Air Blowing Across Thermocouple Question

[bostonman]

Long story short, I had the "draft" measured on my boiler system - the air temperature and other measurements on the air leaving the boiler. This was done by a plumber who repairs these brand systems and he used a higher end meter that displayed oxygen, temp, etc...

The probe was slide into a small hole on the flue exhaust pipe and he measured 140 degrees F.

Later, after he left, I took a thermocouple along with a temperature meter, stuck the thermocouple inside the same hole, and measured approximately 160 degrees F (twenty-degrees hotter).

I also confirmed that my thermcouple and meter are working correctly by comparing the ambient air temperature to a wall clock that also displays the temperature; the meter was within one-degree of the wall clock.

Since I measured twenty-degrees hotter than the plumber with the fancier meter, my question is: does using a thermocouple with air blowing across it cause incorrect temperature measurements?

[EPAIII]

Could the key words here be, "after he left"?

Perhaps your boiler actually did get hotter.

But other explanations are possible.

The two instruments may have a different accuracy. Perhaps his has drifted or been damaged.

The air movement, within reason, should not change the reading of a thermometer. But a liquid on it can lower the temperature due to evaporation.

I am sure others could be proposed.

[IanB]

Two different measurements at two different times, possibly under two different circumstances? Any particular reason why they should be the same?

(No, air blowing across a thermocouple won't affect the reading (much). It may read a couple of degrees higher, but also will respond faster and reach equilibrium more quickly.)

[bostonman]

The system was shut down during the last several minutes he was there and remained off for some amount of time before I re-powered the unit and took my measurements. If anything, the boiler was colder than when he took measurements.

I'll provide more details, but believe this thread may divert as a result. I've been getting rust chips falling out of the flue pipe over the last few years, and, the inner wall of the flue pipe before it enters the chimney is rusted (I had removed the pipe a few months ago and saw the rust, but the plumber also removed the pipe to inspect it).

The chimney has a 5" liner, but it's completely dry meaning, rain and/or condensation isn't coming from the chimney liner and dripping down. The flue pipe is 4", and the pipe area existing the boiler is hollow underneath (by design). When the boiler turns on, a door opens that's located slightly above the boiler hollow pipe area, and it causes rust chips sitting in the pipe to fall on the floor.

The plumber was baffled why condensation is accumulating on the inner wall of the flue pipe (a few times I've seen tiny condensation puddles on the floor). He called the boiler tech support who was slightly baffled and stated the chimney liner (being 5") is too small and the air leaving the boiler is only 140 degrees F as a result, but should be 200 degrees F; also the oxygen level was off (either too high or low - I don't remember) which is contributing to the condensation).

I'm still baffled by this statement, but the suggestion was to increase the chimney liner size because too small will cause the flue pipe to be colder. To me, it seems a smaller pipe would cause the boiler to accumulate more air and be hotter instead of colder.

In any case, I'm suppose to now get a chimney guy to tell me what they can do.

Meanwhile, I was curious whether the temperature ramps up, remains constant, etc... so I took my own measurement and got 160 degrees F that seemed to remain consistent (with a few degrees variation). Now I'm wondering if the plumber did the measurement correctly - although it's still much cooler than 200 degrees F.

[ejeffrey]

I also confirmed that my thermcouple and meter are working correctly by comparing the ambient air temperature to a wall clock that also displays the temperature; the meter was within one-degree of the wall clock.

This doesn't really test anything except the "cold" reference sensor inside the meter.  Thermocouples measure temperature difference.  If just reading ambient, the output of the thermocouple will be close to zero regardless of the type and the display will just be the reference.

So for instance either you or the plumber could be using the wrong alloy thermocouple for your meter or a damaged thermocouple and it would still read ambient correctly.

If you want to check thermocouple calibration use boiling water and/or an ice bath.

It also could certainly be a variation either with time or the exact location you are measuring.

[thermistor-guy]

...
Since I measured twenty-degrees hotter than the plumber with the fancier meter, my question is: does using a thermocouple with air blowing across it cause incorrect temperature measurements?

Measuring gas temperatures with thermocouples can be tricky. Two sources of error:

* gas flow rate is high enough to cause heating of the thermocouple;
* radiation heat transfer between thermocouple and surroundings (in your case, walls of exhaust pipe),

NMI Australia's monograph #5 covers it; 3th edition (2004): https://shop.measurement.gov.au/products/monograph-5?variant=1042647327

The author, Robin Bentley, covers measurement of gas temperature in section 3.9. Table 3.1 has some error estimates.
He discusses the use of radiation shields to reduce the second error above.

[IanB]

* radiation heat transfer between thermocouple and surroundings (in your case, walls of exhaust pipe)

I think this should not matter as the walls of the exhaust pipe are being heated by the flue gas flowing through it. So when everything is in equilibrium the walls of the pipe will be no hotter than the gas. If the goal is to avoid condensation we want the walls to be hot.

[Kleinstein]

The thermocouple will measure a mix of the actual gas temperature and the radiation. Different thermocouples may measure a slightly different mix.

Chances are the walls are a bit cooler,as there is usually some loss on the outside. On the other side the furnace may radiate with a higher temperature.

Chances are the conditions can be a bit different, though normally the data measured are reasonably consistant from year to year, at least for a good and stable system.
The temperature of the boiler can make a difference and also wind may effect the draft of the chimney - it ideally should not, not everything is ideal. Another factor is the gas to air ratio that can change with draft.

[ajb]

The system was shut down during the last several minutes he was there and remained off for some amount of time before I re-powered the unit and took my measurements. If anything, the boiler was colder than when he took measurements.

This is a heating boiler, like for radiant heat?  Depending on what "some amount of time" means, some of that temperature change could be due to heat soaking back through the system after shutdown.  The radiators and attendant loops form a significant heatsink that basically disappears when the circulating pump shuts down, at which point some parts of the boiler/loop may *increase* in temperature as the system overall cools down towards ambient.  This happens over a few tens of minutes on my boiler, which is a high efficiency direct-vented type with a small heat exchanger.  If yours an older-style boiler with a 500lb cast iron heat exchanger, then there's a significant thermal time constant for it to reach operating equilibrium and then to reach idle equilibrium. 

Regarding your rust chips, one thing you may want to check is your thermostat configuration.  Many modern thermostats will run the heat in relatively frequent short bursts, which helps reduce temperature swings and works just fine for forced-air heating systems.  But since radiant heating takes more time to respond, those short bursts may not be allowing your boiler to reach proper operating temperature before its shut down again.  You may be able to reconfigure your thermostat to a radiant heat mode which will use a more appropriate control strategy. 

If you're using natural gas, make sure the chimney liner is stainless.  The water vapor content + combustion products are very corrosive and will eat steel and aluminum liners in short order. 

Another possibility is a leaky heat exchanger is just leaking heating loop water into the combustion area, and that's both reducing your gas temperature and causing increased condensation in the flue. 

[mzzj]

...
Since I measured twenty-degrees hotter than the plumber with the fancier meter, my question is: does using a thermocouple with air blowing across it cause incorrect temperature measurements?

Measuring gas temperatures with thermocouples can be tricky. Two sources of error:

* gas flow rate is high enough to cause heating of the thermocouple;
* radiation heat transfer between thermocouple and surroundings (in your case, walls of exhaust pipe),

Gas flow rate would need to be pretty substantial to cause any discernible friction heating error in this case.

Third major error source would be stem conduction or immersion error. Depending on conditions and thermocouple diameter this can be a major hassle or next to nothing.

[Terry Bites]

Mass airflow will remove thermal energy from a source thus redcucing its temperature.
Many detectors operate on exactly this principle.

[IanB]

Mass airflow will remove thermal energy from a source thus redcucing its temperature.
Many detectors operate on exactly this principle.

The second law of thermodynamics says that heat will flow from a hotter body to a colder body. Thus, if the air is hotter than the thermocouple probe, then the probe will heat up. If the air is colder than the probe, then the probe will cool down.

[bostonman]

I haven't had a chance to double check my thermocouple in boiling water, however, I'm quite sure it's K type (yellow) and my meter is K type (just looked at the manual to confirm).

I guess this isn't completely deviating, but why would enlarging the chimney liner cause the flue pipe to get hotter?

If air is flowing faster, it seems to me the hot air (fumes exiting the boiler - which is a forced hot water baseboard system powered by natural gas) would be leaving faster and wouldn't increase the flue temperature.

The plumber baffled me by this statement; nor could he explain why. As for the twenty-degree difference between his meter and my thermocouple meter, this is also a mystery.

Maybe I can try holding the thermocouple to a soldering iron set at around 200 degrees F in hopes I can get a good enough reading.

[CaptDon]

Let me share my condensate story with you. First a side note, the 90% efficient furnaces using PVC pipe for the flue always have condensate drip back and require a condensate pump usually sharing a duty with the air conditioning evaporator unit if it is a forced warm air system. My 80% furnace was meant to use also a 5" flue pipe although it leaves the furnace at 4" and necks to 5" as it ascends the inside of my brick chimney. When the furnace shared the flue liner with a 3" pipe from the water heater which had and open air gap draft inducer there was no condensate drip back. I replaced the gas water heater with an electric model and capped the 3" side pipe (which also ended the effect of the open air gap draft inducer on top of the water heater). Ever since then I have condensate drip back, the colder out and the colder the brick chimney the more condensate drips back. I forced a condensate gathering place into the flue pipe and put a drip cup leading to a 1 gallon can. The can will fill to the top in 48 hours during really cold spells. I have decided to just live with it, however I expect a big problem to arise because the 5" chimney liner is galvanized steel and is indeed rusting. My plan is to eventually buy a 90% furnace and run PVC pipe out the side basement wall. Cheers mate, best luck!!