Triac speed controller for circulator motor
We have two Weil Mclain WGO-5 boilers heating our 4-unit condo building in a 100-year-old converted gravity system with cast iron radiators. The boilers are 30 years old and run from a cold start. Due to the cold start (with no bypass) and the high flow rate from the Taco 007 circulators pumping through a low-head converted gravity system, we were getting about 19 gpm of cold water pumped through the boilers on every start. And because the boilers only run for about 30 minutes on a typical cycle, the high volume of water in the converted gravity system never gets really hot, so the boilers usually finish with water temps around 130 degrees.
I’ve been warned by the experts here that the flue gases may be condensing due to the low water temperatures, but after 30 years of these boilers running low temperature water, there’s still no evidence of condensing. The heat exchangers are in good condition and I inspect them annually alongside the boiler tech who cleans them.
The other result of the high water flow rate is a relatively low delta T across the boilers of 12 degrees. The boilers will output the same BTUs/hr regardless of the water flow rate, and they’ve been heating the building this way for 30 years, so the delta T may not be that important. But I’ve learned here that it’s common for pros to design a system with a delta T of 20 degrees, and US Boiler expert Ron Beck even says he like to see a delta T of 30-40 degrees across cast iron boilers.
Another factor related to the high water flow rate is air management. In past years we’ve had to bleed radiators annually because the air management piping wasn’t installed optimally with the boilers. There are cast-in air separator chambers in the boilers, which pipe up to the steel compression tank. But this didn’t keep air out of the radiators, probably because the boiler techs were draining the tank every fall during their annual servicing, and refilling the tank with fresh water full of dissolved air. I tried to improve the situation by installing an Airtrol tank fitting with properly-pitch 3/4” copper pipe from the boilers up to the Airtrol. And I asked the boiler tech not to drain and refill the tank this fall. So far that all seems to be working well.
But the air management system only works if the hot water in the boiler loses its air bubbles to the air separator chamber inside the boiler casting. And that only happens if (1) the water gets hot enough forair bubbles to form, and (2) the water is travelling slow enough to prevent those bubbles from being carried past the air separator chamber.
So I figured I’d try to slow down the water flow through our boilers. This would increase the delta T (even though as I’ve said, we see no evidence of condensing) and hopefully improve the performance of the air management system.
One way to slow the water flow would be to install a throttling valve, but that would involve breaking the pipes open and introducing air, which I want to avoid. The other way is to install a speed controller on the circulators.
In searching the heatinghelp archives, I found a number of threads referencing triac-based speed controllers for circulator motors. The Taco 007 is a permanent split capacitor (PSC) motor, and I found several threads in which people said they had success using triac-based controllers on PSC circulators. @hot_rod has posted several comments over the years about triac controllers, and has said that while these “wave chopper” triac controllers are not as “friendly” to the circulator motors as true PWM control, has has yet to see a motor failure in 8 years (as of the year 2014). For reference, that discussion happened here:
He also mentioned in another thread that he’s used a triac successfully to control a 5-hp shop vac.
So I figured I’d try installing triac-based controllers on our circulators. @LRCCBJ kindly shared with me a link for a triac-based controller board that he has used succesfully. Ultimately I decided to use a Dayton industrial triac fan controller that I got from my local Grainger (Dayton #48C172). It’s a 6-amp controller, so our circulator draw of less than 1 amp gives plenty of margin.
So I installed these on our two boilers last month, and so far they are working fine. Before the controllers, our delta T was 12 degrees, with an implied flow rate of about 19 gpm. After installing the controllers and turning them to their lowest setting, our delta T is 18 degrees, which implies a new flow rate of 12.6 gpm. So our flow rate has been reduced by about 34% by the controller. I would have liked to get even more reduction, and this would have been possible using the more capable triac board that LRCCBJ recommended.
(Edit: I just re-read the Dayton installation instructions and realized there's an adjustment screw that allows to user to adjust the minimum speed, so apparently I could reduce the flow rate even lower by fiddling with this adjustment screw. Maybe I'll do that later.)
I was curious to see the waveform output of the triac control, so I bought an el cheapo $35 digital oscilloscope from Amazon. The photos below show the 120 volt sine wave output to the circulator motor before installing the controller, and after installing the controller, with the speed dialled all the way down. It’s a nice visual showing how the triac “chops” the waveform to reduce the power delivered to the motor.
Comments
-
draining the system yearly for 30 years? That could put a lot of minerals in the boilers also, with every new fill. I imagine a system like that hold a lot of water?
30 minutes is a good run cycle, is that on design day?
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream0 -
I figure we have about 150 gallons total in each system, but the tech would only drain and refill the tank. The tank is plumbed to both systems, so the tank serves about 300 gallons total. The tank itself is a 40 gallon, but the tech would presumably refill it only about halfway, so say 20 gallons. So each drain/fill would be replacing about 20 gallons out of a total of 300 gallons.
30 minutes run time is on what I'd call an "average" day. On a design day it might run 45 minutes, but then be idle for 90 minutes, for a roughly 33% duty cycle.
0 -
the design day run cycle tells a lot about the over sizing of the boiler. It will be interesting to see how much you can slow flow and provide adequate heat.
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream0 -
Exactly. And the numbers I gave are conservative. When we had the zero-degree polar vortex last February with 40 mph winds, the boilers only ran 40% duty cycles! And that is with downfiring the gph to 80% of boiler rating.
So far there's no measurable difference in boiler run times or heat output with the slower water flow. I guess the logic is that even though the water is flowing more slowly, now it's hotter, and the net-net is no difference in how fast the heat is delivered into the space.
0 -
Most hydronic systems adhere to the golden rule 500 x flow X delta T. Changing any input number changes the btu delivered
At some point if you want the boiler running a 40 delta and the system running 20 delta you might consider hydraulic separation.
A 40 delta in the distribution gives the last radiators on the circuit a much lower output to cover the load.
So primary secondary piping or ideally a multi function hydraulic separator for two different delta design
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream0 -
The ultimate solution will be to wait until these boilers fail, then install smaller properly sized boilers with (1) bypasses and (2) smaller circulators.
With smaller boilers we'll need gpms down in the single digits, so I'd like to get 3-speed circulators so we can select the lowest speed, and maybe carry over these triac controls for even more flexibility.
0 -
depending on how much cranking down you do, the triac may need a heat sink, they can get pretty warm
A new electronic circ is a better option for changing or varying speed. Less power consumption, more adjustment features.
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream0 -
I see the new cast iron boilers are mostly coming with the electronic 007e, but I didn't know you could vary the speed on those manually. I thought they just had some internal self-regulation. How do you manually control the speed on those?
Also, I think I read someone saying the 007e's are more susceptible to failure because the magnets pick up rust particles from the dirty flow in older systems. Have you seen this?
I'll keep an eye on the triacs, but so far they don't seem to be heating up.
0 -
I will leave the pump speed control to others. 20 degree TD has been standard for years.
I would not do a 30-40 degree TD under any circumstances. Too much stress on the boiler and you could crack a back section (or whatever section the return is connected to.
You could convert to primary secondary and run the boilers at a 20 degree td (at a higher temp) and the system water at whatever td you want. This would solve the condensation issues.
But I probably would not do the expense of repiping old boiler that are working well
0 -
Agreed that repiping the old boilers doesn't make sense for us. They've worked fine, albeit under sub-optimal conditions, for 30 years. We can do all the repiping we need when the boilers get replaced.
The delta T debate is an interesting one. I think we've proved that at least SOME cast iron boilers can run low return temps at high flow rate for 30 years without any evident problems. And we have plenty of cast iron radiation to heat at SWT of 120 degrees or less.
So while I understand the logic of designing a system with a delta T of 20 degrees as a target, it seems like there are at least some systems out there running fine with delta T's below 20. I recall someone posting here who said his cast iron boiler was running a delta T of 7 degrees and working fine.
0 -
Lower delta is not an issue. I disagree with the 30-40 degree recommendation. To me say a 130 return and a 170 supply puts too much stress on the boiler sections but that is JMHO
0 -
Understood, and I don't know enough to have an opinion either way. But since cast iron boilers seem to work fine at delta T's of 20 and below, I'm not sure why Ron Beck recommends delta T's that high.
0 -
Older boilers don't have very efficient heat exchangers, so even when supplied by cold water the flue gas is still pretty hot. This is most likely why you have no issues as is.
I don't think reducing the pump speed is buying you anything. With a fixed output boiler, no matter the water flow rate, you get the same net BTU for a fixed burn time. If this was a modcon, the extra delta T would get you a bit of an efficiency bump but I doubt it is doing much here.
On my gravity to circ conversion, I also had issues with flow reversal and water distribution when I tried to run at too low of a flow, something to watch.
If you want better, an oudoor reset type setup with primary secondary loop would be your best option. This would get move even heat to the emitters and avoid pipe expansion noises at each firing.
P.S. If you do upgrade to ECM pumps, make sure to remove the speed controller. The two are not compatible.
Sample of 1, but I had 3 ECM circs running without any magnetic separator for about a decade with plenty of sludge in the old rads.
0 -
@Kaos said:
"Older boilers don't have very efficient heat exchangers, so even when supplied by cold water the flue gas is still pretty hot. This is most likely why you have no issues as is."
I think this is the right answer. These WGO heat exchangers are probably like, what, inch-thick cast iron? So the combustion gas side is always going to be 400+ degrees, which is the flue gas temp downstream of the HX, regardless of the water temp on the other side of the HX. The return water temp could be 100 or it could be 180, but the hot side of the HX is always going to be 400+ when the boiler is running.
The only exception will be on startup, when both sides of the HX are cold, but even if any condensation could occur on startup, it would quickly be burned off by the 400+ degree flue gas as the boiler heats up.
"I don't think reducing the pump speed is buying you anything. With a fixed output boiler, no matter the water flow rate, you get the same net BTU for a fixed burn time."
I agree, this won't change anything as far as efficiency or heat output. The one area I'm hoping will improve is air management. Lower flow rates mean the water travels more slowly through the boiler and heats up more. This gives the water more time to form air bubbles, and more time for those bubbles to migrate up into the air separation chamber. This would be a good place to use @ethicalpaul 's sight glass in the supply line right above the boiler, to compare before and after and see if the lower flow rate corresponds to a reduction in air bubbles escaping into the supply.
Another change I made is to program in a night setback of 3 degrees, so in the morning when the boiler recovers, it burns 60-90 minutes, raising the supply water temp much more than usual and giving the boiler a chance to "burn out" any dissolved air that didn't get eliminated in the shorter, lower temperature burns of the previous day.
So I'm hoping that the combination of the long recovery burn and the lower water flow rate (along with the previously installed Airtrols) will finally solve our air management problems.
"On my gravity to circ conversion, I also had issues with flow reversal and water distribution when I tried to run at too low of a flow, something to watch."
Interesting, I will watch out for that.
"If you do upgrade to ECM pumps, make sure to remove the speed controller. The two are not compatible. Sample of 1, but I had 3 ECM circs running without any magnetic separator for about a decade with plenty of sludge in the old rads."
OK, thanks. I understand triacs won't work on the ECM pumps. If you do want to electronically reduce speed of an ECM pump, how do you do that?
0 -
@jesmed1 , Are you still stressing over those 2 WGO-5s that are so oversized, you could heat the next three buildings on the block? The air that is dissolved in any new water that is added to the system will separate when you increase the temperature. If you increase it to 180° it will separate faster than if you have 130° as the highest temperature. But regardless of the temperature, the air will separate. You don't need to have a higher temperature.It’s not like a given temperature eliminates a given percentage of air. All the air will be released eventually. Just as air will be absorbed when the temperature is lowered. That is why you want to put the air in the tank and have the tank fitting so the lower temperature water in the tank cant absorb any air and leave the tank in order to deposit that air in the radiators.
The next thing to consider is the built in air scoop in the WGO boiler front section. (the one with the 3/4" hole next to the supply tapping). The gallon per minute is not the factor that will cause the bubbles to get to the air scoop and miss the lower supply opening. It is the foot per second speed at which the water is flowing. If you have 19 GPM going through a 1-1/2" pipe, that may move at about 3.45 feet per second. When you put that same 19 GPM of water through the boiler section it will slow to less than 1/2 foot per second. That slower velocity is what will cause the bubbles to rise to the top and get caught up in the path of the expansion tank vent pipe connected to the internal scoop.
I believe that you are done with this project until you decide (along with the other condo owners) that it is time to install a new boiler. That is when you can “Fix” all the things wrong with what you have. Pumping away from the expansion tank, a more efficient ∆T, selecting the correct size boiler, and better air removal design.
One of those boilers will be failing within the next 20 years, and I would get the Condo Docs to read that if one of the boilers fail, then you will be connecting the correct size boiler for all 4 condos. And perhaps a back up boiler that will be able to handle the entire building at 20° above design. That way all the owners will be able to realize the savings of the new system. Two-stage with two boilers with outdoor reset will cut the operating cost by up to 50%
Only use 2 boilers when you are within 20° of your design temperature. But that is not my problem, You need to work that out with the other owners.
Edward Young Retired
After you make that expensive repair and you still have the same problem, What will you check next?
0 -
"Are you still stressing over those 2 WGO-5s that are so oversized, you could heat the next three buildings on the block?"
Can you tell that I have too much free time on my hands? I won't say I'm stressing. I'll just say I consider this a learning opportunity. The triac circulator controls may not be helping much, but for $100 I learned about triacs, PSC motors, and how to add another dial to to twist to give me the illusion that I'm doing something useful. 😀
But seriously, thank you for the expert comments. I do think the 2-boiler solution with one boiler running as a primary and one as a second stage in colder weather is an excellent idea. I will start pre-planning that with the other owners.
The other factor is DHW. We now have a 75,000 BTU atmospheric natural gas water heater that I'd love to eliminate, because it's sucking warmed air up the flue. That and the atmospheric boilers which have no vent dampers keep our basement cold all winter.
So I'd like to have one of those two new boilers running an indirect water heater. Then with automatic vent dampers on the new boilers, and no flue pipe for the indirect water heater, we eliminate a major source of air leakage in the building. I imagine that changes the sizing calculations for the boilers, but I'm not sure how. I guess we'd want the indirect DHW running off the primary boiler, so we could purge the end of each heat call into the indirect tank?
0 -
If you have constant air issues, somewhere air is getting into the system. For example, I had a problem with a slightly water logged compression tank and an open auto vent on the 3rd floor. When the boiler cycled off, the pressure dropped enough that the 3rd floor piping dropped bellow atmospheric and air was pulled in through the auto vent.
Find where the air is getting into your system.
0 -
If you follow this type of information on this forum, you would know that an automatic Vent and a standard compression tank are never supposed to be on the same system. A pre-charged expansion tank that has a membrane between the air cushion and the water is the only type of expansion tank that you should have automatic vents on. the standard air over water compression tanks NEVER have automatic air vents. Properly designed the air from the radiators ends up in the compression tank. Improperly designed you get a water logged expansion tank
Edward Young Retired
After you make that expensive repair and you still have the same problem, What will you check next?
0 -
compression tanks are air “management” system
Diaphragm or bladder type expansion tanks are air “elimination” systems
If you put any air vent on a compression tank that removed air needs to go back to the tank. Keep or manage the air or the tank will water-log
Since you are into learning, download the Amtrol Engineering Handbook. Excellent info about the various expansion methods, sizing, tips and tricks.
Auto air vents up high suck air in when you pump at the expansion tank PONPC, and pull negative pressure out in the piping system, usually at the hydraulically remote section.
Absolutely no air elimination on air management systems unless it is sent back to the tank. The two part AirTrol fitting is an example
I highly recommend magnetic separators on iron systems, all systems actually with ECM circs. 90% of the returned ECMs are bound tight with magnetic. According to one pump manufacturer.
This graph shows where air comes out of solution best. The hottest point, and the lowest pressure point. Yes, the hotter the water the more and faster it releases entrained and dissolved air
Only a micro bubble resorber will catch those microbubbles, smaller than the eye can see. A ramp purger will not, a boiler section will not
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream1 -
Auto vents are standard item on most manifolds. They are great for the initial fill to get lot of the air out, doesn't matter if compression or expansion tank. Just have to remember to close the cap once initial fill is done.
My point is somewhere air is getting into the system. Figure out where that is.
0 -
OK, thanks. Air was getting into the system before this year because the boiler techs had been draining the 40 gallon expansion tank and refilling it every year. So if the tank was half full, say, the techs were draining and adding 20 gallons of air-enriched fresh water. I did not know that was wrong until last year when I learned it here.
Now that I've installed an Airtrol, I've told them to not drain and refill the tank. So in future I don't expect to have air getting into the system. From here on out it's just a matter of bleeding whatever air is left in anyone's radiators (there are 3 other owners and I can't always get them to bleed their rads properly) and keeping the system on autopilot.
Ultimately I do want to switch to bladder tanks and air eliminators, but we'll wait until a boiler fails and needs replacing.
0 -
That's an odd waveform for a TRIAC, it like backwards, maybe it is the scope. How do you get a TRIAC to turn off in the middle of the alternation ?
https://eepower.com/technical-articles/an-introduction-to-triac-basics/#
National - U.S. Gas Boiler 45+ Years Old
Steam 300 SQ. FT. - EDR 347
One Pipe System0 -
You have a good eye! Yes, this is interesting.
You are correct that when the triac is not connected to the circulator, you'll get a waveform with the leading half of each wave-half chopped off just like your Figure 2.
But the waveform I showed is the voltage trace you get when the triac is connected to the circulator with the circulator running. In that configuration, with the circulator motor running, you get a mirror image of FIgure 2, with the leading half of each wave remaining and the trailing half chopped off.
I don't understand the electrical theory, but I think it may have something to do with "back EMF," the resisting electrical force generated by electric motors that self-limits their speed. So maybe what my waveform shows is the back EMF being produced by the circulator.
0 -
If you want to vary the speed of an ECM circulator you need to get one designed for that. A Taco VT2218 variable delta T circulator or a variable delta P circulator like the Taco Viridian or Grundfos Alpha.
1
Categories
- All Categories
- 86.4K THE MAIN WALL
- 3.1K A-C, Heat Pumps & Refrigeration
- 53 Biomass
- 423 Carbon Monoxide Awareness
- 94 Chimneys & Flues
- 2K Domestic Hot Water
- 5.5K Gas Heating
- 101 Geothermal
- 156 Indoor-Air Quality
- 3.5K Oil Heating
- 64 Pipe Deterioration
- 925 Plumbing
- 6.1K Radiant Heating
- 383 Solar
- 15.1K Strictly Steam
- 3.3K Thermostats and Controls
- 54 Water Quality
- 41 Industry Classes
- 48 Job Opportunities
- 17 Recall Announcements