Hey Guys,

I'm interested to see what some experts think about my schematic before I get the circuit board made.

I have previously tested the photo-voltaic MOSFET drivers and the low-power ZCD circuit. They operate well, and give me really good dimming waveforms.

I've now added power monitoring using a DC-DC converter to isolate the power supply required for these ICs, and added opto-isolation. The power monitoring ICs output in pulses.

Is there any simplifications that could be done? Things that could be done better? Note I am REALLY out of space on my circuit board, so I'm limited to what I can add (unless a clever idea can remove some components).

Thanks.

 

It is more like a wiring diagram than a schematic. What area are you concerned with?

 

Not sure I understand the difference. A quick Google shows examples that makes me believe mine is a ‘schematic’, and not a ‘wiring diagram’.

But, I don’t really care what it’s called

I’m interested in

• The circuit for the power monitoring and the isolation/5V power. Any glaring issues? (As this is the new addition to my previously tested device)
• Anything seem dangerous? Have I isolated the HV side from the LV side enough? Too much/unnecessary?
• Any tips on simplifying any of it?
• Any helpful ideas at all?
• Any constructive comments?

Thanks.

 

There certainly is a whole lot more functionality to the circuit than "just a dual dimmer" would have. I did not evaluate or analyze what it is doing, but certainly it is a lot. A basic single dimmer has no more than five components, unless it is controlled by an analog voltage circuit or a digital computer.
So to even have an opinion about this circuit a description of the intended functionality will be needed.

I would not call the drawing a wiring diagram, but rather a circuit schematic. There are places that might be improved a small bit, but not worth doing a redraw. The signal/power flow is reasonable as well.
Overall it is a medium complexity circuit that needs to be explained because it is much more than just a dimmer.

 

A quick look at the DataSheet for HLW8012 shows 6 x 470k resistors connected to Neutral, your circuit shows only 5 that are connected to Neutral, Is this intentional ?
If there is enough room on the PCB, I would recommend each HLW8012 has its own set of 470k resistors rather than commoned.

 

Otherwise the Circuit seems OK.

 

A quick look at the DataSheet for HLW8012 shows 6 x 470k resistors connected to Neutral, your circuit shows only 5 that are connected to Neutral, Is this intentional ?
If there is enough room on the PCB, I would recommend each HLW8012 has its own set of 470k resistors rather than commoned.

Yes it is. Sonoff only use 5. When you do the math, I think 5 gives you a measurable range of ~960VAC. More (way more) than is required.
Why would you go separate? Considering Neutral is common between the two..?

 

Have you looked at power MOSFET dissipation during turn-on and turn-off times? Note that the data sheet specs those times as typical at Cl=1000pF. Worst case over production and temp could be considerably longer, and the effective MOSFET input capacitance will surely be different. Not a simple task, unfortunately, but you don't want the power transistors to let any magic smoke out.

 

I have chosen MOSFETs with Ciss of 555pF and low Rds, have limited total current to 1A (fuse is 2A and I have power monitoring), and am using it as a trailing-edge dimmer; the turn-on time (the slowest part due to the photovoltaic drivers) doesn't matter as it turns on at zero crossing. The turn-off time isn't too bad; ~30-40uS. So, all up, I think I should be safe from overheating.
During operation, these are barely warm to touch on the 10W LED drivers I have attached. I haven't tried a 1A/240W load yet however.

 

Take a look at the TI INA237 and friends. I2C and includes auto-measure of current, voltage, and power.