Later that year, my friend found a Pentium 4 PC hanging out of a dumpster. The machine was in relatively good shape, and was likely ditched because of a perceived obsolescence on the part of the owner. I suppose one man’s trash, is another man’s treasure, right? Well to make this poor PCs fate worse, the same day, my friends demonic cat was walking on the counter and knocked a pot of dirt right onto the open motherboard. We quickly cleaned off the board, and it was decided I would take this cursed machine. As I had a machine that had gone through hell, I chose to do what is often done to old machines, I made a Linux Server out of it.
As of that time, I had played with Linux, but never actually took the time to learn it. At the behest of my Linux loving coworker, Grace, I decided to give it a go. I made it a point that the server was going to be a hacker’s delight and made out of components thrown away, or hanging out of trash bins. Shortly after, my company threw out a logitech webcam..
I played around, and ultimately decided to run Zoneminder off of it. As I wanted the camera to be able to move, I started looking around for cheap ways of moving it, and found a blog from the user “zoomkat” that showcased a cheap way of driving servos with the “ezservo1″ chip from Kronos Robotics.
I ordered the then renamed “zpservo1″ chip, and an “EZ” Rs-232 module. I received the chip, wired everything up, and found that I couldn’t drive a servo. I emailed Kronos, and they ended up mailing a new chip. I came to learn later that the chip was based on a PIC16f627, and was the original was mistakenly not programmed. This peaked my interest, and I decided it was about time I started learning about microcontrollers.
I received the new chip, learned the protocol it used, and coded a routine for moving the servos in python. I mounted them, and liked the idea of controlling hardware from an I/O port, so I went back to the LPT LEDs I had breadboarded. I connected it to the server, and found a program called “lptest” I could use to drive individual pins, and hence drive individual lights.
I had bought some opto-isolated solid state relays and connected one to an IO pin. I measured 13mA draw from a multimeter, which was on the edge of the Printer Port spec, but I felt was safe enough. I connected a string of Christmas lights to the SSR, and was excited by the fact I had command line control over an AC socket.
I knew I wanted to do more automation, but wanted a better way to drive hardware. Enter Microcontrollers…
Hi.
Does the transformer induce some phase (delay) to the zero cross detection?
Juan, I’d imagine the Sine Wave looks the same after being stepped down, only with less power, so I don’t think a delay is being caused here, or the wave is being pushed out of phase.
The wave becomes rectified and drives a Transistor that in turn inverts the signal into a a square wave. i.e. when voltage is 0, the transistor passes no power, and a pull up drives the interrupt high. What I believe is happening is once the wave hits a low threshold, say 4v, for example, the transistor no longer functions, and the interrupt is triggered. I’d like to find a solution to this.
Hope that helps.
I know this has been idle for a few months but just to give a possible solution.
You have a processor, which allows you more flexibility that a transistor circuit monitoring a full wave rectifier.
If you use a single separate diode as a half wave rectifier,with some resistors you’ll get a square wave at the line frequency, that you could read with logic input.
Now if life were perfect,you’d be set, because all you would have to do is monitor that input for an edge. Rising or falling. That would be a zero crossing. It may even work just that simple. But possibly there is line noise on the AC that would give you several as the signal is near Vinputhigh. There is also a symmetry issue, where voltage crossing would be different on one phase from the other. Its the kind of thing you’d use a Schmidt trigger to read if you were using pure logic.
This is where the processor comes in. You can calculate how deep into the wave form the input crosses Vih, and when it will crass back again on the way down. From there you can calculate where the real zero crossing is.
With further elaboration you should be able to use that data make a phase locked loop in software that locks into the 60 Hz signal of the AC line. Using that as a time base, then you can determine your triac firing times.
On additional thought, you can really factor out the asymmetrical aspect of the half wave detector. Whatever Vih is, the peak of the AC powerline waveform is going to be half way between the rising edge and the falling edge. Likewise the nadir of the low part of the cycle will be half way between the falling edge and the next rising edge.
This PIC has a timer. So if (after deglitching) you take the times of the rising and falling edges, you should be able to calculate where the peaks are, and thereby the zero crossings.
I’d probably set it up so that a timer caused an interrupt when it rolled over, intending that to be a zero crossing. Reloading the timer count should take into account the error based on the the expected peak times. If the peak is not at 1/4 of a cycle, and the low at 3/4 of a cycle, calculate the error, and add that in when setting the timer count back 1/2 a cycle in the interrupt service routine.
You can even use an initialization process at the start of day to figure out how many timer ticks a cycle is, and not worry about an absolute scaled time base. just get an average for a few cycles.
Therefore with a bitof deglitching, you should be able to know
You can use another optocoupler on the transformer line in place of the transistor circuit instead. Look up the 4N26. The internal diode will half rectify the wave and the transistor will cause a more precise square wave you can base your timing from.
You must be Psychic. The last comment here was over a month ago, and I JUST had gotten done purchasing it’s sister chip, the H11AA1 when you commented. I was researching this earlier. The Renard PIC controller uses it, and I think I’ll adapt it also.