Jump to content
The Corroboree
Slybacon

Solar Powered Bitcoin Mining Design Project

Recommended Posts

Goal - To create a sustainable and finacially viable digital coin mining system powered by a combination of Solar/Grid. Then seek funding to purchase and install system. With the introduction of high end mining cards high GH/s makes mining more profitable then ever. There is also the opportunity to hedge your efforts accross several emerging digital currencys. The viability of the project hinges on the stability of digital currency exchange prices and the difficulty of solving a blockchain.

Current Bitcoin Difficulty -

609,482,680

speed-small-lin.png

Current Bitcoin Price at 2pm Friday 22nd November 2013-

$748

Reward For Solving One Block @ Current Exchange Rate ($748)

25BC ($18700 USD)

Time Frame Required to Solve One Block @ 50 GH/s (Giga Hash per Second) @ Current Difficulty

244 days to 1 Year

To be accurate we will need to prodict the rate of difficulty increase in order to get the true time to solve one block. This can be somewhat predicted using the above metrics. Even at Double the calculation difficulty we would still be able to solve one block in 3 years @ 50GH.

Dedicated mining hardware can now produce up to 600 GH/s from a single card. The card can run on 350W with and extra 100W for the rest of the computer. So we can achieve 600 GH/s for a mere 450W. This means we can power this system with a 600W solar system by day and switch to off peak grid at night. This would make the running costs pretty cheap, but what about the initial outlay?

COMPUTER

Motherboard $150

CPU $150

Bitcoin Mining Card 600 GH/s $4680

80plus Gold PSU $250

Case $150

HDD $100

Operating System $100

SOLAR SYSTEM

Solar Panels 200W x 3 $1185

Frames $250

108 Ah 12v Batteries x 2 $750

SB3024 $575

XTS 1200 / 24 $1600

Relay Set up $50

Remote $379

Cables and Accesories $200

Enclosure $350

Labour $500

Extra Expenses $500

TOTAL COSTS

$11919

OUTPUT

600 GH/s

CONSUMPTION

450W

Time to Solve One block @ 609,482,680

11 Hours to 50 Days

I will factor in some more detailed calculations soon, and include analysis of other digital currency but with recent advancements in hardware and soaring digital currency values I can see this as being a very viable venture. We need now to do some more detailed calculations and predictions and come up with a sensible proposal. We should be able to get a prototype proof of concept together with cheaper componants and stuff we already own, but Im very excited to see if this can work. Stay tuned :)

Edited by Slybacon
  • Like 3

Share this post


Link to post
Share on other sites

Sounds great, though I know very little about what you speak of. If it's so easy to make such good money with current technology, why isn't everyone doing it?

Share this post


Link to post
Share on other sites

Have you considered using FPGA's or ASICS? The power savings would outweigh the initial cost of the device, especially since you're considering solar...

https://en.bitcoin.it/wiki/Mining_hardware_comparison

Edit: Ah, I see you have 'Bitcoin Mining Card 600 GH/s $4680' - What sort?

Edited by IndianDreaming

Share this post


Link to post
Share on other sites

This study is based on current conditions, which are favourable. Mining Card technology has only just improved to a point where you can produce high hash rates for low power consumption. The market is also very volitile for example, in this month alone saw the market capitalization peak at 8.9 Billion and fell to 6.9 Billion in a metter of days. Without governed backing or anything real assests to value it against big investors aren't interested (yet), and most everyday people have either never heard of it or aren't even interested in it. Try and spark up a converstaion about bitcoins at work or the pub and have a look at the reactions you get. I think its so new and risky people are shying away. I truely don't know if it will be viable, under current conditions it certainly looks that way. Its a matter of who is right, in predictions moving forward. Some companies are building rigs and renting under mining contracts. Who would of known, "DIGITAL FUCKING MINING CONTRACTS!!"

Thats some psyfi shit right there, or nerd central whatever way you look at it. Theses contract ensure finacial stubility by leasing by the GH. $10.98 per GH/s. The risk of the market is then on the contract owner whilst the equipment owners can plan with finacial security.

PS - Typing Fail..... Not my highlight!

Edited by Slybacon

Share this post


Link to post
Share on other sites

Have you considered using FPGA's or ASICS? The power savings would outweigh the initial cost of the device, especially since you're considering solar...

https://en.bitcoin.it/wiki/Mining_hardware_comparison

Monarch BPU 600 C

https://products.butterflylabs.com/homepage-new-products/600-gh-bitcoin-mining-card.html

:)

If it went ahead

Edited by Slybacon

Share this post


Link to post
Share on other sites

yes but when will that card ship...

Share this post


Link to post
Share on other sites

ideally we want 4 of those cards, each person buying in has a card each etc :P

and yes it is an ASIC, so its pretty much useless for anything BUT mining. same card i was looking into but there's problems with shipments and people not getting them until many many months later, and some not at all... its a high risk business atm.

but with high risk comes possible reward, just don't put anything you cant afford into it.

Edited by C_T
  • Like 1

Share this post


Link to post
Share on other sites

Sounds like a lot of fun at the very least Sly! I know very little about bitcoins, apart from my friend that used to nail me down at the pub and hammer me with bitcoin talk :)

The variability of the market makes it even more fun IMO, and you might kick yourself in a year if you don't go ahead...

Is it possible to use each of the 4 cards individually? As I understand it, a block takes a variable amount of time - as you mentioned, approx. 0.5 to 50 days. It makes more sense to mine multiple blocks in this case than to pump all your power into a single block.

Thelma? Are you here - you might be able to work out what's best probability wise - focus on a single block or split across 4 blocks...

  • Like 1

Share this post


Link to post
Share on other sites

that comes down to whether you join a mining pool, where lots mine for the same block and its distributed regular pieces of the btc

or you mine your own block individually, that takes balls hehe

more hardware just means better return each day, putting you further up the pyramid of processing power. which is where you kinda need to stay

Edited by C_T

Share this post


Link to post
Share on other sites

Seems like buying coins just after SR went down would have been a wise investment... ah hindsight...

Share this post


Link to post
Share on other sites

How many hours a day would this setup be operating?

I might be missing something here but your battery rating seems under spec to me.

If it was to run off battery power in the dark period, wouldn't it need a much larger capacity (this is if you intend to mine 24/7)?

12 hours(dark period) * 450 watts = 5400 watt hours

5400 watt hours / 12 volt = 450 amp hours. You could halve that with a 24 volt system, with 36 volts you could divide the amp hour rating by 3, then there would be some losses at the conversion back to 12volts or at the inverter etc.

and that's running at 100 % capacity for the full night without any consideration for discharge levels and no consideration for losses at connections & relays ect.

Share this post


Link to post
Share on other sites

Nice Point, and very true. We weighed up the price of batteries and decided its cheaper to use the grid at off peak times. The batteries allow the system to work in cloudy periods or when you are switching to grid power. We haven't finished collecting the data for comparison but once the spreadsheet is finished we should have a clear picture. The Aim is 24hour Mining at the cheapest possible / hour price. We think we can find the balance between battery / solar / grid. Its very preliminary at this stage but I couldn't help but share the data givin the current bitcoin conditions, thought it might inspire others and we could draw inspiration from other people.

  • Like 1

Share this post


Link to post
Share on other sites

Sory about that Sly I misread your OP, I have a tendency to speedread in phrases or paragraphs and sometimes miss a few details.

How would the economics work out if you were to charge the batteries on off peak for use the following day and use the grid during off peak billing and then do away with the solar setup ?

  • Like 1

Share this post


Link to post
Share on other sites

Hey thats not a bad idea.... :) I'll look into that.

Was also considering Grid Feeding the Solar to Offset the Rig as well.

So many scenarios to build spreadsheets for.

Edited by Slybacon

Share this post


Link to post
Share on other sites

Obviously the solar would be the ultimate goal, but it could be a way to cut costs in the interim.

I don't think it's technically legal, but for a relatively low current draw it could fly under the radar, especially if you don't have a smartmeter. even then you could wire into the hot water system circuit (depending on the billing rate) and it would be hard to detect.

Edited by Sally

Share this post


Link to post
Share on other sites

Bitcoins trading at $875 AUD today!

Just calculated that 450W Mining rig will only cost $1200 per year to run. Kinda makes solar un feasable :(

Share this post


Link to post
Share on other sites

Maybe see if you can find the solar gear second hand or free - if you keep an eye out something might come up. Some house being demolished with panels or something...

Share this post


Link to post
Share on other sites

Yeah I was going to say I live off solar, but if I had grid power something that small Id just run it via the grid. Not worth the hassles. Id run it on off peak times. Just look at the variation on power around australia, pick somewhere with a low rate and put it there.

Also charging batteries on off peak to run during peak is inefficient so you probably wont win out. Id just bite the bullet and run it on grid.

  • Like 1

Share this post


Link to post
Share on other sites

Look into wind and making a program that will let you boost hash rate during peak power production.

As it seems cheaper to have more bitcoin mining power than back up battery power . So during the day

when your panels are putting peak out and a bit of wind power on the side . It would be better if you could use

that extra power to boost your mining power . The batteries can be bought on the cheap testra replaces all there

backups every two years they use big 2volt batteries so you will need six to make a 12volt circuit.

Just a question can you get computers to run on 12v dc as it would save heaps on inverting costs

and power loss

Share this post


Link to post
Share on other sites

Something Has to give here. Butterfly Labs offer Cloud Based Bitcoin Mining at $10.83 - 1 GH/s. If you were to compound the earnings back into more GH/s you would see the income start to grow at what appears to be an unsustainable level. With half of all bitcoins already in exsistence, if demand remained constant we should only see a 50% reduction in bitcoin trade price should the other half be mined overnight. All the Bitcoin hardware seems to be sold out in a multi million dollar digital gold rush. As the saying went, the best money to be made in a gold rush is selling shovels!. So what is the risk and the downside? Here is some scenarios to consider-

$1083 Invested in 100 GH/s then Compounded for 6 Months- (@ Current BC Trade Price of 797.98 USD)

Buy In - $1083 for 100 GH/s

https://products.butterflylabs.com/homepage-new-products/1-gh-cloud-hosted-bitcoin-hashing-power.html

Month 1 - 100 GH/s = $2460.52 / $10.83
Month 2 - 227 GH/s = $5585.39
Month 3 - 515 GH/s = $12671.70
Month 4 - 1170 GH/s = $28788.13
Month 5 - 2658 GH/s = $65400.73
Month 6 - 6038 GH/s = $148566.44

TOTAL PROFIT $147484.44 (Plus 6038 GH/s 6 months 50% = $30190)

The above scenario is the best case, I can't see the difficulty increasing to much over the 9 month period (15% maybe) and with more and more merchants accepting bitcoins it may very well go up before it goes down. I will run some more numbers and include difficulty increases I just wanted to get a basic idea on paper to see how it looks. Verdit - Very Fucking Tasty!

$1083 Invested in 100 GH/s then Compounded till Break Even- ( After a Crash BC Trading @ $100 USD)

Buy In - $1083 for 100 GH/s

Month 1 - 100 GH/s = $246

Month 2 - 122 GH/s = $300

Month 3 - 149 GH/s = $366

Month 4 - 182 GH/s = $447

Month 5 - 223 GH/s = $548

Month 6 - 273 GH/s = $672

Month 7 - 335 GH/s = $824

Month 8 - 411 GH/s = $1011

Month 9 - 504 GH/s = $1240

9 Months for $157 Profit and 504 GH/s for the next 3 months, and then it would drop by the rate it increased monthly untill your contracts expire considering you stopped compounding at the breakeven spot.... Like this

Month 10 - 504 GH/s

Month 11 - 504 GH/s

Month 12 - 504 GH/s

Month 14 - 404 GH/s

Month 15 - 382 GH/s

Month 16 - 355 GH/s

Month 17 - 322 GH/s

Month 18 - 281 GH/s

Month 19 - 231 GH/s

Month 20 - 169 GH/s

Month 21 - 93 GH/s (CONTRACT EXPIRE)

In this scenario you would have $9072 in your account after 21 months if the bitcoin value remained at a mere $100 and the difficulty did not increase. Currently they are trading for $850 AUD. Virgin just started excepting Bitcoins as payment for some of its services so chances are it will go up before it goes down!

There are so many ways you can look at this. What I am failing to understand is the effect a rush will have on computaional difficulty. If a huge surge in bitcoin miners suddenly got involved what figures and by how much would this effect the viability of mining. With Cloud mining it seems to be a bit easier to manage. The maths on the above is very very rough and If anyone would like to correct me that would be awesome, its very interesting I just wonder for how long this can keep going.

NOTES:

http://www.bitcoinx.com/profit/

Extrapolating bitcoin difficulty or price is pure voodoo. It is much easier to predict the relationship of the two parameters in form of the Mining Factor. The Mining Factor 100 is the value in USD of the bitcoins you can generate if you let a 100MHash/s miner run for 24 hours. If the Mining Factor 100 rises above $2 or so everybody buys mining equipment and thus increases difficulty. If it falls people will stop mining eventually. The estimate starts with the current Mining Factor and decreases it exponentially such that the decrease accounts for the factor decline per year. Please note that a profit/loss by holding the coins is not accounted for in this estimate.

Edited by Slybacon

Share this post


Link to post
Share on other sites

Seems odd to me that if Butterfly Labs has that sort of computing power - why wouldn't they just mine the coins with it themselves and reap the 'profit'

Share this post


Link to post
Share on other sites

Seems odd to me that if Butterfly Labs has that sort of computing power - why wouldn't they just mine the coins with it themselves and reap the 'profit'

Their profit is when they accepted BTC for their orders over 2 years ago when bitcoin was worth MUCH less than it is now. The miners people bought them are basically worthless now, if they've got them yet.

That's something you have to consider aswell, which I'm not sure you have, ASICs have a HUGE lead time. Order now and you'll be lucky to get it in 6 months. By then, the difficulty would have increased by a shitload.

Share this post


Link to post
Share on other sites

ASICs have a HUGE lead time. Order now and you'll be lucky to get it in 6 months. By then, the difficulty would have increased by a shitload.

True, thats why I thought cloud mining would help eleviate that problem. You would have to make sure timelines are structured into your mining contract. As far as difficulty increases, there is a chart that shows the percentage increases over time. The next increase is expected to be about 9 or so %. You could predict the increase in difficulty and factor in worse case scenarios. Thats my next plan.

Edited by Slybacon
  • Like 1

Share this post


Link to post
Share on other sites

Word of warning on Butterfly Labs: their support sucks balls. They have a single email address for their entire company regardless of the enquiry. ie. they don't want to know that the hardware they've shipped after you waited 7 months is faulty. :BANGHEAD2:

So that Monarch card you've had your eye on - is more likely to be only 500GH...

  • Like 1

Share this post


Link to post
Share on other sites

Word of warning on Butterfly Labs: their support sucks balls. They have a single email address for their entire company regardless of the enquiry. ie. they don't want to know that the hardware they've shipped after you waited 7 months is faulty. :BANGHEAD2:

So that Monarch card you've had your eye on - is more likely to be only 500GH...

I heard they are dodgy. To be honest I am using them as an example, out of interest, why would the card only output 500?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×