Optimize this circuit - power supply- help needed

Thread Starter

Tomy983

Joined Nov 22, 2017
9
Hi, I found this instructable https://www.instructables.com/id/Multichannel-Wireless-Light-Dimmer/ with a circuit for a trailing edge dimmer. I would like to replicate it but instead of two channels I would like to go for 12.

I understood all the working principle, but since I will have 12 channels I would like to optimize a little.
This is the original schematic. https://cdn.instructables.com/ORIG/FP6/5RF8/J2MAORX6/FP65RF8J2MAORX6.pdf

In the instructable description the author states that each channel must have its own 12V supply to trigger the mosfets.
I can't understand why.
Is there a way to have a single 12 or something volts (the mosfets I plan to use hame max Vgs of 25V) supply for all of them?
 

dl324

Joined Mar 30, 2015
16,846
Welcome to AAC!

Unfortunately, your post violates the AAC terms of service. The PDF contains a transformerless circuit:
upload_2017-11-22_8-21-51.png
 

Thread Starter

Tomy983

Joined Nov 22, 2017
9
I'm sorry, I haven't read the terms of service. I jave now looked for them but I couldn't find them. Could you please point me to it?

About the circuit.. It is not my circuit and I would like to improve it in efficiency and part count, but of course it can also be improved on safety. If I would add an insolation transformer to it would that then make it suitable for the terms of service? Or, could you suggest any other improvements safety wise before helping me further?
Thanks
 

dl324

Joined Mar 30, 2015
16,846
I'm sorry, I haven't read the terms of service. I jave now looked for them but I couldn't find them. Could you please point me to it?
https://www.allaboutcircuits.com/user-agreement/
About the circuit.. It is not my circuit and I would like to improve it in efficiency and part count, but of course it can also be improved on safety. If I would add an insolation transformer to it would that then make it suitable for the terms of service? Or, could you suggest any other improvements safety wise before helping me further?
The user agreement specifically mentions transformerless power supplies, but I believe the intent is to protect people from being exposed to high voltages. So, even if you added an isolation transformer, there would still be high voltages present.
 

Thread Starter

Tomy983

Joined Nov 22, 2017
9
Well, I did put a transformer and modified the supply voltage levels. If you could have a look..
What I would like to do is combine the power supply for the mosfet triggering to only one... In this schematic only two output are represented, my intention is to put 12..
 

Attachments

dl324

Joined Mar 30, 2015
16,846
What I would like to do is combine the power supply for the mosfet triggering to only one...
They can be combined, but you need to adjust the zener current. The original design is operating them at about 12mA and each optoisolator draws 1.2mA. This is the minimum ratio normally used to get decent voltage regulation.

If you add more loads to the zener, you need to increase the current to keep the zener past the knee voltage.

If you find that the zener can't handle the current, you can use an NPN transistor as a current amplifier.
 
Last edited:

Thread Starter

Tomy983

Joined Nov 22, 2017
9
They can be combined, but you need to adjust the zener current. The original design is operating them at about 12mA and each optoisolator draws 1.2mA. This is the minimum ratio normally used to get decent voltage regulation.

If you add more loads to the zener, you need to increase the current to keep the zener past the knee voltage.

If you find that the zener can't handle the current, you can use an NPN transistor as a current amplifier.
Thank you, dl324.

What I was actually thinking, if possible, is to use another power brick to supply the 12V rail needed to trigger the mosfets. This will also eliminate the transformerless power supplies rendering everything safer and possibly more reliable.
The mosfet I will be using are the STB15NM60ND
Do you think it will work as I draw it?
I saw that one of the safety concerns is the rectified mains for the zero crossing detection section. A small transformer could take care of that but I have read somewhere that it would be less energy efficient during stand-by. I don't like to have those large resistors dissipating heat either.
I have now modified the zero sense part to use an AC input optoisolator, removing the bridge rectifier. Is there anything better?
 

AnalogKid

Joined Aug 1, 2013
10,987
The circuit
The 12 V -Vout, which you show tied to the GND symbol, must be kept isolated from the +3V3 -Vout, which you show tied to the same GND symbol. You need to create a separate ground symbol and net name for anything referenced to the power line in any way. As shown, the 3.3V circuits are *not* isolated from the power line.

ak
 

Thread Starter

Tomy983

Joined Nov 22, 2017
9
The 12 V -Vout, which you show tied to the GND symbol, must be kept isolated from the +3V3 -Vout, which you show tied to the same GND symbol. You need to create a separate ground symbol and net name for anything referenced to the power line in any way. As shown, the 3.3V circuits are *not* isolated from the power line.

ak
I see. Thank you, I overlooked that particular. If you have any other suggestions or critics, please, I'm all hears!
Once the circuit will be ok I will lay down the PCB and post it here.
 

Thread Starter

Tomy983

Joined Nov 22, 2017
9
There is something wrong. The transistor I checked them again, they look ok to me.
Trying to simulate the circuit with LtSpice and it does not work as intended. it looks like it only works with separate 12V supplies...
 

Thread Starter

Tomy983

Joined Nov 22, 2017
9
Please update the schematic with a unique reference designator for each component.

ak
I updated the schematic. I am also trying to simulate the circuit with LtSpice but with no luck.
It works OK during the positive half cycle of the AC supply (which I set to 30V for now).
During the negative half, if one of the mosfet pair is switched, the other pair will follow. I tried different things, but I guess I really don't know what to look out for. I see, in the simulation, voltage and current spikes on the turned off mosfet and I don't know how to get rid of them.
I tried a couple of circuits, one I found here, but without luck. I tried to leave one of the two common sources floating, Putting diodes and capacitors.. I don't have any more clues and certainly because I don't understand what is going on..
 

Attachments

Top