Two pumps before a Y?

[zlojack]

I've been thinking about ways of taking advantage of the pressure of dual DDC pumps. Most say just to put the pumps in series with some good tops and then have them go directly into the block (EK Supreme or other impingement style block)

Maybe this is a crazy idea, but what would happen if you put two pumps feeding into a Y-fitting right before the block? I realize the flow would be lessened but would that not give a nice boost in pressure going into the block, or would the Y just kill it? Obviously this would be for impingement style blocks and the pumps would have to be set up at equal distances from the Y to equalize the pressure.

Am I crazy?

[Sniper]

I think your crazy. But that's generally speaking mind you lol.

Why not try it with the Y & see what it does for you? It's only tubing to cut & would be a nice little experiment I think.

[RedRaider]

I'm glad your thinking, but you're focusing your efforts on something that is pointless, IMO.

Two pumps feeding into a "Y" ?? Run the pumps in series and don't bother with the added bullshit. Have the outlet of one DDC feeding the input of the second DDC.

If you want to do a "Y" though, do what you please and be sure to post your results.:)

Seems to me that you must make sure the tubing from the each pump to the "Y" is exactly equal in length. I would tend to believe you would induce unwanted turbulance and losses at the "Y"...:shrug:

I believe you should always employ the K.I.S.S. method---- Keep It Simple Stupid

A "Y" sure as hell ain't falling within the KISS frame...

[zlojack]

I agree there, Red. It was more a sudden idea that I thought I'd run by you guys.

Turbulence and stuff at the Y...true, I suppose.

I will probably go the series route, but I'm tempted to try this crazy Y idea just for fun.

[Xilikon]

Don't forget to :crack: or :coke: some before trying...

[zlojack]

Alright...

I know it might not be a good idea, but I'd like to know why.

So far, Red has proposed that the added turbulence at the Y would cause problems.

What else? Maybe I'm on crack, but I'm trying to think a little outside the box here.

[rubidium]

I've been thinking about ways of taking advantage of the pressure of dual DDC pumps. Most say just to put the pumps in series with some good tops and then have them go directly into the block (EK Supreme or other impingement style block)

Maybe this is a crazy idea, but what would happen if you put two pumps feeding into a Y-fitting right before the block? I realize the flow would be lessened but would that not give a nice boost in pressure going into the block, or would the Y just kill it? Obviously this would be for impingement style blocks and the pumps would have to be set up at equal distances from the Y to equalize the pressure.

Am I crazy?

Consider the analogy between a fluid circuit and an electrical circuit. Pressure is analogous to voltage, flow rate is analogous to current. The pressure difference between the outlet and inlet of a pump is analogous to the potential difference across a battery.

Now if you consider an ideal pump as you would an ideal battery, then two ideal pumps in series increase pressure. Two ideal pumps in parallel - assuming they each generate identical pressures - result in the pressure of a single pump. If the load is the same, then the flow will remain unaltered as well. Period.

In the real world, however, real batteries have their voltages "sag" as you increase the load on them. This is dictated by the so-called characteristic curve" of the battery, which gives the voltage of the battery as a function of the current draw. (The ideal battery has a flat characteristic curve - voltage is independent of current draw.) Just put a volt meter across your car battery and watch how much the voltage drops when you crank the engine, and you'll witness the non-ideal characteristic curve big time!! The same is so with pumps. Considering the restrictive nature of the EK block, the flow rate offered by an unloaded pump will more or less become reduced by the loading presented by the block - depending on the pump's flow vs backpressure characteristic curve. Putting two such pumps in parallel will not increase the pressure, but the combination will present a different, flatter, characteristic curve.

Connect them in series ...

rubidium

[zlojack]

Interesting, Rubidium and thanks for that. I think I'll stick with what's proven and go for series.


But does electricity behave the same as water?

[rubidium]

If you have the characteristic performance curve for the pump (head vs flow rate), and the performance curve for the net load (pressure drop vs flow rate), the intersection of these is your system operating point. Some empirical performance curves are are provided in Martin's spreadsheet, which also plots the intersection and shows the operating point.

You can extend Martin's spreadsheet to compare the benefit of hooking up two pumps in series or in parallel as follows: For two identical pumps in series, the resulting pump performance curve for the pair is obtained by doubling the head for a single pump at each flow rate. For two identical pumps in parallel, the resulting pump performance curve for the pair is obtained by doubling the flow rate for a single pump at each head value. Thus you will get two different characteristic curves for the different pairings, and be able to see for yourself which performs better in your circuit.

I've calculated an example for you. Assume your circuit has a single EK Supreme, a single PA120.3, 5 feet of 1/2"ID tubing, and a single DDC-3.2 with EK G1/4 X-Top. The intersection of curves looks like this:

http://farm4.static.flickr.com/3104/3211675430_4e97d00d3d_o.jpg

Now add another DDC-3.2 with EK G1/4 X-Top in series:

http://farm4.static.flickr.com/3409/3210836053_52850749e6_o.jpg

Compare the above to what would happen when adding the second pump in parallel with the first:

http://farm4.static.flickr.com/3259/3211685902_7a93cc5540_o.jpg

Notice how the blue curve for the parallel pump case became flatter (as I predicted in my last post) - especially on the right side - indicating that the second pump in parallel has "eased the strain" on the single pump at higher flow rates, making the combination behave more like an unloaded pump. The performance hasn't significantly improved, however, because for the load in this example, the single pump was not operating in the "strained" right region in the first place. Said differently, the gain of adding a second pump in parallel occurs at upper flow rates - a regime denied to you by the restrictive nature of the system load.

In general, if you want higher performance by adding a second pump, add it in series.

rubidium

[zlojack]

Got it. Excellent stuff rubidium!

Thanks for taking the time to explain this to my very un-scientific mind :D