Topic: Solar array starting to look good. - page 8. (Read 20530 times)

Awesome project Phil!  Best of luck to you on it.  I'm looking forward to seeing and hearing more about it.


You have to think of the solar setup more like power supplies in mining.  Even if he stops mining the solar setup will still generate money for him and still have many other uses.

This is some good info as to why we are not using batteries.

Like I mentioned before I was interested if a solution was available  other then what we are doing.  You have pointed one out. Thanks.

Since this array is being built by forum member buysolar and solar installs are his business it has advantages in price most people can not do.

He uses 3 sets of footings for the rows of panels for demonstration purposes.  If I own a superfund site   in NJ >

Pretend I own a polluted dumping ground maybe an old factory did the pollution. I purchased the ground on spec.  The feds may approve a solar array since the land is ruined for just about anything else.  I could get clean up funds.

That may allow me to build this type of array.

this cement base method below is the most costly but it does not penetrate the ground so it could be used on a super fund site.  Useless polluted land can serve a clean purpose.  So his solar array is also advertising for his business.

Forget about batteries. Lead acid are not usable (only 30-50% capacity can be used for better lifetime, so price per kWh is at least doubled), Lithium batteries have much better cycle life, but they are higher cost. Typical costs for stored 1kWh are about $1USD, etither Pb or Li.

I can suggest you much better and easy solution for your miners to save a lot on efficiency. It need direct connection to your solar array which should have maximum power point between 100 and 400V and already connected MPPT inverter (you have it already).

Solution:

1) If your MPP string voltage is around 400V, you can connect DC voltage directly to PSU DC input line just after PFC. You will have direct path from solar DC to miner PSU and you will eliminate some very small loses of PSU's PFC circuit. If RMS input voltage is higher than PFC set voltage (usually 380-400VDC), PFC controller just switch itself off and let voltage pass through. We have DC line, so RMS=DC voltage, so if you do not care about 3x 0.7V voltage drop across forward diode, you can connect DC line directly to AC main input.

2) If your MPP voltage is less than 350V, you sould connect DC line directly to AC input or inside just after diode bridge (2x 0.7V drop).

3) set your 12V output very little higher than your AC line PSU, so if your AC line PSU has 12.0V, your DC line PSU should be about 12.2 or higher.

You will feed your miners primary from direct DC-DC path, if DC line PSU has not enough power, it will switch itself off and your miners will be fed by your AC line. Your MPPT controller always keeps your solar grid at maximum power point, so you do not care what is your voltage within 300-400V range. In case of 100-400V range, your PFC can go less than 60V, so it will not switch off immediatelly when cloud comes on the sky.

This is general description. Some PSU's like Fortron Raiders, IBM 2980 or Bitmain's needs higher voltage to work and start-up, so your solar grid DC voltage range will be narrowed to 230-400V DC. Maximum DC voltage accepted by PSU is limited by input DC capacitor voltage (the bigges one in PSU).

You need to answer yourselves some questions here.

1) buying batteries: does it have ROI, when every kWh stored is about $1USD and needs additional charge controllers? Can your miners survive voltage variance between discharged and charged battery (11.5-14.5V) or need another DC-DC controller?

2) is it enough to save 5-10% efficiency to buy another high power PSU (MPPT AC inverters are not 97%, they are less depending on actual MPP votage set)?

3) there is NO SOLUTION to connect PV panels directly to miners without power losses. The best way is to use MPPT controlles for 12V or any MPPT combination as shown above. You will always lose some power at conversion.

Buying cheap panels and cheap inverters may do better ROI than thinking about batteries and direct connections... There are 1200W on-grid outdoor micro inverters at $280, so if you have $0.55/W panels, it is $940USD per 1200W complete on grid system with super easy installation and unlimited stackability.

looking at this thread, it got me looking at solar panels, kits and cells(diy ftw!) the most expensive is 0.9W/$1au to the cheapest 2W/$1au for diy.

if I can get a setup to get 1KW off-grid I think I'd be a happy camper.. but I never properly worked out the math to get the 1000W draw to needed cell power (KWh) and battery storage.

If anything, I want to setup my aircon to off-grid by modding it to run on DC and generated solar/wind/thermopile/steam..

Anyway, good luck with getting that solar setup running nicely!

This is top notch project.

I'm inside the photovoltaic universe too. With "all of my love". We got lot of work to do man. But in the end it's always amazing.

I can't sell electricity to anybody nor any company, but it's the reason I'm alive. Long life to solar photovoltaics and bitcoin mining.

your project is very, very, veryyyyyyyyyyyyy nice bro. Congratulations with an A+++ label.

 

Well I'm sure you've though of this but for now just figure overall daily miner kwh usage and subtract how much of that the solar provided each day. Use that figure against the daily kWh rate to see what you pay the electric company each day of mining 24x7x365. No matter what the contribution made by the solar should be substantial.

NJ power is 13 cents in the winter and 17 cents in the summer.

So solar or no mining.  Plus the excess power can be sold back at 5 cents rather then mine.  Note NJ is very good this way we are one of the better states with goverment subsidies for solar.

So very true.  But assuming part of the idea is to allow it to run at night there are of course well established commercial ways around it.
Disclaimer: I cannot say I've delved into this subject deeply. Only have good knowledge of how current large power systems work overall.

To me the easiest is to first ask: What commercial power storage devices are being used for small (yes, even 16kw is still 'small') setups to allow running after dark when one is fully off the grid? Look to there for ideas.

My guess is that it is a DC storage system be it batteries (bulky, insanely heavy, maint & lifetime issues), supercaps (very good but high initial expense), or whatnot rated at what the inverter/panels combo runs at be it 12, 48, or 600v. All involve robbing Peter to pay Paul. Can't forget that the power used to charge up for the night time comes from the daytime power budget.

No, the wire losses would be greater than any inverter inefficiencies, plus you would need ALOT of copper. You have to remember even in a 150w 12v panel you looking at 12.5 Amps of current. 60kw would need copper that can support 5 THOUSAND AMPS of current lol...ill let you do the math on the copper thickness you need for that

This is why most solar setups are wired in series, and some have muti kilovolts running into their main inverters, or simply have an 120v inverter directly on the panel.

Looks awesome...but even building it from scratch with no labor/overhead costs you'd be hard pressed to ROI on this anytime soon. Even if you paid 60k for this setup that would be able to power 20kw worth of miners for free it would take over 4 years to ROI. Guess its worth it if your paying 15 or 20 cent power in jersey, since this will essentially let you mine for the long term and compete with the big players paying near zero cent power.

Batteries = $$  they don't last long. They can be dangerous just think of the energy stored in them.

But I am interested in the answer.

The ibm 2980 server is a real platinum psu say 91-92%  I do not think the best inverters   are more then 98%

this article claims 97%

http://www.electronicproducts.com/Power_Products/Invertors/High-efficiency_solar_inverters_in_North_America.aspx

so 91 x 95 = 86 to 88%  from panel to miner would be best case.

so would batteries and a regulator be better?

I do not know .

Looking really good guys, very excited for you!

This is a question for buysolar, if one were to do a PV setup that wasn't grid tied (I guess you would need batteries to buffer) and mining continuously with it, would it be possible to keep voltage produced in DC and regulate to 12V to prevent inefficiencies from inverters and then again from power supplies?

It can be complex set of rules and regs to follow to know the payback.

But with the correct set of circumstances a solar setup can really help out mining coin.

I cant wait to fully move the gear in and then decide how to expand for 6 kwatts of mining gear to 16 kwatts of mining gear.

Also in some states (usually where solar installation is widespread and popular) utility companies wouldn't let you install more than your last year's usage. They don't want any consumer to be a producer. In my case in SoCal, I had to certify that my projected usage will more than double to install 10.65 kW worth of panels. And Edison agreed to pay for any excess production at a wholesale rate of $.03/kW.

That is buysolar's field  to talk about.

Remember each and every USA power company has different payback rates on excess power.

Here in NJ most companies pay back 5 cents a kwatt

so a 60kwatt setup using 20kwatts  and selling 40 kwatts back ends up at 0 cost for power.  see note 1:

Some states won't pay back at that rate  and you may need a 15kwatt used and sell 45kwatts to end up at 0 power cost.


So if you can install diy at 1 usd   a kwatt in NJ you may do okay, but if you install at 1 usd a kwatt in Ohio you may not do okay.

You need to know what you can sell the power excess for to determine if it is worth doing.


note 1:   average sunlight in NJ year round is on this map

 http://www.wholesalesolar.com/solar-information/sun-hours-us-map

Depending on the state u are in in the USA, if u generate more elec than ur home uses ... the elec company has to pay you.
Guess thats a nice lil perk if you decide to stop mining.

Amazing man, good job! Wish I could do someting like this! 

I think we all agree look amazing.  Do you mind sharing cost of getting up  and running on solar?

Also how  long do you have to run till the solar gear has paid for it's self

There in lies the question.

It is all up front money and then no expenses to speak of.


 Below is hypothetical and does not represent what we did. Since buysolar is a solar contractor and this install is far lower cost then if you hire someone to do it.


But pretend you buy 60kwatts in solar panels + kit  from this dealer on ebay


http://www.ebay.com/itm/SolarWorld-10kW-CR-Solar-Kit-with-Enphase-Inverters-Shingle-Roof-/151972428611?

that would be 6 of these kits or  over 125k usd    more then 2 usd a watt. Even if you do all the install the price is too high


 So at retail it just won't work out.

But if you are in the business or you find a deal on panels and are really good at diy .  the numbers can work.

The key is a mixed  use grid no battery setup.   

 if you put in 10kwatts  ¼  for miners ¾  for sell back 

you get 2 s-7's on free power.

 

A pair of s-7s right now cost 1500 usd

they would mine for 135 days and the ½ ing would hit  .

Say July 3rd   you would be 200 usd in the hole   on the miners.  We just do not know what diff will be, but the miners won't cost anything to run.

So you could sell one  get 250   to be ahead 50 bucks and mine the other.  Just one of many possible things.  You can still use the solar to lower your home power costs. Just say fuck it to mining.

That is one really nice setup. Can't wait to own land so I can set something up like this. How long will it take to break even or even start seeing ROI?