Reservoirs vs T-lines in Flow Rate = Insignificant

[Martinm210]

I was curious about how T-lines compared to Reservoirs and I also needed some equations to plug into my estimator spreadsheet I've been working on, so in spirit of my curiosity, I ran pressure drop curves. The end result is that I found that T lines are less restrictive, but both in general are SOO small, you really shouldn't worry about it.

The contenders:

Don't you like the tractor? And hey...I kind of like the coke bottle, maybe I'll make a good one for real..



And the resulting graphs:


Now for the reality of what this small pressure drop means:
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing = 1.68 GPM
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing + 1 T-line = 1.66 GPM
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing + 1 Reservoir = 1.64 GPM

I already have these worked into the next flow rate estimator release, but the bottom line is...neither one amounts to any noteable loss.

I wouldn't base your decision on flow rate, back to personal preference..

[hecktic]

I nominate another great thread Martinm210 has made to be put on sticky please.

You are da the man Martinm210!

I had always thought T-Lines sucked and cause problems but boy was I wrong, although I do have one concern with a T-line, arent you restricted to the amount of coolant you can add to your loop using a T-Line, or does it even matter if using a good radiator to cool the coolant?

Try and test with a MICRO-mcres if you have one.

[Martinm210]

I nominate another great thread Martinm210 has made to be put on sticky please.

You are da the man Martinm210!

I had always thought T-Lines sucked and cause problems but boy was I wrong, although I do have one concern with a T-line, arent you restricted to the amount of coolant you can add to your loop using a T-Line, or does it even matter if using a good radiator to cool the coolant?

Try and test with a MICRO-mcres if you have one.

Thanks, I had been thinking for a while now that reservoirs were this big flow rate hog...so I had to answer my question...I was wrong.

I REALLY like T-lines for a couple of reasons:

• Easy to monitor coolant level, Put a black zip-tie on your level and you have a great indicator of loss.
• More spill proof for spill happy fillers like myself. I can pull the T-line out away from my case and fill and spill as needed worry free.
• It seconds as a drain line, turn it on end and pull a hose off my rad and the system drains really well.

And no I've never felt restricted by the T-line capacity, actually the other way around. My bay reservoir only allowed maybe 1/4" of drop in coolant level before it started sucking air back into the loop, not a problem with the T-line...lots of elevation there if you leave yourself a long line.

My preference is still a T-line, it's just a simpler setup for me...less tubing, less chance of spill damage, nice and quiet. THe only disadvatage is they take a bit longer to bleed and they are so still at the top, it's hard to see you have flow except on startup when the pressure builds and the level drops because your tubing is expanding a bit. A reservoir give you a much better indication that you have flow, and alot of people like the looks of them.

[NaeKuh]

was the bottle sealed? Im thinkn, if you had possitive pressure inside the bottle, water would need to be pushed harder due to gravity working down on the water level.

It would be nice if you could try this test again on a smaller reservoir, like the microres by swiftech.

[Baleful]

There's to many variables in this to just say "resorviors are super restrictive". I REALLY don't see how they could be restrictive unless you're using a crappy bay drive res. Use a swiftech micro res, or one of the EK's then come back to us. If you're gonna post something like this, and want to be taken seriously, don't use some home made concoction. Use stuff that is used by us every day, ie. micro res and EK res.

[Waterlogged]

Honestly, I don't think it's restriction at all but rather, a release of pressure in the loop. I've often pondered this quandary but never really looked into it.

With a T, the pressure from the outlet eventually forces water into the inlet, thus keeping higher pressure. A res on the other hand, breaks this continuity of perpetual motion.

[ColonelCain]

.02 GPM compared to a t-line? What difference would that make in the long run?
I would have definatly gotten a resivoir if my budget had permited, because it looks better, and is 50x easier to bleed IMO.

[Jedda]

Great to see you laying solid foundations under your flow estimator, Martin!
WTG
I've always thought it was a balance between ease of fitting and ease of filling/bleeding. Nice to know for sure.

[virtualrain]

Doesn't every radiator effectivly have three tanks (one at the inlet, one at the outlet, and one at the other end)? Aren't these equivalent to a res in basic function in that they contain a mass of fluid that comes in one way and exits another? If so, then it would seem silly to worry about adding another res to your system for bleeding purposes when every loop already contains 3+ tanks.

[hecktic]

Well the data is still not deniable. But I did in fact suggest he use a swiftech micro mcres res. It would add more clarity to the data this way and be more convincing.

I still will believe him with or without the swiftech res test.

[Frank M]

My preference is still a T-line, it's just a simpler setup for me...less tubing, less chance of spill damage, nice and quiet. THe only disadvatage is they take a bit longer to bleed

That's why I have a res at a high point, and a T at a low point.
Best of both worlds:
- easy bleed
- easy drain
Adds 0.06GPM restricition according to your calculation? Could'nt care less
(especially since bleeding with 8mm ID tubing takes a lot of time - so much that
after some hours of unsuccesfully filling/bleeding, I put in a res...)

[Martinm210]

Sorry guys, the title was misleading the reality is they are both extremely small:

Now for the reality of what this small pressure drop means:
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing = 1.68 GPM
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing + 1 T-line = 1.66 GPM
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing + 1 Reservoir = 1.64 GPM

I already have these worked into the next flow rate estimator release, but the bottom line is...neither one amounts to any noteable loss.

Both are extremely free flowing and amount to nothing in terms of temps

[sick_g4m3r]

Excellent results! Keep going doc martin, and all these threads should be combined into a nice sticky

[zhangmaster12]

grrrr now i have to get a t-line...

[7499richard]

you should try it with a real reservoir that many people use b4 i pay any attention to this, coke bottle is different imo

[santiagodraco]

There's to many variables in this to just say "resorviors are super restrictive". I REALLY don't see how they could be restrictive unless you're using a crappy bay drive res. Use a swiftech micro res, or one of the EK's then come back to us. If you're gonna post something like this, and want to be taken seriously, don't use some home made concoction. Use stuff that is used by us every day, ie. micro res and EK res.

Who cares? The main point an intelligent person would take away from this is that T lines are not that restrictive after all.

Actually I need to make one edit. Martin, you might try the same tests using the T-lines in a different orientation. Have the flow impact the sidewall and carry on in a horizontal plane if that makes sense. Having the flow go across the top of the T might not have as much impact as perpendicular. Or did you do that already?

[CyberDruid]

I too would be curious to see if anything about the shape/size/plumbing of a reservoir makes a damn bit of difference--I doubt it will.

Having used T-lines and several different types of reservoir I have to say the only thing a reservoir does is give you something fun to look at...

A T line IMO is much more to the point and easier to use in many ways...but a 5.25 bay res is fun to look at.

The "Cyclone" res that Performance-PCs.com sells is identical to Martin's coke bottle in capacity and plumbing..

And for the haterade directed at home made set ups--all test rigs are at some point or another "home made" it's the methodology that is important.

A pair of inlets into a container is not going to be much different if the container cost you $20 or 5 cents as long as the size of the containeer aproximates the containers that would be actually in use in a loop.

Nice work.

[Grinch]

I have been using T-lines on about the last 20 builds

[santiagodraco]

it sounds like your talking about turning a T line into some sort of elbow, is that correct?

if so I'd imagine about the same as any comparable elbow, or at most very little difference.

andyc

Pretty much correct. Just making sure that the test has the fluid changing direction by 90 degrees, and in the case of a T being impacted by that change happening in two directions. For example I'm using more of a T to split my loop into a 1/4 leg (NB) and a 3/8 leg (loop continuation) and combined back together again. The 3/8 leg exits and enters on the perpendicular part of the T while the 1/4 leg follows the straight path along the top of the Ts.

[Martinm210]

grrrr now i have to get a t-line...

No don't do it because of this, my main point was to say they are both so very very very very small it's insignificant. I would not worry about .04 GPM, you should consider this 0.

It was so small I wouldn't have been able to measure it with any pressure guage, I had to measure in inches of water down to the nearest 1/8" just get a smooth reasonable curve and that's not easy.

I also had to run a 4th curve just to remove the pressure drop from the 3' piece of tubing and the T-lines I was using to measure pressure. That's how small a pressure drop these are. And one elbow is more restrictive than an entire reservoir.

[Martinm210]

Who cares? The main point an intelligent person would take away from this is that T lines are not that restrictive after all.

Actually I need to make one edit. Martin, you might try the same tests using the T-lines in a different orientation. Have the flow impact the sidewall and carry on in a horizontal plane if that makes sense. Having the flow go across the top of the T might not have as much impact as perpendicular. Or did you do that already?

This was a test with the T setup such that your flow passes straight through. A branching T fitting where your flow takes a 90 is much much more restrictive, more so than a sharp 90 degree elbow. I would not use T fittings to split flow if you can avoid it, you're much better off with a Y fitting.

A typical fitting loss chart for PVC fitting will say:
90 degree elbow is worth about 1.5' of pipe
T flow through is worth 1.0' of pipe
T branch is worth 3.8' of pipe
Now keep in mind this is for PVC and fitting which are more efficient than tubing and fittings because of the step down from hose to fitting. I find that an elbow is actually worth about 4' of tubing, and a T fitting is closer to 2', so taking those relations and applying it to the 3.8' PVC rule of thumb I would guess you're talking about 8' of tubing or more...that's not good.

I could run a curve for it, but I would have to use two T fittings and then split the result in half.

[Martinm210]

I never considered a T line to be restrictive to any degree at all. now elbow fittings another story, but even those have very little effect on flow rates or temps. if you design your loop right to begin with with the proper flow rate of 1.5 to 2.0 GPM, you can always add a little here or a little there in regards to T lines, elbows, res. etc.

andyc

Yep, that's spot on!

Ok I fixed the name of the thread, I apologize for doing a little bait and switch, I really didn't intend for everyone to think there was some real noteable difference. My end message intent was to say yeah T lines are a little less restrictive, but it's so incredibly small you shouldn't worry about it.

[Boyu]

How about using T line as elbow but the bottom part was only for drain purpose. Coz i'm planning using dangerden delrin T as elbow and drain port as well.

[Martinm210]

lotta great work there Martin, and should be a great help to those designing their builds for some time to come. this and your flow rate calc are really what the intent of XS is IMO.

really great posts and effort, I thank you for it.

andyc

Thank you!
The reaction was my own fault for the baiting topic. It's also a good example that we all tend to skim read and read headlines and forget the rest...I should know better..lol!

[Martinm210]

How about using T line as elbow but the bottom part was only for drain purpose. Coz i'm planning using dangerden delrin T as elbow and drain port as well.

Just one elbow (similar by probably slightly more restrictive than a regualar elbow) won't amount to much, just don't go putting a dozen of them in and you'll be fine.