Will support for bminer be added?

https://bitcointalk.org/index.php?topic=2519271.0

Is is faster than DSTM in my testing.

Thanks

Sure!  Maybe once I get all my miners situated in/near my shop this spring, would be a good time for a follow up.

Thanks Phil!


What are your thoughts on going with a bunch of smaller fans vs. fewer larger ones.  Larger ones are much more efficient when you look at CFM per watt.   The Master Flow 1600 CFM pulls just under 200 watt, so that works out to 8 CFM per watt.  The large 36" fans I was looking at do 10,000 CFM @ 800 watt, so that is 12.5 CFM per watt.

But having more smaller fans allows for better redundancy and air flow can be adjusted by turning some of them off when not needed.  Cutting round hole in metal is a bitch though.


That is a great idea!   The container comes on a roll away 5th wheel trailer, so it will be tricky to get the container up on that base since the driver won't be able to drive up on it to do the unloading, at least not without a lot of temporary support.  They do offer "precision delivery", but that almost doubles the cost since they have to bring out a big crane.  I'll have to check with them on what options I have.  With my equipment I might be able get it up on there myself.  I think a 40ft is around 8,000 lbs.  I mean, I can do this:



So I can probably position the container on the foundation somehow.  Should be fun. 


Another great idea.  I figured I can do mesh skirt all around the container to keep out animals and other large dirt object.  Sort of like a pre-screen for the pleated filters in the floor.


That or have the inside spray foamed.


That link did not come through.  But basically create a slope in the ceiling towards the exhaust fans, right?

The rural county that I live in started offering fiber based internet service about 3 years ago to help develop and attract businesses.  I got together with my homeowners association (16 of us, each with decent sized lots ranging from 6 to 200 acres) and worked out a deal with the county to bring in fiber.  Cost of construction was about $50,000 which the county subsidized so the cost to the homeowners was only 50% of that, and they let us pay it back over 5 years at 0% interest.  All but one homeowner signed up, and we closed on the deal.

That was 2.5 years ago and we now all have nice fast fiber based internet service.  I actually have 2 ONT's (Optical Network Terminators, or fiber modems), at my local, each on different sub nets, for some redundancy.  I'm still working with the county to get them to offer symmetrical GigE service.  The equipment is capable of handling it, but they just don't offer residential rates at that speed.  So I currently have dual 50 mbps down / 10 mbps up for an aggregate of 100/20 service.

The rest of my network equipment is all Ubiquiti Unifi, which is the same company that sells the solar equipment that I use.  So I have a central dashboard from which I can manage router, switches, and access points.  Here's the main dashboard:



A view of the network devices on the network:



The AP down at my shop is currently disconnected as I ran out of ports on the switch, so the above view, as well as the main dashboard, provides an indicator that something is wrong.

Here's a topology view.  Sorry that is is all squished together horizontally.



Here's the actual core network equipment:



24 port path panel on top for hardwired ethernet jacks through the house.

48 port core 1Gbps switch with 500W PoE (powers ip phones, access points, security cameras, etc).

16 port 10Gbps switch for server connections and fiber down links to other switches.

Dual WAN/LAN router/gateway.

2x ONT modems to the service provider.

Great to hear they are running 100 cards in the 20' with no issues.  I should be able to scale a 40' to 200 cards easy then.  So here's the spot I have picked out:



The container is not quite to scale.  It is a little small.  The pole barn opening is about 14'.  North is pretty much straight up in the above pic, so the container will be in the shade during the afternoons.

And I'm only about 100' from the power pole:



Figured I'd have a separate 400A service run to the container.

So this is how I envision the ventilation system:



So air will flow from the low intakes to the high mounted exhaust fans on the opposite wall.  I don't anticipate needing any "helper" fans in the mining racks.

Great topic about air flow and cooling so I figured I'd post about my plans here.

As I prepare for summer, I need to figure out what to do with all my rigs.  I'm at around 16kw right now, but anticipate being well above that by the time summer (and heat) rolls around.

My plan is to get a 40' shipping container (about $2,000 delivered) and ask if I can get it in white.  If not, I'll have to paint it myself.

I plan to install 3 10,000 CFM 36" exhaust fans, evenly space along the top, along one of the long walls.

These guys:

https://www.northerntool.com/shop/tools/product_200422375_200422375

Full specs here:

https://www.jdmfg.com/wall-master-exhaust-fan

They are very efficient and move about 20 CFM per watt.

On the opposite long wall, I'll cut out 6 24x24 inlets, evenly spaced along the bottom.  My mining racks will then go right down the center with a 36" isle on either side.

A 40' container is about 3,000 CF, so 30,000 CFM should be able to replace the air volume 10 times a minute.  That would be enough draft I would think, that I would not have to worry about any additional cooling, even when it is 100 degrees outside and the sun is baking down on top of the container.

Good idea or bad idea?

Afterburner works with 9+ cards.  However, you have to leave the cards in sync.  If you try to apply individual clocks and power, it will only support 8.

Thanks for putting the video together Vosk!  It was a lot of fun.

@ x299  My system is grid tied.  So I draw power from it at night and anytime during the day when I consume more than I use.  On sunny days, I push quite a bit back onto the grid, despite what my miners consume.

The solar gateways allow me to track the production from each array.  Here's a snapshot from around noon on the 15th:

Roof array:



24 panel ground array:



48 panel ground array:



So how much sun there is on a given day has, as you would expect, a huge impact on the power being produced.

Also, I still got a couple of inverters to test on the 48 panel array as only 43 of 48 are reporting as seen above.

I'm from Denmark originally, but reside in the U.S. in central Virginia these days.  Those are the Blue Ridge Mountains you can see in some of the pics.  We don't have mountains like that in Denmark.  The tallest "mountain" if you can even call it that, is 561 feet.  

Yes, they are topped off about twice a year.  Interesting idea about building a blast/fire wall.  Will definitely look into that if I don't end up just removing them.

Reason I might just get rid of them is that I plan to move all my mining rigs down to the shop come spring, and plan build a dedicated mining room within the shop that will have hot and cold isles and vent to the outside.  The inlet will be very close to where those propane tanks are.  While some rigs will go back up to the house next winter, enough will stay behind to keep the shop building frost fee.


Damn those are expensive right now!

I like micro-inverters because they allow me to track individual panel performance.  Also, if a panel is shaded for part of the day, it won't drag the rest of the string it is part of down.

But yes, cost is higher, but you avoid dealing with high voltage DC and the now mandatory rapid shutdown those requires, making the cost closer to equal.

For the 2 most recently installed ground mount arrays, I ordered the panels and mounting hardware from Renvu the last time they ran a special with free shipping.

The micro-inverters, solar panels, end run connectors, and associated cabling and such, I ordered from Streakwave.

Rough ball park on cost:

$200 280W Solar Panel
$100 250W Micro Inverter
$50 Solar Gateway
$20 End Run connectors
$20 Y-Cable

So to build a 24 panel array, you would need:

24 panels
24 inverters
24 Y-cables
1 Solar gateway
2 End Run connectors

So about $ 7,770 for a 6.72 kW array.  To this you have to add the cost of ground or roof mounting, plus conduit, wiring and breakers, to tie into your existing system.  Doing everything myself, I've been able to keep the cost down to about $1.25 - $1.50 per watt.  Most solar companies offering turn-key solutions charge between $3 and $5 per watt.

Do you not have a net metering arrangement with your power company?  I'm basically using the grid as my battery.  Any excess energy I used during the day is pushed into the grid, and I'm credited the exact kWh I push in, and I then pull those back out at night, plus whatever additional kWh I need.  If I push in more than I use over time, I build up a "bank" of kWh with the POCO.  Before I started mining, I had build up a bank of well over 1,000 kWh.

Miners are up and running already.  Please see my updates in the 4th post.

I have a couple of different methods to monitor my energy production and consumption.  Each array cluster has what is known as a solar gateway that collects data from all the micro-inverters to provide stats on how my energy each panel produces in real-time.  Each micro-inverter is equipped with a low power Blu-tooth transmitter to communicate with the solar gateway.  The solar gateway has an Ethernet connection to my LAN.

Here's a view of the initial 24 panel ground mount array, showing the power being generated by the individual panels:



So drilling down to this level of detail is great for trouble shooting performance issues with individual panels and/or inverters.

For my location and with my current setup with 128 panels rated at between 260W and 290W, my monthly production should be something like this:



Here's a look at my actual power usage and cost over the last several years:



My first array came online in December 2016, and you can see from the above chart that by March, 2017, I had reached a point where I was producing more power than I used.  So the only thing I had to pay for was the monthly service charge of $31.73.

Then in July 2017, I started mining and it quickly got to the point where my miners were sucking down way more power than I was able to produce.

I recently installed a system that allows me to track both usage and production.  Here's a look at the summary dashboard from this application:



From here I can drill down into various graphs to look at individual stats.

Here's a view of my solar power graph from yesterday:



Some clouds rolled in during the afternoon as you can see from the drop in power levels starting around 2pm.

Here's another view showing my actual kWh production yesterday:



So just under 120 kWh.

Here's a view of all my power consumption yesterday:



So a little under 15 kW plus whatever household appliances happens to be running.

A view of the net power consumption yesterday with the power produced from the solar arrays subtracted:



Since I have a current transformer connected to each circuits that carries miner related loads, I'm able to track consumption down to that level.

Here's a view of the miner dashboard I created.  It shows daily power, in kWh, consumed by each rig (or rigs connected to a given breaker circuits), and well as the power level usage throughout the day:



In order to track power at the circuit level, I had to install energy monitors at each breaker panel.  Here are the monitors installed at the 2 main panels:



I have similar energy monitors installed at the network/server sub-panel, as well as the shop sub-panel where the solar panels are located.

The original shop roof mount array and the ground mount array installed at the same time, provided an 80 panel system capable of producing 20 kW.  For my latitude, that works out to around 2,200 kWh month averaged out over a year.

This was more than enough to cover all my energy needs, which was pretty high to begin with.  But once I started mining mid summer last year, I quickly found my consumption outstripped my production, so I decided to add another 48 panels via a pair of 24 panel arrays.

The panels have arrived.  These are rated at 280w where the previous ones were 260W.  Not a huge difference, but I'm curious how many kWh they will produce in the real world compared to the older panels.



I got the tube framing in place and the rails on the 1st array:



And about 50% done mounting panels:



All done:



For the power back feed, I decided to get a single exterior rated 200A panel, and consolidate all 128 panels to this one location.  I tapped the 400A shop sub-panel meter base as follows.  Note that ground and neutral are kept separate since this is a sub-panel off the main service entrance at the house:



And the 200A tap runs to the right towards the 200A solar sub-panel:



And here's where the 3/0 Cu cable terminate in the solar sub-panel.  The pic was taken before the ground wire was run.

And shot of the dedicated solar sub-panel with the roof and initial ground mount strings being connected to the grid:

The 24 panel ground array consist of panels identical to what is on the roof array.  Being that the array was located on recent back-fill, I decided to play it safe and do a full footer pour.



With the core frame in place, it was time to add the aluminum rails the panel were to rest on:



Panels all in place:



Backside showing the micro inverters at each panel"



You can see this is just before the end-run connectors were wired in.

This 24 panel ground array was broken into 2 12 panel strings, with each string returning to a 240V 20A breaker.

I figured I'd start a thread detailing my 3 stage (so far) built out to what is currently a 128 panel, 32 kw solar array to assist with offloading the cost of my mining operation.

It all started during the summer of 2016, before I got into mining.  I just wanted to generate my own power, but still be connected to the grid via an arrangement known as net metering.  In a nutshell, that means that when I generate more power than I use, my meter runs backwards, and I'm credited at the same rate I'm billed, for any power I put back into the grid.

I have 2 buildings on my property, the main dwelling with a 400A service, and my shop building with a separate 400A service.  When I inquired about going solar and wanting to connect all the panels to the shop building, I was told that was a no no since the shop building used very little power on an annual basis compared to the main dwelling.  So I had to disconnect the service at the shop, and instead, have the house be the single service entrance, which would then have to feed the shop and have the solar farm connected to it.

So in order to accomplish that, I had to first have the power to the shop building terminated.  POCO come out and drop the lines feeding the service to the shop:



I then dug up the 500 MCM Al cables between the pole and the shop, and dug a new trench from the house to the shop, to drop them back into:



What a mess that was...  500 MCM cable is pretty damn stout, btw:



So in order to meet code, I had to get a 400A safety disconnect and since I wanted to have a whole house backup generator, a 400A transfer switch.  The POCO was kind enough to drop off a new 320A meter base (transient rating is 400A), so that I could do all my prep work prior to the cut over.  So here are all 3 pieces prior to installation:



Here's the new meter base and safety disconnect mounted with the existing service just "flapping in the air":



Called the POCO back out, and it took me about 45 minutes from when the cut the power at the pole, to pull the old meter base and replace it with the 400A transfer switch and get everything wired to the point they could restore power:



After the "ground work" inspection, I was then able to back fill the trenches and seed:



So now I could finally prep for installing the first solar array, which consisted of 56 panels mounted on my shop building roof.

Here are the panels and assorted hardware:





I started by installing all the mounting brackets on the roof and then the bottom skirts:



After a few days, I was about half way there (I'm doing all this solo, btw):



6 days later I was down to the last panel:





Now it was time to do the electrical work.  The 56 panels on the roof consists of 4 strings with 14 panel in each.  I use micro inverters that are rated for 250W, so each string can generate up to 3,500 watts @ 240V.  Here's what each panel looks like before mounting on the roof:



So the DC voltage from the panel if fed into the inverter, which in turn puts out 240A AC which is then connected in series with the other panels in each string:

At the end of each string, there is what is know as an "end-run connector".  It simply terminates the AC voltage to 10 gauge wire, which is feed down to a 20A breaker:



All 4 end run connectors are then combined into a single run down into the breaker panel:



And here are the 4 solar panel strings terminated into 20A breakers in the sub-panel:

Working fine for me as well.  I did have that delay this morning for the first time, but it resolved itself within a few hours.

And counting.  I''m still not seeing anything...