Topic: Exhaust temp of a miner - page 2. (Read 673 times)

You can use Specific Heat Capacity

Q = m * C * ∆T

Q = heat transferred (in joules)

m = the mass of the substance (in kilograms)

C  = the specific heat capacity of the substance (in joules per kilogram per degree Celsius)

ΔT/delta T= the change in temperature (in degrees Celsius


m=density×volume , air density at 20C  = 1.2041 volume = 9000m3 so m = 10,836
c= of air is 1005

Delta T will depend on your room temperature and how much you want to raise the temps, let's say the normal room temp is 20c and you to raise it to 25, it would be 25-20 = 5

plug these in and you get

10,836*1005*5 = 54,450,900 joules

Now we need to figure out the time it takes for this to happen in order to find the power in watts needed

W = Q / T

Where W = watts, Q = Specific Heat Capacity , T = time in seconds

You say 5 S21s, it's 18,250W so in order to calculate time we would just inverse the equation and it becomes

T = Q/W
T = 49 minutes

If the outside temp is 5 and you need to hate the place to 25 then

200 mins for that to happen assuming no energy loss at all which is impossible, in fact, given the heat loss, you may never raise the room temp even by 2 degrees with that little heat, which is why real-life experience is crucial here, a professional who knows your country weather, the insulation you use, how often doors/windows are opened and all details that matter will be more accurate than any formula.

If we talk about real-life scenarios, it's 15c outside now, I run an 4KW AC in the living room, it keeps the living room and two rooms next to it pretty warm at 22c, the total area in cubic meters is roughly 300m3, the AC runs like 15 mins an hour, so you can say I use 1KW AC 24 hours to raise the temp in 300m3 area by 7 degrees if your place had the same condition as mine it would require 30KW running constantly 24/7.

I don't know about 35W per cubic meter seems like a lot, unless it's freezing outside and you want the room to be boiling hot, here is a random space heater calculated (based in the UK)

https://www.puravent.co.uk/calculator/space-heater

I plugged in your numbers which are 10*10*90 = 9000m3, with a delta T of 5c, it says 36KW is needed if your room is well insulated, 126KW if it's poorly insulated, which makes a lot of sense.

But with all seriousness, get a professional who would estimate the exact heat you need to operate your business, and then do the math based on that.

I'm sure its from his experience in the business. I asked him but got no replies since last tuesday or so. Same for the 862 value I mentioned in OP.


Hum, I wasn't thinking about it that way. In fact it's a reasonable way of thinking about it but I was thinking in a way of working with lower numbers, to not get fired instantly. lol

I thought of how many S21 I could buy with the same 35k€. Let's say only 5 to have some room for electricity costs and kinda make values tend in favor of the miners!
35k€ / 5k€ = 7 miners. Let's say 5 only.

5 miners running @ 3650Wh = 18250Wh.
Times 24h = 438kW a day
Times 0.18€/kWh = ~80€ a day

But this is just to compare the costs of running 5 miners per day. Although I also need this, I need to also confirm that these 5 miners can heat up the same space in a reasonable amount of time. The 9000m³ of volume. (This woul be controlled by a Y shape duct with a 3D printed valve connected to an Arduino or so, that was controlling temperature and humidity inside the space and the valve would redirect the heat inside or outside the space according to the data gathered by the sensors connected to the Arduino. But this is another subject. It was just to contextualize).
Something like
Miner1 duct --------\            / (out)
Miner2 duct ----------\      /  /
Miner3 duct -----------\---/-/ (valve)
Miner4 duct ----------/      \   \
Miner5 duct --------/              \(in)

About the boiler, he only told me he needs 35w per m³ and showed me the math I mentioned in the OP and that the cheapest biofuel was 30€ per m³. But didn't tell me how much of this biofuel the machine consumes per time frame. That's why we started talking about BTUs.

Where did he bring that 35w from? did you give any numbers to work with? anyway, to calculate how much heat you need to maintain a certain temp in x*y*z area is going to depend on so many factors including insulation, outside temp, and all of that, it would be very difficult to get an accurate figure if this isn't your field.

But now let's say you did so and you came up with 100KW  (i'd ignore BTU, jouls, and every other thing to avoid confusion) just focus on KW to easily compare the two systems (boilers vs miners).

How much does a 400KW boiler cost?  35,000€, how much does 400KW antminer S21s cost ? 400/3.5 = 114 miners * 4600€ = 524,400€  = your boss is going to fire you  , jokes aside.

next step.

Calculate how much it costs you to run miners vs boilers per day.? let's say 1000€ biofuel for 400KW (just throwing a random number coz I have no clue about biofuel prices in your country)

How much it costs to run 400KW worth of S21s? 400*24*0.18 € = 1,728€

So not that it only costs nearly half a million € to buy, it would cost 728€ more per day to run, but, it would make you some good money, how much is that as per today's standard? nearly 2000€, so you are getting only €300 a day (so 1300€ more profit than running boilers), in other words, every day you don't run your boilers, save 1000€ on boiler fees + make 300€ profit.

Using boiler = -365,000 a year
Using miners = +109,500 a year
Net = 474,500 a year, so basically, in just about 14 months you would break even on the half-a-million-dollar investment you made, and going forward you will be getting "free" heating while making 300 euros a day in profit.


Of course, if the boiler costs 10$ a day to run, everything changes, it becomes
Using boiler = -3650 a year
Using miners = +109,500 a year
Net = 113,150 basically 5 years to ROI = terrible business.

You then need to factor in other things like maintenance, it's probably a lot cheaper and easier to maintain the boilers than to maintain the miners, BTC price is highly volatile, it's 300 euros you make today, next month it may be just a 100, or 400, you just can't tell.

There are other technical difficulties in transferring that heat generated by miners, you can't just put them inside whatever enclosure you are trying to heat, they do output air as hot as 90c but the sure thing is, they won't operate in a room above 40c, there are a lot of details that need to be checked, but generally, repurposing miners heat on an industrial level where a lot cheaper methods exist, especially in a country where the power rates are that high -- doesn't work.


Fun fact, if you had infinite airflow in the miner, the exhaust temp would be equal to the intake.

Yeah, I just got it in the meantime.

Now, I need to find a way of demonstrating that using miners, I can achieve the same goal at roughly the same cost and on top generate some passinve income with Bitcoin!

It comes from being the equation determined in the early 1800's for converting watts to BTU... Just Google it, "convert watts to BTU"...

BTU is the standard unit used to measure energy needed to produce a temperature change. You will find that all fuels have a spec of how many BTU per-unit of volume or weight it produces when burned. The other common unit of energy is the Calorie though that is more for chemical reactions and not heating/cooling.

I'll will take the help because I'm here running round and round and the more I look at it more confused I get.
So, I have contacted a manufacturer of biofuel boiler / buerner (?, not sure it's the correct term) and assuming the following data:
Volume to mantain / heat up = 9000m³
Needed power to heat up a m³ = 35w (not sure where this value comes from, the gyu stopped answering my questions)
Needed power to the given volume = 315000W, plus a 20% margin we get 315000 / 862 = 365kW (also not sure why the 862. 20% would be 800. It's another question he didn't answered to me).
They have boilers / ovens capable of around 400kW to 500kW and they cost something around 35k€ up to 45k€.

Electricity price for the common citizen is 18cents/kW (I know people will say that for mining, it's a no go, but there is the extra of the Waste Heat that I want to monetize and recover).

Then the biofuel data I have is the following:
3 types are possible but let's only consider the cheaper one because this will give me the worse case scenario against the miner
m³ of wood chip = 30€ (it was not told me how many m³ would be needed a year, or even in tons - I could convert m³ of wood chip into tons)

Miner data:
Price: 5k€ (let's assume)
Power input: 3650Wh (rated at 3500 but let's assume that higher value just in case)
Fan CFM: 160 =>  4.53m³ per minute
BTU: 10000 (according to @philipma1957, still need to confirm the value of 3.41. No idea where it comes from)

So, I have to find a way to compare the 2 systems somehow.
And I don't need and don't want to account for variables like outside temperature variation, or miner load (if we set it to 100% or 50%, I'll assume 100%).

Another factor I have to account for is life cycle or I should say something like "for how long I can use theses numbers to justify that a miner is at least as good as any of these other conventional system. I would point for something like 4 or 5 years. Not sure if is non-sense or not. I mean, is a S21 still perfectly viable 4 or 5 years after it's first appearence in market or since its inception?


There is also the possibility to add power generation with PV panels, but one thing at a time!

Yes! Computing is a not a form of energy, heat is the product of computing, passing electricity though the components, or transformation of electricity to heat!


Not an issue!


It's actually pretty simple, you're basically comparing heating via biofuels with heating via electricity!
As I said before, it  all depends what you try to achieve and from which part you want to start:
- you want to know how much it costs to heat a room from 40 to 50 Celsius, you calculate the volume of the room, the amount of BTU needed and then turn that into watts and biofuel!
- you want to know how much heat $1 will give you , you convert that into kwh based on the price of 1kwh and the price of one gallon/liter of biofuel.
The only difference is that for each kwh burned the miner will make up a few cents back!

I can't really throw numbers around without knowing what you want to compare and what's the endgame but even if I would not be capable of doing the math there are enough here that could help you, so just give us a rough description of what you have in mind and we'll try to help!

Not sure that's that straigt forward. The miner is rated at 3500W but not all that power is converted to heat, right? Good part of it will go for the computing itself, I guess. Or am I wrong? Can we consider that the total amount of input power is, in the end, converted to heat?

The other issue I have is that the data I have on conventional systems are like in watts produced. But I think there are too many variables. Because if I want to prove that, in terms of costs, a miner can compete with another system, I have to account things like how much time the same volume takes to reach a specific temperature, if the conventional system is working non-stop, how many tons of biofuel it takes, etc!

Wait a minute, if it's heating, why do you this round-about way?
A 3500W miner will burn the same electricity as a 3500W heater and will throw out as much heat as the other!
It's the law of thermodynamics, videocard, asic, oven, heater, the heat generation is the same, you can't destroy energy.

You have a formula for biodiesel, quick search it's ~127,960 Btu/gal, you know the W/BTU formula, all you have to do is factor the cost of both and add the revenue for the miner.

Yeah, heating stuff, drying out stuff...


The goal is to compare how much it costs to heat up an enclosed space using miners and using other conventional heating systems. Namelly in industrial envirnoments where the conventional systems are quite expensive, from what I could already understand. Biofuel systems to be more precise!
So, I need to get some knowledge how fast a miner can heat up a specific volume up to a specific temperature, calculate costs, ROI, etc and then do the same thing for any other conventional system like the ones using biofuels.
If I can prove that, at least, up to some point, a miner can be as efficient as one of these systems for some time, that would be good!

I'm in toutch with a manufacturer of biofuel heating systems but the guy stop answering me last week, so I'm having a bit of hard time getting some data of such systems to be able to compare!

1W= 3.41 BTUs/hr , standard formula.


I'm a bit curious what you plan on doin with it, if it's other than heating stuff.
Probably I don't have to tell you but capturing back energy from that heat in the air is really beyond what one can call just hard to do.

Yes, I understand that there are quite a few factors that will impact the temperature, but if I can have a value with even -/+5ºC of error, I'm ok with it.
Another curious fact is that manufacturer announces a power consumption of 3500Wh and I've seen several times people talking about 3650Wh. That's fine too. I'll do the math always to the worst case scenario.


Thanks. This might help me a lot. I didn't even considered to work with BTUs but now that you mentioned it, maybe this is what I really need to be able to compare with other heating systems.

BTW, where does that value of 3.41 comes from?

yeah btus are what matter.

so 3500 watts =

3500 x 3.41 = 11,935 btus  are coming out of the gear.  not all leave from the fan exhaust.  Just touch the sides of the unit etc.

So pretend 10,000 btus are being pushed out  by the fans at 100%  temp could be 65c

the same 10,000 btus are being pushed out by the fans at 90%       temp could be 70c

and the same 10,000 btus are being pushed out by the fans at 80% temp could be 72c.

temps will be hotter as fan slows because you have the constant 10,000 btus leaving at a slower pace

since the btus are constant and the fans are slower temps rise.

this could be meaningful if you want to dry something out with the hot air.

The temp at exhaust fans would be irrelevant without considering the airflow at the time too. For example if I set my fans to always 100% in the miner settings, the exhaust temperature will be much lower as there is more airflow.

I always think the temperatures feel relatively low, however put an S21 in a medium sized room with the windows closed, and you'll soon realise these put out a LOT of heat. Low temperature but high volumes due to the powerful fans.

Could it be a better approach to try and work out how much of the ~3650W (S21 consumption) is translated into heat energy? I'd estimate 150W on fans and processing, with the remaining 3500W used by the ASICS

I have one of those at work. Difficult part is to convince boss to aquire a miner for testing purposes. heheh
I am running some calculations to try to demonstrate that in specific scenarios, running up to 4 or even 5 miners can be as much profitable as other systems in the context of the project I'm working in, to present to the big boss.
This is nothing new, maybe it would be pioneer in my country because electricity prices are absurd and we have no slack to play with miners at all. So this project I'm working on is on a very tight leash because everything I good I tell about using a miner to other purposes other than exclusivelly for mining can be countered with arguments like electricity, lack of hardware support in case of hardware failure, people's mindset and a few oter factors that I need to account for so that I can produce a enough well documented report to be able to answer all those challenges!

But I'll take those temp values as a starting point for my report!

Well the miner shuts off when chips go to hot about 80-85c

So if you are running hot I would guess 70-75c seems possible right at the fan.

Buy this and test it out.

https://www.amazon.com/KIZEN-Infrared-Thermometer-Gun-Humans/dp/B07VSHR9M6/ref=sr_1_1?

Hello...

I'm doing a small investigation work on a project and I want to include miners as one factor of this project and I need to know the exhaust temperature of a Bitmain Antminer S21.
I know it can depend on many factors such as current power on which the miner is working at, environment temperature, etc, but I need an estimate.
I know that the current operating temperarute of the miner is between 40ºC and 60ºC but I'm not sure I can use this range to use as the exhaust temperature of the miner!

So, if it is possible for anyone to measure the temperature of te exhaust fan near to the exhaust fan itself in an average environment temperature (like not in cold Russia nor in hot Kazakhstan in the summer), could be a value I would use for my investigation project!

A few days ago I found some information (can't remember where) that the air exiting the miner coul reach temperatures up to 60ºC. I'm not sure how accurate (or not) this can be, but knowing that there are many electronic components that withstand temperatures up to 90ºC or even more for at least a small amount of time, I wouldn't be very surprised that 60ºC could mean electronic components inside the miners being at temps around 70ºC or so.

If anyone could help me with this I woul be very appreciated!

Thanks
dark