An Experiment with Thermal Purging

fentonc

I added a PLC to my oversized (~4x) CI hydronic boiler last season for experimenting, and I finally got around to trying some thermal purging experiments. Very impressive so far!

Setup: I initiated a call for heat from a room-temperature cold-start with the smallest zone in my house (22 feet of high-output fin-tube baseboard). The boiler is nominally 140K BTU/hr input, but actually clocks at 120K BTU/hr. I let the boiler fire until it hit the aquastat high limit at 160F, and then I inhibited the burner and let the circulator run until the SWT hit 80F. I measured the supply and return temps for the zone every second, using the average water temp and linear interpolation based on the fin-tube datasheet to track the overall heat output. I measured the burn time to track the overall heat input.

The boiler burned 9667 BTUs worth of gas before it was shut off by the aquastat high limit, and I then extended the run cycle for a total of 1h45m before it hit my PLC-enforced low-limit of 80F. The 'thermal efficiency' is the sum of delivered BTUs + 'residual BTUs' if the circulator had been halted at that temperature (based on water volume and temperature) divided by the theoretical BTUs of the gas burned (assuming a burn rate of 120k BTU/hr). The boiler has a nominal AFUE of 84%, and I was able to hit 77.56% based on delivered BTUs using a thermal purge all the way down to 80F.

I'm definitely going to incorporate a strong thermal purge feature using the PLC this season.

Big Ed_4

Just like the System2000. ?

fentonc

Yup, I'm just trying to bolt that feature onto my Weil-McClain CGA-2 to try to improve the efficiency.

DCContrarian

Two comments:

First, I don't see how this really increases the efficiency. Where does that heat go if you don't "purge" it? Into the building envelope.

Second, how do you know how much heat the room actually needs? It seem to me that this strategy can only do one thing, deliver ~7500 BTU in 1:45. What if the actual heating load is higher than that? Or lower? You're going to overshoot or undershoot. Remember, the purpose of heating is to deliver comfort. If you're overshooting or undershooting you're not delivering comfort.

fentonc

If my house were hermetically sealed with a centrally located boiler, I would agree. In practice, my house has three floors and reasonable insulation, but the boiler is located in an uninsulated concrete corner of the basement with poor air sealing. As a thought experiment, if I were to just have my boiler maintain its internal temperature all season, I would guess it would burn about ~1/3 as much gas as it normally does while providing nearly no useful heat to the vast majority of the house. At the supply temperatures preferred by non-condensing boilers, my house has something like 4-6x (depending on floor) the radiation required to meet the design-temp heat loss, and the boiler can similarly provide 4x more heat than needed. The result of this is that my heating system spends the vast majority of the heating season on the far left side of this graph (my record in the last 3 years is achieving a 25% burner on-time on 3 separate occasions), and seems to average <50% efficiency:

and then with actual data from my house:

Hence my interest in thermal purging as a means of increasing the fraction of BTUs that make it into the useful part of the house.

I agree that the system also needs to be comfortable, so the design challenge is "How can I optimize both?" To that end, I'm attempting to run some careful-ish experiments to see what might be achievable in terms of efficiency, and then how that might be integrated while maximizing comfort. I have a great deal of control over the system, so I'm hopeful that the present state of things can be improved upon. Overshooting by a degree or two would only be a comfort issue in the bedrooms at night, for instance. Similarly, undershooting by a degree or two in the lower floors at night would also have no impact. I think some kind of modulation scheme might work reasonably well in practice, but I have a lot more experiments to do.

Hot_water_fan

First, I don't see how this really increases the efficiency. Where does that heat go if you don't "purge" it? Into the building envelope.

Ha! DC, this is some peak engineer brain. You’re right. You’re also wrong.

Teemok

"Where does that heat go if you don't "purge" it? Into the building envelope."

There's a still warm dead horse around here related to this question.

fentonc

I’m merely proposing a Rube Goldberg machine to slowly drag that horse around my house until it has finished cooling =)

hot_rod

Interesting data!

What about post purging into the DHW. Water is a much better storage and transfer mechanism and it is something used most all days. No comfort issues with running the tank temperature up.

I think EK has shared info on this site about the recovered energy and efficiency gains. A hot boiler loses a % of that off cycle heat up the flue, not necessarily helping the structure.

Teemok

@fentonc DC's right, comfort is typically the prime directive, efficiency ends up being low on the totem pole. Longer firing times and longer lower temperature off times. Loss reduction that doesn't impact comfort and you get to play with a PLC seems like a win. How much can one can tune a system seeking efficiency without crossing a low performance limit is a home owners game or at least a residents game. Some of my customers love to do that and others don't want to be bothered.

There's low hanging fruit going from %5 to 20% burner on time.

fentonc

@Teemok - I'm well into "bored homeowner with too much time on my hands" territory, but 'fiddly performance tuning with electronics' is basically my day-job, so it's been fun to try to apply it to my home heating system. I think there's a lot of opportunity for situation-specific tuning if you have carefully measured data (which is basically impossible to get in the field if you're working in residential heating).

@hot_rod - The water heater is already enlisted in this fight - It's a heat-pump water heater located next to the boiler (and was almost undetectable in adding to the heating load last winter). Thanks to all my home measuring and tweaking, I've managed to reduce my yearly gas usage from 920 therms the first year we lived here to 414 therms in the 2023-2024 season. I've got high hopes for 2024-2025 =)

EdTheHeaterMan

Some of it goes up the chimney. and that is a large waste of energy. @DCContrarian

Teemok

@fentoc I've had success with buffer tank adds on systems that were really bad. Cost benefit for efficiency is sometimes hard to get very clear. Reducing cycles definitely helps with reliability. The cycle counts on some of these micro zoned combi's are incredibly high. I came across an ill fitted Lochinvar combi that had a control board gas valve relay failure at year 3 and I though that was odd, until I witnessed it run and looked at the cycle count. In addition to being micro zoned it had a constant DHW recirc. Insult to injury. The customer got hostile with me, defending the installer, insisting "this is how it's supposed to work". I said OK. He paid me for my work and the next person with a clue will have to deal with it.

fentonc

This is a neat result - If I estimate the 'delivered BTUs' for my test zone as I continue running the circulator, I pretty consistently dump 975 BTUs per 10F degree drop in supply temp (so 97.5 BTUs per degree F) across 3 separate experiments. In my on-going quest to figure out "Where do the BTUs go?", it had always bothered me that I didn't actually know how much energy was stored in the boiler. The boiler itself only holds 2.7 gallons of water (so 8.33 x 2.7 = 22.5 BTU/degree F), so it looks like the rest of the system (cast iron, pipes + water, etc.) holds about 75 BTU/degree F.

Starting at about 70F (not super precise), I fired the burner in 3 separate experiments up to 140F, 160F and 180F.

If I compare how much gas I burned to hit the aquastat limit, vs how much was delivered + stored at the time the burner shut off, I get:

• 180F: 13,166 BTUs in / 11,789 Delivered+Stored (89.54% Efficiency)
• 160F: 9,667 BTUs in / 9,525 Delivered + Stored (98.53% Efficiency)
• 140F: 8,466 BTUs in / 7,235 Delivered + Stored (85.46% Efficiency)

I think the boiler hadn't fully cooled down for the 160F, but this is quite close to the nominal 84% AFUE number for my boiler (having a model within 5% is actually pretty great!).

hot_rod

https://forum.heatinghelp.com/discussion/comment/1809811#Comment_1809811

the Ek system uses a plate HX as part of their thermal purge efficiency boost. That could be done with a HPWH also

Wouldn’t that be better than purging into a space that doesn’t need or want addition heat? As you increase the room temperature heat loss increases, so chalk up some small additional loss of that purged energy,

I still applaud your chase. As energy costs continue to increase we will see more heat recovery controls and devices, certainly on large scale projects.

Many years ago the Hydronic Institute training showed two stage t stats that would run circ before the boiler fired to purge from a warm block. Cast boilers were bigger heavier and higher water content back then. That was back in the 1980s

Actual data on a rock simple drain water recovery tube always surprised me

I agree with the benefit of reducing the cycling of relays, gas valves, vent dampers, , ignition systems , etc. All those are designed around a cycle life.

Hot_water_fan

Actual data on a rock simple drain water recovery tube always surprised me

@hot_rod do you have the data handy?

hot_rod

https://forum.heatinghelp.com/discussion/comment/1809849#Comment_1809849

I think @Larry Weingarten and Mark Eatherton have posted some test and performance data here over the years. Try the search button above.

Here is a link to a manufacturers site with an independent lab test report.

Screenshot 2024-09-08 at 6.29.59 PM.png

https://renewability.com/wp-content/uploads/2017/07/SpecDirect_Renewability-1.pdf

Larry Weingarten

Hi, What I have backs up the info @hot_rod posted. Essentially, with a longer unit that has plenty of surface area, you can expect to recover 60 to 70% of the heat going down the drain. These are counterflow in design, so can collect that much heat. This really is a simple, elegant idea. Many years ago, Steve Bear of Zomeworks in New Mexico designed a simple copper coil to go in the shower floor. Modern units are double-wall, so contamination of the fresh water is unlikely. Taking the idea a step further SHARC Energy: https://www.sharcenergy.com/ uses wastewater as the heat source, and via a heat pump, captures far more BTUs than could happen otherwise.

I think I've wandered off topic, but finding ways to make low grade heat work for us, can only be good.

Yours, Larry

DCContrarian

https://forum.heatinghelp.com/discussion/comment/1809861#Comment_1809861

Any thoughts as to why these units haven't caught on? It seems like the companies that sell them constantly teeter in and out of business.

Larry Weingarten

Hi, I can only guess. I knew Carmine Vasile, who invented the first shower heat exchanger, GFX, standing for gravity film heat exchanger. His big problem seemed to be that it was a new idea. People resist change and he faced enormous negativity. This idea is far more popular in Canada, where it's called for in building code. Another problem is like what PEX experienced when it was new here. The different types of PEX, A vs B, kept putting each other down. This kept potential customers from buying any PEX. In shower drain heat exchanger land, the vertical crowd and the horizontal crown don't approve of each other, and have tried various tricks to put each other out of business. Finally, as these are usually made of copper, the cost has gone up. People don't know to consider life-cycle cost, but only look at first cost. This has to be hurting sales too. So, lots of forces preventing widespread acceptance.

Yours, Larry

PRR

Copper, the cost has gone up.

I knew that, but not how much. Yowsa!!