Building a DRSSTC Pt. 8 - Interrupter

Hi everyone, and welcome back to my blog!

Today's post will be all about the interrupter for the DRSSTC. The interrupter is the part that effectively turns your Tesla coil on and off at a certain frequency by disabling the driver circuit multiple times per second. The interrupter is probably one of the easiest parts to design and build, which is why I left it for last, but is still a critical piece if you designed your coil to run in non-CW (continuous-wave) mode. "Continuous Wave" means that your Tesla coil is constantly on, which means the output is running at the TC's resonant frequency. These coils tend to be very quiet due to the fact that their operating frequency is above the range that the human ear can detect, and since it is not being switched off for part of the cycle, the current it draws is significantly higher (sometimes hundreds of amps higher) than if you ran it in interrupter mode. This can be great for effect, but it puts a lot of strain on your coil, capacitors, H-bridge, and driver circuitry. Since the IGBTs I will be using really aren't high-end, I would not dare run it in CW as they would surely be destroyed. The interrupter pulses the current in the H-bridge which significantly reduces the average current flowing through the circuit.

If you'll recall from Part 6, I am using two UCC27425 MOSFET driver chips to drive the GDT. These particular chips actually have an "enable" pin that allows you turn them on and off. This is perfect for my DRSSTC since I will be able to have my interrupter control it directly. If you build a DRSSTC and don't use driver chips that have enable pins, you can use some AND and NAND gates (for the inverting/non-inverting drivers) instead, and have one input of each gate connected to your interrupter. Just make sure that when the interrupter is off, NEITHER driver is active. Otherwise you will be drawing current when you shouldn't be, and it can cause some problems in your circuit.

Probably the simplest driver design I have seen to date is a basic 555 timer in astable mode. When designing an interrupter, there are several things you will need to take into consideration. The main thing you'll need to know is the maximum on-time that your IGBTs and capacitors can handle before the average current through them is too high. For this you will need to look at the datasheets, primarily the graphs. Figure out how your IGBTs, especially, are affected by temperature, current, and duty cycle. A very useful tool that I used was written by Matt Giordano, a.k.a. "Sigurthr" on the 4hv.org forums. I have attached a zip file containing his program so that you can give it a try as well.

Once you figure out the burst length, you can figure out the necessary duty cycle for your interrupter.

The following design is one Steve Ward used in his DRSSTC1, which I have been looking at for reference throughout this project. It simply uses a 555 timer and has two adjustments--one for on-time and one for off-time--which allows you to set the burst length yourself directly.

87763

I will be using this circuit for testing. However, I have been working on developing a multi-mode manual/MIDI controller using an Arduino Uno so that I can control my Tesla coil directly from my PC via the serial port. This is an ongoing sub-project, however, and is not yet ready for use on a DRSSTC. Therefore, I will stick with the basics for the time being and will not go into detail on musical Tesla coils (yet). I will mention, though, that a musical DRSSTC is quite easy to create from a regular DRSSTC, as all you need to do is develop an interrupter that turns the TC on and off at a variable frequency. Turning the coil on and off with the interrupter at 440 Hz, for example, will make it sound like you are playing an A4 music note. Since the Tesla coil itself operates at a frequency too high for humans to hear, they mainly hear the 440 Hz tone produced by the interrupter switching the coil on and off at that frequency. Once again, however, the tricky part is to prevent the duty cycle (the on-time, mainly) of any given note from keeping the coil on for too long, which could cause overheating and damage of your IGBTs or capacitors.

One thing I did not mention in Part #6 was that there should be a flip-flop circuit between the interrupter and the MOSFET drivers. This is to ensure what is referred to as "soft-switching" of the IGBTs. In other words, it prevents them from turning off mid-cycle when they're switching dozens, if not hundreds of amps. Instead, it waits to turn them off until the next zero-crossing point of the primary current. This is yet another way we make sure that the H-bridge and MMC will be safe, and won't see voltage and current levels that exceed their maximum ratings.

Now that we have looked at all the different parts of the DRSSTC schematic, we're almost ready to start the build!

Something to remember when designing a Solid State Tesla Coil, however, is that the designs are always changing. Sometimes you'll need to add something here, and remove something there, and you'll often have to recalculate when you do that. There really isn't a such thing as a "Final DRSSTC Design". There are always improvements that can, or will need to, be made. Therefore, try to take everything in this blog up to this point only as theory, and probably not the final design we will be using. For example, I just ordered some better capacitors for a MMC bank, which means my primary coil will have to change as well in order to match the resonant frequency of the secondary. I will also be adding over-current protection, and the J-K flip-flop I mentioned earlier for soft-switching. I will get into this a bit further in a future entry.

That's about it for now. Next time we'll start taking a look at the actual build.

Thanks for reading, and as always, comments, questions, or suggestions are appreciated!