Natural Gas Portable Mining Unit

[Upstream]

Hello fellow bitcoin mining enthusiasts!

Our company is building a portable bitcoin mine and we're very excited about it. Personally, I only got into Bitcoin and mining at the end of 2016 and since then many of you have helped me learn what I needed to know to attempt this project. This is an amazing community so I figured I would pay you all back with some content and updates on how the project is going. Some details I won't be able to answer just yet, so apologies in advance if some questions go unanswered. In time I will be able to answer any and all questions. I do know portable mining containers have been done before, and I know some other versions may appear to be more capitally efficient than ours. Our application is completely different than these previous iterations AFAIK.. but more on that later.

Please follow us on Twitter, @UpstreamDataInc

Here's what we got so far

https://imgur.com/4yYiiVJ
https://imgur.com/sccVAHz
https://imgur.com/w15Hbro

Cheers,
Upstream

[alh]

I've never really understood the appeal of a "portable" mining unit. It's never been obvious to me what would motivate somebody that had the electrical infrastructure in place, and Internet access, and location to park the thing to then decide they just wanted to move it? I understand the idea of deliver a complete "box with mining hardware" to a customer with plenty of money and adequate electrical and Internet. But what would motivate a subsequent move and making that "easy" having a big value?

Just my random thoughts on the idea.

[NotFuzzyWarm]

Don't think the containerized package is so much about relocating as it is about very fast deployment and being a self-contained solution. Really the only building required would be to house the electrical switchgear and transformers. Might be a good place to put the network gear also but personally the further it can be away from main switchgear the better.

Just pour a pad or at least prepare a hard foundation, put in place, open up and hook up power and go.

edit: And of course have very good security for the site....

[Upstream]

I've never really understood the appeal of a "portable" mining unit. It's never been obvious to me what would motivate somebody that had the electrical infrastructure in place, and Internet access, and location to park the thing to then decide they just wanted to move it? I understand the idea of deliver a complete "box with mining hardware" to a customer with plenty of money and adequate electrical and Internet. But what would motivate a subsequent move and making that "easy" having a big value?

Just my random thoughts on the idea.

I agree with you. Past iterations like the Cryptokube (think that's what it was called) didn't really have a customer need requiring a portable mine, and hence they failed. If the source of power is incompatible with a stationary farm however and requires a portable mine, then it makes sense

More to come 

[Upstream]

Today have the power plant in place. Just waiting to fit the gas tree and finish up the electrical on the opposite side of the container, followed by programming and commissioning

 https://imgur.com/gN0jfm0

Edit - how do I post the picture directly and not the link??

[aquapanic]

Is this an "infrastructure" mining unit (network, cabling, power) or is it a "mining" unit (miners included). What sort of final costs are you looking at? I have someone looking at making an infrastructure unit, his costs were looking at 2.5 ish BTC at current prices. Currently have access to a location with very, very cheap power and as NotFuzzyWarm said - rapid deployment and easy scaleability without much physical legwork (running cables, running network, setting up routers) is a critical value-add for this kind of project.

[Upstream]

To start I'm putting in 30 x Antminer S9's, that's about as much as the genset can handle. I've sized the container for a larger load however, and to accommodate a larger genset. So the max I'm thinking is about 72 x S9's, but only after I have this baby up and running and can be confident I won't have issues with condensation (which is my biggest worry honestly).

Is this an "infrastructure" mining unit (network, cabling, power) or is it a "mining" unit (miners included). What sort of final costs are you looking at? I have someone looking at making an infrastructure unit, his costs were looking at 2.5 ish BTC at current prices. Currently have access to a location with very, very cheap power and as NotFuzzyWarm said - rapid deployment and easy scaleability without much physical legwork (running cables, running network, setting up routers) is a critical value-add for this kind of project.

This unit will have a cellular modem, complete wiring and remote circuit control. I'm not planning on building these for sale, but if that was something someone was interested in then obviously we could work something out. The application is obviously not plug-and-play into grid power, as you can see I'm generating power off of natural gas. It is plug-and-play for natural gas.

[Maren]

Some numbers about building costs and efficiency would be great!

ACU would fix any condensation issues.

[Upstream]

Some numbers about building costs and efficiency would be great!

ACU would fix any condensation issues.

As far as I know air conditioning is completely impractical in this case, the air exchange rate is too high. If I didn't use exhaust fans and instead used an industrial AC unit, the additional power draw (as well as the capital cost of the unit) make it impractical from an economics perspective.

[Upstream]

https://i.imgur.com/7nRa1p6.png

Power plant almost plumbed in. Hopefully electrical is completed by the end of this weekend & ready for testing. Will show more electrical pictures later.

[NotFuzzyWarm]

Very interesting project,I think I will watch him in the future.
And I wonder how people in 2017, still do not have Twitter , although it is possible this is old-school and on Facebook . Wish you success in your endeavor!

Screw Twitter, Facebook and other 'social media' schemes.
Never seen the point of 'instant and constant contact' for all the world to see. Their #1 function is to collect data that can then be sold to marketing agencies. What better way to con (many) people into exposing every waking moment of their lives than by having them think everybody in the world needs to know what they had for lunch and other mundane things like that.

That said, looks good! I take it the genset is by a capped off gas well?

[Upstream]

Very interesting project,I think I will watch him in the future.
And I wonder how people in 2017, still do not have Twitter , although it is possible this is old-school and on Facebook . Wish you success in your endeavor!

Screw Twitter, Facebook and other 'social media' schemes.
Never seen the point of 'instant and constant contact' for all the world to see. Their #1 function is to collect data that can then be sold to marketing agencies. What better way to con (many) people into exposing every waking moment of their lives than by having them think everybody in the world needs to know what they had for lunch and other mundane things like that.

That said, looks good! I take it the genset is by a capped off gas well?

Bitcoin twitter is a lot of fun, I don't use any other social media though (unless Reddit is considered social media), just Twitter bitcoin feeds. Get into it

You're close re the capped off gas well, more on that soon when I tie it in 

[NotFuzzyWarm]

Will you be using a single generator which operates 24 hours or will you use 2 generators for 12  hours shift
If you use one that runs 24 hours can you please tell me the cost and brand name of generator for 30 S9 miners you are using

12 hrs shift? Riiiight....
He is not talking about a typical home portable generator. These are what is know as Prime Power generators. That means they are made to run for MONTHS and more between having to be taken off line for inspection.

They are available in most parts of the world, just look for used ones. Good example is http://www.criticalpower.com/generators/480-kw-cummins-natural-gas-generator-gta-28/ Ja that particular one is a standby vs Prime but was a quick grab from their site.

[Upstream]

Yeah this particular genset is a 5.7L GM, pretty common for natural gas.

Cost is a secret

[Upstream]

I doubt I'll ever discuss costs, but it isn't exactly rocket science to price something similar out yourselves. Also, since this is really a prototype I have a lot of optimization to do for the next one. I'd expect I could get costs down by at least 25%.

More important that capital cost is operating cost, that's my major concern. Gensets require maintenance and must be checked regularly etc. Mining doesn't work if operating costs are like $0.10 / kWh or greater, I'd be crushed when the bear market returns.

[Upstream]

Programming the controller today, should hopefully have the mine installed on Friday. I'll post pictures this wknd if all goes well 

[McKaig]

Where are you building this? 

We have extremely cheap Natural Gas in my area, and Im starting to get ideas.  I've already designed a 20ft container capable of housing 54 S9's, however would rely on external power source.  Nat Gas is very interesting.. 

[Merptastic]

Where are you building this?  

We have extremely cheap Natural Gas in my area, and Im starting to get ideas.  I've already designed a 20ft container capable of housing 54 S9's, however would rely on external power source.  Nat Gas is very interesting..  

By cheap, want to give a rough cost per GJ?

Natural gas in my area cost about $4.7 USD per GJ (converts to roughly 278 kWh).

That being said, I have no idea just how efficient these natural gas generators. I've checked Cummins website and it's only specified as "Incredibly Efficient Operation". I'd be nice to have a number.

How about natural oil mate? in the picture you have posted I don't see any miners, it's just a gas pump or what is it exactly?

It just looks like a natural gas fed generator with meters and a rack for miners in the background.

What do you mean by natural oil? Crude oil?

could you convert gas To electricity and spend less money than producing electricity with water?

Are you talking about hydropower? And yes you can use natural gas to make electricity using a natural gas generator.

[McKaig]

Where are you building this?  

We have extremely cheap Natural Gas in my area, and Im starting to get ideas.  I've already designed a 20ft container capable of housing 54 S9's, however would rely on external power source.  Nat Gas is very interesting..  

By cheap, want to give a rough cost per GJ?

Natural gas in my area cost about $4.7 USD per GJ (converts to roughly 278 kWh).

That being said, I have no idea just how efficient these natural gas generators. I've checked Cummins website and it's only specified as "Incredibly Efficient Operation". I'd be nice to have a number.

Our rates are listed in CAD $/m3.  Link to rates:  https://www.uniongas.com/residential/rates/current-rates/rate-m1

How did you determine how many kWh you can generate per GJ? 

I've been looking into Microturbines to see if this method could be viable.  What I do know is that our area has an abundance of Nat Gas, and I have two uncles who work in senior management positions at the utility company, who could help me negotiate larger commercial pricing.

[McKaig]

Our rates are listed in CAD $/m3.  Link to rates:  https://www.uniongas.com/residential/rates/current-rates/rate-m1

I did some math to compare my rate to yours, using these rates from the Gas company in CAD:

Gas used:  13.7312 ¢/m3
Gas price adjustment:  3.4547 ¢/m3
Storage:  0.7153 ¢/m3
Delivery:  8.0131 ¢/m3
Delivery price adjustment:  0.5143 ¢/m3
Monthly charge:  $21.00 (Not used for GJ calculation)

Total = $0.264286 CAD /m3
 
1 GJ = 26.137 m3 (http://callmepower.ca/en/faq/gigajoule-cubic-metre-gas)

Therefore 1 GJ = $6.91 CAD, or $5.47 USD  This is taking into account the highest residential rate, for the first 100m3.  Price drops after 100m3, then again after 250m3.

If 1 GJ = 278 kWh, then

1 kWh = $0.025 CAD or $0.02 USD

Are these calculations correct?  I did not research to confirm the "1 GJ = 278 kWh" part.