Using 0-10 volt control to get zero flow in variable circulators

Nimrod66

Has anyone out there had any experience with using 0-10 volt controlled circulators, Grundfos Magna 3 for instance, where it was possible to actually get zero flow with a zero volt signal? We are finding it isn't possible. "Zero" flow is actually quite high for some models in our experience. What might we be missing?

newagedawn

what type of switching relay are u using?

Steve Thompson (Taco)

Great question. Specific to Medium/Large ECM circs (above 70 Watts), there are normally two ways to stop the impeller from turning (stop flow) using a 0-10Vdc to control speed (typical). Optionally you can control differential head (but not commonly used).

You can enable/disable the pump and control the speed with the same 0-10Vdc signal or enable/disable with a digital input (make/break contact) controlling only the speed via 0-10Vdc. Depends on how the pump is wired and set-up (but is a standard feature with no extra stuff needed as long as the pump can accept the Vdc signal).

Probably a little too much detail but from a functionality standpoint, if you enable/disable with a separate signal the control range is from 2 to 10Vdc. Using the 0-10Vdc to enable/disable is from 3 to 10Vdc. The pump is "dumb" and specifically reacts linearly to the 0-10Vdc input (10 Vdc is max speed, 2 Vdc is minimum speed for example), the PID logic rests where the Vdc signal originates.

You also have a run signal output that is kind of nice if you want to prove the pump gets and/or reacts to the 0-10Vdc input.

Nimrod66

Steve Thompson, thank you. If I understand correctly, you are writing that in addition to a 0-10 Vdc control signal, a seperate enable/disable signal is required to stop flow. The 0-10 Vdc signal will result in a minimum functional control range of 2-10. Am I understanding correctly?

Steve Thompson (Taco)

You could use an additional enable/disable or the 0-10Vdc can enable/disable as well as control speed. You call.

Sequence of operation, voltage increase from 0: 0 to 2 volts pump motor is off. 2 to 3 volts the pump runs at min speed. 3 to 10 volts modulates the pump speed between min and max depending on the corresponding voltage.

On voltage decrease: When the voltage drops to 3 volts the pump runs at min speed and will stay at min speed until 1 volt. Less than 1 volt the pump shuts off. Between 2 and 3 volts is a "dead band" to stop the pump from cycling.

One key thing to consider is what happens if the pump looses the 0-10Vdc signal. Answer is it runs at constant, medium speed.

The 0-10Vdc external control is becoming a lot more common with the Boiler guys to directly communicate with the pump from the boiler control (AKA firing rate). Plus there are delta T to 0-10Vdc controllers that work slick if you want to control the pump speed via delta T (installing the temp sensors anywhere you want to maintain the most efficient delta T).

Paul_Gilbert

Manufacturers of residential circulators do not recognize the market need for a voltage controlled ECM circulator to regulate speed directly. The need goes beyond using the 0-10V output from a boiler control board (for example, a Lochivar WHN085) which tries to control the primary loop pump. I have modified both Bumblebees and VT2218s Delta-T control to react to a Arduino microcontroller PWM output but the control "searches" more than necessary because the controller never knows what the speed of the pump is, which is what a voltage controlled circulator would improve. My residential boiler control system uses thermistors in place of thermostats for 2 space heating zones and for the hot water tank in order to control firing rate and zone valves and circulator flow to achieve excellent temperature control and Delta-T for energy efficiency. This system reduces boiler starts to less than 1000 per year (even with a 8+ month heating season, a small 40 gal DHW tank, a family of 6, and a min boiler firing rate of 20% instead of the newer 10% min which would reduce the number of starts even more). During the heating season with baseboard heat (no radiant heating whatsoever), I average 55F delta-T, which allows me to maintain condensing conditions for my entire Maine heating season. The secret is to place the hot water zone (in my case, a BPHE) and the two space heating zones in series. The first floor temperature controls the boiler temperature while the second floor temperature varies the delta-T setpoint slightly to adjust the flow up or down. The few times that a zone is inactive and starts to run again, the microcontroller handles the transient condition by minimizing the boiler firing rate and maximizing circulator flow before the transition such that the transient delta-T does not exceed the maximum that my boiler allows. Circulator flow is also increased when thermistors around the BPHE (which the cold water feed flows through before entering the DHW tank) detect a strong hot water pull. This eliminates the extremely high Delta-T that would otherwise occur at a later time shutting down the boiler. My point with this long description is defining that the weak spot in my current system is the lack of a direct control of speed in circulators. Armstrong is coming out with a Compass H that is supposed to be 0-10V controllable. When will Taco and other circulator manufacturers come out with similar products but at a more reasonable price?? Similarly, other boiler manufacturers should increase their maximum delta-T to at least 60F to get on the boiler efficiency bandwagon. Even Lochinvar isn't perfect because they only allow a maximum boiler temperature of 180F on firing rates controlled by external inputs even though the same boiler allows 190F when controlled by their internal circuit board. To provide improved energy efficiency, boiler makers and circulator manufacturers need to work together to provide what I discussed above and then work with a company to deliver a plywood board with an entire prefab heating system complete with instrumentation and data collection.

Jamie Hall

The question I always have on more complex control systems -- like the one @Paul_Gilbert is describing above -- other than the obvious cost/benefit and service ones -- is this: what are the possible failure modes, and what does the system do in response to the various failure modes? Granted, with a heating system most of the possible failure outcomes turn out to be relatively benign -- no heat -- but some -- a boiler runaway, for instance -- may be less so. I may be more sensitive to this than some, since I used to fly airplanes and some designers and aircraft are much better at that than others (no names, thank you -- if you're really interested, PM me) and the pilot is always the first to arrive at the crash. But it is still an interesting question (and may become more relevant, with self-driving cars...)

Gordy

Paul_Gilbert said:

Manufacturers of residential circulators do not recognize the market need for a voltage controlled ECM circulator to regulate speed directly. The need goes beyond using the 0-10V output from a boiler control board (for example, a Lochivar WHN085) which tries to control the primary loop pump. I have modified both Bumblebees and VT2218s Delta-T control to react to a Arduino microcontroller PWM output but the control "searches" more than necessary because the controller never knows what the speed of the pump is, which is what a voltage controlled circulator would improve. My residential boiler control system uses thermistors in place of thermostats for 2 space heating zones and for the hot water tank in order to control firing rate and zone valves and circulator flow to achieve excellent temperature control and Delta-T for energy efficiency. This system reduces boiler starts to less than 1000 per year (even with a 8+ month heating season, a small 40 gal DHW tank, a family of 6, and a min boiler firing rate of 20% instead of the newer 10% min which would reduce the number of starts even more). During the heating season with baseboard heat (no radiant heating whatsoever), I average 55F delta-T, which allows me to maintain condensing conditions for my entire Maine heating season. The secret is to place the hot water zone (in my case, a BPHE) and the two space heating zones in series. The first floor temperature controls the boiler temperature while the second floor temperature varies the delta-T setpoint slightly to adjust the flow up or down. The few times that a zone is inactive and starts to run again, the microcontroller handles the transient condition by minimizing the boiler firing rate and maximizing circulator flow before the transition such that the transient delta-T does not exceed the maximum that my boiler allows. Circulator flow is also increased when thermistors around the BPHE (which the cold water feed flows through before entering the DHW tank) detect a strong hot water pull. This eliminates the extremely high Delta-T that would otherwise occur at a later time shutting down the boiler. My point with this long description is defining that the weak spot in my current system is the lack of a direct control of speed in circulators. Armstrong is coming out with a Compass H that is supposed to be 0-10V controllable. When will Taco and other circulator manufacturers come out with similar products but at a more reasonable price?? Similarly, other boiler manufacturers should increase their maximum delta-T to at least 60F to get on the boiler efficiency bandwagon. Even Lochinvar isn't perfect because they only allow a maximum boiler temperature of 180F on firing rates controlled by external inputs even though the same boiler allows 190F when controlled by their internal circuit board. To provide improved energy efficiency, boiler makers and circulator manufacturers need to work together to provide what I discussed above and then work with a company to deliver a plywood board with an entire prefab heating system complete with instrumentation and data collection.

Tell us how is the system piped? Direct, or primary secondary? Where are you getting the 55 delta? I assume you mean at the boiler.

The whn has a 35* max delta for the boiler. There are reasons for this. Mostly to do with the HX longevity. Running a really high delta across the HX will put undue stresses on it.

If you are running 180 to the emitters, and sending 125 back (55 delta) if piped direct. The way you get the return temp down is by increasing emitter size to lower supply water temps thus lowering return water temps.

Paul_Gilbert

I see some interest and some skepticism to my post. About safety, it is important to know that the BMS input is always overridden by the boiler controls. If boiler temperature is exceeded or delta-T is exceeded, the boiler firing rate is reduced or even shut down. So external control of the boiler never sacrifices the manufacturer imposed safety limits. About the maximum Delta-T of the Lochinvar WHN085 boiler, I was assured by the manufacturer and his distributor that the maximum delta-T of the boiler exceeded 55F. I have found the boiler starts to throttle back the firing rate when the delta_T reaches 58 and the boiler shuts down at about 62, which is why an external control is needed to avert these conditions.

The other question was how can I heat my upstairs zone with water cooler than the first floor? The answer is simple. My house is a colonial and the original builder installed the same amount of baseboard for the second floor as for the first floor. The second floor is kept 4 degrees cooler, has updraft from the first floor, and has less heat loss via doors and windows. By feeding the return of the first floor to the second floor, the control seems to satisfy the needs for the second floor by staying at a delta-T about 55F. For houses with less baseboard on the second floor, this scheme would operate with more flow and result in lower deltaT. But unless the second story is truly undersized with baseboard, operating such a scheme will always result in a greater DeltaT than the standard parallel arrangement for zones.

As far as how this boiler is piped, the first circulator pushes the boiler output through the BPHE and back to the boiler if no zones call for heat. If the first zone is calling for heat, a zone valve in my loop closes, forcing the boiler flow to go through a B&G Flo-Control valve and then through Zone 1. The return from zone 1 reenters my loop on the downstream side of the previous mentioned zone valve. I should have piped Zone 2 exactly the same but I was afraid that I would need more circulator power. I piped a second circulator branching from my loop immediately after the return from Zone 1 and I connect my Zone 2 return immediately after a swing check valve located in my loop (between the supply and return of Zone 2). Had I piped my Zone 2 exactly like Zone 1, I could have run my entire house on 1 circulator and two zone valves. How do I know that? I currently run the entire house on just my Zone 2 circulator when both space heating zones are running and I still have to throttle the flow even with this circulator running at minimum speed. In this loop, I have a Taco-4900 air separator and Extrol expansion tank ahead of the first circulator and strainers ahead of both the BPHE and the boiler. The DHW tank does not have a coil. A 006 circulator circulates my DHW tank from a lower port through a check valve and combines with the cold water feed to go through a strainer ahead of the BPHE before it re-enters the tank near its middle. This design was chosen because I already had the tank in my old system but I have to say that I would pick the BPHE and this much less expensive stone-lined tank in a minute over the more conventional tanks with internal coil.

My Arduino outputs about 20 inputs and output values every 15 seconds to my PC which stores this data in a file, which was quite handy for debugging my control logic at the beginning. A power outage does not affect the Arduino. It takes about 1 second to power up after an outage and it restarts the boiler and continues on.

People are right to be concerned about complexity because there is a lot to be said about reliability when heating a home in winter. Having said this, the only failure I have had in 4 years is in one of my circulators that would occasionally decide to not start, even when it had run less than 5 minutes before. My $5 Arduino has not failed, nor any of the relays that run from the Arduino. The thermistors have not failed either and the fact that the boiler system is kept running through most of the winter probably introduces less stress to the HX than having the boiler starting and stopping and running cold water from inactive zones back into the boiler which happens with thermostat operation.

For me, I enjoy the thrill of condensing all winter long while watching the big plume of steam exhaust from my neighbors' systems in the middle of winter. I even go through 20 feet of pipe within a pipe heat exchanger to preheat my incoming air with the boiler exhaust before bringing both of these pipes through my roof. My calculations show this to gain me between 1.5-2% additional efficiency. However, the additional >5% efficiency of my system to a standard condensing boiler with standard controls during the worst part of the heating season has to be discussed with some increased reliability problems that occurs with more automation. The future always points to higher efficiency even at the expense of more automation. The winning formula is introducing this efficiency in a robust and reliable manner with potentially redundant spares installed. To those circulator companies out there listening, the first step in this robust system is introducing an ECM circulator with 0-10V or PWM speed control.

Paul_Gilbert

The boiler manufacturers also have to get on board this efficiency program. They have to get their maximum deltaT to that of Lochinvar or even exceed it. I believe that the Lochinvar Firetube HX is manufactured in Poland and is used in many boilers today. Many of these manufacturers stipulate a lower maximum DeltaT than Lochinvar. Do they have actual data about its effect on reliability or are their DeltaT's chosen by what they think the market will accept. 20, 30, or 35 degree deltaT may sound like more than enough when you consider zones in parallel, but this will never get you to condense in the middle of winter. I believe manufacturers should strive to reach a maximum delta-T of 70 degrees, if possible.

Dear boiler manufacturers: Throttling a BMS input firing rate when boiler temperature exceeds 180F is stupid when you allow your internal controls to operate up to a maximum of 190F. The Arduino controller is programmed to be extremely cautious about increasing firing rate or reducing circulator flow when in the area of maximum boiler temperature - much more so than the controls in my boiler that acts the same in that range as in a lower temperature range. I was told that this 180F limit cannot be changed. Please don't limit us unnecessarily.