how much condensate per therm NG?

Mike T., Swampeast MO

Searched web and books for a while without success for the simple number. Found it for oil, but not natural gas.

"Typical" natural gas composition and molecular weights are everywhere if you can deal with the conversions.

Problem is that the number will be the total amount of water produced during combustion--NOT the amount of water condensed.

There's also the "little" problem of the water already in the air being drawn for combustion.

To even begin to make sense of comparing condensate to gas consumed you'll need reasonably accurate measurements of:

1) Temp and relative humidity of combustion air. Reports say that the Munchkin isn't particuarly well sealed, so it's a stretch to assume that the vast majority of the combustion air comes from the outdoors.

2) Temp of flue gas VERY close to the HX.

3) Temp of flue gas VERY close to the exhaust.

Chuckles_3

Would someone be kind enough to post this number or any related quantity. I can't seem to find anything by googling or by searching here, although I know I have seen the number here before.

Background: I am planning to try to see if the loss of efficiency in my Munchkin is related to reduced production of condensate even when the water temperatures are low enough that there should be close to full condensation. I am going to store and measure the condensate each day and compare it to the gas consumed.

Ed Goldner, NJ

The amount of condensate

varies with the RH of the air entering the burner any condensate from the natural gas would be negligible. You really should allow a longer period (at least a year) to establish a baseline before worring about efficiency spikes. I bet you"ll see that efficiencies will vary seasonally.

Rely_2

combustion test

have you done one

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Ed Goldner, NJ

Example of

> Would someone be kind enough to post this number

> or any related quantity. I can't seem to find

> anything by googling or by searching here,

> although I know I have seen the number here

> before.

>

> Background: I am planning to try to

> see if the loss of efficiency in my Munchkin is

> related to reduced production of condensate even

> when the water temperatures are low enough that

> there should be close to full condensation. I am

> going to store and measure the condensate each

> day and compare it to the gas consumed.

Ed Goldner, NJ

An example of

why you should not use therms per degree day to determine efficiency.
I have a standing pilot 82% afue boiler, I checked some of my old gas bills and the best I could find was this. In both Oct. and Nov. of 2003 the average daily temp was 52, however, in Oct. my gas consumption was 45ccf or 46.71 therms, in Nov. my gas consumption was 64ccf or 66.432 therms. An approximate 42% difference in one month! Although I can't factor in other gas use cooking,HW,etc. My average summer usage is usually between 14-18 ccf per month. Ambient conditions such as length of daylight, intensity of the sun, ground temp., # of cloudy days vs. sunny days etc. all affect gas consumption. Like the one guy said you can't compare Nov. apples to Jan. apples.I hope this helps. By the way .... I'm going to pick up my Munchkin T80M today.

Constantin

One site I checked claimed...

... 1.15 Gallons of condensate per therm. On the higher capacity condensing boilers, it's a indoor stream of condensate. To find the source, do a search here for 1.15, it shouldn't be too common a number. I have a link to the original source in that post. Cheers!

bob_44

Confirm Constantin's #

1.15 gal. per therm looks like a good number(source-Handbook ofAirConditioning Heating and Ventilating Strock Industrial Press). bob

Chuckles_3

Yes. No problem there, but therms/degree-day continues to drop.

Chuckles_3

> In both

> Oct. and Nov. of 2003 the average daily temp was

> 52, however, in Oct. my gas consumption was 45ccf

> or 46.71 therms, in Nov. my gas consumption was

> 64ccf or 66.432 therms. An approximate 42%

> difference in one month! Although I can't factor


I don't know where you are, but in most of North America November is significantly colder than October, so a 42% increase in gas consumption makes sense. Having the same temperature in both months does not make sense.

I don't know how your gas company calculates "average temperature" for an entire month. My bill shows the number of degree-days, and that is much more usueful.

> Ambient conditions such as length of

> daylight, intensity of the sun, ground temp., #

> of cloudy days vs. sunny days etc. all affect gas

> consumption.


But...I have the same records for my old CI boiler, and it never showed this trend.

Chuckles_3

Thanks! I did my own calculation after finding the weight of a cubic foot of NG on the web, and using the fact that 16 lb of methane will produce 36 lb of water (this follows from

CH4 + 2O2 --> CO2 + 2H2O ).

This calculation gave 1.1 lbs of condensate per therm, which is pretty close to your number.

I will measure condensate volume not just to compare it to the ideal value (I do realize the ideal value will not be reached), but also to compare condensate amounts with warmer and colder water, with and without reset. Right now therms/dday is the same with and without reset, even though it is warm enough that the reset temperature is below 125. So one thing I want to know is, is the amount of condensate per therm also the same with and without reset?

Mark Eatherton1

I guess...

you didn't put much faith in my post about the free btus that were residing in the mass of your home from the summer/fall season eh...

ME

.

Sorry about that. I'd prefer real data rather than a theory, though. The condensate volume is something I CAN test, unlike your suggestion. But if I don't see any problem with the condensate volumes, that will mean your theory could still be right.)

Plus and most importantly, I monitored my previous CI boiler's usage the same way as I'm doing now, and never saw any of those free BTUs then.

Right now, my Munchkin is using almost as many therms per degree-day as the CI boiler did at the same time of year. So if the Munchkin is more efficient only when it is given free BTUs (only in the fall) that are somehow not available to a CI boiler, then I've wasted my money thinking that it was the condensing and modulating that would help.

Mark Eatherton1

Sometimes...

you can't see the forest for the trees... But if you have the time, keep looking. You'll find SOMETHING.

YMMV

ME

Dale Pickard

Great Thread

You might check with the gas supplier as to the actual methane content of the gas.

As long as you are collecting data, you might keep a record of the relative humidity of the combustion air.

Reset should make a difference if you have a distribution system that can be driven at low temps.

I recall Viessmann teaching that condensate that occurs in the flue, and not "fire-side" cannot contribute the latent heat of vaporizaton to the heating water; which makes sense.

Deviation between your calculated result and your experimental result could be due to these kind of effects.

Dale

Kal Row

latent heat of flue gas condensation...

depends on flue gas solution separation...
water condenses below 212, but in solution with c02 at around 136 (depends on the other junk in the solution, and is different for oil and gas)
the engineers at weil-mclain wont come clean about it, or maybe they don’t know, as they did not design the aluminum HX for the ultra, but they get an outlet temp of 158 when the water is 190!! – and they have very little excess air, now, how is that possible, unless they are generating enough flue gas solution separation in the twists’n’turns of the HX to get the h2o part to condense at much higher temps???

And then the are the volume issues – a change in the flue volume by upsizing the pipe, – wont that cause more H2o to drop out of the mix – this is a practical question – since an issue with condensing boilers is visible steam out the side of the house (not to mention the frost coating all over the place) – was thinking of going up and down with the pvc, and on each up elbow , upsize the pipe and put a condensate drain at the bottom, then at the end, cone it back down to one pipe size smaller then original to make it jet away from the building,

My thinking is, that even while the total flue pressure is the same (or greater by the jet) – the banging around of the flue gas by the up down reversals will separate the solution, and each upsize, (pressure decrease by volume increase) will cause localized condensing that I can drain – this way I don’t have all that visible moisture going out the side of the house

This it a common complaint and we are all looking for a good solution

Constantin

Pounds or Gallons???

The reason I ask is 1.1 lbs of water is very different from 1.15 gallons. Perhaps you meant to write gallons, not pounds? By my calculations, 1.15 gallons should be on the order of 9.5 lb...

Mike T., Swampeast MO

Found another source "Condensate Production & Treatment"

Claims theoretical production of 1.6 liters per cubic meter @ ph of 4.0.

After conversion that's 1.196 gallons per hundred cubic feet.

Mike T., Swampeast MO

Found an interesting statement from Shell Oil. "The only valid comparison [using degree days] is for year to year in the same home."

.

Sorry about that, gallons.

Dale Pickard

I'm weak on combustion chemistry

And I anticipate that possibly I am just being dense, but, I don't understand how it's possible to get lower outlet flue gas temps than outlet water temps.

I realize that there is a great deal of heat in the latent heat in the water vapor in the gases, but the heat exchanger works both ways. If the water side of the hx is at 190, then so is the flue gas side, (of the HX). I could see where there is very good heat exchange to the water side but very poor heat exchange into the gas side. Possibly the latent heat in the water condensing out of the flue gas is sufficient to raise the outlet water to 190 (and the HX to 190+ but the heat exchange into the flue gases is so poor that they leave at 158?

Where did you get this information?

Viessmann designs flues for the Vertomat boilers with reductions at the terminal piece to increase the velocity so as to get the heavy, soggy gases away from the flue and the building. I think that one wants to be careful how much one necks down the flue though as these gases are also sluggish, having had most of the energy taken from them, event with a power burner.

Dale

Kal Row

temp info comes straight from wm...

energy release by rapid state change if utilized correctly can do amazing things - especially in a flowing fluid – If the fluid is stationary, the laws of hydraulics take over and everything stops, but in a flowing fluid, you "kon du ah-may-zing zings vees mah-nip-u-lay-zhun"

your question if “(and the HX to 190+ but the heat exchange into the flue gases is so poor that they leave at 158?)" – that happens near the bottom of hx and the wet stuff gets carried away and replaced by more flue gas to condense before the heat can travel back from the hx to the condensate - I think - um – maybe

when the ultra first came out, I thought, (logically of course) , that if used on a high temp load – it would loose heat to the load, if the return was hotter than the flue, – boy was I wrong!!! – floyd would argue with people me included, and they would disagree, well, he was right, the facts speak for themselves – but how???


Maybe it is the overshoot of the state change affect, “here grasp this straw….”

they get down to 459.99 below zero with magnetic state change release - done on top of a helium cooled table that has orders of magnitude more btu capability then the thing being cooled, even though the thing we want to cool is below that temp, – when they shut the magnetic aliment coils on the final table, which allows the item to go back to random flux and thus absorb heat (flux change is a refrigeration cycle) – the helium table this flux plate sits on top of, is ready to absorb much more total btu’s than the item it’s absorbing it from when the coil is on – this difference affects an overshoot of the refrigeration affect and allows it to cool something below it’s normal cooling point – at least this is how it was “splaind” to me

happens with state change of water too – common example is a warm front on top of a cold font and the ran falling through the freezing zone but staying liquid because of compression – but when it hits, the freeze is much more rapid than the rate the same drop would freeze when stationary in the freezing atmosphere

same thing happens in a natural draft cold boiler, when the fuel, air mix hits the cold heat exchanger – the impingement causes instantiations compression of the methane and it drops out of the mix and condenses, but of course flashes shortly thereafter – but by then the o2’s have gone by-by – and you get CO cause now there is not enough o2’s to go around – so here we got condensation in a fire by accelerated compression

so speed of state change is an issue, how/if the ultra utilizes it , beats me, we keep going at this and come up empty – so frustrating…I just hate not having information that should be readily available – how’s your sweedish, cause I think it will take a trip over there, to get the truth about this!!!

fyi – for newbees listing in, the 5 common accepted refrigeration cycles are:
latent heat of physical state change: solid <> liquid <> vapor

absorption: as in when water combines with ammonia it also absorbs heat

expansion: typically used in air cycle machines on jets

semi-conductor: PN diode junctions are cold on one side and hot on the other

magnetic: an item removed from a coil and allowed to return to magnetic random – absorbs heat and vise versa – try this boy’n’girls, rub a magnet on a screwdriver it gets hot and magnetized, bang it hard on a vice, it looses it magnetism and gets cooler than ambient temp

Kal Row

ps i asked twice to clarify...

cause i thought 190 meant discharge with 160 return - that would be in line with the 158 flue discharge

but they said no, 158 is the max design flue discharge, and not to worry about using reg schedule 40 pvc!!! - though
in their control supplement document pg17, we see the flue limit has a setpoint of 216 and the hx limit is 203 for return or discharge


who knows - maybe when it gets higher - they modulate down
or shove more air in - they wont say - shoving more air in would dilute the flue mix more and affect more condensation

bah - i give up....

Mike T., Swampeast MO

Degree Days

How is the number you are using calculated? If the common way it's just a simple average between the high and low temperatures experienced during a single day compared to a baseline of 65°.

The only way that time enters the equation is that single day--nothing whatsoever about how long a certain temperature is maintained.

Late in the fall, the temp frequently stays relatively high most of the time--say the low 50s--with a quite quick dip to the mid 30s on clear nights. Opposite during late winter when the temp stays in the mid 30s much of the time with a quick rise to the low 50s during a sunny afternoon. Exact same degree days are possible with very different heating requirements.

For a more accurate degree day measurement you need frequent temperature measurements--at least every hour. You then graph the temp as a wave and find the line where there is the same amount of space above as below the line--that's a MUCH better representation of "average" temp in this case. It gets even more complicated when the outside temperature exceeds the baseline (typically 65°) as you really need to calculate the effect of the additional heat.

Also that standard baseline of 65° may not be accurate. Even in the upper midwest, systems are frequently put into warm-weather shutdown at around 55° without problems. A higher than necessary baseline will have the exact effect you are experiencing--an apparent decrease in efficiency late in the season as compared to early.

Dale Pickard

Thinking out loud

It seems to me, that the flue gas side of the HX has to be at the dewpoint temperature or below in order to cause the water to condense out of it. The latent energy release from the flue gas intot he mass of the heat exchanger occurrs isothermally at the dewpoint temperature.

If the latent energy release into the mass of the HX were to increase the temperature of the HX beyond the dewpoint temp, the condensation process would stop.

"Stacking" the boiler, that is operating the HX with a large DT across it, that is cold in one area and hot in another, doesn't sound like a good thing for a cast exchanger. (one must pause here to appreciate the design of the Vitodens laser welded ss heat exchanger).

Having said that, the Ultra HX sure seems like a class act. As WM doesn't really make it, or the boiler, I don't know how much credence I would give their information. If they can't explain their information, then it's a clue.

Dale

Dale Pickard

Swampy,

You said,

"If the common way it's just a simple average between the high and low temperatures experienced during a single day compared to a baseline of 65°."

Degree day measurements are always taken hourly. Degree hour data is available. Averages wouldn't work at all.

I assumed that he was using published degree day data.

Dale

Mike T., Swampeast MO

Here's just a few references. Note the various sources...

http://www.vectren.com/web/eenablement/get_help/degree_days_i.jsp

http://www.shellenergy.com/home/GA/Res/chgs1/hdd/printer.asp

http://www.northwesternenergy.com/showitem.aspx?M=2&I=88

http://www.usatoday.com/weather/resources/askjack/2004-12-16-heating-degree-days_x.htm

Completely agree that such an average is not a good measure--better if taken over the entire season.

Will check the Weather Underground site I use and see if they're using this simple method as well.

BTW, found what appears to be a nice--if a bit expensive--degree day (on the spot) recorder here:

http://www.specmeters.com/Weather_Tracker_Degree_Day_Recorders/Model_380_-_Heating_Degree_Day_WeatherTracker.html

Mike T., Swampeast MO

Weather Underground

Uses the simple method. 65 - ((High + Low) / 2)

Am understanding now why my monthly degree day vs consumption numbers (when comparing old vs new boiler) are coming out as almost nonsense.

Will dig for National Weather Service data--surely it's calculated the "right" way--but believe the closest info comes from Paducah, KY about 60 miles away. Weather fronts just love to stall out right over us, so temp differences are surprisingly large over just a short distance.

Dale Pickard

Well, that method is wrong.

The averaging being done at the NW site that you lead me to are averages of degree day data taken several places over Montana geography, and averages of data taken over several years in a single location.

That's not the same as estimating degree days from the daily extremes.

There is also TMY data for 239 locations in the US. The Typical Meteorological Year data is made up of averages and projections of measured data. I have a very cool CD that lists hourly TMY data for a variety of variables, from horizontal radiation to dewpoint.

The best way to evaluate energy transactions is with hourly data. That average is almost meaningless.

Some time ago, a meterologist here at MSU, kept and maintained maps of Montana which were calibrated in the date of Lilac Bloom. It takes so many "solar/thermal units" to get a lilac to bloom. Mapping the dates provides a "bio-assay" of the microclimates in the mountains.

Dale

Chuckles_3

Mike T is correct. Published degree-day data are usually mean of high and low. That's what wunderground.com uses, and that's what our gas company uses. It may not be reasonable, but it is customary in this field.

I have also manually averaged wunderground's hourly data. Sometimes there's a 1F or 2F difference, sometimes there isn't. It is relatively easy to "eyeball" the temp vs time graph on wunderground, and see if the high/low average will match the hourly average. When I see the curve is lopsided (so to speak), I do the manual hourly averaging.

But the difference is usually not significant. It's amazing how often the hourly average matches the high-low average to within a few tenths of a degre.

Dale Pickard

But why average at all

Why not use the actual data? What's the point in averaging?

I guess the accuracy of the averaging will depend on the climate. A climate with extreme highs and lows, like mine, may yield more disparate results. Not much time is usually spent between the highs and the lows, averaging them would exagerate the results.

I like to distinquish between day and night as buildings are used differently.

Dale

Chuckles_3

I am sorry, I don't understand this question.

I assume "the actual data" means the hourly temperature readings. How do you propose to use the actual data without explicitly or implicitly averaging them?

>Not much time is usually spent between the highs and the lows,

>averaging them would exagerate the results.


This statment also makes no sense to me. As long as the usual laws for heat flow apply, i.e. heat loss is linear with temperature difference, then averaging is fine. You don't have to spend any time in between the high and the low.

dconnors

We list the production

Our spec sheet will tell you how much condensate will be produced in our boilers.

Dale Pickard

Misunderstanding

I'm sorry too. Am I missing something?

I suppose the 60 temperatures that might taken across the 60 minutes of each hour must be averaged to a single temp for the hour but beyond that....

Use the actual data by simply adding them up. For degree hours, 8760 hours in a year with an ambient temp for each one.
Degree days come from each 24 hour set of degree hours.

Dale

Dale Pickard

Can you shed light on this?

Is it possible for the boiler water output temperature to be hotter than the flue gas temp?

Dale

michael_15

Theoretical limit for condensate - derived

For those that are interested, I can walk through the determination of the theoretical limit of condensate per therm. This will be approximate, at room temperature.

The formula: CH4(g) + 2 O2(g) --> 2 H20(g) + CO2(g)

is known to release 810 kJ/mole, (19.35 MBTU/lb for you older folks) . If the water is allowed to condense, steam freaks all know that the latent heat of water is roughly 970 BTU/lb (40.7 kJ/mole) at atmospheric pressure. Thus, burning natural gas releases 890 kJ/mole of energy if the water is allowed to condense. For reference, one mole of water weighs 18 grams.

Now, 1 therm = 100,000 BTU = 105,500 kJ. So one therm of reaction is equal to (105,500)/(890) = 118.5 molar equivalents. Since each reaction makes 2 little H20's, we have:

237 moles H20 / therm of CH4 burned. 237 moles H20 = 4.266 kilograms.

4.266 kilograms is 1.127 gallons of water.

Of course, this is just theoretical assuming you get everything to condense and that you're burning things at 1 psi, but it's pretty close.

-Michael

Chuckles_3

To average, you must first add, then divide by the number of elements. Adding is just averaging, and averaging is just adding, except for a constant numerical scale factor of 60 or 24 or 365 or whatever. The trends remain exactly the same, it's just that the numbers are bigger or smaller. The scale factor is not a difference worth arguing about.

Mike T., Swampeast MO

Have You Tried Changing the DD \"Baseline\"?

Try 60 degrees and 55 degrees.

Am trying to think in terms of climate--not problems in the equipment resulting in less efficient heat transfer.

Flue temp is probably the most telling regarding the efficiency of a condensing boiler. Any reduction in heat transfer is certain to show as an increase in flue temp. The temps are low enough that they're easy to measure with inexpensive devices.

Mike T., Swampeast MO

As far as I can measure, I'd say most certainly.

Have a sensor in the flue test port of a Vitodens. What? 4" from the HX?

During sustained modulation, the flue temp is ALWAYS below the "boiler temp" reported by the Vitodens. Not just a touch--numerous degrees.

BUT the supply temp I can measure on the surface of pipe about 3' away from the boiler is almost always a very few degrees lower than flue temp. Certainly this surface temp is a very few degrees lower than the water flowing through the pipe.

Which temp matters? The near maximum produced inside the HX? (Presuming that's what the Vitodens reports?) The "true" supply temperature in the system?

Mike T., Swampeast MO

I don't understand how it's possible to get lower outlet flue gas temps than outlet water temps

Don't claim to understand myself but it does seem to happen--at least as we can reasonably measure.

Dale Pickard

Sure but,

that's not what Mike was suggesting. There is a difference between averaging 24 numbers to come up with a 24 hour average and averaging 2 numbers (max and min) to come up with a 24 hour average.

This will make more difference in the mountains and place where the extremes don't reflect the temperature most of the time.

Dale