Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the temperature)?
Also, is there any advantage to minimizing the hysteresis between
the ACTUAL operating temperature extremes in such a control strategy
(given that lower hysteresis usually comes at an increased cost)?
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the temperature)?
Also, is there any advantage to minimizing the hysteresis between
the ACTUAL operating temperature extremes in such a control strategy
(given that lower hysteresis usually comes at an increased cost)?
On 01/07/2024 02:14, Don Y wrote:
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the
temperature)?
There can be for some high performance low level OPamps. Deliberately running them as cold as is allowed helps take the LF noise floor down and by more than
you would predict from Johnson noise. ISTR there was a patent for doing this back in the 1980's. Prior to that they tended to heat the front end to obtain temperature stability and low drift.
https://ppubs.uspto.gov/dirsearch-public/print/downloadPdf/4883957
Made possible with the advent of decent solid state TECs.
Also, is there any advantage to minimizing the hysteresis between
the ACTUAL operating temperature extremes in such a control strategy
(given that lower hysteresis usually comes at an increased cost)?
Depends how temperature sensitive the thing is that you are protecting. The example I recall they were aiming for medium term stable 6 sig fig measurements
with the lowest possible noise.
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the >temperature)?
On Sun, 30 Jun 2024 18:14:32 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:
Assuming you can keep a device in its "normal operating (temperature) >>range", how advantageous is it (think MTBF) to drive that ambientIf all you're thinking of is MTBF, adding the complexity of an active
down? And, is there a sweet spot (as there is a cost to lowering the >>temperature)?
cooling element is a big step in the wrong direction for the system.
Reducing the thermal impedance of the source, to ambient is the
usual way to go, when addressing a specific aging factor.
https://ve3ute.ca/2000a.html
If you're thinking of performance, It's cheaper and more reliable
to concentrate on reducing the temperature of the point source, not
the rest of the planet.
RL
On 01/07/2024 02:14, Don Y wrote:
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the
temperature)?
There can be for some high performance low level OPamps. Deliberately
running them as cold as is allowed helps take the LF noise floor down
and by more than you would predict from Johnson noise. ISTR there was a patent for doing this back in the 1980's. Prior to that they tended to
heat the front end to obtain temperature stability and low drift.
https://ppubs.uspto.gov/dirsearch-public/print/downloadPdf/4883957
Made possible with the advent of decent solid state TECs.
Also, is there any advantage to minimizing the hysteresis between
the ACTUAL operating temperature extremes in such a control strategy
(given that lower hysteresis usually comes at an increased cost)?
Depends how temperature sensitive the thing is that you are protecting.
The example I recall they were aiming for medium term stable 6 sig fig measurements with the lowest possible noise.
On Sun, 30 Jun 2024 18:14:32 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the
temperature)?
If all you're thinking of is MTBF, adding the complexity of an active
cooling element is a big step in the wrong direction for the system.
Reducing the thermal impedance of the source, to ambient is the
usual way to go, when addressing a specific aging factor.
https://ve3ute.ca/2000a.html
If you're thinking of performance, It's cheaper and more reliable
to concentrate on reducing the temperature of the point source, not
the rest of the planet.
RL
On 2024-07-01 07:24, Martin Brown wrote:
On 01/07/2024 02:14, Don Y wrote:
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the
temperature)?
There can be for some high performance low level OPamps. Deliberately
running them as cold as is allowed helps take the LF noise floor down
and by more than you would predict from Johnson noise. ISTR there was
a patent for doing this back in the 1980's. Prior to that they tended
to heat the front end to obtain temperature stability and low drift.
BITD you tended to get popcorn noise from ions migrating around the
surface and in deposited (rather than thermal) oxide. Cooling helped
that a lot. Nowadays processes are generally clean enough that you
don't get a lot of mobile ions.
https://ppubs.uspto.gov/dirsearch-public/print/downloadPdf/4883957
Made possible with the advent of decent solid state TECs.
Also, is there any advantage to minimizing the hysteresis between
the ACTUAL operating temperature extremes in such a control strategy
(given that lower hysteresis usually comes at an increased cost)?
Depends how temperature sensitive the thing is that you are
protecting. The example I recall they were aiming for medium term
stable 6 sig fig measurements with the lowest possible noise.
You don't want to use a thermostat with TECs anyway--they die very
rapidly, especially the soft-solder ones (Laird/Melcor).
On 7/1/2024 7:34 AM, legg wrote:
On Sun, 30 Jun 2024 18:14:32 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the
temperature)?
If all you're thinking of is MTBF, adding the complexity of an active
cooling element is a big step in the wrong direction for the system.
Shifting the reliability burden into a (relatively) low-tech, ubiquitous >subsystem allows failures to be maintained by "non-technical" people.
It also allows for easier redundancy -- you can add another AHU "in parallel" >with an existing unit a lot easier than redesigning the electronic
system to be more reliable over a larger operating range of temperatures.
Reducing the thermal impedance of the source, to ambient is the
usual way to go, when addressing a specific aging factor.
If ambient approaches the limits of the design, then what?
You design something to be able to operate at 50C (on paper).
It gets deployed *at* 50C. What sort of failure rate do you expect
at that elevated temperature vs. operating that same piece of kit
at 30C by introducing active cooling? (the assumption being that said >cooling can be maintained/repaired by a local run-of-the-mill agency)
The question tries to address that issue -- and, the consequences of
how "well" you strive to maintain a "better" operating environment.
E.g., cooling the environment to 10C and then letting it creep back
up to 50C before repeating the cycle would be different than keeping
the device "at" 30C.
Why set a cold aisle temperature of 20C and not 30C? 40C? Why not
operate the devices at their specified ambient limits?
Continuous Operation: 10C to 35C, 10% to 80% relative humidity (RH).
10% of annual operating hours: 5C to 40C, 5% to 85%RH. 1% of annual
operating hours: -5C to 45C, 5% to 90%RH.
I.e., the cited device CAN operate at 45C. But, at what cost
(reliability)?
https://ve3ute.ca/2000a.html
If you're thinking of performance, It's cheaper and more reliable
to concentrate on reducing the temperature of the point source, not
the rest of the planet.
RL
On Mon, 01 Jul 2024 10:34:46 -0400, legg <legg@nospam.magma.ca> wrote:
On Sun, 30 Jun 2024 18:14:32 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:
Assuming you can keep a device in its "normal operating (temperature) >>>range", how advantageous is it (think MTBF) to drive that ambientIf all you're thinking of is MTBF, adding the complexity of an active >>cooling element is a big step in the wrong direction for the system.
down? And, is there a sweet spot (as there is a cost to lowering the >>>temperature)?
Reducing the thermal impedance of the source, to ambient is the
usual way to go, when addressing a specific aging factor.
https://ve3ute.ca/2000a.html
If you're thinking of performance, It's cheaper and more reliable
to concentrate on reducing the temperature of the point source, not
the rest of the planet.
RL
Tubes? The cathodes fail eventually. Reduce filament voltage and
suffer the reduced gain. Better yet, don't use tubes.
But for most parts that dissipate power, the big win is to have some
air flow. A fan can reduce the theta of your parts by 2:1.
Nowadays, parts are very good, with failure rates in the ballpark of
one failure per billion hours, the Bellcore and MIL217 FITS numbers.
On 7/1/2024 4:24 AM, Martin Brown wrote:
On 01/07/2024 02:14, Don Y wrote:
Assuming you can keep a device in its "normal operating (temperature)
range", how advantageous is it (think MTBF) to drive that ambient
down? And, is there a sweet spot (as there is a cost to lowering the
temperature)?
There can be for some high performance low level OPamps. Deliberately running
them as cold as is allowed helps take the LF noise floor down and by more than
you would predict from Johnson noise. ISTR there was a patent for doing this >> back in the 1980's. Prior to that they tended to heat the front end to obtain
temperature stability and low drift.
https://ppubs.uspto.gov/dirsearch-public/print/downloadPdf/4883957
Made possible with the advent of decent solid state TECs.
We don't design products (industrial/consumer) that are "finicky" -- as
it leads to higher TCOs. You don't want to need to control the environment >*or* have "skilled tradesmen" on staff to maintain/assure correct >performance.
The most common example (that I can think of) where temperature is
controlled FBO the electronics would be datacenters. But, from the
research I've done, there, they simply set a desired temperature for
the cold aisle and largely ignore the resulting hot aisle temperature
(except to ensure it doesn't climb out-of-bounds). I.e., they
don't close the loop on the hot aisle to control the cold aisle's
setpoint (cascaded control).
And, they don't get the cold aisle "as cold as possible" so they
acknowledge there are diminishing returns in doing so -- likely
cheaper to just pan on a (potentially) shorter upgrade cycle
than to waste electricity trying to eek out a bit more life.
Interestingly, I can't find anything other than "lore" to
explain why a *particular* cold aisle temperature is chosen.
Amusing to see how much folks DON'T know about the science
they apply!
When I designed my disk sanitizer, I did a fair bit of research
regarding temperature effects on drives -- because we process a
shitload (thousands) of *used* drives, annually and you don't want to
reuse a drive that has an increased chance of failure (based on
its previous environment, SMART data or observations while exercising
it). The old "10 degree C" saw proved to be totally inappropriate,
*there*.
OTOH, I suspect it *is* worth noting for power supplies (as I
see most failures being toasted power supplies in otherwise
"healthy" products). I suspect power *cycling* is a culprit, there
as I've seen failed solder joints where it looked like repeated
thermal expansion had led to the failure.
Also, is there any advantage to minimizing the hysteresis between
the ACTUAL operating temperature extremes in such a control strategy
(given that lower hysteresis usually comes at an increased cost)?
Depends how temperature sensitive the thing is that you are protecting. The >> example I recall they were aiming for medium term stable 6 sig fig measurements
with the lowest possible noise.
I've needed to control temperature in applications where it was
key to the *process* being controlled. E.g., monitoring exhaust
air temperature to determine the "state" of the bed and a cascade
loop on the inlet air handler to drive that to a desired state.
But, there, you have lots of money for the equipment and can buy >good/precise/fast control with things like face-and-bypass as
the primary controlled variable (so the control loop for the
heater/chiller can be cruder and more energy efficient).
"In the small", refrigeration is the only practical means of
lowering ambient temperatures. And, that adds to operating costs.
If you can tolerate a wider deadband then the cooling cost
can be lower (e.g., cool to X degrees and let it *soak*, there,
before letting it warm to Y degrees instead of foolishly
trying to maintain the environment at some Z>X and <Y).
As you likely have LESS ability to precisely size the HVAC
to fit such a small load, deadband becomes a key consequence
of that selection process.
[Gotta wonder why data centers in northern latitudes don't
exploit outside air more agressively during the winter
months!]
What's the mtbf of a fan? a compressor? a pump?
. . . . or a clamp and a block of aluminum?
Ambient and component temperatures are freely obtained and carefully controlled elements in mtbf documentation recording methods.
The former requires $$ equipment.
RL
[Gotta wonder why data centers in northern latitudes don't
exploit outside air more agressively during the winter
months!]
What are the HVAC costs in data processing and server facilities?
That's just to maintain ambient <40C.
On Mon, 01 Jul 2024 07:46:52 -0700, john larkin
<jlarkin_highland_tech> wrote:
On Mon, 01 Jul 2024 10:34:46 -0400, legg <legg@nospam.magma.ca> wrote:
On Sun, 30 Jun 2024 18:14:32 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:
Assuming you can keep a device in its "normal operating (temperature) >>>>range", how advantageous is it (think MTBF) to drive that ambientIf all you're thinking of is MTBF, adding the complexity of an active >>>cooling element is a big step in the wrong direction for the system.
down? And, is there a sweet spot (as there is a cost to lowering the >>>>temperature)?
Reducing the thermal impedance of the source, to ambient is the
usual way to go, when addressing a specific aging factor.
https://ve3ute.ca/2000a.html
If you're thinking of performance, It's cheaper and more reliable
to concentrate on reducing the temperature of the point source, not
the rest of the planet.
RL
Tubes? The cathodes fail eventually. Reduce filament voltage and
suffer the reduced gain. Better yet, don't use tubes.
But for most parts that dissipate power, the big win is to have some
air flow. A fan can reduce the theta of your parts by 2:1.
Nowadays, parts are very good, with failure rates in the ballpark of
one failure per billion hours, the Bellcore and MIL217 FITS numbers.
This was an example of a demonstrated and documented failure mode
in a specific component (glass electrolysis) that is/was largely
ignored by the general user.
If you know what the specific aging mechanism is that you're
trying to address, your methods of improving mtbf will be more
effective.
RL
On 7/2/2024 7:30 AM, legg wrote:
What's the mtbf of a fan? a compressor? a pump?
. . . . or a clamp and a block of aluminum?
As long as it isn't significantly worse than the impact of NOT
having it, you don't care -- because some (relatively unskilled)
local contractor can fix those things. You don't have to
hire a skilled member of staff to be on-hand to deal with the
"more sophisticated" technology's potential failures.
I'd much rather have an HVAC guy come in and repair the AHU in
the datacenter -- even if it was an annual event -- than have
to risk servers crashing or having to be replaced (and the
data recovered). The former is a "cheap", ubiquitous skillset;
the latter considerably costlier and critical.
On Tue, 02 Jul 2024 10:24:58 -0400, legg <legg@nospam.magma.ca> wrote:
On Mon, 01 Jul 2024 07:46:52 -0700, john larkin
<jlarkin_highland_tech> wrote:
On Mon, 01 Jul 2024 10:34:46 -0400, legg <legg@nospam.magma.ca> wrote:
On Sun, 30 Jun 2024 18:14:32 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:
Assuming you can keep a device in its "normal operating (temperature) >>>>>range", how advantageous is it (think MTBF) to drive that ambient >>>>>down? And, is there a sweet spot (as there is a cost to lowering the >>>>>temperature)?If all you're thinking of is MTBF, adding the complexity of an active >>>>cooling element is a big step in the wrong direction for the system.
Reducing the thermal impedance of the source, to ambient is the
usual way to go, when addressing a specific aging factor.
https://ve3ute.ca/2000a.html
If you're thinking of performance, It's cheaper and more reliable
to concentrate on reducing the temperature of the point source, not
the rest of the planet.
RL
Tubes? The cathodes fail eventually. Reduce filament voltage and
suffer the reduced gain. Better yet, don't use tubes.
But for most parts that dissipate power, the big win is to have some
air flow. A fan can reduce the theta of your parts by 2:1.
Nowadays, parts are very good, with failure rates in the ballpark of
one failure per billion hours, the Bellcore and MIL217 FITS numbers.
This was an example of a demonstrated and documented failure mode
in a specific component (glass electrolysis) that is/was largely
ignored by the general user.
If you know what the specific aging mechanism is that you're
trying to address, your methods of improving mtbf will be more
effective.
RL
Given non-junk products from you-know-where, most electronics failures
are not from classic parts failure. Few real products, in the field,
get close to the standard-calculated-method MTBF rates. They die from
bad design, bad packaging and soldering, or external effects like ESD.
Sometimes one of our customers will ask for a calculated MTBF, so we >dutifully crank one out. We both know that the number is prfetty much >fantasy.
On Tue, 2 Jul 2024 08:26:24 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:
On 7/2/2024 7:30 AM, legg wrote:
What's the mtbf of a fan? a compressor? a pump?
. . . . or a clamp and a block of aluminum?
As long as it isn't significantly worse than the impact of NOT
having it, you don't care -- because some (relatively unskilled)
local contractor can fix those things. You don't have to
hire a skilled member of staff to be on-hand to deal with the
"more sophisticated" technology's potential failures.
I'd much rather have an HVAC guy come in and repair the AHU in
the datacenter -- even if it was an annual event -- than have
to risk servers crashing or having to be replaced (and the
data recovered). The former is a "cheap", ubiquitous skillset;
the latter considerably costlier and critical.
You know what a brass tack is?
On 7/3/2024 6:05 AM, legg wrote:
On Tue, 2 Jul 2024 08:26:24 -0700, Don Y <blockedofcourse@foo.invalid>
wrote:
On 7/2/2024 7:30 AM, legg wrote:
What's the mtbf of a fan? a compressor? a pump?
. . . . or a clamp and a block of aluminum?
As long as it isn't significantly worse than the impact of NOT
having it, you don't care -- because some (relatively unskilled)
local contractor can fix those things. You don't have to
hire a skilled member of staff to be on-hand to deal with the
"more sophisticated" technology's potential failures.
I'd much rather have an HVAC guy come in and repair the AHU in
the datacenter -- even if it was an annual event -- than have
to risk servers crashing or having to be replaced (and the
data recovered). The former is a "cheap", ubiquitous skillset;
the latter considerably costlier and critical.
You know what a brass tack is?
Exactly that! You (as an owner of a piece of kit that you RELY on and
have invested considerable time/monies) don't care if it's theoretical >reliability is lowered; what you care about is how *effectively* reliable >that device will be. How costly (time/money/inconvenience) is it to
KEEP it in service?
This is more than just reliability *or* availability.
If you had to replace a server because a cooling system outage allowed it
to experience 50C, you'd likely be significantly inconvenienced.
If, however, it can continue to operate at 50C -- but with some damage
that will eventually manifest in a reduced lifetime/reliability -- then
you can weather the short term "problem" and plan on taking action
to avoid the anticipated problem -- additional maintenance.
If it is the nature of your business to replace items regularly,
then it's likely that your replacement interval has already factor
into it these types of "disturbances".
If, OTOH, you don't expect to be replacing (expensive) kit, then
anything that compromises that assumption wants to be avoided. How
often do you replace major appliances? HVAC systems? How inexpensive >(time/money/inconvenience) would the replacement need to be in order
for you to tolerate a shorter lifespan?
Or, how much MORE would you be willing to pay to avoid that
replacement?
[There are many devices that I would gladly "pay double" for the
ASSURANCE (not some legalistic "warranty" but the genuine
knowledge) that a device *won't* break in a given period of
time. I.e., the equivalent of having a cold spare on hand -- but
without the space required to store it or the effort required
to put it into operation]
If your products have lifespans on the order of a decade or less,
(or, if they are inexpensive to buy/replace) then you likely never
consider these things.
[Our KWHr meter will be replaced this week. Along with every
neighbor's. This is the only way the expense of such an activity
can be reasonably managed -- sending out a linesman to replace ONE
meter would be extremely costly! But, having a crew step-and-repeat
down the block is much more manageable. What added feature would
motivate them to replace them a *second* time while they still
have serviceable life?]
Sysop: | Keyop |
---|---|
Location: | Huddersfield, West Yorkshire, UK |
Users: | 415 |
Nodes: | 16 (2 / 14) |
Uptime: | 41:05:54 |
Calls: | 8,722 |
Calls today: | 5 |
Files: | 13,276 |
Messages: | 5,956,884 |