Go Down

Topic: rectified AC phase control (Read 1 time) previous topic - next topic

aarg

It looks like your mosfet is shunting the supply. That doesn't seem right.
  ... with a transistor and a large sum of money to spend ...
Please don't PM me with technical questions. Post them in the forum.

Wawa

The IRF9z24N is rated for 55volt, the IRF630 is rated for 200volt.
The rectified pulses have peaks of 150*1.4= ~210volt (if I read your blurry diagram correctly).
50Amp?...
Leo..

wvmarle

Much better readable image! Inlined for convenience:


You'll need a logic level MOSFET, the IRL series rather than the IRF series. Otherwise it won't open fully at 5V gate.

That coil L2 is your load? For inductive loads you must add a flyback diode. Every time you switch off the power you get a large reverse voltage peak.

Considering the image of your desired output wave, a TRIAC can not work as you never reach zero.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

TomGeorge

Hi,
What is L1 for?

You might find this interesting;
https://www.homemade-circuits.com/how-to-use-triacs-for-inductive-loads/

What are you trying to make?

Tom.... :)
Everything runs on smoke, let the smoke out, it stops running....

dgueorguiev

Hi  Tom, What I try do do is that I have one vertical coil and second coil is at 45 degree .I need the result of specific interfere between the two coils. The vertical coil  has to be powered by variable AC from  50V to  75V . The second coil need positive pulses synchronized with  pics in firs coil from rectified AC 100v to 150V or DC. The mechanical part is not a problem.I need help with electrical part of the project. May be somebody will give me different idea how to achieve my desire power for this two coils

wvmarle

What is the peak voltage you expect, and peak current? That are two parameters you need to size your MOSFET.

Just looking at your proposed schematic again, you use an optocoupler to get the input voltage (100-200V or so) to the gate of the MOSFET. Bad idea: the maximum allowed Vgs of the IRL630 is 10V. That high voltage on the gate will kill it. Other MOSFETs will have higher Vgs(max) rating but its nowhere near the voltages you want to apply.

You could make a voltage divider with R1, R2 and a pull-down resistor on the gate (which is missing from your schematic) but that only works reliably when you you have a fixed high voltage. You're planning to regulate this.

You'll need some kind of fixed voltage supply to switch your MOSFET gate.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

Wawa

The vertical coil  has to be powered by variable AC from  50V to  75V .
So a variable transformer (variac).
Leo..

TomGeorge

Everything runs on smoke, let the smoke out, it stops running....

wvmarle

that waveform looks easy enough to produce on an Arduino output.
Set an interrupt to trigger on the zero detection.
That interrupt sets a timer interrupt to switch on the signal.
When the timer interrupt is triggered, set the port (direct port write, digitalWrite() may be too slow) and sets the timer again for 2 ms later.
When that second timer interrupt is triggered, switch off the signal, and wait for the next zero crossing.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

dgueorguiev

#24
Dec 08, 2017, 02:18 am Last Edit: Dec 08, 2017, 02:39 am by dgueorguiev
Thank you  to all of YOU for trying to help me.Dear "wvmarle" I do not know how to write the cod for Arduino. I try some programs written on ARDUINO playground, but no success for now. After reading all suggestion  I came to a new schematic  which I  will attach .  Could you  please check it and if you think it will need some changes to work, please give me your opinion.If you think the schematic will work  I will  really appreciate  if some of you write and send me a code to upload on my Arduino

wvmarle

Thank you  to all of YOU for trying to help me.Dear "wvmarle" I do not know how to write the cod for Arduino. I try some programs written on ARDUINO playground, but no success for now. After reading all suggestion  I came to a new schematic  which I  will attach .  Could you  please check it and if you think it will need some changes to work, please give me your opinion

At L2 there's a flyback diode, that's good - make sure it's a type that can handle the current. A common 1N4007 will do fine. But you should not add that resistor there, as it completely negates the purpose of the diode which is to provide an easy path for the reverse current to flow.

The 4N35 optocouplers: why is there an LED in series? Any purpose for that? Also they miss a current limiting resistor.

Why two of them? You just need a pull-down resistor there. But it's getting tricky with the extra voltage divider. That one has the wrong values, 10k & 8750 Ohm is not going to divide 150V into (140+10)V.

The MOSFET: the Irf9z24n is a p-channel type. You need an n-channel. Also maximum VDS is 55V while you have 150V.

Quote
.If you think the schematic will work  I will  really appreciate  if some of you write and send me a code to upload on my Arduino
Can do.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

wvmarle

#26
Dec 08, 2017, 07:27 am Last Edit: Dec 08, 2017, 07:31 am by wvmarle
Drawn out your schematic in KiCAD - using the actual symbols for the components where available.


Added the missing resistors for the optocouplers and the MOSFET.

Can't read the part number for the phase detection, so didn't add that.

The IRG4PC is some kind of transistor, I'm not familiar with this special kind, but it seems to be a DC part that you put in an AC role. It will break right away as the reverse voltage is just 18V according to the data sheet and you're going to apply 150V AC. I think you want to use a TRIAC there. Your phase diagram in #24 suggest this:
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

TomGeorge

#27
Dec 08, 2017, 11:16 am Last Edit: Dec 08, 2017, 11:19 am by TomGeorge
Hi,
@wvmarle good one, I think you got it all.

The zero crossing uses a H11AA1 opto.

The IRG4PC is an IGBT, but the Op has the driver for it completely wrong plus the IGBT has a recovery diode from C to E so it will fully conduct on the negative cycle.

IGBTs do not like AC current.

Tom.... :)

Everything runs on smoke, let the smoke out, it stops running....

wvmarle

Another thing, that H11AA1 opto. I just checked the spec sheet, so now I know how that works. It switches off for a moment every time there's a zero crossing, so that's easy enough to detect.

There is no need for a resistor in both lines, one of R6 or R7 can go.

Then the value. At 150V and total 300 Ohm you get 500 mA. That'll burn the LEDs very fast. A good forward current is 10 mA, at a drop of 1V over the LED (negligible on a total of 150V) that would require a resistor of 15k. So not 2x 150 Ohm.

Then we have to find proper values for the R4/R5 voltage divider, with it the R3 pull-down.

For that, we need to select a good MOSFET Q2, and a replacement for Q1 - a TRIAC. I don't recall any current ratings by OP for these parts, which is needed to make a selection. Fast switching is a requirement (for accurate timing - target <1 us), so gate capacitance becomes important (in combination with R3, R4 and R5).

Well, OP, I think you have your homework here. Looking forward to your input.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

wvmarle

#29
Dec 08, 2017, 02:43 pm Last Edit: Dec 08, 2017, 02:49 pm by wvmarle
Playing with this schematic again, another problem. The LED at the output of the H11AA1. It can not light up, ever (due to far too little current flowing through the 10k resistor), but together with the optocoupler it will prevent the voltage to drop low enough: about 1.8V for a typical red LED plus 0.4V for that phototransistor means the output voltage will vary between +2.2 and +5V. That +2.2 will not reliably register as a LOW.

I've replaced the IGBT for a TRIAC. That should at least work for switching the AC to the rectifier for L2.

Now I don't really understand how that TRIAC can switch on a DC supply, unless you want to switch only half the time. For switching the negative cycle it seems to need a negative pulse, and that's not provided here. So it won't work. When looking for Arduino/TRIAC circuits an optotriac is used. So all that brings me to this circuit diagram:


The zener is for protection of the MOSFET gate. The values as chosen for the voltage divider will allow about 1.5 mA of current to flow, resulting in 15V on the gate. The zener I set to 12V, may change that to 15V or so.

Just realise that I'm calculating at 150V for the DC which of course is wrong as the peak value is much higher, about 210V. However the moment the MOSFET has to switch, the voltage in the circuit is below peak, so we must keep that in mind as well. Maybe a bigger pull-down resistor is needed so the MOSFET can properly switch on when the wave reaches 100V or so (the current requirements are not like that but it's easy to allow for it now, hard to change after building the circuit).

Another question: what is the purpose of this TRIAC switch? It doesn't seem to me that it adds anything to the switching of that coil L2, which is switched fully by MOSFET Q1 already. I don't see any additional effect of it, except adding some serious complexity.

Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

Go Up