Help me get started with mining

[Carouso]

Hello, i'm one of the guys trying to start up a little added passive income by mining from home with electricity included in the rent.

I was thinking about a rather small setup of 2 to 4 S9 that i would cool via immersion cooling in 1 or 2 miner masters and maybe upgrade to 2 s17 if i find a bargain after some time.

Now, i know some of you want to rush in judging me for practically stealing electricity, but with the burned skin and ripped muscles and broken bones and low wages from my terrible industry job i don't care anymore and my power consumtion fortunately can't be traced here. Also, as my landlord refuses to grant access adequate heating, i would welcome a heat source that generates money, instead of the fan heater that actually consumer more energy than at least one of the setups i am planning right now.

The main concern i have in starting up mining, though, is my personal safety and the simple prospect to possibly burn down my appartement.
The reason for this is, that permanently taking a power output of around 3300w (two 1600w S9 + 100w miner master) from a regular continental european circuit that maxes out at around 3600w would seem rather unsafe to me.
I have three circuits in my appartement and wanted to use two of those to set up one miner master with two S9 running in it.

Now, what i understand is, that though the PSU of an S9 has 1600w, the S9 itself uses around 1400w, which would make it much more manageable.
At the same time, with the immersion cooling i want to set up, i am, of course aiming to overclock them, wich could increase the power consumption again.
Checking up on this issue, i found out that the power consumption of regular CPUs can actually be lowered by 30% and more when adequately cooled.

So my questions are:
Can i expect the same drop in power consumption with immersion cooling for ASICs like the S9?
Is taking 1400w 24/7 from one continental european household socket safe?
Is taking 2900w 24/7 from one continental european household circuit of 3600w safe?
How much added hashrate can i expect from such a setup?

Why didn't god emperor Trump win the election despite raising the real wages of low wage earners by over 16%? ;_;

Thanks in advance,

Carouso

[NotFuzzyWarm]

ASIC-based miners are very different from typical electronics. They do NOT like to be ran cold. The chips and circuits were designed to run at fairly high temps and running below said temps usually makes them unstable. Keep chip temps at 50C or higher.

World wide, the standard rule of thumb for running a heavily loaded circuit 24x7 is to de-rate it by 20% so a 30A socket is good for 24A continuous load.

[Carouso]

They do NOT like to be ran cold.

Ooowww, god damned! And i thought i got it planned out so well. But thank you for the insight, really.

Where does all the advertising about higher power efficiency from immersion cooling come from?

I guess, i will have to take a week just to test out the best pump/cooling liquid configuration, then.

World wide, the standard rule of thumb for running a heavily loaded circuit 24x7 is to de-rate it by 20% so a 30A socket is good for 24A continuous load.

Yeah, i guess i will stick to that. Interestingly enough, i should be well under that load with my setup, so i'm glad it looks like i still have some room for overclocking. I will still have to do some calculations, though. Weird thing is, that when researching all of this, it turned out that i have allready stressed one of my curcuits WELL beyond the recomendet 3600w without knowing. I suppose this only went well, because of the high voltage that continental european circuits suppot. Still, i am very very anxious about even comming close to the official recomendations.

Thanks again.

[NotFuzzyWarm]

Where does all the advertising about higher power efficiency from immersion cooling come from?

I've never seen that claim regarding miners directly. Since each fan on a miner pulls around 50W and then much larger fans are needed to move air through the area where miners are and exhaust it, using immersion does quickly lower power used for that because you can move and remove heat from a LOT of coolant more efficiently than air cooling using fans alone.

The fact that miners like to run hot does not change using immersion cooling. Because they want to run substantially hotter than ambient room temp it actually means less power (from lower cfm) is needed for the radiator fan. Just set temp control for the fluid to run at whatever temp works best.

[mikeywith]

Can i expect the same drop in power consumption with immersion cooling for ASICs like the S9?

Not really, well it's a tricky question, with immersion cooling you could save some power while on the other hand, you could actually use more power compared to air-cooling.

As far as the hash boards go, they don't care if they are in the water or anywhere else, whichever volts and frequency there is, the same power consumption will apply on all cooling methods, the only direct cut you get will come from the 2 fans on the S9, or the 4 fans on the 17 series, the average consumption of these fans is 5amps at 12v, assuming they run at 100% which is 60w each, so NFW figure of 50 seems perfect since it's unlikely that the fans will go full blast.

On the other hand, you are going to need to move the cooling liquid, if the power required by the pump and/or the radiator fan is < the total fans consumption then you win, vice versa, you lose, so if you use a pump that consumes 200w and use it to cool a single S9, then well, air-cooling wins, if the same pumps can run 10 S9s, then immersion cooling wins, I have no clue about which pumps are available and their power consumptions, I am simply showing you the proper way of doing the math.

You should also consider the initial cost for immersion cooling, it adds up to the gear cost and makes the business model look terrible.

[Carouso]

You should also consider the initial cost for immersion cooling, it adds up to the gear cost and makes the business model look terrible.

Like i've written. I do not pay electricity, which makes it easy for me.
As for the initial cost, when i just contacted suppliers on Alibaba, they jacket up the price, so as of right now, i am going to build my own immersion cooling system. Without ghettoing it, i would just pay around 150€ for the parts of the immersion cooling, including transformator oil and 130€ for one S9 plus around 20€ for cables and other things. 430€ in total, with an expected hashrate of minimum 22 th/s, i have a daily income of 2,77€ for the current price of the coin.

That's a little bit over five months until i paid off the initial investment and after that no variable costs to produce the output.

I'm not sure what you would consider a good business model, but i'm right now learning at uni that, for example, investments in renewable energy that only break even after 7 to 9 years and turn in an interest of 4 to 5% interest rate over the whole runtime of the project are considered good investments.

The question regarding the power consumption is, as i wrote in the opening, really just a safety concern.

[mikeywith]

Like i've written. I do not pay electricity, which makes it easy for me.

So, your electity is free, and your default is ROI is short, but you want to extend it with extra expenses for immersion cooling? I am not being sarcastic, but why "immersion cooling"? air cooling works just fine, do you have a specific reason for why not stick to the default cooling?

That's a little bit over five months until i paid off the initial investment and after that no variable costs to produce the output.

5 months is somehow the best-case scenario, the hashrate is going to increase big time in the coming few months and if bitcoin price doesn't push a lot higher then all these numbers will not be met, please read this post of mine for further understanding of how these mining calculators are far from accurate.

I'm not sure what you would consider a good business model, but i'm right now learning at uni that, for example, investments in renewable energy that only break even after 7 to 9 years and turn in an interest of 4 to 5% interest rate over the whole runtime of the project are considered good investments.

A good investment as far as mining goes is super fast ROI, you should try to reduce the initial cost to overcome the difficulty increment and any potential price drop which leads to the overall extension of ROI, I don't know about renewable energy investment but I know that those solar panels have a warranty of 15-25 years and you are not competing against other people (there is enough sunlight for everyone), with mining, however, there is no guarantee that the used miner you buying is going to mine for 5 months straight, and if the price doesn't spike while difficulty does, the 5 months ROI could very well be 2 years.

Excuse me if I sound mean, but really, you are making a mistake thinking that just because you have free power you can beat the odds, I have free power myself, and I have lost many times, I lost a large amount of money, time and efforts into learning how to stay profitable when it comes to mining, it's nowhere close to the theory that says ( if you have free power, you are guaranteed to win).

With that being said, if this is just a learning process then I would invest in immersion cooling or anything else without doing the math, knowledge is power after all, but if this is a business plan, I would re-run the number twice.

Good luck.

[Carouso]

So, your electity is free, and your default is ROI is short, but you want to extend it with extra expenses for immersion cooling? I am not being sarcastic, but why "immersion cooling"?

There are two main reasons i am doing this for. First is the prospect of being better able to overclock and second,  the much lower noise in my small appartement. I set up two big radiatosr with much more fans, instead of the four strong ones of the two antminers. This is just my first test setup and if it works well, i will plan up two three of those exact same setups.

I very much understand the risks involved and am personally actually hoping for a very sharp price drop over the next two to four years, to wipe out a lot of hashrate. That way, i can mine on a compareatevly low difficulty level for some years and sell the coin off with the next hype. Still, my test results right now with the rising price show me an actual higher income, even with the ever increasing hashrate.

For this turning into a loss for me, we are really talking for very specific scenarios with very specific values that have to occur,  while most situations will at least turn even for me.

So, nah... I'm still gonna do it and probably will build two setups.

[mikeywith]

There are two main reasons i am doing this for. First is the prospect of being better able to overclock and second,  the much lower noise in my small appartement. I set up two big radiatosr with much more fans, instead of the four strong ones of the two antminers. This is just my first test setup and if it works well, i will plan up two three of those exact same setups.

Well, now it makes sense, see I don't have a problem with immersion cooling (except for the initial setup cost), but if it's done for the purpose of noise reduction, then it makes sense, keep in mind that you will need a custom firmware (usually 3% fees) or fan simulators to be able to run the miners without a fan, too early for this information, but you will eventually get there.

... my test results right now with the rising price show me an actual higher income, even with the ever increasing hashrate.

I would love to see your test results, mining profitability is in a downtrend, it has been since the beginning, so on average your profit today is better than that of tomorrow.



https://bitinfocharts.com/comparison/bitcoin-mining_profitability.html

There are exceptions to this trend of course, but overall, 90% of gears I owned throughout the years have always made the most profit when I have first got them.