Buffer tank formula relationship to boiler firing rate ?

Dave Carpentier

Something that's been confusing me for a while..

In the standard buffer tank formula you input your boiler's minimum firing rate in btu, the delta in the buffer tank that you would like to use, and the time period that you would like the boiler to run. The calc gives you a buffer tank size in gallons.
But dont modulating boilers run at whatever firing rate is required to try and get the supply water temperature leaving at target temperature ?
As the return water starts becoming warmer as it enters the boiler, the boiler will modulate back (dont need as many btu to hit that supply target).
With larger turndowns like 10:1, does the standard buffer tank formula still work ?

Jamie Hall

I can't see why it would. In fact, I have a feeling that combining a buffer tank and a modulating burner might produce some really strange behaviour; I'll have to think about it -- but I think at first glance that the only time I'd use a buffer tank with a modulating burner is if it couldn't modulate down far enough to handle a microzone or two. Let's see what other have to say -- my take is just theory here.

EBEBRATT-Ed

The purpose of the buffer tank is to prevent short cycling.... smooth things out if the boiler output exceeds the load at any given time.

If the boiler modulates down and doesn't short cycle, you don't need a buffer unless you're doing different temps and want mixing in the buffer tank or for hydraulic separation

As you mentioned in your post, their asking for the boilers minimum firing rate

hot_rod

I suppose it depends on the logic in the boiler control. If a system sensor in a tank is calling on the boiler it should run full fire until the return  temperature sensor ramps it down it down as the delta closes. The delta closing would indicate the tank or load is or is about satisfied. Or the sensor sees a satisfied tank and shuts off.

If it is watching a supply temperature increase wouldn’t an increase there tell you the boiler is catching up with the tank or system load? The circulator needs to be able to move the full output of the boiler to the tank or you would get cycling. With a 2 or 3 pipe whatever the load doesn’t require goes into the tank to recover it

The formula for a modulating boiler looks at the smallest load. If the boiler cannot modulate low enough, obviously it will cycle on and off

If that on off cycle duration is within your comfort zone, no need to have a buffer for that purpose.

Additionally a buffer allows loads to be covered for a period of time without calling on the boiler

Its fairly simple, adding mass to a low mass type boiler to make it operate with less cycling

This also eliminates component wear in the boiler itself

Modern cast boilers are seeing a similar issue as they are much smaller in weight and water content, nowadays. With the added penalty of an on/ off burner, no modulation

which is why the tank size changes between modulating and non modulation boilers

The boiler wants to catch up with the load, period. However you want to allow it to is really up to you😉

Dave Carpentier

In my particular situ, I do have a couple of micro loads (2000 and i think it was 3200 ?).
The 110k modcon Im planning on turns down to 11k.
So, Ive become comfortable with planning on a buffer tank (12 gallon water heater with 2 side ports and 2 top ports).
They dont really give much burner strategy in the manual. I had the impression that the burner will operate on a closing delta (inlet/outlet ports) but it also has to consider the remote mounted System Sensor.
If it's based on a delta between inlet port and System Sensor (no outlet sensor), then its likely to spend a fair amount of time at high fire due to the mass ?
Surely this throws off the buffer tank formula ?

hot_rod

Dave Carpentier said:

In my particular situ, I do have a couple of micro loads (2000 and i think it was 3200 ?).
The 110k modcon Im planning on turns down to 11k.
So, Ive become comfortable with planning on a buffer tank (12 gallon water heater with 2 side ports and 2 top ports).
They dont really give much burner strategy in the manual. I had the impression that the burner will operate on a closing delta (inlet/outlet ports) but it also has to consider the remote mounted System Sensor.
If it's based on a delta between inlet port and System Sensor (no outlet sensor), then its likely to spend a fair amount of time at high fire due to the mass ?
Surely this throws off the buffer tank formula ?

Good question, if the boiler output drops (firing rate) as the temperature returning is increasing, it is doing that to keep it in condensing mode possibly? This is the boiler and control version where the boiler pump also slows isn't it?

So then if the desired tank operating temperature was at 180 for example what would happen?

Or maybe it watches return temperature just for a protection, when the delta gets too low the circ would speed up.

And where does the ramp delay fit in? If it is asking the burner to keep stepping up, but also watching return, who wins?

Certainly there is a way to get full output whenever it is needed, the boiler cant just decide to ramp down if it has a high demand.

Good question for any of the Lochinvar experts here. I'll find out Monday if know one has an idea.

Jamie Hall

I'm following this rather eagerly! I know how I would control it -- and I'll be absolutely fascinated to find out what Lochinvar has to say, and why, and how the do it!

Dave Carpentier

The parts of the Epic non-combi manual seem to be;

"By installing the system supply sensor as shown in the
Hydronic Piping section of this manual, the temperature of
the primary supply can be controlled. The boiler control
automatically detects the presence of this sensor, and
controls the boiler firing rate to maintain the system supply
temperature to the set point. "

"Determine controlling sensor
For space heating systems, the temperature control can be
based on the outlet or if installed, the system supply sensor
(factory supplied). The control will automatically switch to the
system supply sensor once it is connected."

"Modulation
The Epic Fire Tube is capable of modulating its firing rate
from a minimum of 10% to a maximum of 100%. The
firing rate is dictated by the call for heat (i.e., space heating
or domestic hot water), the heating load, and various other
temperature limitations.
Boiler outlet temperature sensor
The Epic Fire Tube is factory equipped with a boiler outlet
temperature sensor which is also used as the limit sensor
for the heater. During a space heat call for heat, the outlet
temperature sensor is used to control modulation of the
boiler unless a system supply sensor is field installed.
System supply temperature sensor
A factory supplied, but field installed system supply
temperature sensor may be used with the Epic Fire Tube.
When installed, this sensor is used to control modulation of
the boiler during a space heat call for heating.
Boiler inlet temperature sensor
The Epic Fire Tube is factory equipped with a boiler inlet
temperature sensor. "

So, it does seem to be described like it abandons the outlet sensor in favour of the system sensor.
Without a buffer tank in the mix, that would seem a logical approach.. the firing rate tracking the load.
None of the piping diagrams feature a buffer tank, it's always a low-loss-header or the closely spaced tees.

No variable speed circ control on the Epic.. it's their "basic" model. Just a field supplied circ, sized to fit in the min/max flow rates window.

hot_rod

I have never been a fan of constrained delta T operation, as some recall from the "heated" discussions here If in fact that is what this control is trying to do is keep the boiler within a certain delta.

So in this example a buffer is hungry for BTU, the boiler can transfer them by widening its delta, running up to 40∆ safely in most mod cons. But the stubborn control is preventing that.
Sure it will extend the boiler run cycle, but is that the goal? Longer run, slower heat delivery? Or do you want to get the job done quickly, or at least keep up with a design load. Forced delta T circulation as all the hydronic formulas indicates, limit that potential to move large loads when the system may most need them.

A good read on the basics of how heat energy moves. That formula should be tattooed on the forearm of all heat and, hydronic especially, wranglers. A good project for Erin to get some of those stick on tats with the hydronic formula.

Ironman

Here’s the formula:

A simplified way to look at it: 10 - 1 - 1,000. With 10 minutes run time, you need 1 gallon for every 1,000 btus of difference between the boiler’s minimum firing rate and the smallest load. This assumes the standard 20* delta T.

So, if the boiler’s minimum firing rate is 11k, and the smallest zone is 2k, you need 9 gallons minimum.

hot_rod

Keep in mind that minimum load number you plug in is at a design day condition.
80- 90% of the heating season you are below, sometimes much below, that load. Or above, depending on how you are looking at it. So the buffer really comes into play for a 1000 BTU/hr. or lower load that all systems will see.

Upstate NY as an example, consider maybe a 5° design condition. Reality is 30- 35° is the largest %

Dave Carpentier

@Ironman - The confusion I have with that formula is the Minimum Boiler Firing Rate. If you have a control sensor in the buffer tank and the delta for the tank is set to say, 30deg (I gotta figure out how to make that delta triangle), why would the modulating boiler run at min fire rate ? It could crank up to 100% and then start stepping back as the temperature approaches setpoint, no ? If so, does that not throw off the buffer load calculation ?

Hot_water_fan

why would the modulating boiler run at min fire rate

A lot depends on programming: you might cap output at 50% or delay ramping upon start. Many ways to get the long runtimes. 

hot_rod

Dave Carpentier said:

@Ironman - The confusion I have with that formula is the Minimum Boiler Firing Rate. If you have a control sensor in the buffer tank and the delta for the tank is set to say, 30deg (I gotta figure out how to make that delta triangle), why would the modulating boiler run at min fire rate ? It could crank up to 100% and then start stepping back as the temperature approaches setpoint, no ? If so, does that not throw off the buffer load calculation ?

Read through this issue, learn about thermal equilibrium. The system dictates the boiler operating condition, not vice versa. So if the load matches the boiler output, at any firing rate, then the boiler runs non stop. In a perfect heating world we would, or will have boilers that could modulate infinitely.

In the meantime boiler control designers can add functions and algorithms to change the operating condition to help best match the load condition. They essentially force the boiler to run within a certain range. That may or may not be what you want or need. At some point you need to get inside the brain of the software designer to see how they are accomplishing that.

A typical boiler aquastat on a cast boiler is a control that forces the boiler to run in a safe operating range. A boiler could in fact run to 300F to reach equilibrium, unless we imposed a control condition to prevent that. Which they do, thankfully

The same with the new ECM circulators. It took manufacturers a few, many, tries to get those algorithms to a point where they work properly, or for the majority of applications.

If you have an 85K boiler and a 60K load you could on some controls derate that boiler on the control to 60K. At that exact design condition the boiler would run non stop. Obviously loads change second by second, temperature, internal gains all sort of factors. You may find a condition where you drop below design conditions, now you want the boiler to go beyond that derate. Should it, can it, under what limitations. Does the boost function overide all other control signals and demands? Always, or based on ??

I can pretty much understand and explain heat transfer based on the laws of thermodynamics. And by much practical experience. Asking me how a microprocessor reacts to inputs, is a question for someone above my pay grade. It could be making decisions based on any number, or combination, or what ifs, of inputs from 1/2 dozen sensors and switches on the boiler.

In some cases you may not get an answer on exact control logic as it is propriety information. Most factory labs you visit have a "skunk works" section that you cannot see. You could see a bunch of competitor products in there being reverse engineered. Programmers tend to live in the area of the shop

Ironman

Dave Carpentier said:

@Ironman - The confusion I have with that formula is the Minimum Boiler Firing Rate. If you have a control sensor in the buffer tank and the delta for the tank is set to say, 30deg (I gotta figure out how to make that delta triangle), why would the modulating boiler run at min fire rate ? It could crank up to 100% and then start stepping back as the temperature approaches setpoint, no ? If so, does that not throw off the buffer load calculation ?

Dave, I see your point and it’s a valid one. If that’s an actual scenario, then I’d increase the size of the buffer tank. Remember, it’s short cycling that we’re trying to mitigate, not necessarily the modulation of the boiler, though that’s an important component. 

There are piping and control strategies that can be employed to address it too, but I think that hot rod’s explanation is the best.

hot_rod

I think one possibility to solve the premature ramp down is as I have suggested in the past, put the boiler sensor in the tank. As long as sufficient load, cool return, the boiler should run full speed to cover it. If not that part of the control needs to have the ability to be turned off, IMO. Manufacturers push back on moving that sensor, possible it acts as a multi function sensor, or a safety needed for their listing, and absolutely need to stay in or on the boiler?

As it turns out the very same issue is showing up with the inverter driven, variable speed A2WHP. They want to modulate to keep COP high and this prevents full output to the buffer.

Dave Carpentier

Premature ramp down could be a concern for the load side, depending on how critical the setpoint temperature is for your emitters (in my case, a slow slab.. not so critical).

I think a bigger buffer tank, with the boiler's "system sensor" in it, would help with the run time but the boiler is likely to aggressively burn to try and hit that setpoint and still throw off the "min fire rate" in the buffer calc. Would you use the "max fire rate" instead.. or maybe 60% ?
The boiler doesnt know what the heck it's heating, it just knows that there is a current setpoint and it's going to go for it.

Keeping the boiler-buffer circulation down to a minimal speed (this was touched on early in the thread when hotrod mentioned a variable speed circ) seems like it would help.
Capping the max burner % would help (mentioned earlier). I do plan on capping mine after some monitoring.

It seems a paradox that increasing the delta in the buffer sensor theoretically allows for a smaller buffer tank but at the same time likely causes a higher burn % (initially) . Which wins ?

Boon

@Dave Carpentier said

with the boiler's "system sensor" in it, would help with the run time but the boiler is likely to aggressively burn to try and hit that setpoint and still throw off the "min fire rate" in the buffer calc

Dave, I'm a DIYer who installed a 18 gallon, 2-pipe buffer tank. I needed to add mass to my system to better serve low-load zones in the shoulder seasons. With the system sensor in the buffer tank, my system would 'aggressively' fire trying to satisfy the buffer tank, producing water temperature that was higher than what the emitters needed. Then, it would low-fire when it caught up to the set temperature. Eventually, the temperature dropped and the boiler would lag behind the buffer tank temp leading to aggressively firing, again.

I disconnected the remote system sensor in the buffer tank and started using the boiler's output temperature sensor because the emitters were rarely seeing the supply temperature prescribed by the ODR.

I posted some thoughts & observations about my buffer tank & temps and maybe something in this thread will help you: https://forum.heatinghelp.com/discussion/comment/1460488#Comment_1460488

I've enjoyed following your posts & learning from the comments.

Dave Carpentier

Thanks for the link @Boon
I wonder if primary side circ speed has more of an effect on it than we think ?
A slow primary flow might cause the boiler to start throttling back (getting too close to high limit) or even trip out on safety.
A fast primary flow gets the boiler's heated water to the buffer's sensor quicker, but at the same time wouldn't it allow the burner to burn at high rate and end up cycling too much ? Plus of course, it would churn the buffer tank.

I think a good control solution would be if they could use the outlet sensor as a throttling guide but still use the system sensor for the setpoint control.
I wonder what happens if you set the max burn to some demand-acceptable value like 80%.. would it start stepping down the burn % earlier - sort of a lag time allowance ?