Combining DHW flat plate w/ cascaded boilers & electric tankless (Rube Goldberg?)

chrismarcellino

Hi,
Long time lurker and hydronic enthusiast. In short, due to frequent formation of difficult to clear ice, lawn damage and frequent falls on a very steep long Minnesota driveway that is approaching the end of its concrete service life, we are planning to embed PEX to achieve class I snow-melting (really ice melting--I plan to generally clear it by hand for environmental/sentimental/exercise reasons generally but want to have the option especially as we hopefully age here.)

Thanks to LoopCAD and many Idronics issues, I came up with a plan for the 1800 sqft driveway using 5/8" hePEX (I have the expansion tool) with 19.5 gpm at 278k BTU, with supply temp. 112 F, total head full circle 24 ft. head pumped by a Taco 0034e. I have long term aspirations of remodeling the upstairs of a typical 3 story to include in floor heat as a series of remodels over many decades have removed many of the forced air supply ducts (which were already quite builder grade to begin with.) We live in a primarily heating climate. Hence, a cascade system with 2 units would be ideal to achieve maximum net TDR. The snowmelt would be manually controlled, with a simple Tstat to avoid overheating.

I'm considering two Lochinvar Noble 150k BTU boilers. I have been collecting other new/old stock system components on eBay at a considerable discount, including a very long/large heat exchanger for the snowmelt zone that meets the capacity for this purpose per Xylem's fantastic calculator as I can't stomach the idea of having to deal with glycol indoors every time a boiler needs maintenance or the system is extended etc.

Please see the attached schematic named "System.png." I have an engineering background but this is the first sort of time I have worked on this sort of thing so please keep that in mind. I have done a considerable amount of copper & PEX plumbing and electrical work though am a non-pro. I would have professional combustion analysis and venting from a plumber that I have worked extensively with.

*** My main question however, is that due to limited total gas supply to the house (a bit over 500k BTU per the POCO with a handful of other appliances, limited utility room space with all of this new gear, and occasional depletion of our 40k BTU traditional gas DHW tank, and desire to exercise the boilers in summer and eliminate some vents, I would like to use the boilers for DHW. We have a 1" CSST at 2 psi serving the utility room from a prior remodel fortunately.

Obviously an indirect would be a great solution space and cost aside and can probably fit, but I had some "creative" ideas and am looking for opinions on combining a small *electric* tankless with a flat plate heat exchanger to sort of get the best of both worlds, 1) "unlimited" hot water--though an indirects recovery rate is probably effectively the same, 2) condensing return temps 3) minimal space, 4) slightly lower cost as I have for better or worse already purchased a (smaller) FPHE at a good deal that I can't otherwise use though I could sell it, 4) elimination of standby losses--though from what I can tell these are about $30 annually anyhow for us so we won't really break even.

It would seem that the fast firing performance of electric tankless (like the Chronomite single phase units) would reduce the cold water sandwich effect, and that using two flow sensors and a delay-on-close relays, I can avoid firing the boilers for short low flow draws but fire them immediately for large draws--to help avoid the traditional short draw inefficiencies of standard gas tankless units. I was thinking of using a 30 or 48 amp 240V unit to heat around 0.8 to 1.0 gpm at our 78 F worst case temp rise (42->120 F), while firing the boilers. We have the electrical capacity in every respect (service & feeder load calcs etc.) We have very inexpensive electricity 10p-8am when much of our water usage lies, and the overall electrical expenditure should be low given that it is just a bridge for the boilers to spin up. (Though in other words, this would be reinventing a modern preheating combi boiler without a tank.)

I would have to use a space heating priority zone since the Noble's cannot cascade DHW and a single system temperature. Using 140 F as the supply temp would work well it seems for both applications (and eventually though a mixing valve for in floor radiant upstairs at a lower temp.). Of course, a higher supply temp would heat as much water comes in our 1" service and 5/8th meter (about 10 gpm) but at 140 F this B&G BP412-50 would heat 4 gpm to 120 F though a small FPHE, and 6 gpm to temps that the electric can finish (107->120 F) all though a mixing valve, even accounting for considerable fouling. (Our municipal water quality is excellent however, and it would be softened.)

Obviously the Knight has more options in terms of temperatures and cascading, but for slightly more I can get a single unit with 285k BTU but only 1:10 TDR. Two units around 150k BTU are substantially more and requiring larger venting, so the Noble is ideal and probably sufficient for this I think/hope. Recirculation would be occupancy sensor based. This would be an inspected project of course in case that matters.

If you made it this far, thank you for reading this and I appreciate any advice or criticism on any part of this plan (DHW or otherwise)! The best option is probably just to make the indirect work though I enjoy a challenge. It's too bad there isn't a 2 gallon indirect or stainless electric tank that could serve as a buffer without having to pump the domestic side.)

Chris

Hot_water_fan

A lot going on here. To begin, how did you arrive at the driveway heat loss? 

You’re right, an indirect will be easier. It’ll also have greater capacity. I’d find a way to fit one! Electric tankless is almost never a good idea unless you’re severely short on space. 

hot_rod

What about a boiler dedicated to snowmelt, a simple copper tube, glycolled, no HX? Less $$, less complicated, then the boiler sizes accurately to the home load. I would at least put the cup in the driveway to add a snow sensor later, tekmar or HBX. The next owner may appreciate a full automatic snowmelt control.
Manual operation with a spring wound timed, not an on off switch!

I've done two systems of my own with a plate HX, sized for 2 gpm at a low approach temperature. My last one was able to generate DWH with 120 SWT. It takes a generously sized HX to get there.

We have showed that tankless assist to a plate HX in a few Idronics. I have never tried one, no reason it wouldn't handle you application. The warm up wait time can be frustrating.

A first stage thermostatic mix valve after the FPHX to prevent 180F hitting the electric tankless

This could be a buffer heated by a wood boiler for example. You want a low flow activated type flow switch, I don't think those McDonell Miller can reliable switch a single lav flow at or below .5 gpm. This is a pic of the Harwill that I used, Sika is another common brand used in tankless and combi boilers.

chrismarcellino

Thank you! Fair points. The only McDonnell-Miller valve that I found that gets low is the FS6 (0.12 gpm minimum/factory adjustable to 2.5 gpm. I found some inexpensive old stock.  Will do regarding the driveway cup and timer.

I was thinking that with a 140 F SWT and an electric heater that was rated to both accept and pass up to 160 F water (such as the 3/4” Chronomite), that I could forgo the 1st stage mixing valve since the output water can’t exceed SWT? I would leave a margin since the electric heater has a high temp cutoff that would fire above 160 F. For better or worse with the Noble, I am stuck at one system temperature. Even with that big FPHE, I could go down to 130 F SWT and still achieve 112 F Glycol temperatures, so I would be a passing water a bit warmer than necessary to allow DHW to have a 140 F SWT.

We have a similar (Rheem) POU electric on a distant from the DWH heater sink that fires very fast, maybe 2 seconds so hopefully that is the case, while the FPHE catches up. Here’s hoping.

Thanks again.

chrismarcellino

Hot_water_fan: Thanks! I essentially just used LoopCADs interface to determine a heat load based on class goals, local temps, wind, insulation plan/thickness. Ends up being 155 btu/sqft, which seems fairly typical based on what I’ve read, or even on the light side for our locale.

hot_rod

ASHRAE did modify those Class 1,2, 3 designations with a % of snow free condition

Class 1 was a basic melt category with accumulation possible, but now that % can make a difference

But as you noted snow melting is a large BTU gambit