First power bill for my 6 GH/s rig

[rjk]

There's that "objective discussion" again.    They don't want to learn, they just want their wrong opinion to be right.

Teach me, Master. Tell me the way to cool a rig with 100 degree summer air, WITHOUT making more noise than the fans being at 50%, and without pulling a significant number of extraneous watts.

[sveetsnelda]

Oh, for sure.  I love educating others and sharing knowledge while they treat me like shit.  Let me get right on that.   

[rjk]

Oh, for sure.  I love educating others and sharing knowledge while they treat me like shit.  Let me get right on that.   

Pretty sure you aren't going to be able to come up with a way, but if you can I will certainly retract any statement that I made earlier that is invalid or incorrect.

[chiropteran]

There's that "objective discussion" again.    They don't want to learn, they just want their wrong opinion to be right.

Teach me, Master. Tell me the way to cool a rig with 100 degree summer air, WITHOUT making more noise than the fans being at 50%, and without pulling a significant number of extraneous watts.

I don't think anyone claimed the fans would be silent.  Nobody mentioned the amount of "extraneous watts".

It was simply a question of can you do it, yes or no.  The answer is yes, you can.  Can rjk add some new restrictions or special situations to create an artificial situation where air cooling may not work, sure.  But you are answering a different question at that point, one which is irrelevant to the original thread.

It looks like you realize you have lost the original argument, so you are moving the goalposts.

[rjk]

There's that "objective discussion" again.    They don't want to learn, they just want their wrong opinion to be right.

Teach me, Master. Tell me the way to cool a rig with 100 degree summer air, WITHOUT making more noise than the fans being at 50%, and without pulling a significant number of extraneous watts.

I don't think anyone claimed the fans would be silent.  Nobody mentioned the amount of "extraneous watts".

It was simply a question of can you do it, yes or no.  The answer is yes, you can.  Can rjk add some new restrictions or special situations to create an artificial situation where air cooling may not work, sure.  But you are answering a different question at that point, one which is irrelevant to the original thread.

It looks like you realize you have lost the original argument, so you are moving the goalposts.

Does anyone know the reason that most datacenters specify that ambient air be between 68 and 75 degrees F? Anyone? Don't disappoint me now.

If I wish to cool my rigs to an acceptable temperature with a low outside ambient temp, I use an open window and no additional fans, or perhaps a small fan to direct the air to an appropriate location. If I wish to do the same in the summer, I need massive amounts of air flowing, which requires noise and power. And even then, the cards are not going to remain within a tolerance that I wish to see.

You can cool your rigs with ambient air at close to 90 degrees and low GPU fan speed - OK, what is the core temperature? 75, 80, 85 degrees C?

[PulsedMedia]

Hey guys! I know how to keep my miners within a proper temperature at 100+ degrees Fahrenheit in the summer! I just need one of these per rig:



No actually fuck that, do you have any idea how many watts those things pull? And how loud? Damn.

Depends on model, but i actually have one with over 2kW rating >

[PulsedMedia]

Teach me, Master. Tell me the way to cool a rig with 100 degree summer air, WITHOUT making more noise than the fans being at 50%, and without pulling a significant number of extraneous watts.

Honestly thinking about that, you can get "semi-free" airflow by utilizing the heat ... Yes, utilizing the heat generated
So assume you have a tent, say yurt / pyramid shaped.
The shape of "roof" is like a funnel, constantly accelerating the air flow, on top you have say a 500x500mm hole with that sized fan, but very low wattage, say 15-30W
Below you have an open air opening.
Because heat tends to go upwards, and BIG fans tend to be more efficient moving air around, you are going to get a decent air flow amount ...

Just saying ...

[DeathAndTaxes]

Wasn't the claim that one could cool 20KW of heat load with a pair of window fans in 100F ambient temps?

Funny how the topic has changed to custom building design, 3 square foot holes in the roof, 2000 watt blowers, datacenters in sweden, "anything can be cooled with enough airflow", how houses get cold when you open the window in winter, and other distractions.

What happened to 20KW heat load, pair of window fans and 100F ambient temps?  

[PulsedMedia]

What happened to 20KW of heat load, pair of window fans and 100F ambient temps? 

Who knows .... Wait what load? XD

I thought that was achieved with evaporative cooling?

[phorensic]

Wasn't the claim that one could cool 20KW of heat load with a pair of window fans in 100F ambient temps?

Funny how the topic has changed to custom building design, 3 square foot holes in the roof, 2000 watt blowers, datacenters in sweden, "anything can be cooled with enough airflow", how houses get cold when you open the window in winter, and other distractions.

What happened to 20KW heat load, pair of window fans and 100F ambient temps?  

Best summary of what happened, IMO.

[chiropteran]

Does anyone know the reason that most datacenters specify that ambient air be between 68 and 75 degrees F? Anyone? Don't disappoint me now.

This isn't a data-center.  The beauty of outdoor cooling is that you don't need the ridiculous density of a datacenter.  You can spread things out, you don't need to stuff 20 video cards into a single tower.  Also, given this is just a hobby or personal business for most of us, we are not restricted by customer contractual demands.  Reliability isn't nearly as important or critical.

If I wish to cool my rigs to an acceptable temperature with a low outside ambient temp, I use an open window and no additional fans, or perhaps a small fan to direct the air to an appropriate location. If I wish to do the same in the summer, I need massive amounts of air flowing, which requires noise and power. And even then, the cards are not going to remain within a tolerance that I wish to see.

Like I said, moving the goalposts.  You can cool your rigs with outside air, even in the summer.  Massive amounts of air flowing?  Whatever.  It's all relative.  If that "massive" airflow is cheaper than running the AC 24/7, it's a net gain.  Noise?  Power?  Absoklute power is irrelevant, what matters is the relative power.  If it's less power to cool via blowing air than it is to run an AC unit, the power usage is a gain, not a negative. "A tolerance that I wish to see", lovely.  So it's not even about provable facts anymore, if you decide the numbers aren't good enough you get to win the argument for free?  Nice try, but no.  If the cards work reliably enough than who cares what rfk thinks is the correct temperature.  Your wishes are not important in this discussion.

You can cool your rigs with ambient air at close to 90 degrees and low GPU fan speed -

Is that a small spark of intelligence?  Are you finally starting to understand how heating and cooling works?  If it's possible to cool down to 60C at 90F, how much harder do you think it is to cool to 80C at 110F?

OK, what is the core temperature? 75, 80, 85 degrees C?

Who cares?  He said it could be done, he didn't specify any particular temperature other than the general assumption that the cards were functioning.

Funny how the topic has changed ..."anything can be cooled with enough airflow",

The topic hasn't changed.  That was from the original post that started this discussion.   Nice try though.

[minero1]

bad things happen when you think temperature is heat,

harder to cool anything with lower deltas? of course, but not impossible

swallow it: IT CAN BE DONE.

[rjk]

This isn't a data-center.  The beauty of outdoor cooling is that you don't need the ridiculous density of a datacenter.  You can spread things out, you don't need to stuff 20 video cards into a single tower.  Also, given this is just a hobby or personal business for most of us, we are not restricted by customer contractual demands.  Reliability isn't nearly as important or critical.

You haven't bothered to look at my sig, have you? Density is key and king. Reliability is damn important as well. You might be in it as a hobby, but some folks like to make a business of it. No problem with that.

If I wish to cool my rigs to an acceptable temperature with a low outside ambient temp, I use an open window and no additional fans, or perhaps a small fan to direct the air to an appropriate location. If I wish to do the same in the summer, I need massive amounts of air flowing, which requires noise and power. And even then, the cards are not going to remain within a tolerance that I wish to see.

Like I said, moving the goalposts.  You can cool your rigs with outside air, even in the summer.  Massive amounts of air flowing?  Whatever.  It's all relative.  If that "massive" airflow is cheaper than running the AC 24/7, it's a net gain.  Noise?  Power?  Absoklute power is irrelevant, what matters is the relative power.  If it's less power to cool via blowing air than it is to run an AC unit, the power usage is a gain, not a negative. "A tolerance that I wish to see", lovely.  So it's not even about provable facts anymore, if you decide the numbers aren't good enough you get to win the argument for free?  Nice try, but no.  If the cards work reliably enough than who cares what rfk thinks is the correct temperature.  Your wishes are not important in this discussion.

Do you know what causes a lack of reliability? That's right, heat. 80 C is far too hot to be sufficiently reliable for me, and 60 is just right. "Just working" isn't good enough. Working for years without failure is what is necessary.

You can cool your rigs with ambient air at close to 90 degrees and low GPU fan speed -

Is that a small spark of intelligence?  Are you finally starting to understand how heating and cooling works?  If it's possible to cool down to 60C at 90F, how much harder do you think it is to cool to 80C at 110F?

I have yet to see evidence of actual rigs being cooled to 60C at 90F, and as I said above 80C is completely unacceptable. Since I can't control the temperature in the summer, I may even need to maintain 60C at 110F. Using an existing air conditioning system and making it run continuously is actually more efficient than installing powerful blowers for outside air cooling, since part of the A/C isn't used for this application anyway.

OK, what is the core temperature? 75, 80, 85 degrees C?

Who cares?  He said it could be done, he didn't specify any particular temperature other than the general assumption that the cards were functioning.

I care. I want my cards to last forever. Is that too much to ask?

[RandyFolds]

Cool, bro. You can argue semantics and provide poor analogies.

How about you just explain how 'two window fans' provide an adequate volume of air to compensate for the delta-t of 40F. You do this for a living; I'd love to see the calculations.

It's a delta-t of 65.2F.  I don't run my air-cooled GPUs at 60C.  I'd like the fans to last more than a few months and I don't need the GPUs to keep running for decades.

I'm not doing your homework.  Judging by your previous posts, you'd turn mine into a conspiracy theory.  If you knew enough about the subject to understand the calculations one would post, you'd realize that a temperature differential of 65F isn't really THAT difficult to work with when you have a decent supply of fresh air.  I'm not showing you my birth certificate either.

I grow dope and fish for a living, bro. I have no idea what volume of cross-flow you would need to dissipate 21.5kW of heat with 100F ambient temperatures, nor how to calculate it.

I can say that I have dealt with ~25kw of indoor lighting, and it took a six-ton compressor working it's ass off and fans that no one on earth would call 'window fans' even when the intake temps were in the sixties(F).

Cannabis and fish together or separately?  I'm trying to get my aquaponics setup going right now, but it will just grow leafy greens and maybe peppers.  Hard to find edible fish to grow around here too, so I just have some hobby style fish.

I have set up a cannabis aquaponics system, but it's too finicky, and requires too much space. Peppers get tall and kinda bulky. You will get a lot more productivity sticking with herbs and greens. Where do you live that you can't get edible fish? Tilapia are pretty easy to snag except in Florida, and they are completely fucking indestructible.



It blows my fucking mind how many idiots are in here arguing that two lasko 20" fans can move enough volume to sufficiently cool a 21.5kw heat source to ~70C with 100F intake air. Chiropteran, you don't seem to understand the claims at hand, nor is there any moving of the goalpost. There is no AC or massive airflow involved. Just two window fans, allegedly.

How about sveets just nuts up and posts one (1) photograph of this 21.5kw farm of his, clearly demonstrating the two window fans?

[chiropteran]

You haven't bothered to look at my sig, have you? Density is key and king. Reliability is damn important as well. You might be in it as a hobby, but some folks like to make a business of it. No problem with that.

Again, you are changing the argument to get the results you want.  Nobody said you could cool a super dense mining cluster in 110F summer heat.  (Nobody said you can't either, but that isn't the point).  If that is your situation, that you consider density the most important aspect even if it means you are throwing away efficiency and value, so be it. However, your personal limitations don't apply to the rest of the world.

Do you know what causes a lack of reliability? That's right, heat. 80 C is far too hot to be sufficiently reliable for me, and 60 is just right. "Just working" isn't good enough. Working for years without failure is what is necessary.

It's really not.  Reliability in mining is largely a cost benefits analysis.  All things equal, a more reliably setup is better, sure.  But all things are not equal.  There is a point where the money you are spending to keep your cards at 60 instead of 70 or 80 is greater than the cost of simply replacing or upgrading failing cards or systems.  I'm not about to do the calculations here, but the point is an arbitrary statement that 60C is best isn't true, unless you can back it up.  The guy who runs cards at 70C but saves $500/month on cooling bills could well come out ahead of you in the long run, even if he has to replace cards more often.

I have yet to see evidence of actual rigs being cooled to 60C at 90F, and as I said above 80C is completely unacceptable. Since I can't control the temperature in the summer, I may even need to maintain 60C at 110F. Using an existing air conditioning system and making it run continuously is actually more efficient than installing powerful blowers for outside air cooling, since part of the A/C isn't used for this application anyway.

I've yet to see any evidence at all from you.  How do I know you aren't actually a monkey?  I seriously doubt that running AC 24/7 uses less electricity than running fans 24/7, especially given the fact that AC system uses fans itself, in addition to all the electricity required for the compressors.  Not going to take your word for it, sorry.  Back it up with some proof.

I care. I want my cards to last forever. Is that too much to ask?

You can have what you want, but the rest of us don't have to play by your rules.  I think it's silly to try to baby your hardware that much, since the constant progress in technology means your cards are going to be slow and obsolete compared to the new things in a few years anyway.  I could care less if my cards fail in 2 years, if it means I can save some money on cooling.  FPGA will probably have completely replaced video card mining by then anyway

[chiropteran]

It blows my fucking mind how many idiots are in here arguing that two lasko 20" fans can move enough volume to sufficiently cool a 21.5kw heat source to ~70C with 100F intake air. Chiropteran, you don't seem to understand the claims at hand, nor is there any moving of the goalpost. There is no AC or massive airflow involved. Just two window fans, allegedly.

How about sveets just nuts up and posts one (1) photograph of this 21.5kw farm of his, clearly demonstrating the two window fans?

Where was this claim?

edit: You mean this post?  https://bitcointalk.org/index.php?topic=67826.msg797258#msg797258

What exactly is "I used a small evap cooler last year during the summer because my current layout ends up recirculating a little of the air from one rig to the next." if not an AC?

What does "standalone small household fan for each rig and 2 window fans" mean to you?  1 fan PER RIG plus window fans isn't "just two window fans".  A window fan could potentially be a lot of different things, you don't have any idea how powerful it is or how much airflow it is producing.

[DeathAndTaxes]

It blows my fucking mind how many idiots are in here arguing that two lasko 20" fans can move enough volume to sufficiently cool a 21.5kw heat source to ~70C with 100F intake air. Chiropteran, you don't seem to understand the claims at hand, nor is there any moving of the goalpost. There is no AC or massive airflow involved. Just two window fans, allegedly.

How about sveets just nuts up and posts one (1) photograph of this 21.5kw farm of his, clearly demonstrating the two window fans?

Where was this claim?

edit: You mean this post?  https://bitcointalk.org/index.php?topic=67826.msg797258#msg797258

What exactly is "I used a small evap cooler last year during the summer because my current layout ends up recirculating a little of the air from one rig to the next." if not an AC?

What does "standalone small household fan for each rig and 2 window fans" mean to you?  1 fan PER RIG plus window fans isn't "just two window fans".  A window fan could potentially be a lot of different things, you don't have any idea how powerful it is or how much airflow it is producing.

Evaporative cooler != AC.

Funny you spent 10 posts defending it and now pages into the thread decide to actually read what you were defending.  You say people are trolling because they are blasting a claim of using 2 window fans to cool ultra high density and saying "that isn't what was claimed" and now you realize you have no idea what was claimed because you didn't read it until just now.  Fans on rigs aren't going to get rid of heat.  As an experiment shut all the doors and windows to a room, crank up 21KW of space heaters and use all the fans you want.  Let me know how you plan to keep the temps below 120F+.

Instead of saying "hey I was an ass I had no idea what was going on" you decide to hold the line on the definition of "window fans".

LOLZ.     Unless he has a 2 meter by 2 meter window one would expect window fan is 20" or less.  Since 9 square feet windows are kind rare one would think that unless someone was trying to hide something would explain that.  i.e. "I can cool it with 2 window fans but then again they are 72" across. j/k "  My any normal persons definition a window fan without clarifiction would be a fan one can fit in a window.  20" maybe 36" across. 

The funny thing is the author of the asinine claim has backed away (not refuting it but certainly not defending it).  You just kept going on and on calling people trolls and saying they were changing the topic, defending something that you hadn't even read.

READ MORE &  RANT LESS.  

[RandyFolds]

It blows my fucking mind how many idiots are in here arguing that two lasko 20" fans can move enough volume to sufficiently cool a 21.5kw heat source to ~70C with 100F intake air. Chiropteran, you don't seem to understand the claims at hand, nor is there any moving of the goalpost. There is no AC or massive airflow involved. Just two window fans, allegedly.

How about sveets just nuts up and posts one (1) photograph of this 21.5kw farm of his, clearly demonstrating the two window fans?

Where was this claim?

edit: You mean this post?  https://bitcointalk.org/index.php?topic=67826.msg797258#msg797258

What exactly is "I used a small evap cooler last year during the summer because my current layout ends up recirculating a little of the air from one rig to the next." if not an AC?

What does "standalone small household fan for each rig and 2 window fans" mean to you?  1 fan PER RIG plus window fans isn't "just two window fans".  A window fan could potentially be a lot of different things, you don't have any idea how powerful it is or how much airflow it is producing.

An evaporative cooler is not a compressor AC. One fan per rig is circulating air within the space, the two window fans provide cross-ventilation. You really can't see a difference between the two?

A window fan is a window fan...a 14" MaxFan would be referred to as a 'window fan' by exactly zero people on this planet, and that's the kind of volume he needs to be moving. As I said above, I have extensive experience dissipating heat from lighting setups. You cannot evacuate 21.5kw of heat in a closed space with two window fans when the intake is at 100F.

[chiropteran]

The funny thing is the author of the asinine claim has backed away (not refuting it but certainly not defending it).  You just kept going on and on calling people trolls and saying they were changing the topic, defending something that you hadn't even read.

READ MORE &  RANT LESS.  

Uh, no.  The claims I have been arguing against and the things I have been "defending" didn't include this post.  This is the first time I've responded to a post talking about that specific situation. 

For all I know the exact situation could be BS, which I why I asked for more details, but it doesn't change the basic point- can you cool a GPU with sufficient airflow in 110F air?  Yes you can.

[DeathAndTaxes]

The funny thing is the author of the asinine claim has backed away (not refuting it but certainly not defending it).  You just kept going on and on calling people trolls and saying they were changing the topic, defending something that you hadn't even read.

READ MORE &  RANT LESS.  

Uh, no.  The claims I have been arguing against and the things I have been "defending" didn't include this post.  This is the first time I've responded to a post talking about that specific situation.  

For all I know the exact situation could be BS, which I why I asked for more details, but it doesn't change the basic point- can you cool a GPU with sufficient airflow in 110F air?  Yes you can.

And?  You jumped in when people claimed the setup described above was implausible.  

It would be like:
Kid #1: I can go to the moon for $8.
Various Kids:  Impossible, not even close, etc
chiropteran:  Hey trolls it is possible.
Various kids: WTF?
(50 worthless post later)
chiropteran:  See I was right.  You can get to the moon given ENOUGH money.

In related news chiropteran most of the time you can (fill in blank) given enough (fill in blank).  It is kinda useless claim.

but it doesn't change the basic point- can you cool a GPU with sufficient airflow in 110F air?

That was never the basic claim in anyones mind except yours.  Generally that is called a strawman. Make up your own weak defense so you can defeat it.

Congrats you winz the interents.