farm twisted up like liquorice, heads ripped clean out of the tower where the hold-down bracket was not installed. Not unassailable challenges, of course, but buck converters is more my expertise than steel structures :-)

I have a national approval for the design up to 18M and 16 panels. I did this 100% because of CoCT

9 minutes ago, plonkster said:

No argument there. I'm saying that if you add an MPPT as well you will do even better.

My best day was actually last year when I made almost 12kwh on a particular day. But there is no point in splitting hairs. With a stationary array you can make 5kwh-6kwh for every 1kwp of PV panels you have. With a tracker it's over 7. So none of this is really that surprising.

Sure, it will likely cost another 10k (frames, PV panels, etc) to get to the same level of production... but no moving parts. Less potential for wind damage. I've seen the damage a bad wind does to a windmill. I've seen a 12m tower on a Namibian farm twisted up like liquorice, heads ripped clean out of the tower where the hold-down bracket was not installed. Not unassailable challenges, of course, but buck converters is more my expertise than steel structures :-)

Let me be honest with  did not expect these results. Living on the west cost when the NW is op galop it bends my house windows 2 - 3 cm inwards. Then it is scary. My veranda was destroyed but the solar stood firm to my amazement. I went and have a look could not stand on my feet as the wind was blowing but solar was fine. Trust me I never expected that.

9 minutes ago, plonkster said:

No argument there. I'm saying that if you add an MPPT as well you will do even better.

My best day was actually last year when I made almost 12kwh on a particular day. But there is no point in splitting hairs. With a stationary array you can make 5kwh-6kwh for every 1kwp of PV panels you have. With a tracker it's over 7. So none of this is really that surprising.

Sure, it will likely cost another 10k (frames, PV panels, etc) to get to the same level of production... but no moving parts. Less potential for wind damage. I've seen the damage a bad wind does to a windmill. I've seen a 12m tower on a Namibian farm twisted up like liquorice, heads ripped clean out of the tower where the hold-down bracket was not installed. Not unassailable challenges, of course, but buck converters is more my expertise than steel structures :-)

 

Plonkster thanks for the chats and making me do some exercising would love to meet you.
Next week I should have my new 2Kwat unit which is grid tie. If I am happy  I will order another 2 maybe 6.
Then I will push the councils buttons again and apply to install another 8 panels.
I would love to end with a total of 64 panel 

If you feel like 'n draaitjie by die see come and drink a coffee.

Don't forget to bring the MPPT that I can see what an expensive anchor looks like 

Edited September 18, 20187 yr by Erastus

On 2018/09/18 at 9:51 AM, plonkster said:

I think the concensus here is if you build your own and if you like doing that sort of thing, or if space is at an extreme premium, then it makes sense. For everyone else it is cheaper to just add more panels :-)

Agreed. My own friend built his own, cause he can, and wanted to, and learned that it's cheaper to rather add more panels if possible. But his roof space was limited. And he way-overengineered the frame making it cost more than it could have. 

Another friend also built one, even though he had ample roof space for more panels. He used it for a season and then completely tied it down due to fear of the wind blowing something away. 

Would you guys be interested in more info on this. I am in the last stage of designing some tools to punch own bearings etc. as that is one of the expensive items.
I do not want to irritate people and also I am not trying to sell it to any person. This morning @8H40 I already harvested 20 amps to charge batteries. Did a test by moving the panels due North and I could only get 7amps  from the solar at 09h00.

To understand I am 4 meters below these houses and the sun rises behind these houses. The clear the houses at +/- 7H45 currently.
The sun track:

My solution to this issue is to have multiple arrays. In my case part of the panels point NNE, another partpoints N, and another portion (about half actually) points North West because of the really long Solar days in summer. Using this process I can also get a bit more earlier in the morning. Not going to get into an argument about which is better... for me this is better, even if it costs a bit more :-)

On ‎2018‎/‎09‎/‎18 at 9:10 AM, Erastus said:

I have installed my 8 panels on a solar tracker I built and ..  I find that where I live I get up to 55% extra power from the same panels.
Would like to know if someone less also did this and what they are experiencing.

I designed the bearings, frame, cpu and.
It is amazing to see the difs between non tracker and tracker.

 

Hi @Erastus would it be possible for you to post more pics of your tracker please?

5 minutes ago, ibiza said:

Hi @Erastus would it be possible for you to post more pics of your tracker please?

Will do

tnx. I have some ideas how to construct the tracker but it's always good to compare to other ideas. and implement some good solutions from others.

18H25 still harvesting 8,2 A = 25% of solar panels

On 2018/09/20 at 9:48 AM, plonkster said:

My solution to this issue is to have multiple arrays.

What I am picking up, Erastus has designed and built a solar tracker, at a price the defies belief, compared to what one would pay out there.

Me, I'm interested to know more, for yes, one can add more panels, panels themselves are one cost, the cabling and mounting of them, increases their cost.

Erastus, what did your tracker cost again, what did you pay? 

EDIT: and he designed for the wind ... so that is sorted.

 

Edited September 21, 20187 yr by Guest

I will give you a short overview about how and why I did it.

The frame is R4700. I live opposite the sea so I had to get i galvanized that was another +/- R2000.  The frame I was suggested to use by installers would have been  between R4K and 5K and would have never lasted . I do not have a roof where it must be mounted. The gear and bearings +/- R 2500. Depending what two motors you want it was given to me.  One can use a small motorbike/generator start motor or....  The frame is designed for 16 Panels and I can push it to 24 panels strengthening my vertical mounts and then I need to change my mounting point.
16 Panels = 26 - 28 panels.  I paid R 1300 each for 260 Watt X 8.  My Solar charger (bulk charger and pwm 320 amp) with solar tracker. I designed. That costed me +/- R 1000.  6 panels I bought second hand for R1000 each 18 months this guy was convinced he needed 340Wpanels and I smiled tried to convince him its a waste but then I said thank you sir.

If you want to compare do not look at the tracker only look at the system:

My system                                                                    Suggested

8 panels          R  8 600                                               14 350Watt              R 42 000

controler         R  1 000                                                2        R 1800            R  3 600   and this exclude any other stuff.

Gear                 R  3 000                                                                                     000

Odds               R  2 000                                                                                    R 1 500

Frame             R 7 000                                                                                     R 4 500

Inve 3k5VA     R 3 000                                                                                     R ??????                     

Total              R 24 600                                                                                    R 51 600              

 

And a lot of told offs as people thought I am stupid and crazy ( I can't argue this)             

This compares to 12 panels mppt ......
The controller disconnect the solar from the load and the panels.

A small algorithm then measures the amps from the panels compare it to the Vbat and can check the Voltage Current relation with the total current used from the batteries and put back into the batteries to determine the state of the batteries. Therefore I do not require fancy equipment. If I need to replace this with what is recommended on this forum to make sure all works I would love to know what it will cost me. But I am sure not what I paid for.

In my case I looked at what was offered and recommended and did what makes electronic sense. I also then expanded the controller to control my swimming pump and other equipment.

We bought 20 inverters that is why the price was so nice. For me I do not know much about solar but electronics a little. The equipment what was suggested and the cost did not make sense to me. So I did my own. I took the cct of an old battery charger I dev a few years back adapted it for solar with pwm and never looked back.

You will see I asked many questions and still don't understand the need for all the electronics.

For me I do not believe you need all the equipment people are talking about THAT IS ME.

I took time to do all the bearing designs and casing. Got engineers to look at the design, stability and I have COC for installation, design and certificate for electricity and applied by Nersa as SSEG. Last week I signed the contract with CoCT and I am approved to push back in the grid. I have also now applied to get my inverter SABS and NERSA approved.  The reason is simple. I want to push back double the 10KVA so I sell what I can and reuse so I always get back.  The (new) inverter 2000W cost is +/- R 3000. And yes it has everything build in it and is grid tie also off grid.
If it works well then I will install a few at my home and hopefully help some people to save on electricity.

When I look what is suggested and what it cost me I am embarrassed to compare my system with what is talked on the forum. One thing I can assure you you will not beat my power I harvest and the way I charge my batteries. 

My actual capital lay out was +/- R 45000 because I burned a few wholes in my desk and made a few wrong designs. But now I have a very nice system.  Does not have all the fancy stuff but I do not believe I need it. My system been working for 3 years now. Currently I am redoing my tracker with my own bearings. Should have my bearing and plates. Tuesday I can check that the stainless is been lazer cut. Hopefully Tuesday is Christmas. Little nervous but still excited.

 

I do not believe I can harvest the same power and handle the stuff more cost effective.