Heating circulator pump control

alternety

Member
Join Date
Sep 2005
Location
NW WA
Posts
6
I discovered a problem with my new residential radiant heating system. The system is designed to use differential pressure to control the main circulator pump. The objective is to maintain constant water pressure across the heating circuits as the flow changes due to the opening and closing of zone valves.

The contractor supplied a Grundfos variable speed pump and a box that measures differential pressure. Both devices can be set for 0-10V or 20ma interface. Unfortunately the contractor thought this would yield a fixed pressure. It does not. The motor has a simple triac speed control on what appears to be their standard AC motor. The pressure sensor only measures. It provides 0-10 with either a positive or negative change in output that represents the pressure.

I need to do something to fix that. I have spent a bunch of hours on the web so far and I am having real trouble uncovering data and particularly prices. I think I need a PID controller to look at the presure signal and control the motor speed. This needs to be as inexpensive as possible. It would mount on the wall and would need a case to keep inspectors happy. It does not need a display. Once set up it will just run. The control should also have one additional 24VAC input for looking at the heat demand signal and setting the output voltage to 0 when no heat is needed. Or just drop 120V power to the motor.

Can anyone tell me if this is indeed the approach I need and make some recommendations for a unit. It needs to be very simple to set up. I simply do not have the time to write a lot of code from scratch after learning the language. I am working full time on other things on the way late house.

Any help appreciated.
 
A Red Lion p48 would probably work but they aren't the cheapest.

As far as shuting it off, you may have to add a relay, control it with the t-stat,and run your 0-10v signal throuh the relay.
 
I agree with the previous replies. A single loop process controller should do the trick. You should be able to configure the input signal for 0-10vdc or 0-20mA, and output for either signal also.

You may want to look at getting a controller with an "event input". This input (normally a dry contact) could put the controller in a "stop" mode and set the output to zero. Advantage being that this will stop the PID function and prevent any kind of "wind up" of the output when the heating is needed.

I know there are several manufacturers of these controllers. I am familiar with the Omron offering. You could use the E5CN series with analog input and current output (analog voltage out not offered), and an event input board.

Hope this helps.
 
Simple solution: Have the contractor fix it.

Or take a look at the Grundfos CU301 SmartFlo Kit.

Why is it critical that a given pressure be maintained ? If you're worried about deadheading the pump, put in a shunt line.
 
I discovered a problem with my new residential radiant heating system. The system is designed to use differential pressure to control the main circulator pump. The objective is to maintain constant water pressure across the heating circuits as the flow changes due to the opening and closing of zone valves.

The contractor supplied a Grundfos variable speed pump and a box that measures differential pressure. Both devices can be set for 0-10V or 20ma interface. Unfortunately the contractor thought this would yield a fixed pressure. It does not. The motor has a simple triac speed control on what appears to be their standard AC motor. The pressure sensor only measures. It provides 0-10 with either a positive or negative change in output that represents the pressure.

Running constant differential pressure is OK
IF all the loops (zones) are identical)
They rarely are.
Balancing (whether air or water) zones is a royal pain.

IF it were mine I would use individual pumps one for each zone. Little late in the game and more money BUT you won't be fighting the balancing issue.

Dan Bentler
 
Thanks for all the input.

The smart flow pump is for residential water systems not heating circulation.

The system is buit. Has water in it. Rather too late to make major modifications. There are about 20 zone valves. This approach is the simplest and most economical. There are balancing controls on each zone.

I have looked at the Omega. and talked briefly with their tech support. He recommended the CNI16D52. I am going to read the documantation to see why the D is necessary. I suspect to get setpoint from the front panel, but I will look. This is a bit pricey but seems to be what I need. It does not involve having to learn a language and write a program. I just do not have time to spend on anything except finishing other things in the house.

Any other comments are still welcome.
 
Still looking

I have looked at a bunch of the controllers mentioned. The Omega looked promising but if I read the documentation correctly there is no way to provide the event input recommended by gtsuport. It was a bit pricey but I could have lived with that. This does seem to be the way to go. I was concerned about getting the loop control process upset by interrupting either the input or output to get it stopped. The input seems less likely to upset anything.

I have looked at the AB Micrologic 1000, Automation Direct PC35, Mitsubishi Alpha 2, and Tri logic. The AB looks like it has the electronics to do it (1761-L20AWA-5A) but seems quite expensive and requires buying a software package (not to mention programing).

I would like the whole solution to be in the $250 or less range. I am not in the business and it can be a bit trickey actually finding prices on these things.

The ability to just set it up from the front panel is ideal.

I believe I need 1 analog (0-10 or 0-20 ma) in and one digital input (preferable dry contact) and one analog out (v or ma). Ability to use the digital input to stop processing and drop analog output to 0, or keep analog output on but open a relay that breaks the output being fed to the motor controller.

I suspect that I would not want the motor control voltage to actually start at 0 when it turns on because of noise; but I don't know. If the output went to 0 on stop and then resumed at the last level this might make things smoother. I don't know how the system will react. The new flow demand will not necessairly be the same as the flow demand when heat was last needed. This could make stop and open a relay a better scenario.

What would happen if I used just analog input and output and interrupted the input (differential pressure) with an external relay to stop the pump? I could do this by adding something to the boiler controller. There is a "heat demand" signal there somewhere.

One other thought I had. Would it be better to eliminate the TRIAC control on the 120 VAC motor and use a PWM output with some sort of external amplification.

Any help or evaluation of what I think I need to do would be appreciated. Including a cheap and simple controller.
 
So my mom's house has a cintrifugal pump that is run via a relay from the zone valve, which are operated from the thermostats. No fancy stuff here. Each zone valve opens when the thermostat makes, then the zone valves close a contact which operates the contactor which starts the motor. Other then replacing the motor-pump couplings and seals once every 4 or 5 years, it's been pretty realiable for more then 40 years.
 
I have look at this for a few days before deciding to respond. Like Dave I do not understand why the contractor is not responsible for correcting the situation....unless you designed the system and found it not to work after install. Its also possible they just did not tune the system properly.

That aside Dave also mentioned using a shunt, it sounds like an option.

Option 1
Allow the motor to run continuous as long as there is a demand for heat. Use a pressure bypass valve that regulates the pressure in the line. Example: http://www.rkvalve.com/html/show_products.php?id=11
This should maintain a constant pressure in the lines regardless of the valves open or closed. Simple and cheap assuming the motor is capable of handling full flow at rated pressure with all valves open.

Option 2
You could look at using a VFD to control the motor/pump, this could use the analog input and may offer better control. The issue would be if the pump you have could work with a VFD. AutomationDirect offers VFD's under $200 http://web1.automationdirect.com/adc/Shopping/Catalog/AC_Drives_-z-_Motors/GS1_(120_-z-_230_VAC_V-z-Hz_Control)
Using a VFD you can preset start speed at a pre-set point, parameters set using keypad.

Option 3
Check www.grainger.com rotary pumps with adjustable relief valves. I could not recommend one because I did not know pressure and flow needs. These start around $250 at 20psi and 4.1GPM

I have more but do not know enough about the system in question. I have not determined if you have control for the zones or not, I assume they would be controlled via thermostat. I am not sure what you mean by input from the front panel so have not delved into that aspect. I also do not understand why you want to turn off the motor, if it cycles to often it can actually increase power usage.
 
Please address the problem

Guys, I appreciate your spending the time to respond but - please do not go back and try to change the conditions. The system is done. I need to fix the control.

The contractor has no idea how to fix it. He made a mistake in taking the advice of a vendor tech support person from the pressure sensor company (he thought it WAS a controller). It may well have been a failure to share a sufficient technical vocabulary to really communicate. The contractor is a two man shop who made an honest mistake. He is very good at what he does, but this system approach was different than "common" practice. I am not going to ask him to lose his butt on this.

The system is designed to specifically avoid bypass valves. No alternative to this will be considered. Many $$s have been spent to acomplish this.

The pump is a varaible speed pump using what appears to be a simple TRIAC control. The motor is integral. The pump is specifically designed to circulate hot water and is matched to system requirements. Yes, starting and stopping it has some issues in terms of both mechanical wear and maybe energy costs. Overall I suspect lower operating costs but we are not talking about a whole lot of power. Although, over 20 years, with energy costs doing what they are; who knows. I most likely will never see payback on what it has cost.

The system is built out of lots of nice shiney copper and plastic on a pretty white finished wall. It will not be reworked. It would not be acceptable on many levels, but cost is a paramount one.

When heat is called for by any zone the circulator starts. All other aspects of flow and control are dealt with (there are, I believe, 7 application specific controllers involved). Water goes where it needs to in the volume necessary. When no heat is demanded the circulator stops.

All I need is help with the controller between the pump and differential pressure sensor. There needs to be something that closes the loop to keep a set pressure. This is not what I do, although I have an engineering background. I am spending all of my time trying to get the house completed and this is just one more item I thought had been resolved and it came back to bite me.

I need, and will be extremely grateful, for help resolving my specific control problem. I am sure your time is valuable and limited. You represent what appears to be immense depth of expertise on the subject of PLCs. Please put your creative energies to the problem and not waste your effort on other issues.

Thank you,

Harold
 
Let us compare this to an engine. If the basics of the enginer are poor (burned valves, bad rings, etc) then no amount of tuneups will give you a good engine.

The same goes for your system. If the system design is lousy then no amount of controls is going to overcome bad system design. YOu keep saying the system is in and there is no way in Hxxl you are going to change it (and as I read it evaluate to see if there is a design error)

If you are thinking PLC and VFD and $250 overall cost well you just aint gonna get there - you may get two of the three but not all three.

IN my mind good luck.

Dan Bentler
 
What do I know about heating and air... I offered suggestions. I still do not actually know what the specific problem is, except the system was poorly designed and a cheap fix is wanted. What exactly needs controlled?
 
I, like rsdoran, am a bit confused as too what needs to be controlled at all?

I've had three homes with radiant hot-water heat. The oldest one originally had a large (but silent) motor driving the circulating pump, 4 heating zones, and no pressure bypass/control on the water loop. That ran for over 30 years before the pump failed.

That one was replaced by the newer, tiny, cartrige circulation pumps, and both other houses had the cartrige pump's. Again, no pressure control, no monitoring, nothing.

The last house used a water-loop water heater as well, and including that, had 8 zones on the furnace. I never noticed a problem no matter how many zones decided to be active at a time.

Thermostat called for heat, opened valve to zone, and all 'Valve Open' limit switches were wired in parallel back to control the circulation pump.

What makes you so certain the system is broken, before you start fixing it?
 

Similar Topics

Did anyone please help me with gsd file of krones heating module(make Siemens)
Replies
0
Views
74
Did anyone please help me with gsd file of krones heating module(make Siemens)
Replies
0
Views
85
Is there such a thing? I know you can electrically heat structures to keep them from freezing. But is there a way to cool a structure...
Replies
3
Views
572
Hi, I am trying to have my PLC start a heating fan in a linear manner. A certain percentage at a certain received setpoint. To do this, I use a...
Replies
8
Views
1,521
We've had a number of design discussions and done some extensive testing on enclosure heating/cooling solutions for various panel builds...
Replies
0
Views
675
Back
Top Bottom