Monitoring Boiler Performance

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
  • Super Cedar firestarters 30% discount Use code Hearth2024 Click here
Status
Not open for further replies.

jebatty

Minister of Fire
Hearth Supporter
Jan 1, 2008
5,796
Northern MN
This may be somewhat repetitive of a mixture of posts, but it works well for me, is inexpensive, low tech, and is an easy way to monitor boiler performance, while being adaptable to most systems.

My system: Termovar thermostatic mixing valve and plate hx to pressurized storage/system.

Monitoring sensors: probe type meat or food thermometers. The ones I like the best are the Espresso Milk Frothing Thermometer type, with a range of 40-220F and a 2 inch dial, easy to read. They run about $6 on e-bay.

Monitor points: 1) boiler input to hx and 2) hx return to boiler; 3) hx output to system and 4) system return to hx; 5) boiler return. Cable tie the probe to a section of pipe at each location, wrap with insulation.

Performance: At a glance you can determine all of the following. 1) boiler return water temp – let’s you know whether return is at about the desired 160; aids in adjusting the return protection mixing valve. 2) delta T on both the boiler and system sides of the hx. 3) boiler side of hx delta T – let’s you know actual temp boiler is delivering to the hx and the water temp the hx is returning to the boiler, how the temp rises and falls as boiler may cycle or wood load burns, and how much relative heat is being extracted to the system. 4) system side of hx delta T – let’s you know actual temp hx is delivering to the system and the water temp the system is returning to the hx, how the temp rises and falls as boiler may cycle or wood load burns, how much heat is being used by the system, and the relative hx efficiency at various temps.

Interesting btu calculation: if you know or assume a gal/min flow on the system side of the hx, you can easily determine btu’s being delivered to the system. Formula: gal/min x 60 x 8.35 lbs/gal x delta T system side of hx = btu/hr. Example: assume 5 gallons/minute and delta T of 40. 5 x 60 x 8.35 x 40 = 100,200 btu/hr. If the flow is 8 gal/min, then btus = 160,320. If delta T is 20 and flow is 8 gal/min, then btus = 80,160.

Observation: assume boiler at operating temp range (170-190F). As system input to hx rises, delta T on system side of hx closes. For my system, the maximum system side delta T will be just about 45. That occurs when boiler input to hx is close 190 and system input to hx is 110 (or below). Hx output to system will be 155 in this example. As system input to hx rises above 110, given same boiler input to hx, hx output to system will rise but the delta T will close.

Complication: with the Termovar or other mixing valve for return water protection, it gets complicated to calculate actual boiler btu output to the hx (or system if no hx), as part of the boiler water is being delivered to the system and part is being cycled back to the boiler. Without actually knowing the gal/min on the system side of the mixing valve and the delta T on output to system and system return to the mixing valve, I don’t know how to calculate btu’s being delivered to the system. Maybe someone else has an idea on this.

Hope this is useful.
 
This is a great idea, and I'd recommend it to anyone who is not obsessive enough to put together a computer-based monitoring system. Real data can help a lot, especially if you suspect that something has gone wrong. Comparing to baseline data is very helpful.

There's a way that you can back into GPM calculations. If you can measure the heat delivered into a known load such as a storage tank or a hot tub and you know the temperature drop at the heat exchanger (or boiler, if the load you're looking at is the only load for that period), you can calculate the flow rate.

For instance, I heated a 550 gallon hot tub from 64 to 104 degrees in three hours. It was the only load at the time.

550 gallons is 4565 pounds. At 1 BTU per degree per pound, that's 182,600 BTU, or about 61,000 BTU/hr. The temperature rise through my boiler (and the temperature drop across the heat exchanger) was about 15 degrees.

Divide 61,000 by 60 to get 1017 BTU/min. Divide that by 15 degrees and then by 8.3 pounds per gallon. The result is about 8 gpm.
 
There’s a way that you can back into GPM calculations.

Should have known that. Can always solve for the missing variable if other variables/constants are known.

I actually measured last year the Taco 007 output to storage, including the water filter I have on my system, by disconnecting output and running the output into a large bucket. With a clean filter I was getting 6 gal/min approx. I would like to add a variable area rotameter to watch this over time and verify my result of a year ago. It also would give me a clear indication of when the filter needs to be changes, as flow would decrease, or if any other flow problem is developing. With the system pressurized, it's a pain to depressurize and try to do a test, and I'm not 100% sure how accurate it is with the system depressurized.

In my system I have a Taco 009 on the boiler side, and I want to replace this with a variable speed, like the Grundfor UPS 15-58. I have a strong suspicion that higher throughput on the boiler side will reduce my idling time.
 
Is there a flow meter one can attach to the primary loop to get definitive results? My only known variable is lack of math skills.
 
An update. After cleaning the system side of the 5 x 12 x 30 plate hx, performance improved markedly. Before cleaning delta T on the system side would max out at about 20 as input water approached 130 and above. Now I'm getting a delta T of up to 40 at these higher temps (130 in, 170 out, for example).

Another observation after the cleaning. On the system side the hx is stripping all the heat being input on the boiler side, as revealed by boiler hx return temp = system hx input temp. Obviously, boiler hx return temp cannot be less than system hx input temp, but if boiler hx return temp is higher than system hx input temp, then there are btu's not being transferred from the boiler side to the system side.

I still get boiler idling, which means that not enough boiler hot water is being pulled through the hx on the boiler side to equal boiler btu output capacity. I'm using a Taco 009 on the boiler side (because I had one), and next summer I will replace this with a higher capacity circ. A Taco 009 has a flow rate of about 8 gpm max or less, depending on head. As I'm using 1-1/4 to 1 copper on about the smallest loop possible on the boiler side of the hx, I suspect the head is very low, and a Taco 007 or a variable speed circ should pull much higher gpm through the hx, which should also result in using more of the boiler btu output capacity.
 
Status
Not open for further replies.