Pump question

[Jens]

I would like to increase the amount of water pumped through a heat exchanger loop when my Listeroid is running. I use Taco circulating pumps in my system. There is one loop that runs 24/7 but I would like to increase the amount of water flowing in that loop when the engine runs.
Question: Can I parallel two Taco pumps to significantly increase water flow (the second pump only runs when the engine runs) or would I be better of using a single multi speed pump ? Is there any way of figuring out the difference in water flow between one pump and paralleling two pumps ? Since regular circulating pumps are built to go slow and steady and forever, would I be better of paralleling a regular circulating pump with something that is meant to pump a lot more water but not necessarily in the form of a circulating pump ? The idea being that when the engine runs then the heavy duty pump runs doubling or tripling the water flow but if the engine doesn't run then just the regular circulating pump runs.
I run into a situation where I see boiling in the loop in question - the thought is that if I increase flow I can decrease the delta T across the heat exchanger and avoid boiling until a much higher system temperature is reached.
Has anyone done this kind of thing before ?

[Ronmar]

Where are you getting the boiling/overflow at again, primary or secondary loop?  Which pump runs 24/7?  If I recall from the drawing you posted, the engine and EXheatex are in series.  If you still have it like this, that is probably going to continue to cause you problems.  The biggest reason is that heating already hot water is putting your feedwater temp into the indoor heat exchanger nearer boiling, or perhaps with head pressures, above boiling.  Any little flow or heatexchanger transfer rate issues are going to allow the temp to peak and cause boiling.  IMO, you should be feeding your heat sources(engine and EXheatex) with the coolest water possible(return from indoor heatex).  

At any rate, You need to hand over hand your plumbing and find the smallest fitting in the loop.  That will be a big factor in your flow restriction.  Height change and pipe diameter and length will also play a part.  Circulating pumps do not have very much positive displacement.  IE: they cannot build much pressure and are incapable of pumping to a very high head.  Look at the pump flow graphs for evidence of this.  The rated flow tapers off quickly with an increase in head.  Any choke points, kinks/flat spots in hose bends and the numbers of twists, turns and elbows, as well as height changes all add up to the same thing, an increase in head that the pump must pump against.  

Here is a test for you.  Put a valve and a pressure gauge on the existing pump outlet.  Run the pump, and measure the pressure at the pump outlet while circulating water thru the system.  Then close the valve and deadhead the system and measure the pressure increase at the pump outlet with no flow thru the system.  With a low pump outlet pressure(low head) the pump is producing maximum flow.  As the pressure increases, it is producing less or no flow.  The difference in pressure between the valve being open and closed will tell you how close you are to the pumps head/flow restriction limit.  Another way to do this would be to put a gauge fitting and a garden hose on the pump outlet.  With the pump running and the garden hose discharging water at the same height as the pump outlet, you should have low pressure.  Now raise the end of the hose above the pump till it stops flowing water.  This is the pumps actual head limit.  Measure the pressure at the pump outlet at this point.  If your normal pump outlet pressure while feeding your system is near the measured max head pressure, a second pump or larger pump is probably not going to help much. If I recall, your three secondary loop storage tanks are in series?  Putting those in paralll would probably help a lot with secondary flow. 

Splitting the flow thru your two heat sources should actually help your overall flow thru your primary loop.  Plumbing/hydraulics and electronics have a lot in common.  Resistance in series is additive.  Resistance in parallel is subject to a different formula.  As an example, if I put a 500 ohm and a 500 ohm resistor in series, I would have a total of 1000 ohms of resistance.  If I put the 500 ohm and the 500 ohm resister in parallel, the combined resistance would be LESS, about 250 ohms as the flow is divided down 2 pipes... In the case of your 3 tanks, like three 500 ohm resisters in series = 1500 ohms total, three 500 ohm resisters in parallel = only 166 ohms...

[Crofter]

The problem with paralleling two centrifugal pumps is that if the second pump causes increased discharge pressure into a common line, that is seen as added resistance by the other pump and decreases its output. If they are both pulling from a common intake, the same effect is seen at the intake and is seen as higher lift.

If your exising pump is small in relation to line size (small cross head pressure differential) you could gain by adding another alongside, but If pressure differential is already quite high you would have diminishing returns from the second pump.

Edit, Ronmar got there first with the long answer!

[Ronmar]

How big is the heatex between the primary and secondary loop?

[Lloyd]

Jens

What about installing the second pump in series, far up near the top of the head, as a booster pump? It should not run into the same problems that Ronmar was speaking to...it would even take a little load off the main pump.

Lloyd

[Ronmar]

Jens  With the size of you primary heatex, I don't know if a flow increase is going to help.  I use a 5X12 10 plate for aroung 600 SQ/IN, moving around 17,000 BTU/HR at 3KW of load.  You are probably getting something similar per KW.  If you only have 3KW of electric load, with coolant and exhaust heat combined, you are probably close to 34KBTU/HR.  6KW will probably net you 68KBTU.  3X12X30 plates is around 1080 SQ/IN  Even with the high efficiency of a flat plate heatex, I think that heatex might be a little small... 

I also don't think with your series tank plumbing, you are always feeding the coldest water possible to the heatex secondary side.  This allows the primary loop temp, which has antifreeze? to peak above 212F which in turn boils the secondary water in the heatex.  If you had parallel flow thru the tanks, I think the water in the tanks would have more time to stratify and leave the coldest at the bottom of all 3 tanks to be sent to the heatex input.

I am setting here with an injured shoulder and can't get to my wifes 4Runner project today, so I am a little bored.  From your diagram and descriptions I drew this.  Some of the flow directions are just guesswork.  Give me some feedback and I will refine it.  Like most internet troubleshooting, it is hard to make suggestions without all the facts.  A clear diagram can sure go a long way toward clearing up any confusion...
   

[mobile_bob]

Ron:

i am so impressed with your computer artwork, what do you use?

bob g

[Ronmar]

Well you see, that is just the thing, More flow will not necessarilly get you more heat transfer.  Heat transfer is mainly about temp diff, surface area and time.  You may get a little more temp differential with increased flow, but you will be shortening the time you are exposing the water to the primary loop which = less heat transfer.  I think your net gain from increased flow will be zero.  It may even reduce overall heat transfer, but I guess you won't know till you try.   

How big(GPM) is your secondary loop pump?  Where are your expansion tanks located?

I would say that if you can hear it in the heat exchanger, then it IS boiling.  I would guess it would sound a little like a muted popcorn popping sound...

Thanks Bob, I use the basic microsoft paint program that comes with windows.  I draw by pixels, which allows me to scale.  I typically draw at a 1/8" per pixel scale.  I have found it a very simple way to sketch out and visualize concepts.

[Ronmar]

Hey Jens what is the typical temp increase across the first heat exchanger's secondary loop ports when you are "charging" the storage tanks?