Building a simple heat exchanger


Building a simple heat exchanger

Building a simple heat exchanger


As part of my thermodynamics module I work on a two person team and built a simple heat exchanger in the thermofluid’s laboratory of DIT. We did this to see if a liquid of higher temperature can keep a liquid at a lower temperature constant with an aim of finding the heat transfer of the warmer liquid to the colder liquid. This finds use in brewing of alcohol where temperature change can affect the taste of the alcohol being brewed.

It was found that keeping a liquid at a constant temperature is a lot more difficult than previously thought. Although the temperature at times remained at the desired temperature a lot of fluctuation occurred in the temperature, it is believed that this was attributed to poor hardware wiring and not giving the rig enough time to settle.

The heat transfer was found to be ≈ 1212.2 watts


The following project was conducted to examine how heat exchanged from a liquid of higher temperature can keep a liquid at a lower temperature constant with an aim of finding the heat transfer of the warmer liquid to the colder liquid.

Keeping a liquid constant is useful in the fermentation of mead where the home brewer is trying to keep the batch at around room temperature without any sudden drops or spikes in temperature. If there was a sudden change in temperature the taste of the mead could be affected and yield unwanted results (Winning Homebrew, 2012).

In industry it is important for the temperature to be controlled (HowItsMadeEpisodes, 2009) explains this with fermentation tanks at a brewery being controlled to maintain a specific temperature.

Twelve weeks were available for this project to be completed in, but due to circumstances such as technicians not being available, hot water not working, sensors not working properly or not all material acquired fully the project had some delays.

A rig consisting of a fermentation bucket filled with cold water (to simulate the alcohol being made), this then was placed in a larger bucket and warm water filled the larger bucket.

Temperature sensors connected to a microcontroller recorded the temperature of warm water as it passed around the fermentation bucket filled with cold water. The temperature of the fermentation bucket was also recorded to see the effect of heat being transferred from the warm liquid to the colder liquid.

Designing the project plan

The project plan was divided into six main tasks that needed to be accomplished in order for this project to be successfully completed. These tasks were:

  1. Research
  2. Preparation
  3. Review Laboratories
  4. Laboratory work
  5. Calculations
  6. Report and Presentation

The sixth task was carried out throughout the project as more information was found it was added to the report. The following will be a breakdown into each task and its subtasks explaining how information was found, used and by whom.


During the research phase the method of brewing and fermentation was looked at in detail. Finding that for fermentation to be successful the batch needed to remain at a constant temperature 21°C (John & Elert, 2004). This led to examining different heat exchangers.

Problem 8.26 from a heat transfer problem book (Kreith & Bohn, 2013)  led to a formula for heat convection that could be used to calculate heat transfer from the hotter liquid flowing around the fermentation bucket to the mead being contained in the bucket.

From (Premeaux & Evans, 2011) a way of measuring temperature was discovered by using temperature sensors connected to an ADC (analogy digital converter) from a microcontroller. The ADC reads in an analogy signal in as voltage this meant a linear relationship between temperature and voltage needed to be found by getting the slope of the line  and solving for  the following formula was created:

X = (Y-b)/m

In terms of temperature versus voltage, the slope of the line m is called the temperature coefficient and where Y intercepts b is called the zero degree offset (@0℃ Sensor will output Y milivolts).
The microcontroller being used was an Arduino Uno its ADC divides an analogy input into 1024 pieces. The Arduino’s voltage range is between 0v to 5v so 5/1024=4.8828 mv this represents the voltage portion that each increment of the ADC value represents.
There were two types of temperature sensors reviewed during the research phase National semiconductor’s LM35D and the Microchip MCP9700. With this project’s estimated temperature ranging between-5℃ to 100℃, MCP9700 was chosen because its accuracy was ±2@70℃ compared to the LM35D being accurate ±0.2@25℃ and the MCP9700 sensor had an operating range between -40℃ to+125℃ compared to the LM35D having an operating range between+2 to+150℃. Both sensors had a coefficient of 10 mV/℃ with MCP9700 having an offset of 500 mV @ 0℃
A general equation was then created when the previous equation and values for the temperature sensors were found. This equation was:

Temperature=  (Y-500)/10

Y=(ADC byte code value)*(ADC mv factor)
Temperature sensor offset=500 mV @ 0℃
Temperature Coefficient=10 mV/℃

This formula would be used in the microcontroller’s program when converting voltage to temperature.
The final sub tasks required to complete the research phase was sourcing the materials. This project’s focus was on the home brewer and what they would have access to in terms of material so material selection was kept simple and sourced from local hardware shops and convenient stores.


For the preparation phase the materials were bought and the micro controller program was created. The equation needed to be validated with the sensors. So the first program created was to test the sensor was actually reading in values and displaying them properly. To validate the hardware the following program was created.

This allowed for verifying that the hardware was connected correctly.

This program was used to validate that the sensors and the equation will work successfully together. The full program (which can be found in the appendices) was then created taking into account things like calibration offset and read time. The sensors were compared to an analogy mercury thermometer to calibrate them by means of changing the values of the calibration variables by the difference in the analogy mercury thermometer and the sensors.

The original program was to allow the microcontroller to interact with an LCD screen so that values for temperature could be written down on paper. The LCD screen however broke and with that it was discovered that values could be copied from the serial monitor (a window showing information from the microcontroller) directly into excel allowing for more values to be collected in less time.

Review Laboratories

This experiment required access to water. Laboratories for DT023/3 for the module Thermofluids 2 took place in two different thermo-fluid labs. Lab 201 and lab 180 in DIT Boton Street College were both assessed. Both had access to computers and running water. When discussions with the technicians were being carried out, it was found that lab 201 (supervised by the technician martin) had no hot water available. When discussing the project to James Mahon (lab 180 technician), he suggested using the laboratory taps as there is a hot tap and cold tab available. James said the flow rate in the tap is constant and the temperature can go between . Lab 180 was then chosen as the room to do this experiment.

Laboratory Work

After dates having to be rearranged for various reasons (Jim not available, hot water not working, sensors not working properly to name a few) the rig was assembled in the lab.

The rig was first wrapped with insulation material then set up with pipes correctly fastened to the laboratory taps.

The taps were turned on and a small period of time elapsed with both the hot, cold and a mixture of both sent through the rig to make sure it was working.


The rig was dried and the temperature sensors were taped onto various parts of the rig.
Using a thermocouple connected to a measuring device all sensors were calibrated. On the day the room temperature was 18.3℃.
The fermentation bucket was filled with cold water from the cold tap.

Then both taps were turned on and the water mixed until the temperature values in the fermentation bucket were ≈21℃

The electric stirrer was turned on and the microcontroller program was set to start recording. This procedure was repeated two more times to have three sets of readings to compare findings to. With the absence of an LCD screen the values were copied and pasted into window’s notepad to be analysed in excel.


The values taken from the laboratory tests were imported into excel and arranged to be graphed by (X, Y) scatter. These values will be discussed in the next chapter.

The report was on going with information added to it as work was carried out.

Project Analysis/Detail

First Temperature Readings

When the microcontroller started recording it can be seen that the inlet pipe temperature reached ≈52℃ @ 12 seconds it was noted that the temperature of the fermenting bucket (Taverage) was rising above ≈21℃.
Both hot and cold taps were changed in other to control this from that the temperature began to drop.
The outlet temperature starting at room temperature began to rise until it hit a height of ≈40℃ @ 70 seconds this too dropped until it levelled off at ≈21.5℃ @ 238 seconds.

When the time reached 68 seconds the temperature was at ≈36℃ but then dropped to ≈18℃ at that point it was believed that the inlet pipe had come loose from the tap and needed to be corrected this in turn due to the sudden motion of the pipe caused an issue with the wires on the sensor. The experiment was stopped and this issue resolved.
If electrical tape had been applied more effectively this issue would not have happened. It is also assumed that if this had not have happened the inlet pipe temperature would have remained at ≈36℃. At the time of the first readings the cause of the temperature drop was not fully made apparent and the hot tap was opened more than the cold to increase the temperature. The result of this can be seen on the second graph.

Second Temperature Readings

The spike at the start of the graph was when the microcontroller was restarted it displayed real high values and then started displaying the correct values.

From  the inlet pipe temperature is at a constant . It took approx.  for the outlet temperature to reach . After  the temperature of the fermenting bucket began to rise to , this rise in temperature was counterproductive as the temperature went above the desired temperature ( ).

The experiment was stopped after .

Third Temperature Readings

The third and final readings taken were carried out over 258 seconds. The rig was not given enough time to cool down so the time for the fermentation bucket’s temperature to increase is reduced. The inlet and outlet temperatures were Thi=50℃ and Tho=51℃ the experiment was stopped as the temperature of the fermentation bucket was going out of scope with the desired temperature.

Conclusion of readings taken

The first set of readings proved to be the most interesting.

Values were taken off the first graph. The first two were related to the inlet and outlet temperatures specifically the highest temperature of Thi and the lowest point of Th_o while Taverage remains relatively constant. This was then graphed as T_m (x) and TS

This graph shows how heat is transferred from the inlet pipe to the fermentation bucket this heat transfer can now be calculated as

q_(conv= m ̇ c_p (〖Tm〗_o-〖Tm〗_i ) )
q_(conv=(≈0.01)x(4180)x(23-52) )
q_conv≈-1212.2 watts


In conclusion it was found that keeping a liquid at a constant temperature is a lot more difficult than previously thought. Although the temperature at times remained at the desired temperature a lot of fluctuation occurred in the temperature, it is believed that this was attributed to poor hardware wiring and not giving the rig enough time to settle.

The heat transfer was found to be ≈1212.2 watts

When readings were to be recorded it was agreed that three sets of readings should be recorded as this would give a more detailed set of values than just one set of readings. This however proved to be a futile as the rig was not given enough time to cool down so when the second and third readings were taken it was assumed  because of this that less work was needed to transfer the heat into the liquid contained in the fermentation bucket.

If this project was to be retried the following would be changed:

  • A sensor would have been placed in the centre of the fermenting bucket to identify if a temperature gradient existed.
  • The sensors would have been soldered to a PCB board rather than connected to a bread board this would have stopped wires from coming loose during the experiment.
  • A hole and pipe would be drilled into the larger bucket to allow for the water to escape rather than the rig having to over flow for the water to escape.
  • Securely fix the inlet pipe from the taps to the rig so that less movement occurs and the experiment is not interrupted to connect  the pipes back onto the laboratory taps.
  • Better care would have been taken in water proofing the sensors as some of them got destroyed and needed to be replaced.
  • Proper allocation of tasks would be set and a task completion checklist reviewed every week.
  • More readings would be taken over several weeks to get a good view of the system.

This project proved to be both challenging and rewarding as it allowed for a more practical approach in analysing fluids and their behaviours.


Advantages and disadvantages of plate heat exchangers. (2011, April 24). Retrieved May 06, 2013 from
Anne Marie Helmenstine, P. (2013). What is Fermentation. Retrieved 04 29, 2013 from
Article Castle. (2013). Shell and Tube Heat Exchanger Advantage and disadvantage Specifications. Retrieved May 06, 2013 from
CodeCogs. (2004-2013, December 17). Types of Heat Exchangers. Retrieved May 06, 2013 from
HowItsMadeEpisodes. (2009, 01 18). How It’s Made: Beer @3 mins 29 secs. Retrieved 05 30, 2013 from
John, G., & Elert, G. (2004). Temperature at Which Beer is Fermented. Retrieved 04 09, 2013 from
Kreith & Bohn, B. (2013). Advanced Heat Transfer – Problem 8.26. Retrieved 03 12, 2013 from
Merriam-Webster, Incorporated. (2013). Mead. Retrieved May 05, 2013 from
Premeaux, E., & Evans, B. (2011). Technology in Action. In E. Premeaux, & B. Evans, Arduino Projects to save the world (pp. 22 -23). New York: Apress.
Thomas Publishing Company. (2013, May). Types of Heat Exchangers. Retrieved May 05, 2013 from
Winning Homebrew. (2012, 04 2). Fermentation Temperature Control. Retrieved 04 30, 2013 from


What's your response?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: