Best Position for a Circulator Pump

Hi VT,

Yes, it is better to "push". You always want to regulate flow with a valve at the pump discharge. Also, if you don't have enough inlet/suction pressure, you will get cavitation.

Mike

It's important to avoid air entry into the system to minimise corrosion, by ensuring there are no points of negative pressure. I prefer arrangement, boiler - open vent - cold fill - pump - rads, with open vent and cold fill not > 6" apart. It's the cold fill point that determines the pressure starting point in the system, and this arrangement keeps the pressures high.

It's possible to combine the vent and fill in one pipe but I'm not keen on it.

At one time it was common to pipe the cold fill to the boiler return. Due to headloss across the boiler and pipework, this causes "see-sawing" of aerated water in and out of the tank when the pump stops and starts, and increases corrosion risk.

Cheers......Codey

All of the circ pumps I have installed all pull from the boiler, main loop, zones, radiators and pushes back in the boiler return. These little circ pumps do not pike to push. At least all of the little units I use.

I can't speak for the hydronic system but conventional hot water heating systems almost always have the circulation pump as close to the inlet on the boiler as is humanly possible, consistent with the owner or service technicians ability to provide easy access for maintenance and or removal when needed.

Conventional hot water boilers, regardless of age, return the cooled water to the bottom of the boiler where it naturally rises as it is heated, leaves the boiler and then is routed to the various zones.

Flow through the piping is often regulated at the return end by gate or ball valves that restrict the returning water, just short of the circulation pump.

I've not studied why they do this but the commonality of the installations suggests some method to their madness.

L.j.

Hello LJ

What is the definition of a hydronic system? When I come to think about it, I'm not sure just what it means.

Pump before boiler is OK, and is the norm when the pump is built-in with the boiler. Pump is in the bottom of the boiler case so pumping to the boiler makes sense and allows the open vent and cold feed to be in external pipework, at a location to suit the overall system. If the (built-in) pump were on the boiler flow, putting vent and feed between boiler and pump (if there's room) would mean modding the pipe, and location likely to be unsuitable.

Only disadvantage I can see vs my #2 is that air bubbles released in the boiler have to go round the system to get to the open vent, but that's only on initial start-up.

Cheers......Codey

A pump has energy supplied to the impeller from the motor, this is generally fixed in the case of a constant speed pump. this input shaft energy is converted into two forms of output energy (i am ignoring minor losses in the form of heat etc.).

pumps output 1.) mass flowrate (of the pumped liquid) and 2.) pressure. in various combinations that comprise the pumps operational 'curve' - this curve is the combination of points of flow and pressure that a pump can create and is unique to each pump. With any centrifugal pump, when pressure differences across the pump are low, more input energy is available to move mass, when pressure is high less input energy is available to move mass because more of that avaliable energy is taken to overcome the high pressures delta

So for a given pump at a give flow rate, there is a given and FIXED change in pressure between the inlet and outlet of the pump. When you install a pump on the system and the pump has to pull water to it, because there is not adaquate pressure to feed the pump you risk the pump dropping the pressure on it's inlet side to a point of caviation. Which is that point where a liquid's vapor pressure is exceeded and the liquid goes temporarily to a vapor stage. this will damage the pump.

it is best to push the fluids with centrifical pumps rather than pull, although these are relative terms. make sure that your system provides adaquate supply (inlet) pressure at the inlet eye of the pump for the design flow rate and it will not matter if you say you are pulling or pushing.

"in various combinations that comprise the pumps operational 'curve' - this curve is the combination of points of flow and pressure that a pump can create and is unique to each pump. With any centrifugal pump, when pressure differences across the pump are low, more input energy is available to move mass, when pressure is high less input energy is available to move mass because more of that avaliable energy is taken to overcome the high pressures delta"

i just needed to clarify one question:

suppose i have a pump operating at a mfr of say 50 g/s and delta P of 3 bar, in a closed loop. will there be any difference if the ambient pressure in the loop is

a) 10 bar {ie suction pressure 7 bar and outlet pressure 10 bar( or 10 and 13?)}

b) 20 bar{ie suction pressure 17 bar and outlet pressure 20 bar(or 20 and 23?)}

m a bit confused...

The position of the pump is not critical provided it is not cavitating and this is a function of the static pressure and the fluid temperature. For most low temperature heating systems (80C max) the pump can be in either the flow from the boiler or the return to it. Generally the cooler the pump the longer it lives! Therefore I prefer to put it in the return line. If the system is operating above 100C then the suction pressure at the inlet to the impeller is the determining factor and a sufficiently high margin to avoid cavitation is needed. Typically 1.0 to 2.0 Barg above the saturation pressure would be needed. As these systems tend to be pressurised water there teds to be a rising pressure characteristic as the system heats up and this can be taken into account in achieving the anti-flash margin.

Provided there's enough "Net Positive Suction Head", it won't matter, as with any centrifugal pump.

Two of these blighters are installed on the first floor of a building nearby, with the boiler on the ground floor. There is 2m of pressure feeding the system from an expansion tank in the loft space. No problem with the system, though it pays to carry out routine air bleeding as periodical preventative maintenance ("PPM"). If the pump "roars" in operation occasionally as entrained bubbles pass through it, increase the PPM frequency.

VT -

I don't know if you're still curious, but I just saw the link to this post pop up while looking at other things, and had to browse the responses to see what came up.

Most of the posts have kernels of good info; some terms may lead to confusion if you don't already know what they are trying to say, and hopefully in the mix you have had your questions answered.

For anyone not familiar with exactly what is meant by a "closed hydronic heating system" be careful not to provide suggestions that only apply to domestic hot water return lines and pumps, not space-heating hydronics (through coils, radiators, and/or pex radiant heating zones).

In general, it is not only most efficient for the system to have the pump "pushing" fluid into the load (made up of all zones and 'uses' of that heated water), but the clues above about the point of zero-pressure-change in the system are important. Due to many factors, if you have a primary loop circulator, it is properly installed in the supply piping right after the Expansion Tank. This means in the supply line coming out of the boiler; the air/dirt separator, Expansion Tank, and makeup water connection should be between the Boiler and Pump. The order is: 1) Boiler(s), 2) Air Separator, 3) Tee-off for Expansion Tank, typically with makeup water as well, 4) Main loop Circulator Pump, 5) Supply header off of which loads are taken. This may include branch loops, fed by circulator pumps pushing into zones, or routed in reverse-return long loop format with a flow/control valve after each load coming back. This all then ends up connecting into 6) a common return header that gets larger as each return joins, and which then leads back to the boiler plant.

IF the boiler(s) are not condensing-mode type, it is possible that there may be a larger return piping off of which there is a pair of closely spaced "Tees" serving the boiler plant, and if so there will be an "injection pump" at or built-into each boiler. When this is present, the small injection pump may then push "in" to the boiler, at exactly the flowrate the manufacturer selected for proper boiler capacity. That heated water will then effectively be "injected" back into the main loop through that second "tee" of the pair. The water then is mixed before going through #2 & #3, then #4 pumps, then out to the #5 loads again.

The manual air removal at the top (highest physical point) of the piping system is only for filling the system. The airscoop should be removing the entrained air automatically, and 'may' have the Expansion Tank attached at the same point.

btw-If you ever find yourself ready to invest in a new boiler, do NOT allow yourself to be talked into "saving" money on a cheap boiler, GO CONDENSING unless you're just getting the house ready to dump on the market. Oh - and you can likely apply for a 30% energy-saving-investment tax rebate, if available at that time.

To play with a nice computer model of your system, as well as back-checking the sizing and design of your system, you can download the free hydronic-only version of HVAC Solutions that is provided through TACO pumps, downloaded here:
http://www.taco-hvac.com/en/products.html?current_category=84

Free video tutorials are available right online and you should be playing with a real model within an hour. It's unique diagrammatic software, and easy. You can fill in actual pipe diameters and lengths for each section, and get real-life calculations for what is going on in YOUR system.

For the best full-explanation of all hydronic designs I've found - and use as a reference text - get a copy of this:
http://www.amazon.com/gp/product/0766816370