The University of Akron

IdeaExchange@UAkron

Williams Honors College, Honors Research The Dr. Gary B. and Pamela S. Williams Honors
Projects College

Spring 2022

Induction Brazing
Austin Squire
[email protected]

Scott Compton
The University of Akron, [email protected]

Logan Hathaway
The University of Akron, [email protected]

Michael Fleming
The University of Akron, [email protected]

Follow this and additional works at: https://ideaexchange.uakron.edu/honors_research_projects

Part of the Energy Systems Commons, Heat Transfer, Combustion Commons, Metallurgy Commons,
and the Structural Materials Commons
Please take a moment to share how this work helps you through this survey. Your feedback will
be important as we plan further development of our repository.

Recommended Citation
Squire, Austin; Compton, Scott; Hathaway, Logan; and Fleming, Michael, "Induction Brazing" (2022).
Williams Honors College, Honors Research Projects. 1585.
https://ideaexchange.uakron.edu/honors_research_projects/1585

This Dissertation/Thesis is brought to you for free and open access by The Dr. Gary B. and Pamela
S. Williams Honors College at IdeaExchange@UAkron, the institutional repository of The University
of Akron in Akron, Ohio, USA. It has been accepted for inclusion in Williams Honors College,
Honors Research Projects by an authorized administrator of IdeaExchange@UAkron. For more
information, please contact [email protected], [email protected].
 INDUCTION BRAZING

By

Austin Squire

Logan Hathaway

Michael Fleming

Scott Compton

Final Report for 4600:XXX Senior/Honor Design, Spring 2022

Faculty Advisor: Dr. Gopal Nadkarni

Faculty/Honors Advisor: Dr. Scott Sawyer

Faculty/Honors Reader 1: Dr. Alper Buldum

Faculty/Honors Reader 2: Dr. Christopher Daniels

01 May 2022

Project No. 30
Abstract

Induction heating technology is used in a wide range of applications. From cooking to

blacksmithing, induction coils can commonly be found in commercially available products. This

design study will investigate the application of induction heating technology into vehicle

exhaust repair. Of the various joining methods of exhausts, brazing is the best suited for

induction heating using copper and silver fillers. In this study tubing made of mild and stainless

steel will be considered since they are the most used in the automotive industry. This purpose

of this study is to prove that it is possible to do the work of a welder using a compact,
transportable induction brazing unit and at a comparatively lower cost than the average

branded MIG welder. The keys to success lie in the proper electronic components and

configurations and the ideal coil geometry.

ii
Contents

1. Introduction .............................................................................................................................................. 1
1.1 Research ........................................................................................................................................ 3
1.2 Objectives...................................................................................................................................... 5
2 Design......................................................................................................................................................... 6
2.1 Design Procedure ................................................................................................................................ 6
2.1.1 Theory and Governing Physics: .................................................................................................... 6
2.1.2 Design Strategy .......................................................................................................................... 10
2.1.3 Conceptual Designs .................................................................................................................... 12
2.1.4 From Conceptual Design to Proof of Concept ........................................................................... 14
2.1.5 Learning Outcomes from Proof of Concept ............................................................................... 15
2.2 Design Details.................................................................................................................................... 16
2.2.1 Redesign ..................................................................................................................................... 16
2.2.2 Coil Design .................................................................................................................................. 18
2.2.3 Building the Final Model ............................................................................................................ 19
2.2.4 Roadblocks with the Final Model............................................................................................... 20
3. Design Verification .................................................................................................................................. 22
3.1 Testing Results .................................................................................................................................. 22
4. Costs ........................................................................................................................................................ 24
4.1 Parts .................................................................................................................................................. 24
4.2 Labor ................................................................................................................................................. 24
5. Conclusion ............................................................................................................................................... 26
5.1 Accomplishments .............................................................................................................................. 26
5.2 Future work ....................................................................................................................................... 27
References .................................................................................................................................................. 28
Appendix A Requirement and Verification Table ................................................................................... 28
Appendix B: Conceptual Designs ................................................................................................................ 29
Appendix C: Final Design............................................................................................................................... 1

iii
1. Introduction

Imagine being tasked with welding new exhaust piping on a car next to a fuel line or fuel

tank and how nerve racking it may be. One single spark can cause a catastrophic event resulting

in injury or death. All it takes is a slight leak in any of the lines or a faulty seal allowing small

amounts of fuel or fumes to escape that is not detected before welding, resulting in extreme

danger. This is what many welders in auto mechanic shops face when installing new exhaust

pipes or mufflers on cars. One slight mistake can cause a hazard for anyone in the immediate

area. What if there was another way to connect two exhaust pipes together without using an

arc welder? What if there was something much safer that could easily connect two exhaust

pipes together, even in hard-to-reach places that a welder cannot get to without taking the

whole exhaust system off the car? Our team’s proposed idea is to combine the concept of

induction heating with brazing to create an induction coil that can go over top of two pieces of

pipe and join them together around the circumference of the pipes. If this is successful, then

mechanic shops can use this to connect to pipes without welding experience and without the

danger of igniting any possible gas fumes. It could also prevent some loss of sight due to arc

flash of a metal inert gas (MIG) welder. Furthermore, this design could be used when installing

duct work or steel piping in various buildings.

The reasoning behind choosing brazing over welding for joining the two exhaust pipes is

that with welding, a high amount of compression force is required between the two pipes to

allow for the metal exhaust pipes to fuse together. Alternatively, with brazing all the induction

generators would have to do is heat the filler metal up to the melting point to allow for the

1
filler metal to bond the two exhaust pipes through capillary action, requiring no compressive

forces to make the joint. This makes the process much simpler and more effective because if

the joint was welded rather than brazed, then the welded joint would need to be cleaned and

finished, whereas brazed joints do not. Also, brazed joints are more corrosion resistant than

welded joints. Brazed joints may not be as strong as welded joints because brazed joints yield

strengths are weaker than the base metal but stronger than the filler metal, so the yield

strengths would be somewhere in between. For an exhaust system this does not pose a

significant issue, as exhaust systems do not see or undergo any high loading or cyclic loading,

allowing for the yield strength of the joint to be a little lower.

Lastly, unlike welding, brazing can join two dissimilar metals together with much more

ease, less skill, and with less complex machinery than welding. Brazing is also more cost

effective and efficient, making brazing more ideal for auto-mechanic shops and the common

do-it-yourself consumer who does not know how to weld. Not to mention many shops do not

have welding equipment, due to shop insurance rates increasing for having that equipment.

Many outsource their muffler work or other exhaust work to nearby muffler shops. If the

concept of induction heating can be combined with brazing to join exhaust systems, then this

could change how exhaust work is done and make it safer for the workers and open the

exhaust repair market up. After two semesters of working on this concept and constructing

different models, our group believes that combining induction heating and brazing is possible

and the equipment to do so will be roughly the same cost as some of the most popular welders

out there. This makes it possible to be successful in the market as well as providing a much

2
safer working environment for muffler and automotive shops around the world, which is our

goal in this process.

1.1 Research

Induction heating works by using an induction coil (made of copper tubing) and a power

generator to generate alternating current through the coil. The alternating current creates a

magnetic field and the direction of that magnetic field changes at the same frequency as the

current. This alternating current and magnetic field change in the metal creates eddy currents

(localized currents) which flow against the metal’s electrical resistance causing metal to heat

up. There is also hysteresis that occurs due to the metal’s resistance to the changing magnetic

field causing additional heating, but this contribution is negligible in comparison to the heating

from eddy currents.

Preliminary research was carried out in the form of scholarly research articles, meetings

with professionals, and interviews with potential end-users. Each member dedicated time to

becoming an expert of the “cans and cannots” of brazing, the common materials used in

muffler and exhaust systems, their alloying elements and how effective they are at forming

bonds with various brazing filler materials, and the melting temperatures of said materials were

determined to hone in on select materials and brazing fillers for testing. Consequently, mild,

and stainless-steel exhaust/muffler systems, along with a silver brazing filler were deemed to

be the most appropriate for testing of the prototype. These materials were chosen based off

their frequent use for exhaust systems and availability, as well as cost-effectiveness and

3
reliability. Also, a silver alloyed brazing filler has a lower melting point than the other popular

choice of copper, which will lead to shorter brazing times per joint.

The temperature of the air flow passing through the exhaust is also of particular

importance. As stated above, a silver brazing filler melts at a lower temperature than many

other common brazing materials. This can be as low as 1145 °F and up to 1650 °F. A typical car

exhausts has air passing through it at approximately 400 °F, but in some non-ideal conditions

the air can reach upwards of 1000 °F. This is the downside to using a filler with a lower melting

temperature, so other fillers will still need to be tested.

Our group also went and conducted research on our potential market to get their

feedback of what they think about the concept, what they would like to see, and what they are

doing today to join exhaust system components and if there was a need for any improvements

with the current method. Our group interviewed 15-20 different muffler/automotive repair

shops. 90% of the exhaust shops that were interviewed utilize the metal inert gas arc (MIG)

welding method for joining pipes, while the other 10% utilize muffler clamps to secure the

sections of pipe together. The average repair took around an hour to complete from start to

finish, and the cost per joint was approximately $2.50. Most exhaust shops were content with

the current process but were open to the idea of testing a new system to join the pipe if given

the opportunity. One stated that after thirty years of welding exhaust systems together using

the MIG welding process his eyesight was destroyed.

Our group also found that most automotive repair facilities typically outsourced their

exhaust repairs to nearby muffler shops since most mechanics are not welders by trade and the

4
welding process requires a great deal of skill. A few of these automotive repair facilities even

stated that they do not allow welding in their facilities due to safety concerns and liability

issues. The repair facilities stated that if there was a system that was simple enough for any

mechanic to use and did not produce the harmful ultraviolet light or sparks that welding

machine have, they would utilize that in their facilities rather than outsourcing the work.

1.2 Objectives

Our main objective is to design an induction power supply unit for the same initial cost

as a new high end metal inert gas (MIG) welding unit. Through our internal consumer interest

study mentioned in the Research section, we found that the most common way to join exhaust

components in the industry is MIG welding. MIG welding is a process that requires an extensive

skill level to achieve sufficient results and has a lot of hazards associated with it as well as

health issues to the user. Even with the best personal protective equipment, the high intensity

ultraviolet rays still can cause extensive damage to one’s eyesight over years of continuously

working as a MIG welder. With our proposed design, it will allow users with no welding skills

such as automotive technicians to make a solid, reliable joint between two base metals of an

exhaust system. While also reducing the hazards and health concerns of the user and those

around them. In doing so, our goal was to construct a unit that is easy to operate, easy to

disassemble and repair any broken or aged components in the unit, and cost as much as a high

end MIG welding unit.

5
2 Design
2.1 Design Procedure
2.1.1 Theory and Governing Physics:

This design is going to use the property of heat transfer. For this design to work in

automotive shops, the exhaust system components to be joined together will have to be a slip

fit (one piece [male pipe] has an outer diameter that is smaller than the inner diameter of the

other [female pipe]). The brazing material be a ribbon design will be wrapped around the outer

diameter of the male pipe then connected end to end to hold in place around the pipe. Then

the male pipe is slid into the female pipe. This will be the basis of the heat transfer circuit seen

below:

TE

Rconvection

TD

Rmale

TB = TC

TA
Figure 1: Heat Transfer Diagram and Thermal Circuit of Exhaust System Components.

In 1-D pipe diagram, the outside pipe that is red (female pipe) is considered a heat

source because the induction unit and coil is heating up that pipe through eddy currents

6
induced in that pipe. These eddy currents then pass heat to the next layer from B to C which is

the braze (grey) ribbon, then the next layer is the male pipe from C to D. Air is inside the pipe at

room temperature. We are assuming steady state 1-D heat transfer from the heat source to the

inner layers. Assuming the thermal resistance of the braze ribbon is negligible due to the

thickness of the braze ribbon being small (roughly R2 = R3). By assuming the thermal resistance

of the braze is negligible, then TB equals Tc in the circuit above. TB will need to be equal to or

higher than the melting point of the braze. The amount of heat transfer (q) is the same

throughout the pipe because the thermal resistors are all in series with each other. Heat

produced (q) by the induction module and coil can be found by the following equation.

𝜋𝑑𝜌 (1)
𝑞 = 𝑃𝑤 = (𝑁 ∗ 𝐼 ∗ 𝐾𝐶 )2 ∗ 𝑄𝑟𝑜𝑑
𝛿𝐿

Where Pw equals q, N is the number of turns in the coil, I is the current/turn. Kc is a

dimensionless factor that can be found based on coil and pipe dimensions; d/D, d/δ, and L/Lc.

Where D is the diameter of the coil shape, d is the diameter of rod/thickness of pipe, L is the

length of the pipe, and Lc is the length of the pipe enclosed by the coil. Qrod is also a

dimensionless term that is found based on the ratio of pipe thickness over skin depth. δ is a

characteristic of a pipe called skin depth. The chart to find Qrod and the formula to find skin

depth is as follows.

7
 Figure 2: Flux Factors Q and P for rods and slabs

𝛿 = √2𝜌/𝜔𝜇 (2)

Where ρ is the electrical resistivity of the coil, and ω is the angular frequency of the

coil’s current, and μ is the absolute magnetic permeability of the pipe. ω and μ can be found by

the following equations.

𝜔 = 2𝜋𝑓 (3)

𝜇 = 𝜇0 𝜇𝑟 (4)

𝜇0 = 4𝜋 ∗ 10−7 (𝐻/𝑚) (5)

Where μ0 is a constant value, and μr is the relative magnetic permeability and is a

function of the applied magnetic field’s strength for magnetic materials. Also f in the angular

frequency is the frequency of the induction module in Hz, in our case, our induction unit

operates at a constant 60 Hz.

8
 Now that q is defined and the thermal circuit is developed, the equation to derive the

temperature the brazing material will see based on that heat output. The formulas for the

thermal resistances are given below.

1 (6)
𝑅𝑐𝑜𝑛𝑣𝑒𝑐𝑡𝑖𝑜𝑛 =
ℎ(2𝜋𝑅1 𝐿)

𝑅 (7)
ln (𝑅2 )
1
𝑅𝑒𝑥ℎ𝑎𝑢𝑠𝑡 =
2𝜋𝑘𝑒𝑥ℎ𝑎𝑢𝑠𝑡 𝐿

Where R1 and R2 are seen in the figure above, L is the length of the pipe enclosed by the

coil. The thermal conductivity of the exhaust is kexhaust. The convection heat transfer coefficient

of air is given by h. Knowing that q is constant throughout the thermal circuit so long as each

element is in series. Heat will equal the following.

𝑇𝐷 − 𝑇𝐸 𝑇𝐶 − 𝑇𝐷 (8)
𝑞 = 𝑃𝑤 = =
𝑅𝑐𝑜𝑛𝑣𝑒𝑐𝑡𝑖𝑜𝑛 𝑅𝑒𝑥ℎ𝑎𝑢𝑠𝑡

Knowing TE equals to room temperature 20 degrees C. Then TD can be found.

𝜋𝑑𝜌 (9)
𝑞 (𝑁 ∗ 𝐼 ∗ 𝐾𝐶 )2 ∗ 𝑄𝑟𝑜𝑑
𝑇𝐷 = + 𝑇𝐸 = 𝛿𝐿 + 𝑇𝐸
ℎ(2𝜋𝑅1 𝐿) ℎ(2𝜋𝑅1 𝐿)

From TD then TC (braze temperature from the heat input of the coil) can be found.

𝜋𝑑𝜌 𝜋𝑑𝜌 (10)

(𝑁 ∗ 𝐼 ∗ 𝐾𝐶 )2 ∗ 𝑄𝑟𝑜𝑑 (𝑁 ∗ 𝐼 ∗ 𝐾𝐶 )2 ∗ 𝑄𝑟𝑜𝑑
𝑇𝐶 = 𝛿𝐿 + 𝑇𝐷 = 𝛿𝐿 + 𝑇𝐷
𝑅𝑒𝑥ℎ𝑎𝑢𝑠𝑡 𝑅
ln ( 2 )
𝑅1
2𝜋𝑘𝑒𝑥ℎ𝑎𝑢𝑠𝑡 𝐿

9
 By finding TC based on the induction module/coil package heat production, we can

ensure that the induction module and coil we design will heat up the pipe enough to get the

braze material to reach its melting point. If it cannot get the braze material up to its melting

point then we cannot order that and must order another induction unit or build another coil.

2.1.2 Design Strategy

We first looked at current designs and how they are being implemented. The current

coils used for induction heating vary but the most common types are the multi turn channel,

the multiturn ID, Multi turn OD, open-end C-coil, and the pancake coil. From these choices we

narrowed it down to the open-end C-coil. The multi turn channel would not suit our need with

exhaust pipe since there would be no way to get the currents to flow properly around the joint.

The multi turn ID and OD would work for the current flow, but once the exhaust is joined

together there would be no way to remove the coil and with the cost for a coil it would not be

feasible to have it as a sacrificial item. The pancake coil like the multi turn channel would not

get the adequate flow of current to the pipe to braze.

We then looked at the machine design as a system and there were two main options,

one having the power supply separate from the induction module. This design would be the

easiest to manufacture since the induction coil could be mounted directly off the module and

there would be no loss of induction currents through any cables. The water pump could also be

housed with the module requiring less consideration for cooling efficiencies. The downside to

this design is the bulkiness of the induction module, which would not work well with our

application having to lift a big unit up and support it while under a car. The other option has the

10
induction module and power supply inside the same case and utilizes cables that can carry

cooling water to a handle where the coil attaches. This design is the best for any work in tight

places like a vehicle and minimizes the stress put on the coils from having to support the bulky

weight of the induction module. The issues with this design though are the cooling cables can

be expensive and some induction currents are lost through the power cables before they get to

the coil end. We decided that it was more important to have ease of use within our system

since we are trying to appeal to shops that do not have welders on staff.

Next, we had a decision to make with the input power, whether to make it 120 volts or

220 volts. With the 120 volts the system would not be as efficient, but it would be readily

accessible to all shops as 120-volt outlets are abundant and there would be no need for the

user to have a special outlet installed. With 220 volts the system is a lot more efficient and

doubles the power of the induction currents. With the cost of induction modules increasing

exponentially with their rated power output the cost of having an effective system utilizing 120

volts would triple effectively taking the machine out of the market due to cost. We have

decided to move forward with 220 volts for the time being but have discussed adding a 120 volt

plug option that could be used if needed, however the times to perform the induction braze

would increase. There are welders that have dual voltage functions on the market today so that

option is still viable, and we will keep looking into it.

We discussed different timer options from simple dial timers to complex computer

systems that can be programmed to set the time based on pipe sizes being brazed. After much

consideration we decided within the scope and time that we had we would set up our machine

11
like a lower end welder that had a chart setup with the pipe sizes and the recommended times

to set for the induction cycle. The controller we chose is a digital timer that allows the user to

program in different time settings, this allows the user to save time from looking it up on a

chart if it is common pipe sizes that they are brazing a lot. One issue we found with this design

is that it needs a solid-state relay to control the high current and DC voltage switching. Solid-

state relays are cheap, but they tend to get extremely hot as they operate. This system requires

a large heat sink for any extended operation, to mitigate this we have made our cooling system

holding tank out of 3/16” thick aluminum and the entire tank will act like a heat sink as the

relay will be mounted directly to the tank with thermal paste.

Our team had different ideas on the main power switch for the unit, but we settled on a

sleek lighted push button design. This design was not meant to carry high voltage or current

and with our system rated at 220 volts and 9 amps another solid-state relay was needed. Again,

we decided to mount the relay to the cooling tank to act as a large heat sink but decided to

implement a small radiator with two fans that blow across it to help dissipate heat buildup

within the system. With the addition of the radiator comes in more plumbing within the system

for water to flow adding cost, but it gives us the ability to have the system run with no down

time due to overheating issues which is important in an exhaust shop where time is money.

2.1.3 Conceptual Designs

At the start of the design process towards the end of the fall semester, our group began

to brainstorm different ideas, and hand drawn different conceptual designs. These drawings

can be seen in Appendix B of this report. These drawings helped to get our group started in

12
thinking of ideas and ways to encase the induction unit for ease and use for our target

consumers automotive repair shops and do it yourself individuals. These concepts are not

detailed drawings or designs, they are lily pads that were used to leapfrog from so our group

could determine the overall picture of the unit and end goal. Concept 1 in Appendix B is a

housing with the induction module, power supply, and cooling system all inside with dial knobs

to set the length of time and power draw of the unit based on thickness of the piping system

and the material of the pipe. With gratings in the casing to allow air flow to come in and cool

down the electronics. The handle would be magnetic so that the user to mount the handle

underneath a car where the coil needs to be to create the brazed joint. With two different coil

designs, a C-coil, or a hook coil. Concept 2 is the same sort of design, except instead of analog

knobs, there is an electronic display on the casing.

Concept 3 is an armature design that has feet with wheels that would roll around on the

shop floor, with the power supply housing that contains the induction module, power supply,

electronics, and the cooling system at one of the feet. The armature has two rotating locking

hinges with the C Coil at the end of the armature. The hinges allow the user to angle the arm

and coil up to the right height and position underneath the car on the lift. There are hoses

running along the armature from the coil to keep the coil cooled and transmit the current to

the coil.

Ultimately our group went with a mixture of concepts 1 and 2 because of the simplicity

and compactness of the whole unit. Concept 3 would take up too much space in a shop, and

would require a shop setting with a lift, which many DIY mechanics may not have.

13
2.1.4 From Conceptual Design to Proof of Concept

After brainstorming ideas and narrowing down what our group was looking for, we

decided to buy an induction unit and power supply from Ebay and begin to build a mock-up.

This mock-up was not the final design, it was just one that we wanted to build to test the theory

to see if the 3000-watt induction module would be enough to heat the pipe up to make brazed

joints. The induction module came with a 220-volt AC to 53.5V DC power supply and a multi

turn OD coil. We constructed the mock-up model in CAD, then manufactured the shell and

ordered the parts. The shell was made from 14-gauge mild steel with perforated 304 stainless

steel vents that were welded to the shell pieces. The vents are used to provide convection

cooling into the case from the surrounding air. The shell pieces were cut out of a CNC plasma

cutter out of Scott’s machine shop and then welded together. The cooling system contained a

12-volt water pump to pump coolant through the coils from the giant cooling tank at the

bottom of the prototype and then used silicon rubber hosing to connect the tank to the pump

and the pump outlet to the coil, then put the coil into a radiator with dual cooling fans to cool

down the coolant and then back into the tank.

The electronic components in the casing consisted of AC and DC solid state relays, 220-

VAC to 53.5-VDC power supply, 120-VAC to 12-VDC power supply, power switch, DC power

switch, 50-Amp DC breaker, DC Power Readout, thermal paste, programmable timer, 220-volt

power cord, and flyback induction module. The 220-VAC to 53.5-VDC power supply is what

powers the flyback induction module and provides the current for coil. The DC solid state relay

is used to transmit power from the 220-VAC to 53.5-VDC power supply to the induction model

14
whenever it is signaled to do so by the programmable timer. The 220-VAC to 12-VDC power

supply powers the timer, the DC power readout display, the pump, and the fans on the

radiator. The breaker switch was used to cut the power to the induction unit in case of

emergencies through the main 53.5-VDC circuit.

We decided to use the coil that came with the induction unit solely because we wanted

to prove the theory out by heating some pipe and getting some temperature readings to see if

this power supply and induction module combo can heat up the pipe hot enough to the melting

temperature of silver braze material. We are not going with this coil design because once it is

around the two pipe connections underneath the car there would be no way of getting it out

once the joint is made. Images of the proof-of-concept model can be found in Appendix B.

2.1.5 Learning Outcomes from Proof of Concept

Upon testing the unit and running several types and sizes of pipes through the coil. We

tested stainless steel, aluminized steel, and steel. We found that the stainless-steel exhaust

pipe heated up as expected, it maxed out our temperature reader at 1200 degrees Fahrenheit

in 30 seconds, so we know the temperature is at least that which is higher than the lowest

melting point for silver braze (1145 degrees Fahrenheit). However, our group did run into a few

issues. One being that after 30 test runs the AC solid state relay burned out, upon researching

and contacting professionals about the issue. We found that when the induction coil is powered

off, the breakdown of the magnetic field would create a reversal current spike through the

power lines directly into the solid-state relay and short circuit it. We found that to counteract

that, there must be a suppression diode connected in parallel to the inductive load if a solid-

15
state relay is used to power it on. The suppression diode allows the backflow of current from

the collapsing magnetic field to loop itself through the coil until it dissipates due to the

resistance in the coil.

We also found that the spacing between the coil and the pipe can play a key role in the

efficiency of the magnetic field and heating. The smaller the spacing the smaller amount of heat

is induced to the work piece.

2.2 Design Details

2.2.1 Redesign
After proofing out the concept and ordering a suppression diode for the induction

module, we began to do a redesign of the housing unit. The original design was all right, but the

coil had to be attached to the housing. Not to mention the housing was fully enclosed and too

compact, making it difficult for the user to take it a part to get to different components if

anything went wrong with the part or to do any preventative maintenance of the parts. Overall,

the original model was messy. Our group decided to make a new model. Something that can be

easily moved across the shop floor and make the coil able to move freely rather than be fixed to

the housing.

We began drawing the model up in CAD. We began with the housing first; the housing is

made of 3 pieces of aluminum sheet metal cut to assorted sizes and then bent. One piece was

bent into a 90-degree bend that makes up the left and top side. On this side there will be a

water neck welded to the top to provide a fill tube for the coolant to the tank, and on the left

side there are two rectangular cutouts to mount ABS plastic louvers for air ventilation. The

16
other sheet was a flat sheet with a step bend towards the top that fits under the first sheet for

the right side. On this side there are two rectangular cutouts for ABS plastic louvers to be

mounted in for air ventilation. Then the last piece was bent on both sides to fit inside the first

two pieces to make the shell. Then holes were drilled out along the edges of the first two pieces

to mount and fasten with bolts to the plastic face plates. The bends of the bottom piece have

holes drilled out with bolt bosses press fitted into the hole to allow the bolts along the bottom

to fasten each plate into it. The plastic face plates for the front and the back were made of ABS

plastic, the sheet metal shell slides and fits into the plastic face plates, then there are bolt holes

along the side edges to fasten screws through to hold the whole case together. The front face

plate with the has a cut out rectangle at the bottom to fasten ABS plastic louvers in to allow

more air ventilation in the case to keep the electronic components cool. A cutout rectangle in

the middle for the DC current/voltage readout display. Then there were hex bolt pattern bores

on the top left corner there to mount and attach the TIG (Tungsten Inert Gas) welding cables to

pass coolant and electric current to the coil. To the top right of the front plate there is a

rectangle cutout for the timer display, for the user to program a time for when the alternating

current for when power transmits to the induction module, how long the induction module

runs. Finally on the front plate there is a circular cut out for the power button to be placed in to

power the unit on. Now, on the back plate, there is a long rectangular cutout to the top left of

the plate for the coolant tank to sit in to allow the user to see the coolant level of the unit. Then

another rectangular cutout below that for the DC breaker, then to the right of that there is a

circular cutout for the 220VAC-power cable to go into.

17
 After the housing was made, next was placing the components inside the unit on CAD to

ensure everything was going to fit. We put a bigger pump in this new design from the last

design because the coolant is going to have to travel longer distances to cool the copper coil.

We mounted the pump to the back left side on the bottom plate to have it closer to the

reservoir tank. The 120VAC-12VDC power supply is mounted in front of the pump. It powers

the pump, the switching, and timer in the unit. The 220VAC-53.5VDC power supply is mounted

next to the 120VAC power supply. On top of the 120VAC power supply, there are plates that

are made of aluminum plates that mount the coolant blocks and AC and DC solid-state relays.

That is all the base level components in the housing. Now above the base level there is a metal

mounting bracket that bolts to the right-side plate the holds the induction module up. And

above that plate the radiator with dual fans is mounted to the right-side plate where the top

vent is. Pictures of the CAD model, and each part model is in Appendix C: Final Design.

2.2.2 Coil Design

Our group met with induction specialist Bill Steuhr, to learn how to design and bend

coils. After meeting with him, we bought some copper line and began to make our different C-

coil designs. He told us that because our induction unit does not have the option of adjusting

the frequency of the alternating current, that we will have to make assorted sizes to provide

the right heating. If the induction module was able to adjust the frequency, then our group

would only need to make one coil design and adjust the frequency to find the right amount of

output heat.

18
 Our group had to create a pipe bender for making our coil designs because we could not

find one small enough to make the bends that we needed. We modeled one up in CAD and

printed it out to do out bends. The coils made along with our modeled pipe bender and coils

can be seen in Appendix C: Final Design.

2.2.3 Building the Final Model

This process took the longest, it took a few weeks to get the materials and get them

machined and processed for us to build. We had Logan’s company, HP, laser cut the aluminum

pieces out of stock sheet of aluminum, then had Technibus bend the pieces to the specs we

need. This took a few weeks due to a backlog of work at each company. Then it took about a

month for the front plate, back plate, housing handle, and the vent louvers 3D printed in ABS

from the University of Akron. Once we had all the pieces by mid-March we began building.

From the CAD model to the design there were a few things machined and created that

we found we needed. We had to machine a custom brass fitting with ribs on one end, that will

fit into rubber hosing from the pump outlet and flared inward at the other to fit the TIG cable

connections in place to create a seal. This fitting was then mounted to the induction module to

carry current through the TIG cables to the coil. We also had to adapt and flare out the copper

coils to adapt the coils to the TIG cables. A lot of various fittings were used. A handle was also

made, and 3D printed that the coil and all the fittings fit on and inside of. There is a switch on

the handle to activate the timer on the unit at the flip of it, instead of having to walk to the unit

to start the timer if the user is away from it.

19
 We used much of the same components from the previous design besides the pump, we

ordered a bigger pump and a new timer that was simpler and easier. The cooling system is

plumbed from the reservoir tank to the pump then from the pump directly to one end of the

one of the TIG cables then through that cable and though all the adapter fittings in the handle

to one end of the coil. Then around the coil and out the other end through the radiator

mounted on the top louver vent on the right aluminum plate. Then the radiator cools the

coolant and goes out of the radiator into the coolant blocks that are mounted on top of the 220

VAC power supply. Then out of the coolant blocks and back into the reservoir. The wiring

diagram for this model can be found under Appendix C: Final Design, along with pictures of the

final model can be found in Appendix C: Final Design.

2.2.4 Roadblocks with the Final Model

After construction of the model towards the end of March, we ran into an issue of

running at half power. The current going through the induction module was cut in half from the

original mock-up. With the current being cut in half, it took much longer to heat up the pipes.

We were not sure whether it was the induction module or the power supply that was causing

the issue. We began contacting specialists from the electrical engineering department to

diagnose the issue. We learned, as per the advice of the University of Akron Electrical

Engineering department professor Dr. Yilmaz Sozer, that we could hook 4 car batteries in series

to act as the power supply to see if the current goes back to what it should be. If the current

and voltage readouts are as they should be then we will know that it is a bad power supply, if

not then the induction module is bad. After conducting this test, the power did not go back to

20
what it was, so we concluded the induction module was bad and we ordered a new induction

module. When we received the new induction module and replaced it, we still had the same

issue. What we believe happened was that both the induction module and power supply went

bad due to a bad solid-state relay, the solid-state relay continuously send out low voltage

causing the mosfets of induction module to hit their linear range creating a high frequency that

burned out a mosfet. Not to mention the solid-state relay was sending current through the

body which was mounted on aluminum blocks which was mounted to the 220 VAC power

supply which damaged the power supply.

21
3. Design Verification

To verify the strength of the brazes, tensile strength testing will be done on multiple

samples of brazed tubing. This testing will verify that the strength of the brazes will be able to

withstand stresses that will be applied to them during vehicle operation. This will also help in

determining the quality of the braze. A braze with low capillary action should not be as strong

as a braze with appropriate capillary action. Therefore, the tensile should be an indicator of

proper braze temperature and heating time.

Hydrostatic pressure testing will be conducted on brazed pipe samples. This testing will

test for leaks in the brazed joint. Leaks may occur if the braze is porous or weak. A porous braze

may be the result of poor joint tolerancing or inadequate capillary action during the brazing

process. This may provide data on the range of acceptable tolerances when expanding pipes for

slip fit. The data pulled from this may also indicate a problem in the brazing process. Problems

including non-ideal heating time or inappropriate brazing temperatures.

Salt spray testing may be conducted on brazed joints and welded joints. Brazing material

is naturally corrosion resistant, materials like copper and silver do not corrode as fast as steel or

stainless steel. Since joints must be cleaned with a grinding tool before welding the passivation

on materials is ruined leading to corrosion. While surfaces should be cleaned for brazing the

joint itself should be more corrosion resistant because of the materials used. Prolonged

exposure to a salt mist environment simulates exposure to a highly corrosive environment like

driving on a salted road. So, any extra corrosion resistance may be verified through this type of

testing.

3.1 Testing Results

Due to unforeseen electrical issues and delays no samples were brazed by the final

prototype. This is because the final prototype is currently only running at half power and may

22
not be able to heat the metal quickly enough for a good braze. This issue made getting samples

made in time impossible. So, no testing was carried out on any brazed or welded samples.

23
4. Costs
4.1 Parts
Table X Parts Costs
Part Manufacturer Retail Cost ($) Bulk Purchase Actual Cost ($)
Cost ($)
220VAC to 53.5VDC ZVS $100.00 - $100.00
Induction Module ZVS $60.00 - $60.00
Radiator with fans Clyxgs $40.00 - $40.00
Water pump Bayrite $21.00 - $21.00
220VAC to 12VDC Alitove $20.00 - $20.00
Rubber hoses LDR Industries $10.00 - $10.00
Copper tubing Uxcell $25.00 - $25.00
50amp breaker Tiayang $25.00 - $25.00
DC switch Mxuteuk $10.00 - $10.00
Water tank Clyxgs $30.00 - $30.00
DC power readout DROK $10.00 - $10.00
220V power cord Bitmain $30.00 - $30.00
14 gage aluminum Online metals $200.00 - $200.00
Water cooled tig SUA $70.00 - $70.00
cables
Rivnuts Lokman $5.00 - $5.00
Screws and nuts McMaster Carr $25.00 - $25.00
Low conductivity Miller $30.00 - $30.00
coolant
20ft 8 gage wire Teuofeng $30.00 - $30.00
20ft 16 gage wire Teuofeng $15.00 - $15.00
60VDC SSR Crydom $150.00 - $150.00
220VAC SSR CGELE $22.00 - $22.00
Thermal paste Artic $10.00 - $10.00
Plastic Prints University $20.00 - $20.00
Reverse Diode 1N-4048 $40.00 - $40.00
Total $998.00 $998.00

4.2 Labor

Prototyping and manufacturing are completely different things so we cannot figure out

the labor cost at the time it took to prototype. However, we can estimate the labor time to

produce per unit at an assembly line operation.

24
25
5. Conclusion

After two semesters of challenging work, the group is proud of what has been

accomplished. We had designed and built a portable induction brazing unit that is capable of

making automotive exhaust repairs, as well as free any stuck or corroded bolts. There was a lot

learned about induction heating, how the magnetic fields create the currents, and how to

intelligently design an electrical circuit. By using low quality electronic components, we caused

cascading failures in other components. This set us back the most, as the bulk of the time and

work spent went into troubleshooting where our loss of power was coming from. By the time

we replaced the original faulty piece of equipment, the AC solid state relay, it had already

caused damage to the power supply and induction module. However, as we were working on

building our new prototype and our initial testing was completed, we did not realize the

damage that had been done until we were ready to test again. After meeting with faculty in the

electrical engineering department at UA (University of Akron) and running power tests on the

induction module with four 12-V car batteries connected in series, we found that the induction

module was still only drawing half the amperage required. We replaced the induction module

with another one, but with our original power supply being faulty we still were not able to draw

the power we once had with the original testing.

5.1 Accomplishments

Although our finished product has suffered some setbacks with electrical components

getting burnt out and we were not able to produce the brazing samples we wanted, we were

able to prove the concept early on with the original prototype. We proved that by using

26
induction heating and copper coils, you can produce a magnetic field around the turns of the

coils that will generate enough heat to melt brazing material and join two pieces of tubing

together. With the proper coil geometry based off the frequency of the induction module, this

induction heating would also be sufficient to use in a hand-held form, utilizing TIG welding

cables, as the resistance through the cables is small enough to consider negligible. Also, even

with the unit currently operating at half power, using a standard coil will generate enough heat

to successfully loosen any stuck or corroded bolts from their fittings.

5.2 Future work

A new 53.5V or 60V power supply should be the answer to getting this unit to operate

back to its full capacity. Every other original component in the brazing unit has been replaced

with new ones, but due to costs we have not replaced this item since we were short on time.

However, the desire to see this design work to its full capacity still exists, so there is still work to

be done. Firstly, the new induction module must be tested with the car batteries in series to

verify whether the power supply really is the problem. If we are able to achieve full capacity

with the batteries, then we will know for sure the problem lies with our power supply. If we are

unable to get full capacity, it will be back to the drawing boards again, something we have

become remarkably familiar with during this challenging project.

27
References

[1] Dong, H., Zhao, Y., Yuan, H., Hu, X., & Yang, Z. (2019). A simplified calculation method of
heat source model for induction heating. Materials, 12(18), 2938.
https://doi.org/10.3390/ma12182938

[2] Matthew Way, Jack Willingham & Russell Goodall (2020) Brazing filler metals, International
Materials [3] Reviews, 65:5, 257-285, DOI: 10.1080/09506608.2019.1613311

[4] Valery Rudnev, Don Loveless, Raymond L. Cook. 11 Jul 2017, Theoretical Background from:
Handbook of Induction Heating CRC Press

[5] Stuehr, Bill. Interview. Conducted by Scott Compton and Logan Hathaway. Feb 11, 2022.

[6] Harvey, Ian. “Induction Heating.” THERMOPEDIA, Begel House Inc.,

https://www.thermopedia.com/content/874/#INDUCTION_HEATING_FIG1.

Appendix A Requirement and Verification Table

Table 1 System Requirements and Verifications

Requirement Verification Verification
status
(Y or N)
1. Reach a brazing temp of 1200 1. Tested with first prototype Y
a. Meet power requirements a. Tested with new prototype N
b. Locale heat as needed b. Tested with new SSR Y
c. Tested with batteries Y
2. Coil For Brazing Exhaust 2. Tested coils while hooked to unit N
a. Must fit over exhaust a. Fitment over different size Y
b. Provide adequate current exhausts
c. Work with induction b. Checked coolant flow Y
module used (matched adequate
frequency) c. Multiple designs made Y
3. Case to house electronics 3. CAD Model all electronics Y

28
 a. Must fit all electronics a. Find space needed for Y
b. Provide adequate cooling ventilation
c. Easy to carry b. CAD model case Y
d. Protects user from hazards c. CAD model handle Y
e. Easy to open and do d. Isolate electronic Y
preventative maintenance connections from aluminum
enclosure
e. Easily removable side cover Y

4. Cables and handle to coil 4. Used cables for tig welding Y

a. Must be safe for user a. Designed sleeve cover Y
b. Carry enough power b. Tested with power hookup Y
c. Carry coolant to coil c. Tested with coolant pump Y
d. Operate machine from d. Cycled switch on handle Y
handle

Appendix B: Conceptual Designs

Concept 1: Has a casing for the induction unit, power supply, and cooling system with adjustable knobs for selecting the
material of the pipe and thickness of the pipe.

29
Concept 2: Has a casing for power supply, induction module, cooling system with an electronic display to change settings on
the unit.

Concept 3: This concept is an armature that can be wheeled across a shop floor. The power supply, cooling system, and
induction unit is inside the power housing box. The armature has two hinges to adjust the height and where the coil is
located.

30
Figure 3: Contents inside the initial proof of concept model.

Figure 4: Initial model

31
Appendix C: Final Design

Figure 8 Mocked Power Supply

Figure 5 12V 60W Water Pump

Figure 9 Coolant Block

Figure 6 12V 10A Power Supply

Figure 10 Cord Grip

Figure 7 12V Cycle Delay Timer Relay

1
 Figure 14 Black-Oxide Alloy Steel Socket Head Screw

Figure 11 Hose Fitting for Compressed Gas

Figure 15 Black-Oxide Alloy Steel Socket Head Screw

Figure 12 Low Pressure Threaded Adapter

Figure 16 18-8 Stainless Steel Hex Nut

Figure 17 18-8 Stainless Steel Button Head Hex Drive

Screw
Figure 13 Brass Barbed Hose Fitting

2
 Figure 22 Bottom of Case
Figure 18 Low Strength Steel Square Nut

Figure 19 Medium Strength Steel Hex Nut

Figure 23 Circuit Breaker

Figure 20 Nut for Barbed Hose Fittings

Figure 24 Computer Fan

Figure 21 AC Solid State Relay

3
 Figure 29 Induction Module
Figure 25 Coolant Fill Cap

Figure 30 Front of Case

Figure 26 Cooling Tank

Figure 31 Induction Prototype Radiator

Figure 27 DC Solid State Relay

Figure 28 Case Handle

4
 Figure 36 Louvers for Radiator

Figure 32 Rear of Case

Figure 37 Case Louvers

Figure 33 Top and Side of Case

Figure 38 Power Button

Figure 34 Side of Case

Figure 39 Power Hose Connection

Figure 35 Light Up DC Power Detector

5
Figure 40 Power Hose Connector

Figure 43: Inside the model

Figure 41: Timer Relay

Figure 42: Full CAD assembly model

Figure 44: Final built model

6
 Figure 45: Inside final built model

Figure 46: Final model heating up a screwdriver with one of our coil designs. The coil can be detached and replaced with a c-
coil attachment with much ease. This coil is for automotive shops to use to heat up rusted seized up bolts.

7
Figure 47: Design Day with all our design work with different coils we made. Left to right: Scott Compton, Logan Hathaway,
Michael Fleming, and Austin Squire.

Figure 48: The black pipe bender it was modeled after the blue one. We needed one small enough to bend the copper line to
make our own C-Coils.

2
Figure 49 Wiring Diagram

3
Figure 50 Bend Drawings for Case Bottom

4
Figure 51 Side and Top Bend Drawings

5
Figure 52 Side Bend Drawings

1
Figure 53 Bill of Materials

2
Induction Brazing WHC HRP Final Signature1.pdf

3
 Induction Brazing

Austin Squire, Logan Hathaway, Michael Fleming, Scott Compton

Department of Mechanical Engineering

Honors Research Project

Submitted to
The Williams Honors College
The University of Akron

Approved: Accepted:

Date: Date:5/2/2022
5/9/2022 Honors Department Advisor (signed)
Honors Project Sponsor (signed)
Scott Sawyer
Scott Compton (Metal Significance LLC.) Honors Department Advisor (printed)
Honors Project Sponsor (printed)

Date:
Date:5/2/2022
Reader (signed) Department Chair (signed)
Scott Sawyer
Reader (printed) Department Chair (printed)

4
 Induction Brazing

Austin Squire, Logan Hathaway, Michael Fleming, Scott Compton

Department of Mechanical Engineering

Honors Research Project

Submitted to
The Williams Honors College
The University of Akron

Approved: Accepted:

Date: Date:
Honors Project Sponsor (signed) Honors Department Advisor (signed)

Honors Project Sponsor (printed) Honors Department Advisor (printed)

Date: Date:
Reader (signed) Department Chair (signed)

Reader (printed) Department Chair (printed)

Date: 5/2/2022
Reader (signed)
Christopher C. Daniels

5
6