Confused about "pumping away" and Knight WH installation

indevrede

First, "Hi." First time poster. I'm trying to understand and apply the principal of pumping away from the PONPC, but a figure in the Lochinvar Knight WH installation manual has me confused. (If the figure doesn't attach, it's Figure 6-10, page 45 in the WH installation manual). The figure shows piping for "full flow" (i.e. not primary/secondary) with a DHW priority loop. The confusion I have is that both the Boiler (=System) and DHW circulators are pumping into the boiler, which then feeds directly into the expansion tank. Since the boiler has such low head loss (0.5 ft/hd @ 7 gpm for the WHN-055), doesn't that break the "pumping away" rule?



Wouldn't it make more sense to move the expansion tank and the two circulators, so the flow would go from boiler --> expansion tank --> tee --> one circulator on each leg of the tee? That way both pumps are immediately after the expansion tank. I asked Lochinvar tech support but they said that the installation manual shows the configuration that's been "tested for years" and that the boiler needs to be pumped into to ensure flow (i.e. preventing overheating).



What do you guys think? Is the manual right, or not? And if not, do I risk voiding the warranty if I install it against Lochinvar's recommendations?

Snowmelt

Pump away

When I am doing an application like that I put the expansion tank before the pump and the seperator after

indevrede

Wouldn't you need two tanks?

OK, leave the air separator where it is, and move the expansion tank to the inlet of the boiler/system pump. But doesn't the DHW pump also need to pump away from the PONPC?

Gordy

You are pumping away

The air separator is getting the hottest water for better air removal. That's why it's in its location in the diagram. They want you to pump into the boiler.

indevrede

Right, leave the air separator where it is

I understand that the air separator goes after the boiler, I'm not proposing to move it. My question/concern/confusion is that the diagram shows both pumps going into a low-head-loss boiler, then right into the PONPC (expansion tank). So essentially they illustrate pumping toward the PONPC.



The fix seems to be either 1) move the expansion tank upstream of the tee that splits off the DHW loop and move both circulators immediately downstream (one on each leg), or 2) don't move the circulators, but move the expansion tank upstream of the boiler circulator and add a 2nd expansion tank upstream of the DHW circulator.



I've never seen a diagram illustrating 2), and 1) is apparently against the manufacturer's recommendations...

Gordy

per diagram

Are you using zone valves?



Are you piping with differential bypass as suggested?



IF yes to both then the primary loop will be pumping away from the PONPC

Zman

Respectfully disagree

If this was the kbn series, I would agree with the drawing and Gordy.

The wh series firetube boiler has the headloss of a bath tub.The circulator is essentially pumping directly into the expansion tank. The bipass valve or the zones would have far more resistance than the boiler.

Carl

heatpro02920

I would follow the manufacturers instructions

Install it how the people who designed the boiler recommend. Is this a new install or are you hooking to an existing system?



I would use separate circs for each zone with a properly sized header.. Also I like to pipe mod cons primary secondary, even the models that they say don't NEED to have pri/sec, I find it leads to less issues...

Mark Eatherton

Yeah, that's a dilemma for sure...

The CH pump IS pumping towards the PONPC, and it COULD lead to air/noise issues. It is critical that both pumps have checks, and that they face each other (as they are shown). If the PAB were installed, then the CH pump WOULD be pumping away from the PONPC. Without it, or another intentional bypass, then you are correct, it is not.



As a solution, if a PAB is NOT incorporated, one could take a nickel ($.05) and drill a 1/4" hole in the very center of it, drop it into the bell of a 3/4" CXC ball valve, insert copper pipe to hold it in place and solder it in between the supply and return manifolds, then close the BV for purging, and once purged, open the BV so that CH pump can "see" the PONPC, and it will produce all positive pressures. I have had to do this numerous times and it works perfectly. Pressure potential lost due to minor 1/4" bypass is negligible.



ME

Zman

Bipass

Mark,

Wouldn't the system "see" either the bipass, a zone, or the the setup you describe as resistance? Wouldn't that resistance be far greater than the boilers resistance? I don't think the circulator will know the difference. The pressure will drop on the intake side regardless.

This post brings up some interesting points.

Carl

Gordy

Carl

I agree with you. I was thinking wbn.



Look at the schematics for wbn verses whn



The wbn is p/s with x tank in the heating loop. Whn is

Identical schematic eliminating p/s. Hmmm seems that ponpc may be variable depending on how many zones are calling no.

SWEI

Fire-tube HXes

In primary/secondary arrangements work quite well with multi-function hydro separators. 



Be sure to size the primary pump carefully (it will end up one or more sizes smaller than anything listed in the manual.)  Look at the curves on p.36 of the IOM.  For a 35ºF ∆T (maximum efficiency) the 055 needs only 3 GPM.  Given that the HX presents 0.1 feet of head at this flow, even a 003 is oversized for most installations.  Take a look at the B&G ecocirc e3 auto.

Mark Eatherton

Think tunnel vision Carl...

It can see the PONPC, but it is a long ways away, and the further away it physically is, the more "droop" it will see.





ME

indevrede

Proposed changes: OK?

Thanks for all of your responses!



After thinking some more about Snowmelt's suggestion (2nd post) I now agree that a single expansion tank located immediately upstream of the system/boiler pump should work. My initial concern that the DHW pump needed a 2nd expansion tank isn't warranted because the upstream connection to the system loop should act as a PONPC for the DHW loop.



In the figure below, I've moved the expansion tank, and also moved the Air Separator (AS) and Dirt Strainer (DS) so that both the System and DHW loops get the benefit of both filterings. Note that the best place (IMO) to put the DHW circulator would be immediately downstream of the "Virtual PONPC", right before the DHW tank. But, since I'll be using a DHW with very low head loss (Turbomax 23, thanks mostly to Mark's advocacy!), the DHW circulator can go after the DHW tank.



So, have I struck a good balance between satisfying Lochinvar's requirements for the circulator locations (both feeding the boiler), and also satisfying the "Pumping away" principal?

Zman

That works

I don't see a problem with what you are proposing.

Personally, I don't think lochinvar has spent time updating the drawing since they took on the firetube boiler. If it were me, I would put both circs as well as the expansion tank and fill valve on the supply side and call it a day.

I understand you wanting to pipe per manufactures spec.

Carl

M Lane

Deleted

Second thoughts.

hot_rod

Ergomax

builds a multi-port tank now. It is designed to be a separator, buffer and DHW source. That would simplify the pumping/ piping issues.



Add a multi purpose air/ dirt sep and away you go.

Pughie1

Ergomax

Thanks Hot Rod - Glad to know about this.

John Pughe

remodel

PONPC

Guys, I see with the bypass this sucker is pumping away but only towards the boiler and DHW tank and lets say PONPC = 25' of head less droop you have maybe 24' of head pump kicks on we are at 29'.  Without bypass pump is starting to subtract from PONPC.  But out to the system how could it be pumping away, by definition that is the fixed fill head, aren't those zones seeing less head than PONPC when pump is on?  How could pump add to that PONPC out to the system when boiler supply goes right into the PONPC.  This was a good post and a good catch. 

Jason_13

Nickel?

ME hope you bid that nickel into the job.