Switching high AC currents in the mains?

[cupoftea]

Hi,
We went into an office and they had been using trailing edge dimmer units with incandescants.
Now they want us to replace them but for use with (literally!!!) whatever lamps the customer may decide to connect up.
(eg LED, CFL, whatever)
They cannot say what lamps the customer may connect up, cannot say how much input capacitance those
lamps may have.! But they tell us that sometimes the customer may severely overload the lighting circuit.!
They also tell us that the customer may simply short circuit the lighting circuit.!

The power level wanted for each dimmer circuit is 3kW.

Now..they took the spec to a consultancy and asked for "cheap as possible" solution.
I have that solution in front of me now.
Its just using back to back 650V TO247 FETs in the mains line.
They switch these on and off with opto's. That's pretty much it!

There is no snubber, no turn-off diversion, nothing to protect against instances where the circuit
may (due to noise) accidentally get switched ON as a leading edge dimmer at mains peak.
In fact, there's pretty well no protection against anything!
What do you say about this?

There is a "cheap as chips" overcurrent circuit involving a small thyristor whose gate-to-cathode
is across a 0.025R sense resistor, and the thyristor should pull the gate down in case of overcurrent.
One thyristor protector for each of the back-to-back FETs ..and one works for positive
half cycle, one works for negative half cycle.

The whole thing absolutely reaks of "cheap and cheerful and may the failure returns build high".

What do you say about these lack of protection levels? What do you say about not even knowing what

the lamps (loads) are going to be.?

 

[KlausST]

They cannot say what lamps the customer may connect up

This is nonsense. Some lamps may be dimmable, some not, some may even be destroyed by "dimmed input".

My recommendation: don´t do it.

Klaus

 

[cupoftea]

Thanks, if they arent dimmable , then the customer needs to buy dimmable lamps.....that wouldnt be our fault in this case.
If lamps are destroyed, again, thats the customers fault in this case.
What worries me, is damage to the dimming unit...because that would be "our fault".

Its trailing edge so ovbiously always starts from the zero cross of the mains (unless it tunrs it on mid cycle due to a noisy signal of course) so inrush "should" thus be limited.

 

[danadakk]

@OP, Lawyers get a hold of this in some convoluted case, by by your company.....
and you as designer, I can see you subpoenaed on the stand now.....

Fires, death, damage......this is a customer to avoid or design for failsafe, no
"in between" design methodology allowed.

 

[BradtheRad]

There's an undercurrent here. Someone is trying to force a business entity to take responsibility for anything that can happen with customers (new generation of office workers?) who are accustomed to lights coming on when they flip a switch. Aren't we ordinary human beings used to that?

For decades the success of mains AC has been the fact it's at a reliable volt level. The population understood they could buy incandescent bulbs of a variety of wattages. One year dimmers came along. Incandescents still worked but warnings went the dimmers: Customer caution: Dimmer may not work with fluorescent, 'energy-saver' bulbs, appliances, etc. These alerts are forgotten today as we've seen CFL, led...

Are the alerts still printed on the packages? Mains is still mains. We can't hope to dial brightness by 'cheap and simple'. Aren't incandescents on the way out (by decree of certain governments)? Unless a universal dimmer design is developed, unmodified mains is dependable.

 

[D.A.(Tony)Stewart]

That’s how many SSR’s are made now with 5 mohm FETs and RF-C coupled PWM from rectified AC yet still have OptoSCRs shown on the case.