Pyrodigital - Advanced Understanding


Advanced Theory of Operation


How does the Pyrodigital system work?
The Pyrodigital (PD) system is essentially a computer. It contains a cpu, memory, input and output ports, most everything your standard PC would contain with a few small differences. Instead of a monitor, it has an LCD (Liquid Crystal Display) screen. Instead of a hard drive, it contains flash memory. Instead of a keyboard, it contains a keypad. These are some of the basic differences between your PC and the PD system. The main difference between a computerized firing system and your typical "manual" firing system is that the computer processes discrete information, and the "manual" system uses analog. The real world is all in analog, so in the end (at the field module), an analog signal must be sent through the electric match. Hugh? What?! Keep reading.

What do you mean by discrete and analog?
Analog = continuous. Discrete = Either on or off. Everything in the real world is continuous. There are an infinite number of different values between 0 and 1 in the real world. For example, 0.01, 0.02, or even smaller 0.001, 0.002, so on and so forth. You can keep chopping it down into smaller and smaller numbers, meaning there is a continuous flow from 0 to 1, and not just a sudden jump. However, in a computer system, the computer only understands "on" or "off". Which is represented in the binary number system as "0" = "off" and "1" = "on". This would constitute the sudden "jump" from 0 to 1 and is what we mean by "discrete".

So how does all this apply to the Pyrodigital system?
So now that we understand that a computer only deals with "1's" and "0's", we can say that the computer sends out streams of these values to the slave modules. Each of these values represents 1 bit of data. Eight of these "bits" are known as a byte. The computer itself is not sending an analog voltage to the electric match since that is not possible due to it being a discrete system. So how does the computer fire the electric match? The computer sends out the data, which in our case is the address of a particular shot to be fired. This data is then interpreted by the field module. If the cue to be fired is on that field module, a discrete in / analog out type device such as a transistor is triggered by the discrete components within the slave and will flow analog current out through the electric match. In other words, the computer is turning the cue "on" and "off". Each field module will activate a firing pulse of a duration long enough to ensure the electric match fires. This pulse duration is also dependent on the system design.

What exactly is Time Code?
Now that you understand what "data" is, time code is simply a stream of data (usually a byte worth), which represents a time value. In our case, such as for Pyrodigital FSK, it represents a sequential number, which the Field Controller (FC) interprets as a time. Each number in the PD FSK time code is unique, and therefore, can only represent one value of time. That is the whole reason behind being able to plug the time code into the FC at any point during the show, and the FC will be able to sync up to the current point in the soundtrack, without having had to be sync 'd up at the beginning. Makes sense? This is what separates the Pyrodigital FSK from the other forms of time code, such as SMPTE. So for a simple example (these numbers are made up and do not represent the actual PD values), the byte 00000001 (binary) would equal time 00001 (decimal), and byte 00000010 would equal 00002, so on and so forth. Each of these different decimal values would represent a unique time to the FC. Learn how to check the quality of the time code you are receiving with the pyrodigital system.

What exactly does FSK mean?
FSK is short for Frequency-Shift-Keying. It is a type of Frequency Modulation (FM). Remember, the soundtrack (we shall assume is playing from a CD) is sending the time code data (the 1's and 0's) to the FC. How do we send these bits of data to the FC through an audio system? Simple (figuratively speaking), we modulate an audio carrier with our information. A CD player and all associated audio equipment, filters all frequencies that are not within the audio spectrum (roughly 40Hz to 20kHz), hence the reason we need to modulate our information signal (the time code) onto a carrier frequency which falls in the audio spectrum. This also allows us to broadcast the time code as a "voice" over a simple hand held radio (yet another great feature of the PD system).

frequency shift keying - fskOk, so then how exactly does FSK work?
Let's say we have an audio carrier (pure sinusoid) of 10KHz. And we have data (our 1's and 0's) coming at 1200bps (bits per second). If we Frequency Modulate our 10KHz carrier with our data, we get the time code signal. To "Frequency Modulate" simply means to vary the frequency of the carrier signal according to our data signal. So, our 1's and 0's (on's and off's) represent only two different values or "changes" that we need to make to the carrier signal. So, we can say that we need our carrier signal to change back and forth between two different frequencies depending on which data value (a 1 or a 0) is currently present. See the three different graphs for clarification on FSK modulation. As you can see from the Modulated Signal, the frequency of the signal changes depending on the data value being a 0 or a 1. Any further explanation of this is above the scope of this tutorial.

So how do you debug a short somewhere in the system?
A short can occur in a variety of places. The most common is in one of the field modules, the XLR cable, or the XLR connector. A short can occur for a number of reasons. If your lucky, they are seen during continuity. In this case, you can take your time systematically tracing out the problem component. If it happens during the show, unless you have time code coming through (not a show fired with internal clock), it's not very practical to fix. With time code, as soon as you fix the short, you can immediately reacquire the current place in the soundtrack, and fire the remainder of the show in sync. You can see that with internal clock, it would take a massive amount of coordination between the operator and audio technician to re-sync a show half way through, after fixing a short.


Training Video!

Debug a Pyrodigital Short

So how do we actually find the short?
You must approach this systematically. If you setup your show with the same type of structure as shown on our pyrodigital setup page, then you will have a main splitter which the rest of the show branches out from. Let's assume we have 3 lines coming out of this splitter. As soon as the short occurs, you instruct the member of your crew you had waiting for such a problem to occur to begin pulling out one line at a time to see if the "short" indicator light on the FC goes out. If not, he plugs that line back in to the splitter, and removes the second. Let's now assume that the short light goes out when the second cable is pulled, so now you know the short is somewhere down that line. So you plug that cable back in and follow it to the next point. The next point could be a field module or a splitter depending on your setup. If it's a splitter, follow the same procedure as you did for the first one. If it's a slave, you will need to start by unplugging each field module, one a time, in order, checking each time to see if the short light goes out on the FC. When it does, you know that you have found the shorted field module (or cable) and can just leave that field module unplugged. Assuming your receiving time code, you can now re-start the FC in Auto Fire mode (making sure you reset the fuse) to continue the show from it's current point in the soundtrack. Short found, problem solved, now the show must go on!