VS Circulators in both primary (boiler) and secondary (emitter) loops

nosirra1Arrison

I see Lochinvar Knight and other boilers have a variable speed (VS) circulator in the boiler loop. I am used to seeing the VS circulator on the emitter side to deal with varying flows needed. Is there a strategy needed if VS is used in both loops (boiler and emitters- two circulators). I am imagining the situation where they would endlessly readjust against each other. Would one have to be set on constant speed? Is there another strategy for this scenario?

Colorado_Dave

What kind of variable speed?  Like auto sensing? Three speed?  Delta P? Delta T?

I mean, all four types could be used effectively on a proper primary loop to help accommodate different lengths of primary. 

Too often I see cookie cutter installations because the installer didn't know how to read a pump curve..... using a UPS26-99 on high speed for a 5 foot primary loop.   More is not always better. 

Robert_25

Assuming things are piped correctly and you have hydraulic separation between the circuits, there should not be an issue.

I have VS ECM circulators in the primary and secondary loops on my system, and even when running them all in "Auto" I have not observed one influencing the other.

Dave H_2

The variable speed circs used with the Lochinvar boilers are specific. The boiler is sending out a 0-10v signal to the circ (which needs to have that input to control speed) in order to add efficiency to the boiler based upon alot of parameters that are internal.

You are correct depending upon the situation and the components used that they would be constantly adjusting. It would be like having one car with two drivers.

Now a delta p circ installed on the boiler loop will never adjust its speed. The pressure drop across the heat exchanger and the associated piping is constant. Even in "auto" mode. The circ will try to flow as much water as possible through.

A 0-10v circ will adjust its speed based upon the signal from the controller, in this case the boiler. A 10v signal tells the circ to run 100%, a 5v signal tells it to run at 50% speed and so on and so forth.

As long as the systems are piped primary/secondary or LLH and such, the circs are not going to influence flow in the other loop and vise-versa. Thats the idea of that type of piping, they are hydraulically isolated from each other.

Dave H.

HomerJSmith

I read on this forum that the high capacitive boiler pumps, such as Grundfos's UPML. etc., pumps, can damage the system boards because of excessive current flow at startup. I take this seriously and have decided to use an isolation relay on my installations. Why take chances.

I talked to a Lochinvar technical person and he seemed to be in agreement, although, I would expect if it was a problem in the field, there would be a notice about it.

System boards are expensive and I don't want to replace one at my cost.

I would like to know your thoughts on this matter.

Dave H_2

With some of the ECM circs, there can be an inrush current (less than 1 seconds worth) higher than the safety relays built into the on-board controller of the boiler.

However with that being said, the 0-10v circs can still be wired however you want, switching relay or directly to the boiler. The 0-101 volt signal is a separate control wire.

I was only aware of one on-board controller with NTI, they had recognized it and notified their installers of best practices but it has been since resolved.

BigRob

I second Dave's comment. I installed a Taco VR on our DHW system and assumed because it was lower watts it would be fine to use the Lochinvar integrated pump relays (rated 1.8A on my units). The relay started sticking closed not too long after and I caught it. After a little fiddling and temp cycling I was able to get the relay contact un-welded, but I switched over to another relay and enabled it in software to be safe. The Taco VRs require a 2.4A or something (from memory) relay, so I used a Functional Devices brand interposing relay. I like FD - they have a line with built in enclosure nipples, so an easy install. They make dry contact monitoring relays, too.

I use the Armstrong compass H on our heating system with the 0-10V input. It uses a PID loop that isn't too fast or slow, so it's fine and saves power. I suppose a constant delta across the heat exchanger helps things, but it's probably in the noise. The electric savings are real from a % point of view.

Before the compass I used a triac control to a regular grundfos for years with no problems. The only problem with the compass was its first season- it wouldn't startup sometimes and I got some high limits. Unclear if it was a wiring issue or what was responsible. I re-wired it and bumped up the Lochinvar min voltage to 3v and it seems to be ok now. Not sure what was up since the compass shuts down below 2v. It was set to 2.5v originally and should have been fine. I confirmed everything with a meter when I set it to 3v.

EBEBRATT-Ed

@HomerJSmith

That's a common practice in some areas. I used to build some automation panels and the controllers had built in relays so they would put additional panel mounted relays for all the outputs so if something shorted in the field it wouldn't take out the controller.

Good idea

hot_rod

If the two circulators are separated via a hydraulic separator, or buffer tank there should not be a problem. The boiler side pump would modulate from the boiler control. The system pump could be a dealt P with zone valves for example. You would not want to run both circs off the variable speed boiler function inside the boiler control. That logic and circulator are designed to work together to maximize boiler efficiency.

I'm building this Knight for my shop. I'm curious how the VS pump modulated so I added a Quicksetter to check flow rates as the VS boiler pump modulates. A delta P circ will be on the system side for a zoned radiant system.

BigRob

I have a SEP4 as well - really grabs air and dirt, well. The SEP4 connects nicely to the Knight.

CMadatMe

Dave H_2 said:

The variable speed circs used with the Lochinvar boilers are specific. The boiler is sending out a 0-10v signal to the circ (which needs to have that input to control speed) in order to add efficiency to the boiler based upon alot of parameters that are internal.

You are correct depending upon the situation and the components used that they would be constantly adjusting. It would be like having one car with two drivers.

Now a delta p circ installed on the boiler loop will never adjust its speed. The pressure drop across the heat exchanger and the associated piping is constant. Even in "auto" mode. The circ will try to flow as much water as possible through.

A 0-10v circ will adjust its speed based upon the signal from the controller, in this case the boiler. A 10v signal tells the circ to run 100%, a 5v signal tells it to run at 50% speed and so on and so forth.

As long as the systems are piped primary/secondary or LLH and such, the circs are not going to influence flow in the other loop and vise-versa. Thats the idea of that type of piping, they are hydraulically isolated from each other.

Dave H.

Hi Dave,

Hope you are well. Been a while since we've spoke. I'd like to throw a curve ball at ya just as a means of discussion - If my heat loss is only 40K - I have a typical 110K Btu/hr Condensing Boiler that has a flow rate of 5gpm across the HX, would I even need to be pri/sec piped? I know fthd plays a role but with data on residual head availability in most if not all install manuals I'm pretty confident there would be no role for pri/sec piping. I look at it like back in the day when you were designing 3/8" joist radiant systems when nobody else was. Chris.

hot_rod

Is it a zoned system? The issue is making sure the boiler always has the minimum gpm required. The manual should indicate min and max gpm though the HX.

Primary secondary, better yet a hydraulic separator assures you have adequate flow in the boiler and system.

If it is zoned, a delta P circ works nicely with P/S or separators.

Zman

The 0-10 VDC circ on the Lochinvar boiler loop is really slick. It optimizes the delta t to assure low return temps and optimal efficiency. There is virtually no risk of thermally stressing the boiler because the boiler is controlling its own supply and return temps.
I too am under the impression that the ECM inrush issue is fairly isolated. My guess is that they either cheaped out on the board relays or made a design error.