Steam Boiler Efficiency Ranges

3 posts / 0 new
Last post

Dear Bldg-Sim list,

We are trying to determine the efficiency of an old gas-fired steam boiler that we plan to replace with a high-efficiency hot water boiler. We asked a company that tunes burners to take two combustion efficiency measurements - one before tuning and one after. Unfortunately the burner was operating so poorly prior to tuning that it messed up the combustion efficiency analyzer; it gave no readings and our opportunity to obtain a pre-implementation combustion efficiency value was lost. We do know that the burner was being starved of combustion air, but we aren't sure by how much, or how low the efficiency of a gas-fired boiler can drop before its safeties shut it off.

If anyone has successfully measured the efficiency of an operating boiler that is starved of combustion air, and is willing to share this efficiency, I would be quite grateful. I am looking your 'worst of the worst' efficiency measurements. The rest of our energy model Is nicely aligned with facility usage, but we would need to drop the overall efficiency of this boiler to less than 50% to align with utility data, and even this value assumes that we are generous in terms of modeling warm winter temperature setpoints, high infiltration rates, etc.

For reference the steam boiler we looked at is about 45 years old, made by the Federal Boiler Co. (Model AST 6 350).


Thomas A. Davidson, PE

DLB Associates | 265 Industrial Way West | Eatontown, NJ 07724
tdavidson at | 732-927-5108 | f: 732-774-5000
DLB Website | YouTube | LinkedIn | Twitter | Facebook

Davidson, Tom's picture
Joined: 2015-07-22
Reputation: 0


A Combustion analyzer tells only part of the story.

Some boilers cycle frequently due to being oversized (a common situation)
and (of course) when operating at part load. The cycling brings some
inefficiency and it can be substantial. Hot water boilers, as I understand
it, are more impacted by these cycling losses than steam boilers.

For one of our projects, a steam boiler was losing water regularly via
condensate line leaks in a tunnel. Our energy model, with pretty good
measurements of equipment power and other operating parameters, would only
match their utility meter readings when we used 52% peak boiler
efficiency. The leaks apparently had big impact on efficiency.

James V Dirkes II, PE's picture
Joined: 2011-10-02
Reputation: 203


That's why you should make a spot test beforehand (though it's more
applicable to oil burners it doesn't hurt to do it) and always wear a CO
monitor in the boiler room. If CO went so high as to damage the sensor in
the combustion analyzer, you've got a significant risk of poisoning should
there be any spillage of combustion air in the boiler room.

As Jim said, the combustion efficiency is only part of the story. And I
would add that you should do your combustion test at different operating
point if you have a modulating burner, at least at low and high fire.
On 'small' (=non industrial boilers), and this is probably your case here
since your boiler is 350 HP and you're in the Northeast, you tend to have a
single jack shaft that adjusts both combustion air and fuel valve via
linkages and cam trims, and that's not very accurate and hard to tune (if
the guy tried to tune it at all). What I pretty much always see if that
people tune for proper excess air at high fire only (around 15%), and
excess air is MUCH higher (50% to 150% sometimes) at low fire so that
they're at least on the safe side to avoid incomplete combustion and its
associated health and safety risks, and this degrades the efficiency quite
a bit.

As far as efficiencies are concerned, combustion efficiency is definitely
only part of the problem (and you're measuring it at one point in time
too), especially on steam systems. The time of the 'Dead men' to quote Dan
Holohan is no longer, and steam systems are never working as best as they
could due to improper knowledge of its inner working and bad 'upgrade' or
hot fixes throughout their lifetime.

Short cycling could and pretty much always is a huge problem, a lot more
than on water systems, and even more so when there's a coil in the steam
boiler providing hot water.
The pressuretrols are rarely tuned properly, even on ON/OFF burners, and if
you add a second ptrol (either Lo/Hi/Lo or modulating) things get even
worse and mains and differentials are very often incoherent leading to
never modulating the burners,cycling on pressure very often, etc. Layer the
aquastat controlling boiler water temp and therefore DHW production and
it's a mess.

If you have DHW, summer efficiency is always really bad: your boiler is
tremendously oversized for the task, even with a high differential on the
aquastat (which is never the case, it's either fixed at 5F or the
differential is set at 5F too...), so you get really bad short cycling on
top of really high standby losses. Expect a summer energy factor of around

Now for heating season: typically energy factors are around .6 to .7. Could
be lower if short cycling, etc.
I would also check: proper weighting of the barometric damper, jacket and
back of the boiler insulation (back of the boiler insulation on scotch
marine boilers is on the inside of the back door and tends to fall off
based on my experience), air leakage at burner mounting and boiler doors
(rope gaskets usually), and I'd inquire about when they last cleaned the
boiler and tuned the burner, and whether they add chemicals for scale,
bleed the boiler, etc.

The boiler is only one part of the equation. The distribution can also be
very problematic. Among things I've experienced first hand:

The classics:
- The good old "double-hung zone valve" = people using windows as
temperature control (walk in the streets of NYC on a cold day and count the
number of open windows, it's horrendous), though that's more on the air
infiltration side... (long ago EDR were sized to provide enough heating for
apartments with open windows, as people once were scared of diseases from
foul air, and I think it has remained this way)
- Uninsulated piping (sometimes thousands of feet of it in large garden
style complexes)
- Busted air vents on one-pipe steam spitting everywhere, busted steam
traps on two pipe

The more exotic ones:
- Cut-off condensate line going directly to the drain (I'm not even
kidding: the boiler was taking so much city water that it was completely
rotted and almost not working anymore)
- Automatic water feeder + long time lag distribution + improper piping
arrangement on the return (the height between the main and water line
wasn't enough to push the water back to the boiler): Condensate can't get
back, boiler line drops too low => the boiler takes in city water, at end
of cycle condensate would start to come back, overflow the condensate tank
and discharge 180F condensate to the drain

When you say you need 50% efficiency to calibrate, if you're not modeling
part load efficiency nor standby/short cycling losses, I'm not surprised at
all, though I would make damn sure I don't have another hidden problem in
my model before and investigate more (especially the above) in the field to
see if I can justify this dramatic efficiency.

I hope this helps,

Julien Marrec, EBCP, BPI MFBA
Energy&Sustainability Engineer
T: +33 6 95 14 42 13

LinkedIn (en) :
LinkedIn (fr) :

2015-07-21 17:21 GMT+02:00 Jim Dirkes :

jmarrec's picture
Joined: 2013-01-09
Reputation: 0