View Full Version : Does the reservior has to be at the top?
I've heard this somewhere, actually many places. I'm just wondering if I have to put it at the top (or just higher that the majority of the loop) The reason being is I plan to put it in the front bays but if I were to put it the top most bays, if there was a leak, my HDDs and ODDs below would have a nice shower. :eek: So because of that I'd like to put it near the middle if not the bottom of the case (top mounted rad for reference).
Nebulous
06-29-08, 10:24 AM
Well the consensus is to put the res at the highest point. This being that it will be easier to bleed/feed. It's also advised to leaktest the parts before installing.
As long as the res is higher than the pump, you're fine :)
This being that it will be easier to bleed/feed.
I understand the bleeding, but what about the feeding? I don't understand why it would be harder.
Conumdrum
06-29-08, 10:40 AM
When you fill the loop ya use the res. It will flow down to the pump inlet, thats pretty much it, just gravity.
Nebulous
06-29-08, 10:40 AM
Gravity will work better as the pump won't run dry when feeding. Running the pump dry when feeding/starting will damage the impeller.
Spawn-Inc
06-29-08, 03:17 PM
and another bit of advice, don't leak test over night, i did that and woke up to water all over my desk and the pump still running, though since its the lowest point in my loop there was water in it still.
It does need to be at the top of the loop or have a fillport going to the top of the loop when filling or else when you open it up, water goes down with gravity and it will of course come right out of your now open res.
Otherwise, a res can be put anywhere as long as it feeds the pump inlet. If it is lower in the loop, most people run a line from the fill hole in the res to a fillport at the top of the case.
For reference my build is the most similar to this one (http://www.ocforums.com/showthread.php?t=558039), the only major difference is the res will beeven lower than that. Can the res be that low for use without causing issues? I have an idea that to fill I could just put the case MOBO facing up, unscrew the res, pull it out and up, then it will be the highest point in the loop. The possible issue with this is it's not the highest for use.
For reference my build is the most similar to this one (http://www.ocforums.com/showthread.php?t=558039), the only major difference is the res will beeven lower than that. Can the res be that low for use without causing issues? I have an idea that to fill I could just put the case MOBO facing up, unscrew the res, pull it out and up, then it will be the highest point in the loop. The possible issue with this is it's not the highest for use.
It will still work as long as you can have it at the highest part of the loop when bleeding/filling. Just realize that one of the conveniences of a res is usually ease of filling, and that whenever you want to top off your loop, it will be a bit inconvenient. I would recommend finding a way to run a line to a fillport at the top of the case as it would make things a whole lot easier.
Does the fillport/t-line loop order matter? With this build, the ideal location for me is between the rad and pump or the rad and CPU block. Then again, I could put it anywhere, but those would be ideal.
You really want the res feeding the pump inlet to give it a ready supply of water so that it doesn't run dry when filling, and to prevent negative pressure at the pump inlet.
t4runnr
07-02-08, 08:36 AM
It does need to be at the top of the loop or have a fillport going to the top of the loop when filling or else when you open it up, water goes down with gravity and it will of course come right out of your now open res.
Otherwise, a res can be put anywhere as long as it feeds the pump inlet. If it is lower in the loop, most people run a line from the fill hole in the res to a fillport at the top of the case.
That didn't happen to my first watercooling system. It was an all Asetek a long time ago with the little Asetek cylindrical reservoir. Anyway, it was at the bottom just in front of the pump. Both were at the bottom of the loop. When the lines were completely full and leak-free, I could open the top lid of the reservoir without the water at the top of the loop draining out.
maxspam
07-02-08, 11:07 PM
A little off topic but why is the impeller damaged when ran dry?
Blazing fire
07-03-08, 02:48 AM
and another bit of advice, don't leak test over night, i did that and woke up to water all over my desk and the pump still running, though since its the lowest point in my loop there was water in it still.
lol :santa:. What about running say 9am and checking every hour subsequently and after 10pm, go to bed when there are no leaks?
A little off topic but why is the impeller damaged when ran dry?
The fluid acts as lubrication for the impeller between it and the shaft. Running it dry can damage the impeller rather quickly.
4GHZ_or_bust
07-03-08, 11:49 PM
A large bath tower or a beach towel wrapped around the base of your PC should help with any leaks if you're going to sleep.
If I may.
Voigts: After our discussing on my thread I did some testing, and I have to contradict you. The res does -not- need to be atop the loop at any times. However this has a few contingencies.
One. It must be higher than the pump.
Two. The pump -must- be running.
I found that as long as the pump is running and pumping faster than you are pouring, you can fill the loop effectively. More tests will proceed to see if I can get it completely bubble free ect with a low res, but as for just filling? It seems to work just fine.
It's pretty simple as to why, I can have a more detailed explination if anyone wants.
Yes please, expand upon this. I don't care about filling difficulties, but topping off is wut really matter to me. I know that filling a top rad will already be difficult, so holding a res in the air while doing it will be even more fun ;)
Does the pump have to be on for topping off?
Anything regarding topping off?
With the pump on, I'll explain how it works, I can only do the rest of the tests when my rad actually comes in the mail.
Here's how it works though. The pump is running. When you pour water into the res,it will progress to the lowest possible point, and will then be forced to the other side of the pump, making, for quite some time, the pump the lowest possible point, the water level rising on the -other- side of the pump. However, eventually the water level will reach the top of the loop and actually start to circulate. However again, the tubing will only be "full" of water on the upward side of the loop from the pump, there still being a bit of room on the downward side of the loop for water to be added, therefore, as you keep pouring, it keeps filling this extra area, and the pump is able to move more and more water over the top of the loop, and more and more of the downward side of the loop is now being filled with moving liquid. The res doesn't overflow, because water preferrs to move down, and even if the top of the res is in the middle of the loop, as long as there is a uniform amount of water, and it is being uniformly circulated, it will still always attempt to move down. Wow, that was a terrible explination, when my rad comes I'm taking pictures once and for all -.-
However I do have doubts for what will happen when the pump turns off. If your res/line is capped, I think you'll be safe (assuming the cap is airtight) but if you just have a loose line, I do think your loop would overflow. The key to what I'm saying is that the pump needs to be on. (Or you need to have an airtight seal.)
No I understand it. Basically the pump keeps the water flowing, so the extra overhead water pressure doesn't influence the res.
And in one sentence you do what my two paragraphs couldn't. -claps- -headdesks-
orthomofo
07-04-08, 08:06 PM
i am biting my finger nails, waiting for the pics and measurements.
since we are going to get the same case, which cooling parts are you getting
res
pump
rad
blocks...
cheers:beer:
If I may.
Voigts: After our discussing on my thread I did some testing, and I have to contradict you. The res does -not- need to be atop the loop at any times. However this has a few contingencies.
One. It must be higher than the pump.
Two. The pump -must- be running.
I found that as long as the pump is running and pumping faster than you are pouring, you can fill the loop effectively. More tests will proceed to see if I can get it completely bubble free ect with a low res, but as for just filling? It seems to work just fine.
It's pretty simple as to why, I can have a more detailed explination if anyone wants.
I hadn't considered keeping the pump running. How are you going to fill a system with the pump running without running it dry? You would have to be doing some mighty fast pouring.
Anyway, why would you want to add complexity to things? The easiest way is to have the res at the highest point, or have a filltube from the res going to a fillport at the highest point.
I hadn't considered keeping the pump running. How are you going to fill a system with the pump running without running it dry? You would have to be doing some mighty fast pouring.
Anyway, why would you want to add complexity to things? The easiest way is to have the res at the highest point, or have a filltube from the res going to a fillport at the highest point.
I want the ease of a reservoir and with regards to placing it at the bottom as mentioned in my original post:
The reason being is I plan to put it in the front bays but if I were to put it the top most bays, if there was a leak, my HDDs and ODDs below would have a nice shower. :eek:
I want the ease of a reservoir and with regards to placing it at the bottom as mentioned in my original post:
I know about you putting it in the bottom. You can always run a line up to a fillport at the top of the case to make things easier. The res will still bleed out the bubbles.
I'll see about the T-line. I'll do some testing to see what I like and if I have extra tubing, I'll at least try it.
metloaf
07-05-08, 12:56 PM
My reservoir is on the bottom just next to the pump. So no it does not have to be on top.
4GHZ_or_bust
07-05-08, 07:03 PM
It's not hard to keep up with filling while running a new WC set. Most pump won't drain the reservoir in a blink. The termaltake pump/res combo I have holds roughly 250mL full and takes about 5 seconds of running before the water level falls below the intake. A funnel on top of the res and pour at about the same rate as the pump pulls in until the water comes through at the other end. Then let it run for a bit to flush out any trapped air and fill again as needed.
Not a problem.
Vagabond102
07-05-08, 08:39 PM
My res is lower then my rad and my pump is on the bottom. (MCP655+) My pump can deffinitely take in far more then I can feed it with the ST Micro-res and a funnel.
Bleeding my loop was simple stupid. Cycle the pump while feeding the res. Once loop is full, top off and don't touch it for a month. :-)
And what you all are say, your res is still higher then you pumps, in all examples given. That's all that matters. The intake _cannot_ go dry, the rest is simple.
Fill ports and the like just make it even more simple.
I've only built my one loop, but now that I've done it once, it only gets easier. Though unlike others, I detest the 7/16 ID tubing. I'd much rather have the 1/2 ID tubing and ensure it's secure with clamps. I really hate having to cut my tubing any time I want to drain the loop for any reason.
Vagabond: Mind putting up some pictures of your loop? I'm just interested to see some examples and yours seems like mine in some degrees.
Voigts: The thing is, I'm going to have a top mounted rad, and wanted to attach my res to the side of the case, and a fillport is expensive, and requires more drilling ><
4GHZ_or_bust
07-05-08, 10:13 PM
Fill port doesn't have to be expensive. Just run a tube up to the top, glue em in place (hot glue for easy removal later), and a cork in the port when not filling.
:) nice looking ports aren't cheap unfortunately and yes they do require drilling.
repilce
07-05-08, 10:26 PM
I was thinking about this earlier today, and the way i see it is.. (see if this works out :D )
With a "full loop" you could have the res at the second lowest point in the loop with cap off and with the pump on or OFF.
Because.. this whole thing of "gravity" and "closed loop" is simple..
IF the water were to "rise" in the res.. you would have to have a LEAK somewhere in the system.. because when the loop is full , nothing but water can move.. and water has to move to make water move.. keeping all things equal.
So even if the line feeding your res was pouring into the res (pump off) it would be draining water at the same time to feed the water coming in. Assuming you dont have any leaks and are nice and water tight you could have it as low as you want, but above the pump just for "pump on" issues.
Vagabond102
07-06-08, 02:14 PM
Vagabond: Mind putting up some pictures of your loop? I'm just interested to see some examples and yours seems like mine in some degrees.
My loop is really no different then others...hang the rad off the back, place the res about in the middle. The flow is the same. I don't have a camera at my disposal at the moment.
Ah. My plan was similar to what was seen in the watercooled rocketfish thread, whereas the rad is top mounted through a hole drilled in the roof of the case. And you don't have any problems? Pics would be nice if you could get em at some point, but thanks for the reassurance ><
so what are the benefits to having the res at the highest point? ease of filling, access and it won't have the pressure of the fluid above it, compared to if it was lower than any other part it would have to be running to fill? I will be doing a top mount rad in my cosmos, but I can cut a hole in the top and have 1/2 the res above the rad, would this be worth the hassle? everything will still be inside the case or hidden anyways. wasn't even thinking about this till I started reading this thread.
Vagabond102
07-06-08, 05:24 PM
Ah. My plan was similar to what was seen in the watercooled rocketfish thread, whereas the rad is top mounted through a hole drilled in the roof of the case. And you don't have any problems? Pics would be nice if you could get em at some point, but thanks for the reassurance ><
http://img301.imageshack.us/img301/9453/water2aj4.th.jpg (http://img301.imageshack.us/my.php?image=water2aj4.jpg)
Pic of when I was testing.
Keep in mind, I know my tubing lenth is too much, and I should have mounted the rad with the barbs down. I had no problem filling this loop and it was completely bubble free in less then a half hour.
LOL, I love how you use the TRUE as a paper weight. I'll be doing some of my own testing, (my only concern being topping off, as I feel that filling is a one time thing [okay may be not once] so it'll be fine regardless of difficulty) to see techniques and may be factors that will make it easier to fill.
The whole idea for using a res is to 1. bleed bubbles fast 2. ease of filling the loop. When setting up a loop, regardless of how you put your components, you are going yourself a favor to keep these as simple as possible. The higher up in your loop you can put your res, the better. And using a fillport isn't a must, but it sure can make filling easier. The easier you make your loop to fill/bleed/drain/maintain, the better off you are.
Air rises. There is no escaping that, no avoiding that. If you have a rad at the top of your loop, you are going to have some issues with air wanting to collect at the highest part of your loop. You can get the air out by tipping your case around so that it goes into your res, and that is what you have to do if your res is lower in your case. Those of you who are saying that your bubbles are bled out in no time without doing this are making up stories. Over time, some water evaporates out of your loop through the tubing and some air gets in. This air will tend to collect at the highest point. So just realize that if your res isn't at or near the highest point, it is easy to get air pockets at the highest point.
I personally got tired of doing case gymnastics to get all of the air worked out of the system, and made my loop this time so that the res is the highest point, and I have a tube going to a fillport. I also have an easy to access drain in the bottom most part of the loop. This makes filling/bleeding/draining so easy it isn't funny.
Keep in mind, I know my tubing lenth is too much, and I should have mounted the rad with the barbs down. I had no problem filling this loop and it was completely bubble free in less then a half hour.
If you had mounted your rad with the barbs down, you would have real problems with air collecting in the top of your rad. It is because you have the barbs up that the pump flow is able to push the air into the res.
Voigts: I still have doubts about some things you're saying, but I won't contradict them untill I've done tests. Now the people selling me the wc things need to get back to me so I can get the goddamn parts... >< (namely doubts with the air collecting in the radiator part. It holds in theory, but when you have excess water in the low-pressure half of the loop due to tline or res, methinks it would force it out, but then again I could just be failing at fluid mechanics.)
Voigts: I still have doubts about some things you're saying, but I won't contradict them untill I've done tests. Now the people selling me the wc things need to get back to me so I can get the goddamn parts... >< (namely doubts with the air collecting in the radiator part. It holds in theory, but when you have excess water in the low-pressure half of the loop due to tline or res, methinks it would force it out, but then again I could just be failing at fluid mechanics.)
I'm no fluid mechanics expert, but I do know that when I used a top mounted rad for about a year with a res of course that was mounted a bit lower, I had issues every so often where I had to tip the entire case to get the air out of the rad as an air pocket would inevitably form in the rad.
Vagabond102
07-07-08, 10:00 PM
Voigts,
I just flat disagree with you on the bleeding thing. Really, it took less then a half hour and no bubbles. Period. I did play gymnastics with it later and nothing new came out. Evidently I was lucky. All three times I have filled it.
As for the rest of what you say, you are spot on, air rises. Though, something about the MCR 320 rad makes it bleed well even without a bleed screw. While I agree with you, my experience contradicts my rational thought that it should have taken far longer. And perhaps the barbs at the top did allow the flow to push the air out and that's why I didn't have any problems.
OP,
I would STRONGLY recommend not only a fillport, but also a DRAIN! Don't forget to think about how you will drain this thing. I overlooked this, and it wasn't a pleasent experience.
Don't worry about the draining and filling!! I have my ways. I'll take pictures of the process. Basically, my res will be on extra long lines so it can be positioned (be it high for filling or low for draining) to easily complete the task at hand. If worse comes to worst, I will add both fillport and drain, but they are not only more costly, but they increase the potential for leaks.
Vagabond102
07-08-08, 09:29 PM
Don't worry about the draining and filling!! I have my ways. I'll take pictures of the process. Basically, my res will be on extra long lines so it can be positioned (be it high for filling or low for draining) to easily complete the task at hand. If worse comes to worst, I will add both fillport and drain, but they are not only more costly, but they increase the potential for leaks.
If the extra 15 bucks or so concerns you, watercooling is probably not your cup of tea...
It's more so for the latter reason, but the reason I mentioned $$ was because when combined, a fillport/drain effectively remove the need for a res. This is of course besides the bleeding, but even that can be done with a fillport. Basically I'm not going to pay $$ to have my res "castrated".
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.