AC Dimmer

Post here first, or if you can't find a relevant section!
pokemon99
Posts: 50
Joined: Wed Jun 07, 2017 9:09 am

Re: AC Dimmer

Post by pokemon99 » Mon Dec 11, 2017 8:24 pm

No. You do not understand correctly, the encoder manages the menu. Everything is protected there.

User avatar
RogerClark
Posts: 7691
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: AC Dimmer

Post by RogerClark » Mon Dec 11, 2017 8:26 pm

Ok.

Sounds possible, but I would not build something that was unisolated for a personal project.

User avatar
RogerClark
Posts: 7691
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: AC Dimmer

Post by RogerClark » Mon Dec 11, 2017 8:28 pm

Pito wrote:
Mon Dec 11, 2017 5:15 pm
Afaik the MOC304(6,8)x devices include the zero switching detector, so you must not mess with a special circuit..

I can’t find a MOC3046 or MOC3048 listed on mouser or Digikey
Are you sure those are the correct part numbers

Edit. There is an opto with 2 back to back LEDs inside which will conduct in either direction, which removes the need for the external bridge rectifier, but it has the same issues with high wattage resistors, and is a lot more expensive than using a single LED opto and a cheap bridge ( or 4 diodes)
The bridge does not need to be high voltage as the resistors drop all the voltage.

BTW.
probably using 4 resistors is better than 2, because most resistors that are readily available do not guarantee to 240V RMS, peak is 350V and spikes can be even higher.

So using 4 resistors gives a better operating margin.

Also be aware that you should not use normal strip-board to build the zero crossing detector, as it may “track”across the insulation
Or if you do, then feed the mains to it via the resistors, so the voltage drop is done before it gets to the board

pokemon99
Posts: 50
Joined: Wed Jun 07, 2017 9:09 am

Re: AC Dimmer

Post by pokemon99 » Mon Dec 11, 2017 8:36 pm

By reason of the caps, you are right. Here I plan to order
Attachments
746629387417292725-1.jpg
746629387417292725-1.jpg (123.76 KiB) Viewed 152 times

victor_pv
Posts: 1866
Joined: Mon Apr 27, 2015 12:12 pm

Re: AC Dimmer

Post by victor_pv » Mon Dec 11, 2017 8:43 pm

MOC3043 for example has zero cross, is the same as MOC3023 + Zero cross.

Pito may have misstyped 6/8

@Pokemon:

This is my zero crossging ISR:
https://github.com/victorpv/reflowOvenC ... w.ino#L155

But as you see here, I am simulating it with a timer to use it without zero cross detection:
https://github.com/victorpv/reflowOvenC ... #L223-L228

The code is a fork from someone else, I only wrote the part where I fake the zerocrossing.

You could use a second timer, or do like I did and call it from the ISR for another timer every N cycles.

I use an SSR with zero cross detection for the highest load (15 amps) and a triac with no zero cross detection for the smallest (a fan).
As for the heater switchint on an off for long periods is ok (my range is 0 to 100, so slightly shorter than 120 ac half cycles I get in the US with 60 hz main, so the switching period is a bit under 1 second, and duty from 0 to 100%), and the fan is either full on or full off, works fine for me.
Obviously this would not work for lighting, because it would flicker, since the on/off wouldn't be fully synchronized with the AC waves (on top on mine period is almost 1 second...)

User avatar
RogerClark
Posts: 7691
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: AC Dimmer

Post by RogerClark » Mon Dec 11, 2017 8:49 pm

Victor

I looked at the data sheet for the moc3043 but the Zero crossing part if the chip something to do with only turning on the triac at zero crossing and not part of the detection as far as I can see

I think it’s for a different use, where you have a highly inductive load and don’t want to turn on half way though a cycle ( but I could be wrong)

pokemon99
Posts: 50
Joined: Wed Jun 07, 2017 9:09 am

Re: AC Dimmer

Post by pokemon99 » Mon Dec 11, 2017 8:51 pm

Thank you. There will be free time I will try.

victor_pv
Posts: 1866
Joined: Mon Apr 27, 2015 12:12 pm

Re: AC Dimmer

Post by victor_pv » Mon Dec 11, 2017 11:42 pm

RogerClark wrote:
Mon Dec 11, 2017 8:49 pm
Victor

I looked at the data sheet for the moc3043 but the Zero crossing part if the chip something to do with only turning on the triac at zero crossing and not part of the detection as far as I can see

I think it’s for a different use, where you have a highly inductive load and don’t want to turn on half way though a cycle ( but I could be wrong)
That's right Roger, it's not for the MCU to detect zero cross, but to ensure the load doesn't turn on mid cycle.
So it's helpful if you want regulate the load to be on for a number of cycles, then off for a number of cycles, like with a heater.
It doesn't help with loads like a light bulb in which you want to regulate by using a turning on during part of each cycle to avoid flicker.

I was just pointing out that the MOC3043 is one that Pito was referring to with integrated zero cross.

User avatar
RogerClark
Posts: 7691
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: AC Dimmer

Post by RogerClark » Tue Dec 12, 2017 12:41 am

Thanks Victor

User avatar
RogerClark
Posts: 7691
Joined: Mon Apr 27, 2015 10:36 am
Location: Melbourne, Australia
Contact:

Re: AC Dimmer

Post by RogerClark » Tue Dec 12, 2017 5:02 am

Not the most compact of boards, but this was my prototype power controller board

IMG_20171212_153930.jpg
IMG_20171212_153930.jpg (86.63 KiB) Viewed 119 times
It does not include the BluePill, as that attached via the pin header on the far right

The zero crossing detector is along the bottom, the power FET stuff is on the left of the board. There is a Hall effect current sensor almost in the middle of the board.

I made a mistake with the PCB footprint for the HALL current sensor, I forgot to update it, so its much bigger than the IC.
Luckily I was able to spread the SMD pins apart on the IC and make them fit the larger pads.

Once it was working, I planned to do a SMD version, but the large resistors and also the distance between SMD pads even on 1206 resistors are a problem, with potential tracking.

As you can see I was cautious about isolating the HT and LT sides of the board, so there are grooves cut to enhance isolation, and also there isnt a ground plane.

The PCB is designed to take connectors with 2 pins per post, but I didnt need to fit those for testing, as I was only going to test 2kW not 4 or 6Kw

Post Reply