Electronics > Beginners
Need somebody to talk to about AC concept
<< < (8/11) > >>
Zero999:
This is about your soldering iron controller project, isn't it?

Attached is a way to control the AC to the soldering iron, without an opto-coupler. It will have lower switching losses, than running the soldering iron off rectified DC and switching it with a MOSFET. The TRIAC needs to be the type which will fire, when the gate current is positive and anode is negative but most small TRIACs full-fill this criteria. If not, the TRIAC can be AC coupled to the MCU circuit and fired on with negative pulses, when the anode is negative, but cross that bridge, if you get to it.

Note that the neutral of the soldering iron, is different than the 0V reference of the 5V supply: it's on the phase node of the transformer, but this doesn't matter: all voltages are relative. The TRIAC is switching the phase conductor of the soldering iron, yet the 0V for the DC supply is the other side of the transformer's winding. What's ground/earth in one context, is completely different in another. You'll see this quite often with circuits run off a capacitive voltage dropper, running from the mains.

No part of the circuit, other than the soldering iron's metal case is connected to PE, the Protective Earth conductor, which is there to stop the soldering iron's tip from floating at high voltages, due to capacitive coupling through the transformer and the element insulation. The PE conductor carries no current and should never do so, other than a tiny leakage, through any RFI suppression components, but that's another matter.

I/O:1 is the zero crossing input. R4 will limit the current to a safe level and the MCU's internal ESD clamping diodes, will limit the voltage to a safe level. The MCU will see a square wave, at the mains frequency. To trigger the TRIAC, I/O:2 should go high, for a few ms, after I/O:1 has gone high, then high again for a few ms, after I/O:2 has gone low, as that will be when the mains has crossed over the zero point and gone negative. If this doesn't make sense, I can post a timing diagram.

The only downside to this circuit is, a half wave rectifier, results in DC flux in the transformer, which can cause saturation and overheating, if this is too high, however it won't be a problem if the transformer is overrated and the DC current taken by the MCU circuit is low, compared to the overall rating of the transformer. Suppose the iron uses 60W maximum: it's unlikely to be a problem the MCU circuit uses under 100mA and the transformer is rated to 75VA. R5 helps to alleviate this to some degree, by limiting the high current surges, taken from the transformer, every time the smoothing capacitor charges. It should be rated to >1W, as it will dissipate a reasonable amount of power.

tester43:
i did this project in the past in a very similar way to what you are showing, maybe some details were different like 7805 instead of lm317, maybe logic level at 3,3 not 5V etc. But I started to freak out with AC GND concept. Could not wrap my brain around the paradigm "there is nowhere in ac a point with CONSTANT potential that I could make my reference". In such case i did not know what is the meaning of GND symbol on the diagrams.

edit: and it's obvious that I used an optoisolator :) I bought them several years ago and they must be used ;D
pwlps:
I'm sorry if my comment will add confusion to your understanding, but actually the statement that any point of a circuit is as good as another to choose the ground is a bit misleading: there is also a "true" ground in the sense of earth ground.  This is important if there are stray electromagnetic fields generated by your circuit because the earth ground acts as an infinite conducting plane i.e. a conducting plane with infinite capacitance.  Consider a half-dipole antenna feeded by a coaxial line. You can connect the ground of the line (outer shield) to the earth ground but if you connect the antenna to the ground it will not radiate (then the shield of the coaxial line will radiate instead of the antenna but the radiation impedance and pattern will be different).  In RF circuits where connecting wires can radiate the choice of ground is more important
tester43:
it's all good. Actually that's the first time in my life that I am tempted to buy a better oscilloscope instead of new graphic card for my pc
CatalinaWOW:
This thread reminds me of the old story about blind men describing an elephant.  There is nothing I would point my finger to as wrong, but certainly many different ways of coming at the subject. 

Good for all of us to remember that there are a lot of ways to skin a cat, and that all of our understanding of the universe is a working approximation.  Some just have more or less detail, or emphasize on aspect over another.  Electricians will have one focus, RF guys another, the guys operating in the solid state/quantum world will add their own details, the hobby guys do just fine with some simple models.   Horses for courses.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod