Air flow

we have 5020 and 2148 switches - we have them in the top of racks with the port side in the rear of cabinet (makes cableing to and from servers more managable) but the fans are drawing air in from the rear of cabinet and that is my hot air exhaust from my servers - my question is can I turn the fans around to pull cold air in from the rear of the switch (which would be from my server cabinet cold aisle?

The air-flow is Front-to-Back, but the back is considered where the ports are. "Air enters the chassis through the fan trays and power supplies mounted at the front of the chassis and exits the chassis through perforations on the rear of the chassis."
Could you past a 'sh environment fan detail' It should show something similar to this below. If it does, then the airflow is blowing air 'out' the ports and 'in' from the power supplies.
switch5k(config)# sh environment fan detail
Fan:
Module Fan Airflow        Speed(%) Speed(RPM)
             Direction
1       1    Front-to-Back 40        6699
1       2    Front-to-Back 40        6708
2       1    Front-to-Back 40        6569
2       2    Front-to-Back 40        6585
3       1    Front-to-Back 40        6650
3       2    Front-to-Back 40        6569
4       1    Front-to-Back 40        6666
4       2    Front-to-Back 40        6666

Similar Messages

How to measure a VXI board's thermal / air flow performance?

I am looking for information on requirements and testing procedures as they relate to the VXI bus and cooling. Specifically, I would like to make some temperature measurements on some VXI cards to see if the increase in airflow temperature (temperatuer of exhaust minus temperature of intake) is within 10 degrees C.
Quite a few documents reference a document called VXI-8 (Cooling Characterization Methodology Specification). I have not been able to locate this document. Would you happen to have it?
Also, any information you can provide on the best procedures for measuring temperature drop, pressure drop, air flow, etc. across a VXI card would be much appreciated.
Thank you.

Thanks. But the document didn't really answer my question. Perhaps an example would be better.
The NI VXIpc-872B embedded controller, for example, has an ambient operating range of 5 to 50 °C. What exactly does this mean?
If you were performing a test to determine if the temperature of the VXI card is being exceeded, how would you do it?
Let's say you installed the controller in a VXI chassis (with integral cooling fan), put the chassis+card in an environmental chamber, and turned the unit on. One thermocouple measures the temperature entering the card (at the bottom), while another thermocouple measures the air temperature exiting the card (at the top). The chamber temperature's setpoint can be varied up and down.
In this example, would the intake air temperature (i.e. air temperature entering the bottom of the card) have to be less than 50 °C? Or would the exhaust air temperature (i.e. air temperature exiting the top of the card) have to be less than 50 °C? Or does it mean the surface temperature of any component on the board must be less than 50 °C?
Thanks,
Michael

T5120 cable management arm and exhaust air flow obstruction

Hello Folks
Concerning Sun servers of 1-U in height... There is debate as to whether the effectiveness of heat transfer out of the chassis is actually impaired by the use of a cable management arm at the rear of a rack mounted server.
From my experience I see this as more of an issue with I-U servers rather than those of 2-U or higher, where enough of the air exhaust flows over the cable management arm.
We have a fair variety of server models, including a T5220 and 14 T5120's. We have used the supplied cable management arm on the T5220 and two T5120's. I notice that the surface temperature of the cable management arm on the T5120's that faces the exhaust is quite hot to the touch, (I'm guessing over 60 C). The metal of the cable management arm conducts the heat and the insulation of the cables does not seem to be effected. However the heat content in the exhaust seems high enough to keep the surface temperature of the cable management arm fairly hot, and much of that heat does not dissipate well into the frame of the rack as there is a mix of materials, metal with insulators like plastic, paint, and air.
Although this type of cable management arm is supplied with the T5120, I am thinking that its use for these 1-U models of servers may be detrimental. The cable management arm was installed in accordance to the instructions in the Sun SPARC Enterprise T5120 and T5220 Servers Installation Guide, doc 820-2178-12 July 2009, Revision A.
The "Sun SPARC Enterprise T5120 and T5220 Servers Site Planning Guide", doc 820-2177-12 July 2009, Revision A, states Allow a minimum of 5 mm (0.2 in) clearance at the front of the system and 80 mm (3.1 in) at the rear of the server when mounted. However the installed cable management arm reduces the rear clearance by about one half.
The focus is to remove heat from the server, whether that is achieved via conduction or convection. Conduction does not seem too effective in this case. Convection of exhaust air flow is obstructed. The Site Planning Guide also does state Manage cables to minimize interfering with the server exhaust vent., which may lead one to deduce not to use the cable management arm at all and route the cables straight out.
What do you folks out there prefer? Why would Sun even supply these cable management arms with these 1-U servers?
Cheers, Roger

Good one, John! Been waiting for your impressions for a while now...
We seem to share the overall opinion about the 180; it's ok, but could definately be improved upon!
My major concern right now is NOISE. Tried the Zalman Fan Mate to quiet the intake fan for the HS, but that seems to affect the CPU-fan as PC Alert reports it as running at 0 rpm from time to time, even though I'm sure it doesn't. And a few days ago the intake fan got a lot noisier than it was from the start, so it's probably giving up the ghost already
Currently awaiting a Papst 60x60x15 to replace it with, hopefully that'll fix the noise problem too!
Noticed your recommendations as to placing quiet fans on the inside of the chassis, and although that is a nice idea it could be difficult on the Mega considering the side panels and top panel are one piece of metal. Could actually be difficult to put the cover back on! If you try something in the way of replacing the HS and/or fans you're more than welcome to share your experiences in this forum!
The few of us who are still clinging to our Mega 180's will need all the help we can give each other, since official support from MSI seems to be held at a minimum...
Cheers!

Imac Intel Fan Speed/Air flow - Whats normal?

Hi,
I have had a intel imac for about a month now, and it is quiet. The air flow coming out of the back slot is warm air, BUT you can hardly feel the flow..
and now i have set up a another imac for someone at work and have a few more to do.
Anyway this other imac when started the fan was loud, not tooo loud but you could really hear it, BUT the airflow was cool air and it felt right that it should be cool air coming out and you could really feel the air. But now that i have moved it to another desk and put ram in it, and started it, the fan has quieted down like my imac and the air flow is warm and much warmer on the top left of the air vent slot.
- So what is right?, should we be able to hear the fan, or not?
- Should the air flow be cool or warm?
- Should you be able to feel the flow on your hand?
- Is the fan on a sensor, ie if it gets hot it works harder...
Both imacs are now quiet, one is more warm/hot than the other (as it is brand new), i think over time it will cool down....
- or it is pot luck what happens with them?
- Also if i get a temperature widget what would be normal
regarding the reading ?
thank you
rai

for temps this program works great
http://www.bresink.com/osx/TemperatureMonitor.html
and heres another thread
http://discussions.apple.com/message.jspa?messageID=2963306#2963306
it sound like your ram was the problem ( perhaps not seated right )
as far as the other thing check out the link above
good luck

Changing the air flow in Nexus 5596UP without interuption.

We have a Nexus 5596UP and we want to change the air flow to the opposite direction.
I think it is possible to change it without an interruption but have no lab to test it.
I think when I remove all the fans and change 1 power supply the systems keeps running.
(don’t know if the Nexus can work with 2 different power supply’s, this is the catch !!!)
After the new power supply is inserted I will remove the other one (the old one) and push in the new fans.
Hopefully all within 30 seconds to prevent a shut down or mismatch errors.
Is this possible?
Please advise
Kind regards,
Remko van Gilst

Hi Remko,
The airflow through the other Cisco Nexus 5000 Series switches is from front to back, but you can change the airflow back-to-front
Note: The Cisco Nexus 5596 has front to back airflow so you must position it with the front facing a cold isle.
Cisco Nexus 5000 Platform switch with one power supply, removing the power supply causes the switch to shut down. If you are using two power supplies and you remove one of them, the switch can continue to operate.
The fan module is designed to be removed and replaced while the system is operating without presenting an electrical hazard or damage to the system, if the replacement is performed promptly.
I hope you will need some down time since you need flipped around physically to align the airflow anyway.
Please try the below steps,
Power Supplies replacement:
Pull one power supply (System operates normally).
Insert one new power supply (60 second timer will begin).
Pull the second power supply (60 second timer stops)
Insert the second new power supply
Fan replacement:
Pull one fan (System operates normally with 3 fans – N+1)
Insert one new fan
Remove second fan.
Insert second new fan.
Remove third fan.
Insert third new fan.
Remove fourth fan.
Insert final new fan.
Regards,
http://www.cisco.com/en/US/docs/switches/datacenter/nexus5000/hw/installation/guide/install5500.html#wp1258897
Regards,
Aru
*** Please rate if the post is useful ***

Averaging Data stream from air flow meter...

Hello,
I'm fairly new to Labview, so this may seem elementary to some. I am pulling real-time data from an airflow sensor and I want to manipulate the data in the following way:
1) after starting the program, I want to release one of 3 valves to trigger the sequence
2) delay for 10 seconds
3) grab one data point at a given interval (every second?) over period of 10 seconds (20 seconds total time elapsed)
4) average the data and spit out a value
I've been trying for a while to get this working for a while. Any tips would be much appreciated. Once again, I'm new at this, so please bare with me.
Thank you.
Mark
Attachments:
Air Flow venturi.vi ‏223 KB

Basically the test setup is this...I am testing air pressure through a mechanical venturi. There are 3 venturis in the machine at the same time. The DAQ assistant communicates with the NI USB-6009 only (no data collection). THis is used to activate the solenoids that clamp and pressurize the test pieces. When one of the nozzles (nozzle1, 2, 3) is released, air flow through and the Omega FMA-1611A flow sensor picks up pressure and flow rate. This is what is connected to the serial port.
What I'm trying to do is streamline the process. Right now, I am reading a number off a visible indicator, which is causing user error. I'm trying to make it so that as soon as I release one of the nozzles (boolean nozzle1, 2, 3), a sub-routine will execute that waits 10 seconds (for the air flow to normalize) then takes a reading once every second for 10 seconds. Then I want to display and record the average of these data points. I want to be able to do this without ending the program (the flowmeter should still be reading continuously). I've gotten to the point where the program works perfectly with the just the visible displays. All I need to do is add the secondary sub-routine I explained.
I did try highlight execution per reccomendation and I now see what you are saying about data flow. I tried just taking data points without averaging or time delays and I still seem to be bogging down the program somehow. Do I need to build another case structure outside of the main loop? I think I can get through the delays and the mathematics...my biggest issue is taking a reading at a set interval (i.e. 1 second) and saving the readings so that I can calculate the average and display it. Do I use an array? The "for loop" uses bracketing, but as you stated, only takes one reading.
I'm an ME guy, so while I can build the test setups without issue...obviously programming is another story
I will continue to play around with the programming. I have gotten far enough to make the program work in its basic form, so I think I can do this, but any tips you can provide would be extremely helpful.
Thanks again,
Mark

Antec SL450 - very little air flow

So, I got my new PSU, the Antec SL450.
Something a little funny, when i put my hand in front of the rear exhaust fan, it doesn't seem like much air is coming out at all, I figured especially with the 90mm intake fan it has on the bottom of the unit, it would be pumping out quite a bit. both fans are spinning, so it doesn't make any sense to me.
if anyone else out there has the SL450, stick your hand in front of the rear exhaust fan and tell me how much air flow you actually feel coming out of it.

As I said in my previous post - airflow direction/fan position is important.
The case I am using now was set up for my K7S5A and - with that board the side fan was right opposite to the CPU so all I had was just a simple piece of card wrapped around the fan shroud to direct the airflow striaght on to the CPU. The little duct was cut to a lenght so that when the case side was on properly the duct was about 1/8"(3-4 mm) clear of the CPU fan.
Unfortunately when I changed mobo a few weeks ago - the CPU position on the KT6V is about 1.5 inches further back towards the rear of the case and my fans didn't line up. The CPU temp was ( to me) unacceptably high at around 48-49C under load even after restting all other fans to get the best advantage. So I made a new duct - out of modellers plasticard(thin ,stiff and best of all cheap). This duct had to be cranked to the rear to get the exhaust end right over the CPU. Result - temps dropped around 5-6C.
Re the airflow around the video card - I have a small director vane on the front low input fan to send some of the airflow upwards but also have a rear bottom input fan immediately below the AGP slot .
On my case there is a gap of about 2.5 inches between the top edge of the PCI/AGP slots and the case left hand edge( as you look from the front). THere is eniough space here to fit a 60mm fan( from old CPU cooler) tho I had to cut a hole for it to fit into. This one brings more air into the bottom of the case- right below the AGP slot(good for the video card and also helps force the incoming air from the front to rise).
Don't know where you are skiltrip but here in the UK Plasticard for the duct is very cheap and can be got at most good model shops. If you can't get that then thin stiff card( such a used for postcards ) would probably do.
Good luck
Dave

Fan air flow feels weak, replace?

hi all
Fan Air flow from my G4 MDD is not as strong as my G4(AGP) Mac's fan. When I put my hand in back of my MDD, air flow is light. Should I be looking for a stronger flow? It maybe why my MDD shuts off sometimes without warning.
your help is appreciated.
Web dude

The air flow out the back of the MDD is meant to be slow, it has a large fan inside that only has to spin relatively slowly. Can you feel the air intake at the bottom of the front ?
Sudden shut off could be overheating, get a temperature monitor program, like "Hardware Monitor" (use google). and check the CPU temp, should float around 58-59, up to 62 when busy busy.
When you have the chance, check inside and clear out any dust build-up.

DC In Proximity Air Flow - sensor is red in iStat

Hi,
I have a MacBook Pro Retina that is almost two years old. Have 768GB SSD and 16GB.
I run iStat and look at it every once in a while (it is a very good tool for those that do not know). I have noticed the "DC In Proximity Air Flow" stat is red and fluctuates between 117 and 119, but it stays red. I manually set fan speed to to high see if I could lower it and was able to get it to 116 and it turned black. As soon as I let the MacBook go back to normal operation, it jumped back to 117 and is red again.
What is the "DC In Proximity Air Flow" sensor and how can I make sure the airway is clean? I was thinking about opening up the back and cleaning out what I can, but I would prefer not to open her up.
Please let me know what you think, where it is, and what it does!
Thanks!
-Kirk

Same exact issue here - iStat 5 (latest version), DC In Proximity Air Flow 118F (48C). Showing red. I don't know if this has been going on for a while, or if it's associated with the iStat upgrade I just bought, or the update I just downloaded today. It's the first time I've ever seen red in the sensors section.
MBP Retina (mid 2012), 768GB, 16GB, Mavericks

Air Flow Characteristics

Hi,
I am a Hardware planner in Fujitsu's UK Data Centres. I am trying to find documentation that describes the air flow charateristics of Cisco kit. Some devices flow from side to side, some from front to rear, some from rear to front. The Data Sheets that I have seen do not appear to contain this information.
Can anyone point me in the right direction to look for this information?
Many thanks
Rgds
Nick Ward

It's not clear what you actually want to do, but why don't you simply use a picture indicator?
Why is there no loop in the toplevel VI? Are you using continuous run?
The sequence structure in the subVI is useless. Dataflow already determines the execution order.
Why is there a disconnected control over a diagram constant?
Changing the boolean from true to false make no difference if both states look the same.
The positions control needs to go outside the for loop. It will not change during the duration of the loop, so reading it once is enough.
What is the purpose of the hidden air indicator of the subVI?
Why the stop button and case structure in the subVI?
LabVIEW Champion . Do more with less code and in less time .

NX7600GT-VT2D256EHD - direction of air flow

Is cool air flowing through the metal grill from outside the chassis to the card, or is warm air flowing through the metal grill to outside the chassis away from the card?

I got this reply from MSI Tech Support:
Cool air is flowing through the metal grill from outside the chassis to the card.
This is the opposite of the conclusion reached by holding the piece of paper against the grill.
Perhaps the fan was running at a very low speed when tested with the piece of paper, and the chassis was pressurized by a system fan, with the end result that little if any air was entering via the grill.
I have a card on its way, so I'll take a close look at the air flow after I install the card.
I occasionally see other graphics cards taking up two slots and having a grill for air flow. I wonder what the air flow direction is for these other cards.

Air flow Direction VI

Hello to all members ,
here is another problem for me
i want to move an object first some position towards left then towards top .
is any way to programtically do it(using property node ----> position )
here is attached VI please see and help me
thanks in advance
Attachments:
Air flow direction.vi ‏11 KB

It's not clear what you actually want to do, but why don't you simply use a picture indicator?
Why is there no loop in the toplevel VI? Are you using continuous run?
The sequence structure in the subVI is useless. Dataflow already determines the execution order.
Why is there a disconnected control over a diagram constant?
Changing the boolean from true to false make no difference if both states look the same.
The positions control needs to go outside the for loop. It will not change during the duration of the loop, so reading it once is enough.
What is the purpose of the hidden air indicator of the subVI?
Why the stop button and case structure in the subVI?
LabVIEW Champion . Do more with less code and in less time .

Administrators of MSI - Air flow

Hello msi, iam having a problem with my VGA 560 GTX TI HAWK, i cant find the correct air flow to my case, the new system airflow of MSI " in air " Is i think complicating my GPU AMBIENT TEMP
because the new better system is sending the hot air flow to 3 directions, and the old system is only sending to one way the hot air
so, at the first time i reached the 96º grades celsius with my card, and a gpu ambient of 115º for the fault of the system air flow of the twin frozr III
the noise is very bad sometimes at 80% of the fan speed 4500 RPM
improving my air flow iam getting a temps of 76º celsius with a 3970 and 4000 RPM at this config and a gpu ambient of 60º
clock 880
voltage 1025
shader 1760
but where is the problem? if the fan is in for example 60% of the RPM, the gpu ambient increase to 99º ( 20 minutes of playing in ultra )
so the problem of this card is the AIR FLOW IN of the system of twin frozr II and III, not the chips
i say this because iam madding test for more of one week ) in this test of my case, i broken one hard disk of 1 terabyte.... all my info, for one error... whatever...
i need the help of the administrators, to tell me in the REVIEWS of the card what system air flow is used on the temps of the pack of the card
i dont wanna RMA the card, because the problem is in the system air flow, the new system air flow of the twin frozr III
this system air flow is not made for the convencional users
so well, i need to know the system air flow and the ambient temp of the test of the card to say the max temps is 65º 17- than the basic system 560 gtx ti

Quote from: _Rick_ on 13-November-11, 08:36:28
GPU-Ambient sounds like a misinterpreted sensor-reading.
Wouldn't worry much about that.
Plus, there's a known issue with one of the card's sensors getting stuck - in Afterburner it's assigned the memory temperature.
67 sounds pretty good, hope I can get there as well, with under 70% fan-RPM.
the gpu ambient is the most important sensor, because u can play for 20 minutes with 70º gpu and gpu diode, but you dont know if the gpu ambient is 70 / 50 or if is 115
if is 115 the problem is, the temp of the GPU and gpu diode, can go in 5 minutes to 91º from 70º and more, and for protection maybe in 10 or more minutes the system is shut down
at now my gpu ambient is 58 ( 2 hours playing in gpu usage 99% )
the problem is solved
what can i say?
if u have the same problem try this
1) change the thermal compound, from the GPU and removing the IHS ( u lost the warranty ) before do this, check the card some days, if u dont see blue screen or anything bad with the card, go ahead with the thermal compound
2) make a good air flow to your card, the twin frozr III system is not so good like others system of anothers cards, if u are using Asus cards for example, your air flow and your case is adapted to the asus air flow system, the twin frozr III if dont have a good air flow, or bad thermal paste the GPU ambient can go to 115º and shut down your PC
3) play with the voltages, this help too to slow the temps, but if u have a bad gpu ambient, u are in the same, maybe with less voltage u stay playing for 40 minutes, and with default voltage u stay playing for 20 minutes
at now my card is in default, the average temp is 68, max temp in hard scenes of the game, the temp go to 69 with just one second, the fan speed max 70 ( silent mode ) 3810 rpm max, average 3000/3300
now i can play with the sound of the game, and not with the creepy sound of the coolers of twin frozr III
my problem now? i dont have warranty, LOL.
next time i buy a card i go to read all the manual.

NYKO Air Flow EX and Lego Star Wars

I read the README file that comes on the CD for Lego Star Wars to find out about game controllers. The Apple store had a NYKO Air Flo EX, so I bought it. This controller is specifically listed in the doc and says that no third party support software is required (supposedly works right from the OS drivers).
"The controllers listed above function using the built-in Mac OS X drivers. They do not require any device specific or 3rd party services."
When I start Lego Star Wars, it still only recognizes my keyboard contros. The NYKO controls do nothing. I know the controller is getting power (the fan works) and the NYKO key mapper sees the controller.
Do I have to map keys for this game? I wouldn't think so, from the README. If I do have to map keys, I feel like I've wasted my money on a game controller.
iMac G5 20 Mac OS X (10.4.4)

Ok, I answered my own question. The Readme file gives you directions for how to uninstall the application. I wondered if there wasn't a setting in a file somewhere that said I had no controller.
I tried just removing the .plist file from ~/Library/Preferences, but that did not help. The Readme then referenced removing a directory in ~/Library/Application Support. This in error; the directory is actually in your documents directory.
I removed that directory and the application and reinstalled (I think just removing the directory would have been enough). Now the game sees my controller and plays more fun.
iMac G5 20 Mac OS X (10.4.4)

Test Requirements to test an airfoil in a small air flow tunnel using Labview

I am actually working to automate the experiment titled "surface pressure measurement on an airfoil"in a small air tunnel[wind tunnel by Plint Industries.,UK].I would like to know the basic test requirements like Data acquisition stuff.

HI,
1. You are acquiring an external signal first of all. That is a pressure signal. That must come into a system that acquires physical signal. It can a SCB-68. But for serious applications when you need to remove noise, or in other words, you need to condition the acquired signals, you need Signal Conditioning System. The devices are called SCXI (Signal Conditioning Extensions for Instrumentation). There are different types of SCXI devices such as SCXI-1100, SCXI-1102, SCXI-1140, SCXI-1141, SCXI-1520 etc. You use one or the other depending on your application such a current measurement, voltage measurement, frequency measurement, filter block etc. I am providing some nice links that I would suggest you to take a look at for better understanding.
2. Once yo
u acquire the physical signals and process them using SCXI, you send the processed signals into computer. Here, in the computer, there is something called Data Acquisition card (DAQ). DAQ cards acquire the signals, converts the analog signal into digital and stores into computer memory. Some DAQ cards are PCI-MIO cards.
Please also do take a look at the following link:
Signal Conditioning
Signal Conditioning for Data Acquisition
Analog I/O
Measurement Hardware
Please feel free to give Nati
onal Instruments Support for more information on products for your system set up.
Regards.

Air flow

Similar Messages

Maybe you are looking for