Frequency Counter Project: An amplifier fiasco

[Abstr7ct]

I've this problem that puts a big weight on my chest, which consequently affects my mental performance and therefore I feel that I wouldn't be good a this. Let me stop here before I turn this into a medical consultation topic.

I've decided to start working on a meaningful electronics project during my two weeks vacation from engineering college. I thought of building a frequency counter circuit. I don't have much experience in hands-on electronics, so the first idea would be to search for a ready-to-use build, understands how it works in detail while building the circuit in parallel. I found this detailed design in the link below with the schematic the link just below it.

http://www.apogeekits.com/counter_article.htm
http://www.apogeekits.com/PDF_Files/frequency_counter.pdf

I've been biased toward a mcu solution from the start as a unit for computing the frequency in software and driving a display to show the numbers. I can easily do that after thinking in software without help, but the problem, that I couldn't think about how to solve without a reference design, was the analog circuit that can take an input analog signal (sine, triangular, square...etc) and turn into a stable 5V TTL-level signal at the output which can then be fed into the mcu. Initially, I imagined a circuit which has a combination of resistors and capacitors for filtering, op-amps and possibly some transistors and maybe some digital digital sampling. However, I didn't know how start thinking putting all these combinations together let alone start building it, so I ran into that link above.

With that design, I looked at the frequency-dependent amplifier stage. What's in there? three transistors and a mind-confusing network of resistors and capacitors. aaaaah, where do I start from? I started reading trying to understand the explanation of this part and each time I stumble upon electronic terms, I get more confused and mentally fatigued thinking that this is just above my comprehension. Constant current biasing? BF244 was selected because of low range of drain current IDSS? variation in variation in transconductance? and the list of confusing terms go on and on?

I know what a FET is as an electronic component and so is the case for a BJT. The thing is, how to start thinking in biasing and configuring these transistors in such a way to give the desired output? are these standard circuits, i.e. just copy and paste, or do I need to follow a couple of datasheet functions and parameters and write the transfer function of each stage?

What you suggest me to do? go ahead and build the circuit and then start trying to understand how everything works by scoping the signals? sit down and read some pages on the internet explaining how to build transistor amplifiers and how to realize the same circuits using op-amps? I really need help here.

[Abstr7ct]

I realize that the inverting amplifier can be realized by an op-amp or two, and by this I can avoid the difficulties of building discrete transistor amplifiers. However, by looking at this interconnected circuit, I can't see how to do this realization myself. I think that there's a much simpler solution for turning an input analog signal into a fixed 5V TTL output. it would be a simple inverting comparator realized by an op-amp. I don't think it would be the best practical choice, but at least it would get something to work.

[madshaman]

Sounds like you need to take a step back and start building some background knowledge; maybe start with looking at some of the basic circuits one can build from opamps (oscillator, amplifier, buffer, comparator, integrator, etc. ).

Then maybe look a little into analog control loops.

Another circuit to start googling is a schmidt trigger: http://en.m.wikipedia.org/wiki/Schmidt_trigger

After that, start looking into the difference between FETs and BJTs and some simple circuits elements that can be built from both.

I recommend starting with opamp circuits because I feel they are much more approachable than jumping right into (non-trivial) transistor circuits.

Another approach is to download the service manuals for some older frequency counters and read through the theory of operation.

[KJDS]

There is an old adage that says that if you design an amplifier it will oscillate. If you build that one then it will.

As a start, just use a schmitt trigger IC and the first FET stage of that design and it will give you reasonable functionality. Once you've digested that then come back to a more complex analog circuit to give a wider range of input signals.

[Abstr7ct]

It looks like that what I should do. Although I've an idea about these circuits and their applications since we've covered them in college, I don't have any hands-on experience in regard to them. It was just theory and ideal cases with no enough time to build, scope and troubleshoot. I will have to take a big step back, as you said with some amplification from my side (big), and start building fundamental analog blocks realized by op-amps. As far as I know, any circuit or most of them, that can be built using transistors, can be realized by an op-amp. It looks like that I'm trying to run away from discrete transistors in the meantime and I admit that.

Anyway, before I do anything, I've to get some testing gear to aid me in practical learning. A power supply, an oscilloscope and a function generator. After getting these, I will start working on the basic analog blocks.

[codeboy2k]

The circuit description doesn't match the schematic at all.

The text describes a FET source follower input stage for high-impedance, followed by an emitter follower as a buffer to the next common collector stage, followed by a schmitt trigger stage which gives the 5 transistors and a voltage gain of 80. That's 50mV to 4V    But the PDF circuit has none of this, it only has a FET source follower , followed by a two transistor schmitt trigger.

As KJDS says, you can start with the FET input and a schmitt trigger IC, it will work reasonably well.

In modern times, you'd likely use a FET input op-amp followed by some additional op-amps to give gain over the needed bandwidth.

[Abstr7ct]

yeah I noticed that the description doesn't match with the schematic. Anyhow, I believe it would be better to take a step back into the basics. It doesn't feel good when I just connect components without knowing how they work in this or that configuration. it would better to take time and experiment with opamp-based amplifiers, filters, oscillators and get to know the non-ideal behavior and the terms surrounding all these circuits. I believe that this would be the natural progression although it wouldn't be as satisfying as building a complete meaningful system. An audio amplifier on the fly would be a nice circuit to build on the way! it seems that frustration has driven me mad lately, but I'm just overloading myself although I've been into practical electronics since just two years ago.

[PA0PBZ]

I don't think there is anything wrong with buiding this frequency counter without fully understanding the principles. You can always work on that later, and with a working model it will be much easier to measure and tweak it. You are probably going to get and use an oscilloscope and a function generator without fully understanding what is happening inside, and that should not stop you from having fun either.