run time

Paul S_3

I have a job with a heat loss of about 95k. system is an old 2 pipe system. 3 in supply and return mains with big old radiators. my question is would I benefit from adding a buffer tank or would I achieve long run times with the condensing boiler just from the mass of system and do away with the buffer? boiler will be a Lochnivar KHN110

njtommy

With 3" mains I think you would have more then enough mass especially with cast iron rads. My only concern would be making sure you can heat all that extra water with having 3" mains.

Paul S_3

I did a heat loss and came up with about 95k...and that's with bad insulation.system that's being replaced is a 250k oil fired with a tankless coil...are you saying I may need a bigger boiler??? mains are only about 20 ft or so

SWEI

njtommy said:

With 3" mains I think you would have more then enough mass especially with cast iron rads.

Completely agreed. Those pipes comprise a built-in buffer tank. Make sure they get re-insulated if the original stuff has been removed.

My only concern would be making sure you can heat all that extra water with having 3" mains.

You lost me on that one. Unless you're trying to use on/off zone valves, of course...

Paul S_3

no....one zone with delta T circ

SWEI

Tune the ODR right and it will hum!

The Lochinvar controls support three ODR curves as I recall. You can make that work for you using conventional thermostats. Instead of turning the boiler on and off, just change curves. Once you tune the ODR for normal conditions, program a higher curve and a lower curve, then use a two stage stat to shift. If you get it right, the system will respond to a windy day by shifting to the higher curve and to a sunny day by shifting to the lower curve. This works far better than turning the boiler on and off with a stat.

I'm a bit less certain about the ΔT circ and the mod/con playing nice together. I'd probably install a 0-10V conventional circ for now and upgrade it once we have suitable ECM circs available. You should need a small circ -- possibly something like a 006.

Paul S_3

so constant circulation on the system pump....use a 2 stage thermostat to operate the 2 curves 2nd stage higher curve 1st stage lower curve....the 0-10v circ would be system pump or boiler pump? I think grundfos has them right??

njtommy

I was talking to a rep the other day he was going to look at a job where the house had big 3" mains. The contractor installed a new boiler that was sized for the load of the home. But was having problems with keep the house up to temp on colder days. He then told me it wasn't the first time he saw this. He said sometimes with larger old homes and big 3" mains you need to up size your boiler a bit to compensate for the amount of water the system holds.

If your main is only 20' or 3" I highly doubt it will be a concern.

Now with that nice fancy Lochinvar boiler in my Humble opinion it would be silly not to install there 0-10vdc pump in the system. I know it's not cheap, but could really make the system play nice with the modulation of the burner.

SWEI

Just one pump, it's a FT boiler. Don't remember if they call that the system pump or the boiler pump -- it may be possible to configure either set of terminals to work.

njtommy said:

I was talking to a rep the other day he was going to look at a job where the house had big 3" mains. The contractor installed a new boiler that was sized for the load of the home. But was having problems with keep the house up to temp on colder days. He then told me it wasn't the first time he saw this. He said sometimes with larger old homes and big 3" mains you need to up size your boiler a bit to compensate for the amount of water the system holds.

The only way the water volume would pose a problem is if the boiler actually turns off. That could be a big problem if the boiler shuts the circ off. Set the boiler controls for constant circulation so the pump only turns off if the OAT hits WWSD.

Paul S_3

The cost may allow for that type of pump do you have any links?...also i will doing the setup below with obviously 1 heating zone and the indirect on priority....

Paul S_3

I think this pump would work fine...what u guys think?

SWEI

The boiler will control it, but it's quite likely twice as much pump as you need on that system. It's also more than twice as expensive as one of these plus a conventional circ. I'd probably stick a used pump on it (15-58, 15-42, 008, whatever) for startup and see what the flows are like, then purchase the correct sized circ for the system.

Paul S_3

ok thanks guys...much appreciated

SWEI

Paul S said:

i will doing the setup below with obviously 1 heating zone and the indirect on priority....

That's a rather complex primary/secondary diagram. You should be able to pipe & pump it direct, like this

Paul S_3

are they smaller pumps that take the 0-10v ....I cant seem to find any...I would like to use it....since the head loss is low I should skip the p/s piping and just pipe direct

Paul S_3

Ill use this

Paul S_3

thanks Hatterasguy... but just for my knowledge if I wasn't piping it direct and piping it p/s...the variable voltage 006 boiler pump would be just fine?

Paul S_3

yes that too...but the pic above of the taco 006 says it can take a 0-10v signal I guess it has the controller built in already

Paul S_3

thank you for all your help

Paul S_3

I will this job is going in sometime within the next 2 weeks....do you mean using the 006 piping the system direct or as a boiler pump in p/s piping....I never piped a modcon direct...pump a little too small?

Paul S_3

I think we will pipe it p/s ....I already put it in pricing

Paul S_3

if I install the viridian on system side...set to delta t mode say 30f? or leave it at default 20f

SWEI

Pipe and pump it direct. 95k needs only 7.6 GPM at 25°F ΔT. The system will probably work fine at a higher ΔT than that.

The Nimbus does provide 0-10V variable speed control. That's the point.

SWEI

Hatterasguy said:

Provides all the benefits of the $450 pump at a fraction of the cost?

Not all of the benefits, but all of the relevant ones for this application. You won't get the vaunted 300:1 turndown that Taco claims, but you will get enough to span the boiler's operating range.

njtommy

My only question is when will Lochinvar being doing 10 to 1 turn down ratio on the wall mount fire tube boiler.

Paul S_3

I really dont know...i do not understand why they dont...this customer originally wanted the whn but i explained the min firing rate ....khn 10k.....whn 20k....for the 110.....that was the selling piont

EricAune

Just some thoughts.

You can try to dial this in with an external control, use multi-stage stats and ramp delay, a VV circ or even pri/sec piping but stepping over a dollar to save a dime seems a bit out there.

The 0-10v circ and the Knight control were developed with system operations like this on the table. Spending all the extra time needed to "dial it in" will quickly eat up anything you might have saved not using the engineered hardware right out of the gate.

Either way you'll probably build a great system but keep in mind replacement down the road would likely be more simple if you chose the VV circ over the external control. I guess I'm always considering local parts availability, ease of install etc.

Great discussion and info shared here, guys. I always spend too much time here, I gotta get back to work!

Paul S_3

Can you guys explain ramp delay in more detail...i/o manual doesnt really explain

njtommy

Ramp delay is a basically a time delay for 100% firing rate.

Once boiler starts in low fire it will idol there for say 5 mins then start to stage up to the next firing rate say 25% it will stay there for 5 mins and keep repeating this till unit gets to high fire or call for heat is over. It really does a nice job keeping the boiler from short cycling.

SWEI

Conventional phase-chopping using TRIACs has a limited amount of turndown range on single phase induction motors.
The Taco controls switch to a PWM algorithm below the cutoff in order to achieve higher turndown ratios. Pulse the power on for a second or two, then let the motor coast, then pulse it again.

The Nimbus is essentially just a smart light dimmer. It applies full power for a couple of seconds on startup to get the motor spinning, then falls back to whatever speed was set. The DIP switches allow you to set minimum speeds and control voltage thresholds to align the operating range of the control signal with the physics of the motor/fan/pump.

The Nimbus comes in a J-box mounted version that takes all of 10 minutes to wire up. For us, as controls integrators, it's the same amount of work as wiring a relay and we get far more flexibility using just one analog output.

As far as failures, you either replace the circ with an easily available off-the-shelf model (instead of waiting for a special-ordered VVF version to show up) or you replace the sub-$50 Nimbus board. At that price, we keep spares on the shelf. The looming availability of 0-10V controllable small ECM circs makes me view this as a temporary solution either way.

Leon82

Delay number 6 will limit the firing rate over the entire burn also.

njtommy

@SWEI

Ok so your basically using a very cheap control board to run a $75 pump instead of a $500 pump.

SWEI

njtommy said:

so your basically using a very cheap control board to run a $75 pump instead of a $500 pump.

Yes, and the same board can control pretty much any wet-rotor circ, including ones previously installed by others. It's a band-aid until we have access to ECM circs in the sizes we need. If the pump is large enough that the small (3A) Nimbus can't handle it, there are ECM replacements (MAGNA, Stratos, Viridian) where we have 0-10V and BACnet control options.

EricAune

While I agree that ramp delay can be looked at as a time delay to 100% firing I have to admit that it took me a while to wrap my mind around all that it can do for controlling a system.

First, most calls for heat in a high mass radiant system originate to satisfy a 1-2F drop in temperature at the thermostat or sensor. Ramp delay in this instance will slowly introduce heat energy into the slab while promoting low fire and longer run times. This is very helpful with smaller zone calls on multi-zone systems.

Second, for systems that operate at or near design conditions with water temps just above condensing the ramp delay function will often promote condensing and higher efficiency rates at the boiler.

I set up most of my radiant systems with what a lot of installers might consider a fairly long delay. For example, stage one is often 25% firing rate for 15 minutes, then increasing 15% each stage at intervals of 10 minutes minimum each. The driving factor here is to lower the cycles and up the run times without effecting comfort.

EricAune

The small ecm circs are coming. Flat curves, steep curves, they're coming soon.

SWEI

Hatterasguy said:

It appears that the TRIAC can get the PSC motor down to 30% of full speed.

WIth a TRIAC, it's percentage of full voltage (as opposed to full speed on a VFD.) I'm not sure exactly what the relationship is, since different motors have different amounts of slip. They tend to stall out around 40-45% of design voltage IME.

I would imagine that the flow at 30% speed is quite a bit less than 30% of full speed flow.

Pump affinity laws say RPM and flow are 1:1, but in a closed loop application the head drops as the flow drops, so I'm not entirely clear what the net transfer function looks like.

if the circulator is chosen wisely, it would appear that the TRIAC can function almost as effectively as the PWM found on the VV pump.

For a typical mod/con, I can answer that in the affirmative. I've never tried to achieve 10:1 turndown.