Interfacing serial windows... NOT

I'm a bit pissed. For the last month I've been trying (more or less once a week) to interface my ad converter with my laptop. Hardware is almost ready, but... my software isn't. The reason for this is Windows.

The user friendly interface of windows has some peculiarities when accessing ports. By default the direct reading and writing of ports is disabled. Meaning you cannot do bitbanging (= easiest way to read ad-chip data).

My idea was to use RTS (ready-to-send) and DTR (data-terminal-ready) to emulate the chips ...oh... SLCK (= something something clock) and the other line (Power on?)... I have to forgive me, this is the first time I am using this kind of chips. Unfortunately windows does not allow changing these up and down easilly (i.e. writing bits to serial ports registers).

There is a solution, however install some crappy tweakytweakies written in c#, visual basic etc... The problem is that I know nothing about c# or any other graphical windows programming language. Ordinary c with python user interface would've been my choice. So changing back to linux and those lovely inp() and outp().

It seems that linux is going to be my primary operating system for this project. My touchscreen doesn't have linux drivers though... hoping wine will manage this.

Working (?) PWM circuit by operation amplifiers

Last time I concluded that playing with high voltage AC was a bit too dangerous. Therefore I decided to consentrate low voltage pulse width modulation. Below is a picture-link to a good PWM page.


Yes, I shamelessly copy-pasted other peoples work here. The above picture is a good example of a PWM modulation circuit. The first two operation amplifiers (the leftmost triangles) produce a saw-tooth wave which is thresholded by the bottom op-amp and switched on/off with the fourth. This circuit includes all the basics what you need to know about operation amplifiers. We have inverting and non-inverting amplification, negative, (sort of positive,) fast and delayed feedback.

I was building this to test whether I remember how to use op-amps, but I ran into a problem. I don't have an oscilloscope. I could measure my circuit at work, but I think I don't want to mix business and pleasure just yet. The next step would be then aquiring a cheap oscilloscope...

12V DC to 12V AC: 1st circuit

As mentioned I had an idea of converting DC to AC using MOSFETs. A quick draw (by dia) of the circuit would be somthing like this:


This is a typical pulse width modulator (PWM) circuit. The oscillating control voltage (green) directs the current through blue and red paths creating an AC current which can be modified to look like sine wave. However this is a bit dangerous circuit.

With voltages around 12V there is no danger, but with high voltages (e.g. standard 230V AC) this becomes deadly. The voltage from battery has to be 325V in order to get the sine to have 230V power rating. The negative side is also taken from the same battery which makes grounding difficult, or even impossible. The device has 325V ungrounded DC and 230V ungrounded AC ... nope, I am not going to do this. And if you see this kind of device near you I suggest a few steps back.

To avoid this problem with deadly electronics, I am turning my head towards DC currents below 120V and AC below 50V. These limits obey the regulation of Protective and Safety Extra Low Voltage (PELV and SELV). The difference is that former is (allowed to be) grounded and latter is not. And further along to regurations... if SELV AC current has a power value lower than 25V there does not have to be any cover for the circuits. This sounds good.

After learning this, the project changes its DC to AC goal to PELV compliant DC with PWM controlled power. The idea is the same as the above circuit. Removing the red connectors (and limiting battery voltage to 12V) makes this circuit safe.

12V DC to 12V AC: choosing components

As this project is an been waiting to happen for some time now, I kind a have a good plan where to start. First of all I don't like transformers, they are too heavy, expensive and old. I like power semiconductors.

The best choice for a high power current (and voltage) manipulating would be insulated-gate bipolar transistor a.k.a. IGBT. This is however a bit too new component. My local radioshack doesn't have it in stock. Second best would be thyristor, an older version of IGBT, but there is the problem how to turn thyristor off...

I have chosen my principle power component to be... MOSFET. The four best things I have learned about mosfets are:

• easy to use
• cheap
  
• great for parallel usage (for higher currents)
  
• power mosfets exist (for even higher currents)

Combining all above together, we have a modular system which power we can increase just by adding more identical pieces together. Controlling this multi-component system shouldn't be a problem in these highly computerized times we're living.

Next week I should have a principle diagram for a on/off mosfet switch... I think I have already designed it. Now I just have to find it.

We're open for business

The blog "System up" is finally opened. I am now more than determined to get my hobbies organized and online... and my hobbies are? Electronics, cars and computers. Not so Surprising for a 25-year-old tech student.

If you want to guess where this is all leading, I can give you a hint: I'd like to save the planet for the future generations. I am not saying that I want to build a low cost EV, but the first project will consentrate on how to convert 12V DC to 230V AC. And because I am not formally an electric engineer, I start by converting 12V DC to 12V AC.