Twp pumps

I have a single 3/8" loop system for my CPU and I was finding that my Ehiem 1048 pump was a little underpowered so I added an old thermaltake 500 in series. I placed it before the other pump so it goes:

-> TT500 - Eheim - CPU - rad - res -TT500 ->

Either pump by itself is crappy but together they seem to do a really good job. One for volume and one for pressure with redundancy for safety. I know I am not the only one to do this, but with two unequal pumps, would you think having two pumps A (a low pressure high volume pump) and B (low volume, higher pressure) be better, worse or no change if you swapped places?

(I am thinking no change but I wanted to hear other opinions)

I never have done it but I heard from very experianced folks it makes both pumps work harder than they should. One is overdriven and the other works harder. It's kinda a no-no.

I can see where you could get into trouble but this setup seems to be working very well and temps are dramatically lower. Maybe the two pumps are closer to each other than I thought. I looked up the claimed specs but I don't believe the Thermaltake numbers.

TT P500 = 500 l/h; Hmax 6 feet
Ehiem 1048 = 600 l/h; Hmax = 4.5 feet

i work in a heating ventilation and air conditioning engineering firm and doing a setup like that is strictly taboo. perhaps on the smaller scale it works fine. what we always do is put the pumps in parallel and from what i see the specs on your pumps are that would work cuase one wont overpower the other........

CAUTION i seriously doubt that you will be able to use 3/8 tubing still if you do that cause one pump will try to work against the other if you can bump up to 1/2" id tubing and run them in parallel.

im in northern canada and do a lot of hydronic heating systems so one thing i know about is pumps and piping. with them in parallel each pump would share half the load assuming that they were identical pumps. meaning that you get alot more flow.

what do you mean parallel?

one at the beginning one half way around?

joedymueller said:

what do you mean parallel?

one at the beginning one half way around?

Click to expand...

the squares with p's in them are the pumps. that help?

Thanks Rustyfender. I guess I need to get some Ts and reorganize, or just buy a decent pump.

What exactly is the danger of pumps working against one another or one overpowering another? Is it quicker time to failure? I cannot see getting much better performance than I am getting if I switch to a parallel setup.

The effect of using one pump or two:

Pump configuration; Idle temp; Super_Pi 2M load temp

Just Ehiem; T= 66 deg C; T= 78 deg C
Just P500; T= 65 deg C; T= 78 deg C
Using two pumps in series; T= 59 deg C; T= 71 deg C

Room temperature is 77 deg C, 2x120 rads

o there are pictures? dont see them here.. im sure it will make sense when i get home .. hehe

well more than anything its just inefficient. since your using two pumps to get the same flow or slightly higher flow since each one doesn't have to work as hard but if you put them in parallel then you get more flow works like this:
in parrallel:
gpm(pump 1) +gpm(pump 2) = gpm(total flow)
(thats oversimplified there are some other factors but for comparative purposes it'll do fine)
so with this method each pump is providing its total flow rate and then once the tow flows meet you have the combined flow rates in one... now this is assuming identical pumps so with 2 different ones one pump will deadhead against the other and burn out quickly.

anyways if you run two identical pumps in parallel you get twice the flow and if one ever fails the other will continue to provide flow. all this is assuming your tubeing sizes are large enough


what your experiencing with temp difference with the second pump i think i have an explanation for its one possability but not the only one.

when we use hydronic baseboards in a heating system to heat a room they rely on convection to provide the airflow to take heat from the fluid via the fins but for the water to excrete enough heat in the baseboard we have to keep the velocity of the water in the tube above certain levels so that the water will not have laminar flow (smooth flow) so by keeping velocity up the water flow is very turbulent when it goes through the smaller pipes on the baseboards. this means that the main body of water in the middle of the tube moves very fast but all along thee side the fluid grips the sides and eddies causing the individual molecules to be in contact with the side of the tube longer.

with fan coils (same as a rad with a fan just bigger) we can lower the velocity cause the fan is moving the air so the rad is being actively cooled so to maintain the same 20f temp difference we have to use larger tubing (i wont get into y everything has to be 20f temp diff thats unrelated) anyways when we go somewhere where they have replaced a pump with one that provides to much flow the velocity has gone up in the fan coil and caused turbulent flow causing the delta t to be more than 20f which means the boilers cant maintain water temp at where we want it.

now that the preamble is over here's the good part its possible that the second pump is providing just enough of a boost in flow that the velocity in your rad has become turbulent meaning that more watts of heat are transferred to the fins each pass. and since the fan will keep the fins are damn near room temp that means that your extracting more heat from the water.

srry for the long post but that's about as condensed and dumbed down as i can make it. anyways im not saying that is defiantly the case but it's the only explanation i can think of.

In the automotive world two pumps in series (fuel pumps, generally) lets both pumps do less work for greater pressure and flow, it's very much a good thing.

im gonna revise my last keep posted

k dont put them in parrallel unless they are identical pumps. with them in series you have a evffective pump with way more of a shutoff head and a higher flow then either on their own. this however is not as effective as haveing two identical pumps in parallel. srry i mislead you about putting them in parrallel there but i didn't realise that till i made a pump curve for them. anyways im explanation from b4 about y your temps are lower is still valid

Bobnova said:

In the automotive world two pumps in series (fuel pumps, generally) lets both pumps do less work for greater pressure and flow, it's very much a good thing.

Click to expand...

actually both pumps are doing just as much work if you define work as units of energy used. its all to do with pump curves.

If you have a restrictive loop the extra pressure from them being in series might actually give you more flow then the extra flow potential of parallel pumps.

Typically, putting two non-matched flow rate pumps in series is a bad idea as its lowering your efficiency. If one of your pumps isn't maintaining its proper flow rate and you need a second one to reach it then it means you have a poorly matched pump or to many losses in your system. The two pumps you are using aren't to far off from each other so you can probably get away with it. You will however have a lower efficiency than just using a single good pump.

Bobnova said:

In the automotive world two pumps in series (fuel pumps, generally) lets both pumps do less work for greater pressure and flow, it's very much a good thing.

Click to expand...

I don't think so. Putting two pumps in series increases your total head pressure, while two pumps in parallel increase your flow rate. It's one or the other, not both unless you have poorly matched pumps. Also, most cars only have one fuel pump because having two is expensive, redundant, and usually decreases their efficiency. The only cars I've seen with two fuel pumps are volkswagons and as everyone knows, VW engineers are always drunk when they design their cars.

you never put two mismatched pumps in parrallel. one will bget burned out.

Bobnova said:

If you have a restrictive loop the extra pressure from them being in series might actually give you more flow then the extra flow potential of parallel pumps.

Click to expand...

yes you are correct cuase with them in series you have more head and slightly more flow with them in parrellel if they are identical you get double the flow of a single one assuming that each pump on its own would normally have enough of a shutoff head that it could achive max flow on its own. with them in parrellel you dont have any higher of a max head then you would with one but you have twice the flow. so if its a less restrictive loop definatly go with two identical ones in parrellel there but if that's the case there are better ways of getting more flow. like make it a less restrictive loop.

Tophinater said:

Typically, putting two non-matched flow rate pumps in series is a bad idea as its lowering your efficiency. If one of your pumps isn't maintaining its proper flow rate and you need a second one to reach it then it means you have a poorly matched pump or to many losses in your system. The two pumps you are using aren't to far off from each other so you can probably get away with it. You will however have a lower efficiency than just using a single good pump.

Have you tried running the two pumps in parallel at lower RPMs?



I don't think so. Putting two pumps in series increases your total head pressure, while two pumps in parallel increase your flow rate. It's one or the other, not both unless you have poorly matched pumps. Also, most cars only have one fuel pump because having two is expensive, redundant, and usually decreases their efficiency. The only cars I've seen with two fuel pumps are volkswagons and as everyone knows, VW engineers are always drunk when they design their cars.

Click to expand...

Volvo 2/7/9 series have two pumps, a low pressure pump in the tank feeding a high pressure pump under the car.
Because there is high pressure required a single pump faces a lot of resistance, it has to put out 60psi of pressure from one end of it to the other, AND actually move the fuel.
With two pumps each one only has to manage a 30psi change in pressure from one end to the other. Because of that, neither pump has to work hard and more of the energy they have can be spent on flowing fuel, instead of just pressurizing it.
You get a lot more pressure potential then you get extra flow, but you do get both.

the reason is not the in efficiency itself that makes series a bad thing.
its WHY they are in efficient .
2 mismatched pumps operate on 2 diferent curves . when they do you are going to be out of tolerance on both sides usually ( there are some situations where this wont happen but not in field)

so whats so bad about being outside of tolerance in a closed loop?

stress

your system may work fine but over time that stress will wear on the parts until failure.
it may not happen right away or even for a few years but it will happen .

so if its working for you go ahead and keep it as long as you know that some time in the future ( usually when you dont want it to happen ) it may fail on you and you will end up replacing one or both pumps

Thanks for all the good discussion. Experimentally, it seems to be working so I will keep it like this for the while. Maybe just turn on the seconf pump when I will be gaming or otherwise stressing the CPU. I might have just got lucky and happened to have two pumps that work well together in series because their curves match somehow. These are two cheap pumps. One is an aquarium pump and the other a low end pump that I have already had to replace by RMA and it was just sitting around as a backup. If one or both breaks I will invest in a good computer water cooling pump. I upgrade my entire system once a year anyway.

unintentional double post