Printable View
When reinstalling, just make sure everything lines up correctly.
Just be careful of static as it can kill pc components.
I used canned air to keep the dust out of my computers.
Might want to look real close at that fan housing as I have found where screws have worked out and All I did was replace them.
Just a idea. People replace fans by them selves also.
------------------
SMILE
and post back
[ Book mark this post to find it again]
You are better off just upgrading the processor (check to see if your motherboard supports the upgrade one first) and getting a fan/heat sink to fit. Getting the processor out is quite easy. It should be sitting on a ZIF socket (Zero Insertion Force). UNPLUG the system first, remove the side panel, touch the inside chassis (to discharge any static you've picked up), then go in there. There's a little lever on the side of the ZIF socket that's held in place by a plastic clip. Just push down on on the lever and swing it slightly over to disengage from the clip. swing the lever back as far as it will go, then gently lift out the chip.
To replace the chip, just do the reverse. Look at the hole pattern on the socket; you'll notice one of the corners is different than the rest (corner is "cut off")--just match the pin pattern on the new chip to the socket. Lower the chip carefully down onto the socket, matching pin for hole. Make sure it's flat against the socket. Then lower the lever and clip it in place (there'll be some resistance--that's OK). Attach the heat sink.fan, plug the fan into the connector on the mobo or power supply.
Voila. Fin.
[This message has been edited by bistro (edited 06-08-2001).]
Here is what I found re: processor upgrade. I really dont understand all of it so, Bistro, all , if you have a chance to glance at it , maybe you could give me some advice.
There is also a section that says I can overclock? What do yall think?
Thanks, k
http://home.earthlink.net/~billselk/...slockets2.html
Well...that's a heck of a lot of pain and tribulation to go through just for an upgrade. Strike One. You also mentioned you want Win 2K--the article says it's not compatible with the Celerons. Strike Two.
I'm telling you that you are better off to save up for a new system. Strike Three https://discussions.virtualdr.com/.
I need to sign off and get some ZZZZ's...(the crowd cheers....) Nite All...
[This message has been edited by bistro (edited 06-09-2001).]
I agree Bistro, but thanks so much for your help. This system was nothing but trouble from the start, and will be glad to be rid of it. Am still gonna try to get the heat sink working properly. and all that advice was great thanks e-body. Nite,
kat
Good Morning All, Had to get off the computer last night early as my better half and daughter "needed" to use it. With emphasis on "needed". That meant I had to go to back up communications so I called Train and he kept me up to date. Am I addicted to this "stuff"? https://discussions.virtualdr.com/ https://discussions.virtualdr.com/
katsatxlady, In reading the comments made about your computer, I observed a couple of things that you could do. Use this computer for communications and build a computer, then could use this one for back up. From reading the comments made and the questions you asked, I believe you could do this. "What the mind can conceive so can you achieve".
Reading last nights posts and the other threads in the Hardware Forum , I realized some more potential exists in the forum. We had said previously that we could not help more than one person build a computer at one time. This just is not true. A thread could be started for each computer being built by a member. That member could then go at his/her own pace with the appropriate help. The Weekly Subject and other hardware questions would continue as they are. Think about the possibilities to help others that exist in this scenario. Maybe we ought to start a different thread on this to explore it more if you all think the idea is something that we ought to do.
Ok, now back to cooling. We have not really discussed all the actual causes of the heat and why we have to do the things we do to cool our computers. Please list the components and parts of individual components that cause heat and a simple why. Knowing what and why helps to understand the cooling process more.
Later - time to run a little.
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
Been 'thumbing' through the posts here and find much interesting reading. One thing that caught my eye was a comment made by jtdoom and the st...st...stuttering sound.
Currant machine has 433 Celeron (upgradeable to 500) and I just increased RAM from 96MB to 256MB. Every so often this thing st...st...stutters. So I really do need to get it cooler.
------------------
Si Hoc Legere Scis Nimium Eruditionis Habes.
(translation: If you can read this you're
overeducated)
Yup, hot air we know about, cool air we have to learn about. https://discussions.virtualdr.com/
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
A good example of a system possibly overheating:
https://discussions.virtualdr.com/im.../2001/06/1.jpg
https://discussions.virtualdr.com/ Not really. This is a good shot of a "smoke test" (no components in there by the way--just the fans). System is set up for a slight positive pressure--slightly more air going in than what's being exhausted. Smoke is being fed through the front (red box is the smoke generator) into a 120mm front intake fan--a big honker. https://discussions.virtualdr.com/ Notice the smoke coming out where the CD-ROMs, etc will be. That's good--air flow to those devices. In this example, you see "air" coming out of the side panel vents--IMHO I would tape those up on the inside; more air flow kept inside the case. Emphasis is on the hottest parts: CPU, vid card and mobo. Followed by hard drives, other drives, sound card, modem and power supply (which also exhausts).
[This message has been edited by bistro (edited 06-09-2001).]
kataxlady,
don't know what type of heatsink you have there, but in most cases, the fan can be removed and replaced without getting the heatsink off.
(in this message I talk about the heatsink and mean that cooling plate with fins...)
(you need acrobat reader) http://support.intel.com/support/processors/manuals/
I selected one of those for the pics in it
ftp://download.intel.com/support/pro.../fcpgaman1.pdf
in the latter manual you see they use a fine blade screwdriver as a lever
some clips are different.
a tip I recently posted was about modifying the tip of a blunt old reasonably broad blade screwdriver, where I advised to put a file to the tip to make a rectangular indenture in it.
that was so you won't slip off the clip that easily
¦--_______--¦
Thanks to all who took the time to help and advise.
jt...thanks for the links am d/l right now. I am still going to try to get the heatsink properly installed because of the various system errors I have been having. I have a cd-rw thats uninstalled because I can't figure out whats wrong with it and it (suddenly) freezes up my entire system up, but I will make that a different post/ different day.
greengoose... thanks for the vote of confidence and offer of help :
katsatxlady, In reading the comments made about your computer, I observed a couple of things that you could do. Use this computer for communications and build a computer, then could use this one for back up. From reading the comments made and the questions you asked, I believe you could do this. "What the mind can conceive so can you achieve".
this computer is promised to my grandson, Big Al, he is 8, and I am going to let him trash the entire system if he wants to. Heck I will even help him. I still might do my favorite method (imagined) of destroying this computer which is drop kicking/pushing it off my 2nd story balcony. <kats smiling and grinning just thinking about it>
Help me build a system, online step by step, no way Jose, You could prolly sell tickets to most of my friends to that event. "Lets go see what she does to screw up her system today" Nope, I could do the hardware stuff, but I am "technically deranged" when it comes to OS, and software. Believe what I say, I have only the barest itdea of what my OS does and how it works. Maybe next year.
Anyway I enjoyed this whole thread. Thanks again to all who replied.
And if I get the heatsink replaced I will let you all know.
Thanks,
kat
Kat.... You know where to find us if you need us. Be well. https://discussions.virtualdr.com/
Bistro, that was a really good picture of the case and smoke generator. You get a good idea what would happen if the tower was set in a corner etc. Ever run with a loaded case to see what the pattern is? Seems that would give you an idea of optimum placement for the fans.
One question that needs an answer is this: Ideally where should fans be placed - internally or externally of the case?
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
[This message has been edited by greengoose1 (edited 06-09-2001).]
I'm back...
Goose, not sure I understand the question correctly or not. But my vote would have to go on external. Get the Hot air out of the case is the best Idea. Am I close to what you were asking?
Hey bistro ya know what the Surgeon General says about smoking???
Don't remember where I saw this, but it was a boxed fan (about 12" X 12" X 12") setting on the floor pulling air from the PC through a flex hose???
------------------
Si Hoc Legere Scis Nimium Eruditionis Habes.
(translation: If you can read this you're
overeducated)
Hey Dodge, Glad your're squared away.
Now, would it be better for fans to be mounted outside the case or inside. I would think outside (if you don't mind how it looks).
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
Missed a lot of last weeks thread/topic, and I'm comin' in pretty late on this one. Work can be such an interference https://discussions.virtualdr.com/. I'm playing catch-up here, so bear with me. A lot's been covered. There are things here that I'd like to simply reinforce and/or expand on, things I like to add, and things I'd like to offer maybe an opposing or slightly different view on. I apologize up front for this be so long winded. I just wanted to put in my 2 (ok, maybe 102) cents.
First and foremost: congratulations Bistro on 5K! The number is impressive, but the time, effort and help you've provided is far more impressive.
I see where the issue came up earlier on about (extra) cooling and extending the life of components. There's no question that excess heat can reduce the life of key components and cause system instability, so it's best to ensure there is proper cooling. But as long as temperatures are kept within a certain range (as per manufacturer's specs, though perhaps not at the high side) components will perform in a reliable and stable manner for quite a long time. In most cases this will be much longer than the component will be in service (before upgrading or voluntary replacement). Reducing temperatures beyond a certain point just is not going to extend their life appreciably. So the objective needn't be to strive for maximum cooling. In my opinion the goal with any system should simply be to assure (all) components are operating within a range on the lower side of their specifications (for heat), and providing stable and reliable operation. Going beyond this won't really buy you any tangible benefit. You're system won't run faster, won't run better, and won't last noticeably longer. So, for example, if your Athlon 900 is running at 105F and your friend's runs at 90F, don't give it a second thought. Again, proper cooling should be the goal, not maximum cooling. Granted, unless you cool things to the point of causing condensation there is no harm in minimizing temperature. And, sure, it's better to err on the side of cooler rather than hotter. But once you reach a point you're not really gaining anything either.
Unfortunately there isn't a single answer as to what it takes to properly cool a system, or even (specifically) what proper cooling is. A number of factors come in to play. Some of these have been discussed, but let me go through a list…
- The first thing you need to know is the manufacturer's specifications for the operating heat range of the components, because these are the targets. But different CPUs, graphics chips, hard drives, etc. have different requirements. The proper (and tolerable) operating temp of, say, a Pentium 133 is going to be different than that of an Athlon 1gig. So even on an individual component basis the "goal" will vary, and you need find out what goal it is you need to target for.
- The heat generated (individually) by components varies, so what needs to be done in the base case on a component by component level will vary. The aforementioned P133 could run fat and happy with just a simple heatsink on it, while the Athlon would burn up like debris hitting the atmosphere. Some processors (like some of the AMD K6 series) were/are known to run hot. Therefore, you will need to do more/different things to cool it. That new graphics card may generate enough heat to require its own fan, where as an on board video chip would not. You need to find out and address the heat characteristics of the individual components being used.
- The number and combination of components in a system must be considered. The combined heat generation of the things in the case will effect what needs to be done from a cooling standpoint. Manufacturers specify requirements for the environmental (ie. internal case) temperatures their components should operate in. Depending what all one has will influence, say, whether you need that extra fan or not and how large it may need to be. In addition, the higher the surrounding temperature a component is operating in, the more it may take to cool it individually. So where as with one system a standard heatsink/fan will cool the cpu adequately, on another a beefier combo may be required to compensate for the added demands.
- The size and the construction of the case have an impact, in several ways. The smaller the size of the case the smaller the volume of air inside, and the quicker it will heat up. And the smaller the size of the case the more restricted the airflow can become, both around components and through the case. That mini-tower case you're planning to choose may hold everything you want to put in it, but it may also create a cooling nightmare. On the other hand a smaller case may require less (stealing Bistro's word) fanage to move air in, around and out of it. The case one has/chooses will also determine the size, number and placement of fans. You may want/need an extra exhaust fan, but your case my not accommodate it, or the opening provided for it may not be in an effective place, or it may not accept the 120mm fan you need to use. Then there are the vents and openings in a case. Where are the vents, what are their size? Is the opening(s) through which the front fan is drawing air restrictive? And what about "other" openings? Air will follow the path of least resistance. Some cases have slits and holes all over. The subjects of positive pressure and airflow have been mentioned. It does no good if your front fan draws air in and it is blown right out some open holes in the bottom or side of the case. A big part of cooling is being able to control where air comes in, how it travels through the case, and where it exits. Often there are a number of undesirable places in a case where air can enter and exit. They prevent or inhibit proper cooling, proper operation of the cooling components. Sometimes a bigger fan isn't needed, simply some duct tape to block off a few openings. Bottomline: the case is an important part in the cooling scheme.
- The outside (room) temperature is a factor. It is the base point from which you are trying to cool your system. Short of using the air conditioning cooling add-ons available, you're system is not going to run cooler than room temperature. And the higher this temperature is the greater the demands on the cooling components (to keep things in spec) will be. It will take less to keep your system cool if the room temperature is 60F than if it's 100F. Datacenters and server rooms are not air conditioned for those that work in them. It's done so as part of the overall cooling scheme for the equipment.
- How hard you drive your system has an influence. The heat generated by components is not constant. The harder they work the more heat they generate. Other things being equal the same two systems (and individual components) will run at different temperatures if, say, one person is just surfing the web and the other is doing intense gaming for hours at a time. The latter person will need to take some extra measures to properly keep things cool. The key here, though, is how hard things are run on a sustained basis. Components can operate without harm if they are subjected to short term demands that cause rises in temperature. It's only being subjected to this on a frequent and sustained basis that's a problem and would require any additional cooling steps. So if you see your Athlon's temp rise to 130F and then go back down because you just did a C compile, I wouldn't worry. On the other hand, if you constantly notice it running at 130F then it is worth pursuing.
- Related to how hard you drive components is overclocking. If you are going to overclock then more heat will be generated and additional steps to keep things cool will be needed. How much you will overclock, how you will overclock, and what you are overclocking all must be considered. The more you push your cpu, the more will be required to cool it. That stock heatsink/fan may not (likely won't) be sufficient. Are you going to be overclocking by raising the bus speed (FSB)? If so then you need to consider the fact that you (barring compensation through bios setting adjustments) will not only be overclocking your cpu, but devices on the varies other buses as well (AGP, memory, PCI, ISA). So they will generate more heat and may require additional cooling. Overclocking often requires raising the core voltage, something else that results in more heat generation. And let me throw one other semi-related comment in here also. The subject of OEM processors was mention. Among the other things mentioned (I think by Jaak), in some cases OEM processors are also "re-marked". What this means is that the processor you are buying is a pre-overclocked. cpu. It's, say, an Athlon 800 that has been overclocked and being sold as a 950. In such cases (and you can usually tell if this is what you're getting) then it will, by default, run hotter than a "real" 950, which you will need to consider in cooling it.
Now I'll back up a bit to some basic/general things to consider and observations when it comes to cooling. Most have been mentioned, but are worth restating and perhaps expanding on. Again I'll resort to a list type format.
-You want a good flow and path of air to the components and through your case, so definately keep cabling neat. A rat's nest of cables will deflect air from getting where needed/desired, restrict flow, and fight the actions of the fans. Rounded cables can help. So can just organizing your cables, pushing/pulling them to the sides. Plastic or velcro ties can be used to hold things out of the way. And don't use cables that are longer than you need.
- The preferred path for air through a case is to enter near the bottom in the front, travel up past the major components, and exhaust near the top rear. Heat rises and will tend to collect at the top of the case, so that's a good place to get rid of it. Place fans (if necessary) to contribute to this flow, not defeat it. For instance, putting a rear exhaust fan too low may well pull hot air from the case top or out of the power supply and direct it past components. What you really want is the coolest air possible circulating past them. Or you might want to think twice about putting an exhaust fan in the front top of your case where it will likely result in a flow in from the bottom intake, straight up and out. Similarly, a card slot fan may suck air from the intake fan straight out the back. In both these cases the fans are keeping the air from doing its job, namely circulating past heat producing components, picking up heat, and carrying and directing the heated air out of the case.
- There should be a pretty even match between the volume of intake and exhaust air, with perhaps a slight bias towards positive pressurization. Not allot though. Trying to blow your case up like a balloon will defeat proper cooling, as will trying to create a vacuum inside. The idea is to bring enough fresh (cooler) air in to the case, let it stay inside long enough to absorb heat, and then get it out of there. So use the appropriate number of and size fans to accomplish this. And don't forget to factor the (generally) exhaust fan in the power supply in to the equation.
- There is no cut and dry answer as to how many case fans (intake and/or exhaust) one needs or what their size should be. It will vary. On an older lower powered system there may be no need for any fan other than the one in the power supply. Most relatively current and/or beefier systems will need more. But again, exactly what the needs are will vary. Some current pre-built vendor systems today run with no cooling problems with just the exhaust fan in power supply (though it is usually oversized). Personally, I would tend to say most new systems should have at least an intake fan, plus what's in the PS for exhaust, and frequently an additional exhaust fan. But it will be based on what all is in your system, the amount of heat being generated, the size of the case and the amount of air you need to move through to achieve the desired temperatures. And in selecting fans, base the decision on the volume of air they move (cubic feet per minute, or CFM) not the number of millimeters or inches they measure. Yes, there is a relationship between size and CFM. But the CFM is the key and it does vary.
- With the exception of some of the overclockers and maybe a few others, most people do not need case modifications to accomplish proper cooling. That blowhole may look cool, but if everything else is done/sized correctly then they generally won't be needed. There is one case mod that I have found that often is needed however. Frequently the fan openings in cases will not be true "openings", but instead will have a built in (metal) grillwork (for lack of a better way of saying it). And this grillwork can be quite restrictive to the airflow. In some cases if you take the size of the opening then factor out the area consumed by this grillwork you end up, in fact, with very little area through which air can move. One may have a fan that is fine, just that it can't pull in our push out the air volume it's capable of. So it often can be advantageous cut this grillwork out.
- When selecting a fan I don't know that brand is necessarily important. Sunon, Jidec and Panoflex are popular and good brands. But it is more important to get ones with the correct specs, moves the required air. And it is better to pick a fan that draws less amperage given the choice. Why consume more power than you have to. Not to mention that the more power drawn, the more heat that component will generate.
- As mentioned above, air will follow the path of least resistance. A case is meant to be an enclosure, with specific points for air to enter and exit. Again, you want a controlled flow of air. So while it may not seem like a big deal having that drive bay face plate or card slot cover off, you are impacting your case cooling, and not necessarily in a positive way. And except for perhaps experimenting/diagnosing a cooling problem, leave the case cover on. With the proper cooling in place there is no reason the case should or needs to be open. If you have to have your case cover off to achieve adequate cooling then something else is wrong. In addition, having it off lets in more dirt/dust (more said on this below).
- Reinforcing what others said, where you place your case is important. Provide adequate space around the case for air to get in and out. Don't stick it next to a radiator or heating vent. An attic room is likely going to be hotter than a lower level room. Don't have in front of a window that the sun beats through. On the other hand, if the room is hot then putting near on open window may help. Don't tuck it away in one of those workstation desk cubby-holes, and definitely don't leave the door shut if you do. And elevate it off the floor and/or carpet (why is discussed below).
- Make sure vents, fan openings and blades, heatsinks, even components/cards are clean. Dust, hair, etc. will reduce heat dissipation, restrict flow, and impede/prevent proper operation of cooling components. I've seen a lot of systems where the problem wasn't the cooling setup, simply that dirt and dust had created a problem by hampering/preventing the components from doing the job. I mentioned elevating a PC above the floor or carpet. The reason being that more debris will be sucked in there. And a carpet may even block an air inlet. I've seen PCs with half the fibers from a carpet or half the hair off the family cat inside. PCs should definitely be opened up periodically to be checked (including making sure fans are operating properly) and cleaned. Use a stray or something to block fans from rotating (their rotation can generate electricity) then spray around some compressed air. Use q-tips where appropriate/needed to remove dirt. You can use a vacuum device, but carefully. You can cause static discharge with these.
- I have mixed feelings on the use of filters and grills on fan openings. They definitely can go a long way in keeping things inside clean and operating more efficiently. But they can also quickly clog up or become less efficient at letting air through. If they are used they should be checked and cleaned frequently. I'll also add that a filter is not a filter. Some inherently are better and worse at letting air through. You want a filter to filter, and hamper the air flow as little as possible. As I said, it's kind of a double edge sword with filters.
- When attaching or reattaching a heatsink/fan remember that improper contact reduces cooling efficiency. Make sure it is properly/squarely situated and any retaining clips are on properly. If thermal paste (or a pad) is called for, then use it. Make sure the mating surfaces are first cleaned (nail polish remover can be used if needed), including removing any traces of any prior compound/tape that was there. If using paste then apply it properly, according to directions. You want as thin and even a layer as possible. The paste is primarily there to bridge the microscopic imperfections between the surfaces. In many (probably most) cases you don't need to use a fancy paste. And thermal pads can be perfectly adequate. But in those cases where you need a little better cooling action using a paste with better thermal conducting properties (Artic Silver for example) may be called for.
- System changes or upgrades necessitate that cooling requirements be reevaluated, and that perhaps your cooling will need to be beefed up. A system that is fine today may not be fine tomorrow. And it's not just a cpu upgrade that can have an impact. That new graphics card or 7200 rpm hard drive may generate more heat, and push things over the edge. One of these components may have/add its own individual cooling requirement. Or it may just be the combination of added components that lead to cooling problems. Maybe things were fine, but after that new graphics card, added hard drive, cd burner …
- If your buying a pre-built vendor system, and not adding any components yourself afterwards, then it is generally safe to assume the system has (at that point) the proper cooling setup. Not always, but generally. By all means monitor the readings. But generally the cooling design (case, fans, placement, etc) will have been tested and certified to properly accommodate and cool the installed components.
- If your building a system or upgrading the cpu and buy a boxed-retail (as it's often referred to) version of a cpu, the fan/heatsink combo supplied with it will be certified/approved and generally provide adequate cooling. If you're buying an OEM processor than make sure the fan/heatsink combo approved by the processor manufacturer. Most people don't need special or fancy heatsinks/fans to cool their cpus properly. In typical/average use there will be no problem. Of course if you're going to overclock then it's a different story. Likewise, if the cpu is going to be driven hard for sustained periods (as mentioned above), or you know that other components/factors will give rise to increased internal temperatures that will make cooling more difficult, then you may need more than a stock combo.
- It's not only cases or CPUs that may need/benefit from a fan or a heatsink. Modern graphics cards/chips frequently come with some cooling. It may be a heatsink, maybe a fan as well. Nonetheless, they may require one or both, or upgrades to the ones they have. And those 7200 rpm drive run hot. Perhaps a drive cooler is in order.
- If you have an existing system and you're motherboard has the built-in capability to monitor temperatures then you should (at least periodically) check the readings to ensure they're at the proper levels. ). Also look for rising trends in readings (absent any changes). They may indicate a need for cleaning, or a failing component. Except in the extreme case, there is no need for additional or enhanced temperature monitoring devices. The ones built in to your board will be more than sufficient (if used).
- For systems that don't have built in means to monitor temperature all is not lost. You can generally get an adequate assessment by simply using your hand to feel the air being exhausted. If it's hot (not just warm) then there might be something to be done. You can also place a thermometer near the exhaust to get a more accurate reading.
- If you have a heat related problem you'll see the evidence. Some people may not recognize or understand the signs, but they'll be there. Granted, not all heat related problems are constant/chronic. They may not manifest themselves until you're playing a game, or doing something else that utilizes one or more components intensely for a period of time. But when the overheating occurs it won't play peek-a-boo with ya. For certain components, and if your system is so equipped you'll get an audible alarm (assuming it's on) or the system may shutdown. Absent a motherboard with these facilities it will still be pretty obvious. When one or more components reach the point of overheating they won't produce a subtle or momentary stutter, glitch or hiccup. There may be a relatively small range of temperatures where certain components will slow down without actually producing noticeable errors, crashes or failures. But in an overheating situation the temperature will hit that range and move right on through it in to the abyss very quickly. So quickly that you won't even notice. There will be a definite sign indicating there's a problem, often you'll get a slap up side the head. In the worst case a component will fail and be rendered unusable. Short of that, if it's the right component you'll get crippling lockups, abends, reboots, boot up failures. In other cases the manifestation may be less widespread and not bring down the system. But there will still be errors, application abends, component level lockups and freezes, sustained component level loss of or abnormal operation. Granted, some these same things can happen for reasons having nothing to do with overheating. But with some diagnosis or analysis of the messages and circumstances it usually becomes clear what the problem is. Bottomline: if you have a cooling/overheating problem you'll know it (if the signs are read and understood). And if you're not seeing tangible evidence then there's nothing to worry about.
I think I'll stop now (he says as the applause rises to a deafening tone). I'll just summarize some of my thoughts by saying that proper cooling is important, and needs to be properly considered and addressed whether building/upgrading or dealing with an existing system. But achieving proper cooling does not mean achieving maximum cooling. And in many cases achieving proper cooling doesn't require any fancy, elaborate, extraordinary or exotic means or solutions. Sometimes following some basic principles, doing or not doing some basic things goes a long way towards achieving the goal. And in addressing requirements and a setup to address them each situation is different and needs to be handled as such. The benefit and beauty of building or revamping one's own system is that it is custom, not a clone. But that also means there is no set group, type or size of cooling components that fits or is best for every need. There are guidelines, principles and specifications that can be used. These then need to be applied/used on a case by case basis to determine if something's needed, what's needed, what size, how many, etc, etc.
Wheeew! Well I'm tired and burn out, and I've surely burned out any one reading this also. I apologize if I rambled, made no sense, or confused things. It's been a long week. I might have missed some things I was gonna say or respond to, but that's enough for tonight (for many, probably more than enough period).
So far these threads seem to be working out ok GG1. Keep it going…
------------------
Bookmark your post to make it easier to find.
And remember, please post back whether the problem is resolved or not.
[This message has been edited by DrMDJ (edited 06-09-2001).]
DrMDJ: Bantam Books just called...they want permission to break your post down into chapters and bind it in leather.... https://discussions.virtualdr.com/
bistro you beat me to that comment.
DrMDJ, that was some great reading. Think I need to change my eyes though, getting a wittle leak from reading all that....LOL just kidding, good information......
Bistro
Are you sure it was not science Tech Magazine.
DrMDJ
Now that will print into a very good book.
------------------
SMILE
and post back
[ Book mark this post to find it again]
DrMDJ, Welcome back, was wondering where you got off to. Would your fingers be any shorter than they were? https://discussions.virtualdr.com/
As you can see by the other comments, that was one great post. That effort will get you at least a 25% raise over what you now make at VirDr. Thank you very much for the time you put into it. I think a book - yes a book. https://discussions.virtualdr.com/
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
Alright guys, go ahead and poke fun at me. It's ok, I can take it. As a mature adult I will simply say I know you are, but what am I? Na, Na, Na, Na, Na!
PS. By the way Bistro, I think they're selling your rectal thermometer probes at 2CoolTek now. There listed under the Anal-log device section.
------------------
Bookmark your post to make it easier to find.
And remember, please post back whether the problem is resolved or not.
You are bad DrMDJ. Sheeeesh https://discussions.virtualdr.com/
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
Excellent posts from DrMDJ and Bristo! Learned a lot this week.
Is there a list of all weekly subject discussion? What's on for next week?
defung, We started cooling a bit early and when it finishes we will start "Considerations Before Building". This is the first time we have attempted something like this and are feeling our way so to speak. Try to give the next subject before we start it. As you have seen tonight we have some great talent contributing. Thanks for your comments and do jump in whenever the fancy strikes you. https://discussions.virtualdr.com/
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
DrMDJ: https://discussions.virtualdr.com/ https://discussions.virtualdr.com/ I was wondering when they'd get those.....
Went looking for this, so copy/paste. cooling sites from the Langalist I got several months ago.
Sure a passel of them. http://www.3dfxcool.com/ http://www.pcpowercooling.com/home.htm http://www.heatsinkfactory.com/ http://www.inflowdirect.com/ http://www.coolerguys.com/ http://www.overclockin.com/ http://www.ocworkbench.com/ http://www.overclockers.com/ http://www.hardocp.com/ http://www.overclockershideout.com/ http://www.athlonoc.com/ http://bxboards.com/
If you are overheated Try this KOOOLLLL!
------------------
SMILE
and post back
[ Book mark this post to find it again]
Good Morning World, Oh me. More reading. Be in and out today. https://discussions.virtualdr.com/
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
Hi all
DrMDj just about made a summarisation.
https://discussions.virtualdr.com/ COOL https://discussions.virtualdr.com/
He even mentions power supply fans don't always blow out.
And yes, once upon a time I mentioned rebadged CPU.
I too heard this practice still exists.
(sometimes they even advertise them as overclocked... in that teeny weeny small font and low contrast, but it happens, and I actually came across a few such adverts.)
in one of the tips I saved I read about the power consumption of a single CPU.
well, I was surprised when I found out how many watts it has to dissipate.
there is a lot left to research.
power consumption of various brands and types of CPU, graphics cards, motherboard (or the mobo chipsets),
different speed hard drives,
burners,
one could say, if you need a 400W power supply unit (aka PSU), you gonna have to get rid of 400W of heat...
how true would this be?
anyone know how many cubic feet of air one has to expell when one has to dissipate 300W, and ambient is 23C, and acceptable exhaust is 35C ?
I bet that's ever harder in british thermal units, and fahrenheit.
it can get complicated.
watt and joulle and calori confuse even me
also note I already used cubic FEET/m
we should talk of cubic meter nominal to stay in the metric system.
an example
to increase temp of one gramme of water with one centigrade, one uses 1 calori
(If I recall correct, to evaporate it takes 570 calori)
I also believe one cubic meter of air at nominal has a mass of around 80 grammes
I don't say these numbers are correct.
anyone interested in a science project?
well, tackle this one.
------------------
You help someone with a similar problem if you tell us what worked.
When you do, please post solution in the original thread?
Kind regards, Jaak
Sounds like a good project jaak. But, not mine https://discussions.virtualdr.com/
------------------
SMILE
and post back
[ Book mark this post to find it again]
no kidding?
didn't trains run on steam?
and gosh, Charles WATT invented them
no wonder they went electric... https://discussions.virtualdr.com/
------------------
You help someone with a similar problem if you tell us what worked.
When you do, please post solution in the original thread?
Kind regards, Jaak
jaak
my name sake http://www.neumedia.net/~cassrr/graphics/num_6a.jpg
More here
------------------
SMILE
and post back
[ Book mark this post to find it again]
[This message has been edited by Train (edited 06-10-2001).]
ahh,
don't let my unscientifically construed suggestion for a research project "derail" this "thread".
https://discussions.virtualdr.com/
...and I'm working on developing a quantum computer made from sugar water and baling wire. Think I have a clear idea on how to do it.
Uh....by the way...what's a "watt"?
Watts on second Bistro. Who's on first.
------------------
Bookmark your post to make it easier to find.
And remember, please post back whether the problem is resolved or not.
Ya'all have gone over the edge. Just not sure which one. https://discussions.virtualdr.com/ https://discussions.virtualdr.com/
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
Good Morning All, There was a wealth of information contributed on the subject of cooling in the past few days. For those of you reading and have a question or two about any part or a question about a problem you may have, ask away.
Does anyone see anything we have missed on cooling?
Should we need to keep going on this subject, I will start a part two tonight.
------------------
Seek knowledge and all else will follow
Please post back results - Press Ctl D to bookmark
Information
I think that is enough about cooling, look forward to next discussion.Quote:
Originally posted by greengoose1:
Should we need to keep going on this subject, I will start a part two tonight.
Well...I was gonna cover how to make a real neato case out of a Frigidaire deep freezer, but yeah...that about covers cooling for now.
https://discussions.virtualdr.com/
science is wonderful
and wonderfully simple in its complexity
and extremely complex in its research of wonders.
anyway
been looking for something about printed circuit boards (aka PCB)
http://www.mayahtt.com/tmwiz/convect/forced/pcb/pcb.htm
and I said something about the mass of air, and how much air would need to be moved thru the case to stay within a acceptable limit of increased air temperature.
Air at room temp will always contain water vapour.
To increase its temp takes more energy than it would for dry air.
Dry air is heavier than damp air at same temp .
Exerpt from "Thermal properties of damp air"
The heat content, usually called the enthalpy, of air rises with increasing water content.
The enthalpy of dry air is not known. Air at zero degrees celsius is defined to have zero enthalpy. The enthalpy, in kJ/kg, at any temperature, t, between 0 and 60C is approximately:
h = 1.007t - 0.026 below zero: h = 1.005t
The enthalpy of liquid water is also defined to be zero at zero degrees celsius. To turn liquid water to vapour at the same temperature requires a very considerable amount of heat energy: 2501 kJ/kg at 0C
At temperature t the heat content of water vapour is: hw = 2501 + 1.84t
Notice that water vapour, once generated, also requires more heat than dry air to raise its temperature further: 1.84 kJ/kg.C against about 1 kJ/kg.C for dry air.
The enthalpy of moist air, in kJ/kg, is therefore:
h = (1.007*t - 0.026) + g*(2501 + 1.84*t)
g is the water content in kg/kg of dry air
When I used the calculator found at http://www.natmus.dk/cons/tp/atmcalc/atmocalc.htm
I saw I remembered at least something half right.
to wit, mass of air can be way different in winter and high summer,
Two factors play their role.
air temp, and humidity.
cold humid air is lighter than cold dry air
I've oft told people riding a pushbike why its harder to ride in 30 klicks headwind during winter than during summer.
(I first read about this in a book by Isaac Asimov)
http://www.natmus.dk/cons/tp/atmcalc/atmoclc1.htm
extremes I calculated ranged from around 55 to 140 gram/cubic Meter
btw, don't let this fool you
but for completeness, you have to consider a fan has a limited capacity.
Your car top speed will be higher if air density is around 70, than it would be if it's around 110 g/cM. Its fuel consumption will be a lot higher while plowing through denser air at same speed.
a fan has to work to move air.
5 Watt will only move so many grammes/s
10 Watt will move more
the air contact surface and material its made of (the heatsink) will play a role in heat transfer.
And the thermal interface between the heatsink and die is truly important.
As a matter of fact, I am gonna get me some of that artic silver.
The stuff I got is adequate, but I'd like to see lower temp at full load.
(I already mentioned this, what use is a heatsink if one finds an insulator on the die...)
I just came across this last link I'll provide, since it looks like it belongs here. http://www.benchtest.com/clinks.html
Greengoose1,
in a way I have to thank you for "coercing" me to partake into these here discissions.
I am learning a ton.
https://discussions.virtualdr.com/
[This message has been edited by jtdoom (edited 06-11-2001).]
[This message has been edited by jtdoom (edited 07-04-2001).]
*Oh, Lordy...somebody's just GOTTA take jtdoom's Scientific Calculator OFF his system...* https://discussions.virtualdr.com/