DIY 5KW Induction Heater

HomeBrewTalk.com - Beer, Wine, Mead, & Cider Brewing Discussion Community.

Help Support Homebrew Talk:

brewman !

Well-Known Member
Joined
Nov 27, 2006
Messages
2,080
Reaction score
217
I built a 5 KW induction heater system for Thing1 (link in my sig) 2 years ago. Since then I've brewed a dozen trouble free batches with it.

I've found induction heating to be ideal for brewing. It heats rapidly. It heats gently, there is no scorching of the wort or mash. There is no intrusion into the kettle it is heating. It is quiet. It doesn't generate any combustion gases. You never run out of fuel. It is inexpensive to operate. It is very easy to clean.

Direct heating a mash tun with induction is much, much simpler and more effective than heating with RIMS or HERMS. And there is little chance of scorching unlike with a conventional electric heat element like in a Grainfather or with propane heat.

After brewing 12 batches, the bottom of my brewing kettle looks like new and has never required more than a wash cloth to clean. All my beers have turned out excellent.

Several people have requested instructions on how to build the heater. I held off because I was worried about safety, complexity, reliability, etc. After 2 years of using it, I feel comfortable sharing what I did.

WARNING: this project involves high voltage electricity. You are responsible for your own safety. DO NOT PROCEED WITH THIS PROJECT UNLESS YOU ARE PROFICIENT IN WORKING SAFELY WITH HIGH VOLTAGE ELECTRICITY. Do not take anything that I've done as gospel. My ideas and implementation may have flaws that I have not discovered. PROCEED AT YOUR OWN RISK. I AM NOT LIABLE OR RESPONSIBLE FOR ANYTHING YOU DO. If you do not have the skills and knowledge to safely build this system on your own, it is your responsibility to find SOMEONE ELSE to help you.

WARNING: I am not going to explain every little detail about induction heating or how to build a 5KW induction heater. I will provide the general details and leave the theory and implementation up to the builder(s). There are a lot of smart people on HomebrewTalk.com. Collectively I'm sure you guys can figure it out.

WARNING: Always operate this unit from a GFCI supply. Even when testing and building. NO EXCEPTIONS !

This induction heating system is a bit finicky/technical to build. But once you think it through it is straight forward. It might look intimidating but it is really quite simple if you pay attention to the details. Basically the whole project is mounting the induction driver board and winding a coil for it.

There are 2 main components to building a 5 KW induction heater - the induction coil driver and the coil itself.

20190215_195216.jpg


Driver Board

This is the induction coil driver I used.

Warning: there is another induction heater driver board out there with a smaller heat sink that is stated to output 5KW but really outputs 3.5KW.

The driver board I used is available from a number of websites. Some of these websites may have better pricing or shipping options.

NOTE: there are a number of commercially available induction driver modules with outputs higher and lower than 5KW. The principles contained in this build should be applicable to most other induction systems. This is the second induction heater unit I have built. The first one output 3KW.

This driver unit was designed to heat the melting chamber of a plastic extrusion machine. Having said that, if you understand the principles of induction heating, the application of the driver board is irrelevant. You could use the same driver board for a wide variety of heating applications.

I have the contact email for the sales and technical support person for this driver board. The after sales support is terrible. Do not expect any after sales support. The unit does not ship with any documentation.

Here are the specs of the unit: (copied from the website)

============================================================

5.0KW electromagnetic heater manual

First, the main technical parameters of 5.0kw heater:

1: Size: 219*160*160 (length * width * height)

2: Working voltage: 220V

3: Power (adjustable): 3500W-5000W

4: Load inductance: 65±5uH

5: Conversion efficiency >90%

6: Working frequency: 20-25kHZ

7: Working temperature: -10 degrees to +50 degrees

8: Working mode: continuous heating

9: Wave protection - +1500V

10: Anti-clutter interference: 4000V

11: The installation distance of the induction coil to the heating element is 17mm (the thickness of the epoxy board is added after the insulation cotton is pressed)

12: Multiple sets of coils are wound around the same heating element, and the distance between the coils is greater than 2 cm.

13: Current 19-21A, 8-10 square lines, about 9-11 meters, insulation cotton 15-20 mm, inductance is only one parameter, (current has been debugged)

It is necessary to use a clamp-type ammeter to measure the input current, and whether it reaches the rated input current. If it is not possible, adjust it counterclockwise through the potentiometer. Note that if the current display value becomes smaller, it cannot be adjusted. It may be that the inductance is too large. Therefore, it is necessary to reduce the inductance. (The inductance is reduced, that is, the number of turns of the coil is reduced, the incoming current is large, and the number of turns of the coil is increased.)

Second, the wiring instructions

1. Three indicator lights are “Power Light”, “Work Light” and “Fault Light”. The "Power Light" and "Work Light" are on during normal operation. The "Fault Light" light is off. The "fault light" light is flashing when the fault occurs in the circuit.

2. The power supply 220V is connected to the column of “connected to 220V AC”, and can be accessed without any fire line.

3. The coil is connected to the two terminals on the ends of the electromagnetic coil, and the screws should be tightened.

4. The board is factory set to supply power and can work. At this time, there is a short-circuit line (black) in the “switch control” socket to short-circuit the socket. You can also gently remove the short-circuit wire (black) on the “switch control” socket by hand, insert one end of a two-core cable into the “switch control” socket, and connect the other two plugs to the thermostat. The two normally open contacts can be connected to the test machine.

1. The soft switch wire, power cable and electromagnetic coil wire cannot be connected to each other or have any connection with the outer casing.

2. Do not electrify or pour water into the water after power-on.

3, note: the radiator, the fan can not be grounded!

================================================================

A few comments about the driver board:

1) My unit does not put out, nor draw 5KW. I am not completely sure why, but I suspect the pot I use (Bayou Classic 1044) is only marginally suitable for this application, ie the bottom is too thin to be magnetically resistive enough to allow the unit to deliver full power. My pot is also further away from the induction coil (3/8") than is optimal. Closer would be better.

My setup will draw between 16 and 18 amps at 240VAC, depending on how warm the wort is. I suspect it would deliver full power (20A) on a better pot. FWIW, the power factor is near unity on my unit.

I have not adjusted the potentiometer on my unit. The instructions have been improved since I purchased my board ! If the pot bottom is magnetically restrictive enough (ie thick enough) there should be no reason why the driver won't deliver the power.

In spite of delivering a bit less than 5KW, I am very, very happy with this heating system.

2) The "switch control" doesn't work reliably on my board. There are supposedly 2 ways to control the output of the driver - 1) By turning the power feed to it on and off and 2) By connecting or disconnecting the two "switch control" pins while the driver board is powered up.

I have not been able to get method #2 to work reliably. I've tried using a manual switch between the pins, a relay, an SCR, etc. It will turn the board on and off a couple times and then the board remains off and will not turn on. Thus I control my board by leaving the jumper on the "switch control" port and turning the power feed on and off.

I suspect that the "switch control" method has a maximum off time before it will not turn on again, but I am guessing and have not received any official documentation to confirm that.

Note there is a delay of several seconds between turning the power feed to the unit on and it powering the coil.

3) The heat sink on this driver board is electrically HOT during operation. It is charged up to a high DC voltage. It must be electrically isolated from all other voltages, including ground.

I have my heat sink exposed under Thing1. Thing1 is protected by a 20A GFCI, which trips if there is any contact whatsoever with the heat sink. You may wish to totally enclose your heat sink or at least put a cage around it. I have accidentally touched the heat sink and my GFCI trips immediately. The resulting shock feels like touching a wet 9V battery. YOUR MILEAGE MAY VARY. PROCEED ACCORDINGLY.

I don't think the heat sink fan has ever turned on on my board.

My board did not ship with the fan to cool the PCB itself. If you look at my implementation, you'll notice a number of small fans on one side. The purpose of those fans is to draw air through the cavity between the PCB and the coil. The coil will produce some heat while in operation.

4) My driver board emits no significant electro magnetic interference (EMI) when totally enclosed within Thing1. I can listen to an AM radio while operating Thing1. I cannot do that with most treadmills.

Coil

The induction driver board is designed to drive a coil wound around a round steel chamber on a plastic injection machine.

There are a number of off the shelf prewound stove top type induction coils available from various sellers. However, I did not find any that were suitable for my brewing needs due to:

1) Unsuitable inductance
2) Unsuitable size (diameter)
3) Unsuitable power handling

Because of this, I had to wind my own coil.

The selected induction driver needs a coil of 65 +/- 5 uH with a certain cross sectional area in order to handle the resonant currents generated by driving the board. The cross sectional area isn't listed in the specs on the website, but it was listed elsewhere. I'll have to find that spec.

The induction coil also needs to be made of many individual small insulated conductors wound into a single conductor. This is so that no eddy currents are generated in the coil itself.

The small insulated conductors are generally enameled magnet wire. My coil is wound from 17? x 20 gauge magnet wire. Smaller gauge equals less eddy current heating in the coil. More conductors (more conductor area) results in less resistive heating in the coil. If you use conductors that are too large or you don't use enough conductors, the heating coil will get very hot. Edit: I need to check my notes on the number of conductors.

The tuned resonant circuit on this driver board appears to have a high Q and thus will only drive a load with a narrow inductance range, ie 65 +/- 5 uH.

The inductance of the load includes not only the coil itself, but also the mutual inductance generated when the pot is sitting on top of the coil.

To determine the length (number of turns) of the coil, you must make up a length of the coil conductor and then wrap a length of it and measure the total load inductance by placing the coil over the bottom of the kettle, as it will be in the final application. Do not wind the coil itself to have an inductance of 65 +/-5 uH because the load inductance will then be out of range when the pot is placed on it.

20190122_213048.jpg


Remember to place the coil on the bottom of the kettle when measuring the inductance !

To determine the inductance of the coil and of the load, I purchased a Victor 6243 Digital Inductance Meter.


You can probably use any inductance meter, but I chose the 6243 because I found out it was the instrument that the driver board manufacturer used to measure their coils. The 6243 measures inductance at 200Hz. While this is much lower than 20-25 KHz the board uses, it seems to work OK for this application.

Note that inductance can vary somewhat with frequency so it is possible for a coil to measure correctly at 200Hz but be off at 20KHz. However, I did not have this problem. The load inductance (coil plus kettle) measures 65 uH and the board seems to have no trouble driving it.

I initially wound about 48 feet of coil. In the end I believe my coil was about 32 feet long. I think my coil by itself has an inductance about about 55 uH. I think my pot adds about 10 uH of mutual inductance to the magnetic circuit.

Here is a flat coil inductance calculator:

I think this calculator under estimates the inductance of a coil.

Hint: the coil conductor has almost no inductance if held in a straight line. So you can wind a bit, then string out the coil conductor and measure the inductance to see where you are at. Then add or remove some conductor from the coil, etc. until you get the right inductance.

My 5KW coil is 12 inches in diameter. This spreads the heat out over a wide area on the bottom of the pot. Most commercial induction burners have a much smaller coil diameter. This concentrates the heat onto a smaller area of the pot and may cause warping.

The Bayou Classic 1044 pot I use holds 11 gallons and has a diameter of a bit over 13 inches.

The induction driver board is self protecting. It senses the load induction every time it is powered up. If the load inductance is wrong, the yellow LED will light up and no power will be provided to the coil.

The induction coil must be electrically and probably thermally isolated from the brew kettle. Theoretically the enamel coating on the magnetic wire would insulate the coil, but you never want to rely on that as that coating is fragile compared to normal wire insulation. With time, chaffing and heat, it may deteriorate and thus the induction coil should be electrically isolated.

I mounted my coil to the underside of a 3/8" piece of exterior grade plywood. You could use glass as well. I attempted to use cutting board plastic and it did not withstand the heat of the pot, the weight of the pot and the heat of the coil.

I insulated my coil from the plywood with a piece of Teflon baking sheet. Not sure this was necessary, but I did it anyway.

My coil is held to the underside of the plywood with a piece of fiberglass dielectric board. This material is commonly used in high voltage cabinets, transformers and motors. It is strong, electrically non conductive and it withstands heat well. Other materials may work, I have not tested them.

The coil conductors must be held firmly or they will vibrate against each other and chaff off the insulation causing a short circuit between conductors. A few short circuits are OK, but many are not.

You must cut ventilation holes in whatever you use to retain the coil in place.

The leads from the driver board to the coil should be kept as short as possible. And they should be insulated so they don't short out against the frame, the enclosure or the driver board itself.

There must be no ferro magnetic materials anywhere close to the driver board or the heating coil or they will heat up ! The induction board enclosure on Thing1 is aluminum. The stand is stainless steel. The fasteners are stainless steel. You'll notice the fasteners on the driver board are all brass.
 
Last edited:
OP
B

brewman !

Well-Known Member
Joined
Nov 27, 2006
Messages
2,080
Reaction score
217
The easiest way to build a "chassis" for an induction system would be to find a SS table that is not the least bit magnetic and put the coil under the surface and an induction friendly pot on top, on an insulator pad. Wala, invisible induction heating.

Or one could cut a circle out of the table top for the coil and place a piece of tempered glass over it and set the pot on top of the tempered glass.

For those wanting a sub frame under the table top, solder or weld some SS bolts to it. And run 2 1x1" SS or aluminum tubes the length of the table, hanging down 3-6" as necessary to hold the driver box in the correct position. Pretty easy to do.

You could have several coils heating pots on the same table top.

The induction driver board could be mounted in a box mounted to the under side of the table top.
 
Last edited:
Top