Help with Great Room heating strategy and the wrong thermostat

schreib

OK, I hear you Gordy. I will report back with data and target 10°F dT. Yes, I understand the relationship, just a matte of what you want: higher flow and low water temp or lower flow and high temp. Your main point being it's advantageous for floor uniformity to go the high flow way. I have some work to do. . . !

Gordy

And the efficiency of the boiler. Lower the return water temps. That boiler isn't running in the 90's for efficiency more like upper 80's

schreib

Here is efficiency curve for 20°F dT
Notes from my brother in Denver, designed the system.

schreib

NEW reset curve uploaded here, dropped it 10°F as a first trial. This prevents Boiler output from forcing dT too high. Presently, dT is 27°F, 10°F outside, 3 zones calling for heat, 70% boiler output rate.

schreib

update: with WILO pump at 12, 3 zones in demand, reset curve st point asking for 115°F, boiler at 60% output, return manifold temp at 94°F-- seeing dT of 115-94 = 21°F. Accumulated pump flow from flow meters is about 3 gpm out to zones.
NOW: plan to increase Stat temps in all zones to forces all to open and obtain flow rates for each one and total for WILO at 12.

Gordy

You were at 133 rwt.......so 88%

schreib

and with direction you have been guiding me will slide her up the efficiency curve. . . all makes sense. thanks.

Gordy

Remember the delta will narrow more as slab stabilizes takes time to settle

schreib

Gordy: took lots of data, made progress reducing dT(16°F now vs 50°F this morning) and increasing distribution flows. Probably above 90% boiler efficiency now. Have yet to convert zones besides Great rm to slab feedback control.
Please See att pdf file with temps, flows, etc.
I'm done for the day! whew.

Gordy

Lol okay, but that wasn’t what I was asking for. However it seems to be progressing, with the data given. Let’s give the settings time to stabilize.

Some pics too.

RxRoy

"However, I can have up to 3 pump fed sub-systems with the Lochinvar Knight unit. Each one can have its own space heating set point(SH1,SH2, SH3) but the highest one's reset curve is used to feedback to boiler. Potentially, I could gut the piping and add another pump just for Great room,"

The WHN cannot run three separate system side circulators. You would need a separate switching relay. It has circulator outputs for DHW, boiler pump, and one system pump. It does have the ability to provide 3 different heating set points as you state. Just FYI, in case you decide to go that route.

schreib

Gordy: Sorry, thought I got it all! What else did I not provide?
Anyway, I relented and went around and reprogrammed all the thermostats. Note that this brand is not capable of controlling above 90°F and some slab areas appear to require more than that for required comfort needs. I will probably have to get a Tekmar 519 or two to allow for that anyway.
Note also that with only 3 zones in demand now my dT has nearly doubled from 15 or 16°F to 29°F-- RWT vs SWT. Basically half the secondary loop flow to the exchanger dropped out so we double or capability for heat transfer.

See final data sheet for today's assignment attached. Hope you approve. Let me know what I missed. thx.

Gordy

Those are not actual loop deltas, and manifold deltas

Your giving me slab temp from probe I presume
Air temp
Surface temp

Those all look good however for the outdoor temp. -10?

schreib

Per spreadsheet the data comes from the T'stat itself. Air temp is from the thermistor probe inside stat head on the wall, Slab temp is from the other probe, Set point is stat's set point set by me. For top of floor, I just shoved the T/C into the carpet pile or used a piece of thin plastic to hold the T/C against a hard floor. NOTE: regarding slab probes. Mounted inside 1/2" PEX buried in conc slab. Probe tip is held into contact against concrete with pressure but location chosen by big talker plumber so no guarantees as to location equidistant between tubing!

NO, not individual LOOP dT's. I must have misunderstood what dT was of interest. Since my RWT is sensing a point(always) on the manifold that feeds the accumulated water from ALL return manifolds it is the average of ALL return water. Likewise the average of ALL outgoing supply water is the "System Sensor" in Lochinvar nomenclature. So the dT I have always been using is average outgoing vs average incoming of the Distribution(secondary) loop. The system sensor is about 7 ft from the close spaced T's after the WILO pump(SWT) and the RWT used is from my T/C taped to feed pipe about one ft from the input to close spaced T's. Tell me exactly what you would like differently and I will try to get it. dT between an individual loop S and R will be pretty cumbersome since I would have to tape the T/C and remove it each time. T/C broke this morning and no spare so I will have to cut it off and twist in a new junction. . . All the usage kinked and broke it.

I have more data from this morning but not publishable yet. Temps of slabs mostly pretty close together. Kitchen is problematic but I believe I no the cure. Great room cannot be heated beyond 90°F to satisfy more comfort and that would be a Tekmar 519 stat solution likely, in addition to adding in the two 500Watt baseboard units(on order) and peeling back carpet a bit near the window walls.

schreib

RxRoy said:

"


The WHN cannot run three separate system side circulators. You would need a separate switching relay. It has circulator outputs for DHW, boiler pump, and one system pump. It does have the ability to provide 3 different heating set points as you state. Just FYI, in case you decide to go that route.

Thanks for the input, RxRoy. OK, I was just reflecting the Lochinvar manual showing different diagrams' capabilities. I did NOT know, however, that it could not provide controls to handle them. Basically, from what you state I need something like one more Taco 4 channel control unit, ie like what I have now. Since you appear to be a Lochinvar trained tech. . . I was wondering if you can provide any insight to smoother control than their 6 level ramp anti-cycling control to prevent output fluctuation. I am not happy at all with how poorly that works. I am used to working with PID loop controls for industrial stuff and this system of theirs is just simply primitive. Wondering: Do they simply provide that for a 1st level simple solution and if you want better you need to use a 3rd party control like BMC stuff?

Gordy

Temperatures, and Deltas on individual loops could show indication of air issues, or higher heat loss areas. It's a more precise measurement.

This helps fine tune individual loops.

Say all loops

Gordy

What you gave me is representative of the whole zone. Slab sensor representative of that confined area. If not well placed means little.

schreib

OK but I have to wait on my T/C order. T/C can't be fixed as is. Next week I will have things levelled out even more and will get that dT data. Each loop. Anything else?

schreib

Gordy said:

What you gave me is representative of the whole zone. Slab sensor representative of that confined area. If not well placed means little.

No, dT I offer represents ALL zones-- whole house. unless I do not catch what you are saying above.

schreib

Are you just saying that I should OBTAIN a dT that represents each loop . . . or ZONE when I am able to get it?

Gordy

Yes both.

Gordy

I think you are close. You made a lot of different changes in a short period with large oat changes. Needs to settle in.

What I ask for gives alertness to loops in which may fall short on flow, or to much flow. Or maybe a loop that has air. Gets flow, but not like it should. Very valuable info. Especially with wildly different loop lengths. The program used gives a starting point for flows. Not real world just theoretical.

schreib

OK. I fixed the T/C and can start measuring again.
1:00 PM
ProOne T755S stat data
area zone Set pt Air Slab
temp temp
craft room 2 89.0 72.0 90.0
dining 3 72.0 70.0 73.0
kitchen 4 79.0 73.0 79.0
Great Rm 5 90.0 70.0 91.0
Guest Suite 6 87.0 67.0 89.0
M Bath 7 89.0 72.5 89.0
M BR 8 88.0 69.2 88.0

air temp not from Stat, T/C in middle of room
Guest suite and Dining mostly unused, want low temps;
The rest we are happy with. . . except would like to see Great Rm capability higher. Top ended with the stat. . .

SOOOOO. strategy for getting the zone and loop dT's and flows.

Please review and suggest options.
usually 2 or 3 zones on each SS manifold. If I run only ONE zone in demand at a time I will get different flow(even with the WILO trying to compensate) than if all zones are demanding. I could turn ON all zones in an individual manifold at a time, just one zone, or all zones ON.
Since this data run will be mostly catching the entire system OUT of demand any zone I place into demand will be an upset to the boiler.
Big zones have two loops, small zones only one.
Boiler startup: Hopefully, I can have a situation where I do not have the boiler cycling up and down in output-- which it will do if the set point value of boiler loop(from reset curve) is exceeded by 10°F. Boiler is immediately shutdown with a "timeout" delay before next start. If System sensor does not pass st pt + 10°F, it dumps boiler output, slowly until back below and again pumps up output-- which is fine, no cycling just hunting the st pt. Usually though, it overshoots by 10°F, kills the boiler, and while in timeout the system sensor goes way low and we have another cycle starting. Cycling like this continues(!) unless I have my pre-programmed ramp (output vs time delay) anti-cycling enabled. I can do this but it takes 30 minutes often to get into the range where reasonable control is reached. As far as I know Lochinvar is backwards in simple proportional control and this cute "anti-cycling" is their primitive solution.

My plan:
-- Set one manifold worth of zones into demand mode with anticycling ON. Take the data from these, turn them off and select next manifold.
-- For two loop zones, average temps to report one SWT and one RWT each zone.
-- Continue till all zones measured for both dT and flow.

Please let me know if this is a good plan for you. thanks!

Gordy

Don’t over think this.

Take temps on loops with an ir temp gun, or a clamp on meter. Put black tape on area of pex you are shooting temp.

Don’t force anything let the system run on its own demands you just have to be patient, and ready. Forcing it just skews actual numbers.

Flows are on the manifold meters. Not concerned at this point of that until we see anomalies in loop deltas.

Delta ts tell a lot about what’s going on. If you know where, and when to take them, and what can be causing the anomalies.

schreib

OK, no IR gun. T/C is more accurate anyway.
By the way, I are an ENGINEEEEEER. I cannot help but overthink things! It is a hazard I live with daily. Imagine my poor wife.

I'll start the data search.
Just the same. . . imagine smoke coming out. . . overthinking. . .

Gordy

Be one with the water, air, and temperature.

RxRoy

> @schreib said:
>
> Thanks for the input, RxRoy. OK, I was just reflecting the Lochinvar manual showing different diagrams' capabilities. I did NOT know, however, that it could not provide controls to handle them. Basically, from what you state I need something like one more Taco 4 channel control unit, ie like what I have now. Since you appear to be a Lochinvar trained tech. . . I was wondering if you can provide any insight to smoother control than their 6 level ramp anti-cycling control to prevent output fluctuation. I am not happy at all with how poorly that works. I am used to working with PID loop controls for industrial stuff and this system of theirs is just simply primitive. Wondering: Do they simply provide that for a 1st level simple solution and if you want better you need to use a 3rd party control like BMC stuff?

I'm not a Lochinvar tech. I only know because I installed a WHN 055 in my own house, and figured this out the hard way... I installed a Taco switching relay for the one extra system circ I tried to originally run off the DHW output. That didn't work out so great, because the boiler wanted to run balls out (which makes sense for heating an indirect), but no sense for a separate heating zone, which you want to modulate.

I can't comment on the ramp settings other than my first season I used them to maximize boiler run times. A better strategy, which I'm trying to use now, is to set the tstats as an upper limit, and let the ODR do it's job. Plan on taking a while to tweak in the ODR settings. I fully believe the ramp settings are a crutch for a poorly designed (or incorrectly adjusted system). I no longer have the ramp settings enabled and my boiler run times are much higher this year.

Thank God for all the much smarter people who take time to help the rest of us understand this stuff. I look at this forum everyday and constantly learn new things.

Gordy

One of these days the boiler manufacturers will get onboard, and be able to control the onboard circulator in vs mode, and ditch the third party bms option. The third party building management system is to many dollars for a residential setting.

EU viessmann 200’s do it.

schreib

have about 90% of zones done.
will complete and then wait 3 days for settling again and provide all data at once.
I will not be providing individual loop data since they appear to have same basic temps on the 2 loop zones. loop numbers should be same as zone data provided.
thanks

Gordy

Not really.

schreib

OK, ON 2ND set of data I will try to see if big diff. If so, I will get individual loop numbers too.

Gordy

Just a reality check. You could have a loop (s) in a manifold set running cooler than the rest especially a big manifold set. This would not be indicated by the manifold thermometer since all the loops come together, and intermix before hitting the thermometer.

Gordy

I thought engineers liked a lot of data points

schreib

Gordy said:

One of these days the boiler manufacturers will get onboard, and be able to control the onboard circulator in vs mode, and ditch the third party bms option. The third party building management system is to many dollars for a residential setting.

EU viessmann 200’s do it.

Gordy, what is "vs mode"? and are you saying that, in fact, 3rd party BMS would provide PID loop control on boiler output?
I looked at the viessmann and, yeah, nice machine. Too late for me though. Any chance I could retrofit a simple PID stand alone loop without going to a FULL BMS system? Or would that likely cancel any Lochinvar warranty? I have an electrical engineer friend who could handle the retrofit. . .!

Gordy

It's a little over the top for a residential 55k boiler.

VS = Variable speed circulator.

Not to be confused with circulator that operates on delta p, or delta t.

The boiler has a fixed speed on board circulator. Some additional benefits come from VS on board. You are tied to a radiant emitter. That's as good as it gets when properly done to give the boiler its highest efficiency potential through low return water temps.

I prefer K.I.S.S when possible

Gordy

From the manual.

schreib

Yeah. Love all that data, but not taking it by hand! I am retired but was "used to" sensors feeding it to computers and simple dump to a spreadsheet file. The other thing here is I'm flying blind, not having solved this problem before. I agree with RxRoy: "thank God for the much smarter people on this forum." Who are old pros at tweaking things into compliance.

That being said: I believe there is still a thread that folks have hinted at that I STILL(!) do not "get". RxRoy and probably you have both talked about using the stats as an upper limit and letting the outdoor reset curve do its job. Could you please elaborate on that? I am wondering if my system is wired to do that or if I simply am not comprehending how the stats, Taco units, and boiler TRULY interact. thanks!

Gordy

Outdoor reset drives the system based on outdoor temps. The curve it uses (set by you) sets the supply water temp to the system to compensate for the load at what ever the OAT is.

Well tuned there is no need for a thermostat. The ODR drives the system. It's not perfect. ODR can not compensate for wind, and external, internal gains. That's where the t stat comes in as the high limit device for those gains. Hey we are getting enough heat shut it down.

Winds drive infiltration. Infiltration varies on the structure. So even when the ODR is perfectly tuned it may be necessary to bump the curve on super windy, and cold scenarios.

Outdoor reset with indoor feedback is the next step. Now you have two devices communicating on what to do.

schreib

"Well tuned there is no need for a thermostat. " and "Outdoor reset with indoor feedback is the next step." Yep!

Gordy: Ok thanks. I get it: The control strategy is system LOOP set point defined by reset curve using CONSTANT boiler loop ouput with individual zone shutdown for high limit only.

-- I started out on this adventure on the forum by trying to get the blasted reset curve to be met. The only way I could conceive of doing it(at the time) was to overwhelm the flow with the boiler output so the boiler was pumping out nearly 100% with the WILO at near minimum to the system! At subzero temps, still could not get there. Your input moved me in the complete opposite direction luckily, and I get that. However, the point is I HAVE been trying to get the reset (set points) to be met and get to constant output. Just not the right "way"!
-- I have been whining about the boiler fluctuating without "anti-cycling" enabled: Multiple ramps are required (up to 6 time periods) with anti-cycling ON before boiler puts out enough Btu's to come close to satisfying the demand set by the reset curve. With huge demand I can turn OFF anti-cycling on a manual(!) basis and can rely on boiler running at 100% output and not fluctuating / cycling. An inherent problem with the anti-cycling algorithm is that it only works for ONE flow rate of the system pump and if the ODT changes significantly(eg 15° down to zero) the ramps settings are no longer appropriate. I have used probably 10 different ramp layouts trying to accommodate. No joy.
-- If I could simply rely on the boiler running at constant output and being capable of critical damped PID control it would come to temp set by the reset curve and stay there without cycling or overshooting so much that it shuts down. I need constant, reliable boiler output. But without ability to rely on the boiler keeping its primary loop under constant output . . . it seems to me the T'stats cannot be used as high limit for me.

Maybe I have not experienced enough of the system dynamics with my recently converted T'stats to slab control. However, I think the boiler dynamics need controlling before I can do this.

As I see it the strategy assumes boiler loop output is constant regardless of ODT. Without that the whole strategy fails because the fluctuating primary loop drives similar ones in the secondary.

Gordy

I think you have been making many changes with a high mass emitter with wildly fluctuating outdoor temps.

You need to let things run for a while so everything can find equilibrium before making assessments.

When equilibrium is reached "small changes" can be made allowing those changes to reach equilibrium before making more changes.

Thermo equilibrium can take days to reach in high mass systems. So running around changing multiple variables before they have a chance to give full effect can get you chasing your tail. Follow me?

The first change of flow rate, while making a difference still has not probably found equilibrium.