I mentioned in a thread a few months ago that I was working on induction heating for the injection lines in my truck. I've got the circuit worked out and bench-tested. I'm happy to report that the circuit is extremely simple -- within the abilities of anyone that can handle a soldering iron. Furthermore, I've got a lead on some dirt cheap transistors.
I'm putting this out so that anyone that's interested can order some of these transistors and get started building an induction heating system on their vehicle. I'm leaving the country less than two weeks from now and I'll be gone three months. Anybody that's interested needs to act pretty quick. After I leave I won't be available for advice.
Here's the deal. The induction heater requires two TIP35C transistors per injection line. Normally these transistors cost two or three dollars apiece. However, Arrow Electronics has a bunch of them in stock they want to get rid of because they contain lead and don't conform with new environmental regulations. Arrow's TIP35CW transistors were listed for about a dollar apiece on their website, so I ordered 20.
Then this morning I get a call from a guy at Arrow asking me if maybe I would like to order more of them, for even cheaper. Otherwise they will just write them off at the end of the year and send them back to the manufacturer. So here is an opportunity to buy transistors for an induction heater at 50 cents apiece or maybe even less. Just go to arrownac.com, look up TIP35CW and call the 800 number that should be listed on the website for phone orders. Tell them you heard they are closing out the TIP35C made by STMicro and you'd like to negotiate a price. If you don't have any luck, let me know and I'll see if I can get a hold of the guy that offered me the discount. I might be able to put you in touch, if I can find him.
If you have friends interested in induction heating, you can get these transistors by the dozens and pass them around.
Because I'm leaving so soon I don't have time to place an order and distribute the transistors among people that are interested. I get back around New Year's and by then it might be too late. It's up to you to get them if you want them.
I'll post some information here about the circuit so you can get a feel if you want to do this. This is a Royer circuit with its transformer primary wound around the injection line. There is no secondary -- the injection line itself acts as a shorted secondary creating high losses that cause it to get hot.
You need to understand first of all how to wind a center-tapped coil on the injection line. The best way is to do a bifilar winding and cross-connect the ends. I used CAT-5 cable because the induction heater runs on the order of 100kHz, so you need cable that has a low impedance at high frequency, CAT-5 fits the bill and is cheap. Basically, it is the poor man's Litz wire. Look up "skin effect" and "litz wire" on wikipedia for some idea why you need special conductors at high frequency.
Cat-5 has eight conductors in it, so you don't need to wind it twice to get a bifilar winding -- just choose several of the conductors to use as one winding, and several for the other winding. Two of the eight little wires will be reserved for feedback windings, so your bifilar primary will consist of three and three. It's simplest to separate the striped wires from the solid colored wires. Then you have three designated striped wires as one of your windings, and three solid colored wires as the other winding. You connect three striped wires together at one end of the coil and connect them to three solid wires as the other end of the coil for your feedpoint -- that is electrically equivalent to a center tap. It IS a center tap, you just don't have to interrupt the winding.
The remaning three striped and three solid wires that you didn't connect will go to your transistors, one transistor at each end of the coil.
Anyway, that's enough information for starters to see if you think it might be for you. The circuit is called a "Royer" circuit. There's a wiki at 4hv.org with a diagram you can look at. It only requires the coil, a capacitor across the coil, the transistors, and perhaps some protection diodes on the transistor bases. You can determine the amount of power by choosing the size of the resistor that feeds the bases. I didn't need a resistor in the feedback winding, nor were the spikes on the waveform at the transistor bases bad enough to require protection diodes.

How about using this as a fuel pickup heater?

Your induction heating setup is interesting but my knowledge of electronics is very limited ( but I can solder <g> ) . Could you please post some some pictures or video of your bench tests / experimental setup ? More specific information on other components and the coil is appreciated . -

PM as necessary .

Thanks

volvo,
Induction heating would work great on a fuel pickup.
patt,
I got a response by private message from somebody with questions not unlike yours, so I'll paste some of the answer I sent by PM:

Here's the wiki I mentioned
http://wiki.4hv.org/index.php/Royer_oscillator
Do not copy those component values unthinkingly. It's okay to start with them in your experimenting, but you will end up using something other than 470 ohms, you'll likely end up using a different value for the tank cap (the one across the coil), and practical considerations such as how easily accessible the injection lines are will help determine the number of turns on the coil. And if you use CAT-5, the number of turns on the feedback winding will equal the primary turns. And so forth.
and an app note:
http://www.diodes.com/_files/p...eg/ANP005_AP2001.pdf
Diagram 2 shows the Royer and a method of controlling the power with pwm (pulse width modulation, or switch mode control). But you might find it's good enough just to run at some fixed power, achieved through judicious choice of the base resistor or resistors... as you may note, the 4hv wiki shows one resistor feeding both bases, while the app note shows a resistor to each base. There's very little difference between the two choices.
Whether you use pwm or not, you still need the choke. It is an indispensable part of the topology for this kind of circuit. Here's a good cheap choke you can buy (one for each injection line/circuit):
http://www.mpja.com/prodinfo.asp?number=18058+CH

The first thing you need to understand if you're interested in this circuit is what those dots next to the transformer windings on the drawings mean. They show the phasing. Phasing is kind of like polarity, but since we are dealing with ac instead of dc, polarity fluctuates. So it's called phase instead of polarity.
The feedback winding has to be connected in the proper direction to the bases, as shown by the dots on the drawing. Those dots show the direction of the feedback winding compared to the direction of the main winding. All the wires in the CAT-5, if you use CAT-5, will go around the injection line in the same direction, so the dots mean that the feedback winding to the transistor base is in reverse phase with respect to the main winding's connection to the collector of that same transistor. So you have to connect the base to the feedback winding at the end of the coil opposite the connection for that transistor's collector.

I spent a several hours putting the windings on my truck and still didn't get finished. And I found it easy to make mistakes, with all those wires to sort through. You kind of need to get a good understanding of the wiring scheme before trying. It's worth putting together a circuit on the bench to heat a single piece of metal with just one induction coil and two transistors before jumping in and trying to do a whole vehicle. I started by winding CAT-5 around a plastic cigar tube that I could slide a wrench handle in and out of to vary the load.

This is something that I would like to do, but I would need a dummy instruction set. I mean DUMMY.

Sounds like the perfect fuel pickup, and it would be as involved as to ILHs.

Oh, well...

Well, I've been writing some fairly long PM's to canolafunola with more instructions. I won't post it all here because I don't want to turn this thread into an ongoing blog, but for anybody that's interested in working on induction heating I can paste those messages together and send them on. Eventually it might turn into a project with easy, basic instructions. I won't have time until next year to work on something like that.

Whenever you can, that would be a wonderful contribution. Thanks!

Kell have you got any more details, like the transformer details and or a circuit diagram.
How much power does it consume.

Sinbad -- I already posted links to two circuit diagrams.
Take a look at them. Look at the 4hv wiki first, it's simpler. Then look at diagram 2 in ANP005.
And go here
http://www.richieburnett.co.uk/indtitle.jpg
and read up on the theory of induction heating. Our application would use much lower power levels and simpler electronics than Burnett's stuff but the theory he explains about the LCLR topology still applies and it will help you to get a feeling for the basics of induction heating -- how the work coil and capacitor form a resonant "tank circuit," how the oscillating magnetic field creates eddy currents on the surface of the workpiece, and the eddy currents and magnetic hysteresis losses make the workpiece get hot. And why it's a lot easier to heat ferrous items, which means, for anybody that wants an inductively heated fuel pickup, they need to use iron or steel.
I wrapped felt (just cheap weatherstripping) around my injection lines before winding the CAT-5 around them because the vinyl jacket on the CAT-5 only has a 60C rating. But you might be able to use magnet wire and just wind it directly on the injection line, or pickup or whatever you want to heat. You might not have to use multiple filaments if you don't plan to draw a lot of power, which brings up your question about how much power does it use. The component values like the size of the resistor that drives the bases of the transistors, how many windings you have, the value of the tank capacitor and so forth will determine the amount of power it draws and how much heat it will generate in the workpiece. And you can pwm it as explained in the app note for continuous adjustability, or even use a sensor and feedback to control the temperature.

This is the correct link.

quote:

This is the correct link.

Thanks, that's a bit more than a self exciting oscillator.

The Royer is a self-exciting oscillator, Sinbad.

The circuit has a work cap, coil and two transistors. Look at the wiki or the app note linked to in a previous post and see for yourself. It's a dead simple, reliable circuit that works great for driving relatively low power stuff like a CCFL inverter or a small induction heater.

Don't get the low power drive circuit we're talking about here confused with the multi-kilwatt stuff Burnett does. He uses a different driver; but the "LCLR" he discusses, i.e. the topology of the choke and parallel tank applies regardless of power level. The theory is the same. And Burnett's discussion of LCLR is illuminating.

I guess you took one look at his driver and dismissed the whole subject. But the bridges he uses are a whole different animal, by necessity, at such high power levels. And the circuits much more complicated with phase-locked loops and such.

I have built several induction heaters for my injection lines and they work very well.
I have one on my 240 D and the other on the fuel gallery feed lines of my 97 powerstroke.
Both vehicles are setup around fattywagons designs and have looped returns and a aluminum budweiser bottle in the tank where the hot return fuel mixes with the pickup , a kind of looped return????

The frequency range is fairly low 6 to 10 Khz. The lower frequency results in higher current draw and hotter lines. By adjusting the frequency you control the amount of power ( heat) to the injection lines
They have a temperature sensor feedback on the injection line to stabilize the fuel temperature just before the injector.
I have the temp set at 200-220 F and it fluctuates about 10 F between full load and idle because of the cooling effect of the preheated ( HIH) fuel from the IP.
It is very nice because the heating works , by design, immediately, that means that the line is at temperature in about 45 seconds.
The current draw ranges between 10 and 16 Amp.
The electronics has a over temperature shut down

I thought about making kits $ 250 or assembled units $ 300 for other people interested but it seemed that when I made references to injection line induction heating nobody triggered.

If that has changed then let me know, I can post some pictures.

Later J

The timing of this thread is excellent kell. Thanks for coming forward on it.

I just finished an experiment this a.m. using age-old resistive heating by wrapping kapton tape around injector lines, and 29 AWG nichrome wrap around that - for about 2 feet wire length... grounded right to the injector line at the end. It works. Gets up to 100C easy and draw is about right at 1 Amp per line.

But in my heart I know and want induction!
More efficient and just way cooler. Er.. ah.. hotter!

Thanks for breaking ground on this. I will be doing it.

J could you post some pictures?
I'd like to pick up on this and give it a go.

Pretty sure I'd like to do better than cat-5 for temperature handling, what did you use to wrap?

Hi Scott,

I thought I had pictures but they are lost, will take some new ones in the morning.
There is no insulation besides the electrical insulation that the wire has. For the wire you can use multi-strand wire with teflon coating or other high temperature insulation. Teflon is very soft. The wire has to be multi-strand to get a lower resistance because of skin effect. Although with the 6-10Khz this works it should not be to much concern. Just wind it 15 turns around each injection line ( for my 4 cylinder)

Later J