Zero crossing power switch

[Rachie5272]

My table saw uses a single phase capacitor run AC motor, but has a woefully inadequate power switch.  Despite being double pole and rated for 15 A, the contacts have welded themselves shut on me twice.  Needless to say, it's a very bad thing when deadly spinning metal things can't be turned off.

I ameliorated the danger by building a triac circuit.  Now the switch contacts only see milliamps running through them, and the actual switching is done by a beefy heatsinked triac.  This works great, but there's one minor problem: the power can be switched on when the AC line is at maximum voltage.  This means the motor suddenly gets hit with a burst of current, instead of slowly ramping up over a quarter period.

How can I build a switch which only activates at a zero crossing?  Lots of optoisolators have zero crossing detectors built in, but for some reason triacs never do.  I thought about using said optoisolator, but I don't need isolation, and it means I would need to rig up a separate power supply.  Solid state relays seem to have the same problem.

Any idea what to do?

[SArepairman]

can you connect an zero crossing opto isolator to trigger a TRAIC?

What do you mean, you need more voltage rails doing this? I think its the common method of doing it.
http://www.nteinc.com/specs/3000to3099/pdf/nte3097.pdf

[Marco]

You can just buy zero voltage crossing SSRs (generally made with two SCRs rather than a Triac).

[Fank1]

You are getting way too complicated.
Get a 40 A rated relay and switch it.
Follow the KISS principle.

[dannyf]

How can I build a switch which only activates at a zero crossing?

Get a mechanical switch from HomeDepot and flip it on whenever you feel like it, including at zero crossings.

[HackedFridgeMagnet]

With the Zero crossing SSRs, they seem to measure zero crossing on the voltage.
What if you have a lagging load, ideally you would switch on the zero crossing on the current.
Are there any that do this?

[bronson]

Are you talking about switching on or switching off?  Sounds like the OP is worried about switching on, so there's no current.

To the OP: the original switch didn't have a zero crossing detector did it?  Typical AC motors couldn't care less where in the waveform you turn them on.  They have enough inductance that they'll force a ramp onto whatever you give them.

[HackedFridgeMagnet]

Are you talking about switching on or switching off?  Sounds like the OP is worried about switching on, so there's no current.

My table saw uses a single phase capacitor run AC motor, but has a woefully inadequate power switch.  Despite being double pole and rated for 15 A, the contacts have welded themselves shut on me twice.  Needless to say, it's a very bad thing when deadly spinning metal things can't be turned off.

I was talking about switching off, and I thought the OP was too.
Still my question remains, how much does switching off on a voltage zero crossing help on a severely leading lagging load? Can it get the the stage where it actually hinders?

[Marco]

With the Zero crossing SSRs, they seem to measure zero crossing on the voltage.
What if you have a lagging load, ideally you would switch on the zero crossing on the current.
Are there any that do this?

The zero voltage detection is irrelevant for switch off, the SCR's will switch off at zero current.

[HackedFridgeMagnet]

The zero voltage detection is irrelevant for switch off, the SCR's will switch off at zero current.

Ok that's what I was confused about.
I only looked at one data sheet it said.

2. Zero crossing functionary (VOX: MAX. 35V)

I took that to mean they switched on a zero crossing of the voltage. Obviously Zero crossing of the current makes much more sense.

[SArepairman]

I love the whole zero crossing thing... how to avoid playing Russian roulette with your designs.

[calexanian]

How about a small motor contactor? The are not very expensive and can be had with either 24VAC, 12VDC, or 120VAC coils. 30 amp is a pretty easy find. Also you can just use a couple poles of a 3 phase Fwd/Stop/Rev lever switch. Those are beefy.

something like this

http://www.ebay.com/itm/60-Amp-Drum-Switch-Forward-Off-Reverse-Motor-Control-Rain-Proof-Reversing-60A-D-/170710854649

[Rachie5272]

To clarify, I was talking about two different problems.  First was with the crappy old mechanical switch which could be shut off at peak voltage/current, resulting in welded contacts.  I fixed this with a triac, which has no contacts to arc, and inherently shuts off at a zero crossing.

The triac can still turn on at peak voltage though, which dims the lights and makes the motor jump.  The sawblade drive belt drive uses the weight of the motor as a tensioner, so the motor can actually jump quite a bit.  This puts extra stress on the belt, and makes the whole table jerk.

Sometimes I get lucky, and happen to switch it at a zero crossing.  The difference is significant.  No dimming lights, no jerking, and the motor spins up to full speed very quickly and smoothly.  Definitely something I want to be repeatable.

Another reason for not wanting another mechanical switch is the original one is designed to switch off just by bumping it.

How exactly would I wire up a optoisolator or solid state relay?  They require a voltage input to drive the internal LEDs, not just closing a switch.  I would have to use a separate low voltage power supply, which seems crazy.

[miceuz]

You always can make a simple nonisolated low volgate supply by dropping voltage via series capacitor, rectifying and regulating it (you might be able to skip this step).

You just have a mains rated switch to provide line voltage to this contraption that will ligth up the led in the optocoupler.

[mjkuwp]

Is there any chance the start-up of the motor is related to the angle it lands at?

otherwise- seems like zero-crossing SSR could work.  If I were doing it I would have a couple concerns. 

1. how reliable is the power-off of an SSR.  If it were me I would build in a relay to the circuit somehow for turn it off - or other safety rated mechanical switch.

2. How much heat is dropped through the SSR - but of course this is related to the size that you purchase.

I have seen others go to ebay/china for inexpensive packaged SSRs

Personally, I used bare component Sharp SSR, zero crossing and connected them to a heatsink.  However!  I was driving a resistive load and this is nothing at all like a motor as you know.

[Marco]

There are SSRs with AC control.

http://www.digikey.nl/product-detail/en/A2425/CC1014-ND/139499

[turbo!]

You do not want zero crossing for inductive load. That's the worst possible inrush. You want a peak turn on.  Better option is to just use a "motor starter" as commonly used on larger machines.

[TerminalJack505]

They make such a thing as a triac driver with zero-crossing (for example.)  The output is designed to drive a triac.  The input looks like an optoisolator.

The datasheets mention AC motor drive as one of the applications.  Whether or not you want to switch an inductive load at zero-crossing, I can't say.  I don't know why it would be a problem, personally.

[Rachie5272]

Another project of mine uses a big SSR to drive a bunch of solenoids, and so far it's been completely reliable with next to no heat dissipation.

Is a low voltage DC circuit the only good way to drive these things?  I know they can be small and simple unisolated things, but it seems rather strange to do it this way.

Why exactly would a motor need to be switched at peak voltage instead of zero?  I know the current in inductive loads is out of phase, but the current has to be zero when it's first starting.

[TerminalJack505]

A capacitive power supply might be the way to go if you use a zero-crossing TRIAC driver or an optocoupler.  They make TRIAC drivers and optocouplers with anti-parallel LEDs too so using one of them might help reduce component count.

You're right, though, there should be (and probably is) a component that does it all in one package.

[richard.cs]

You do not want zero crossing for inductive load. That's the worst possible inrush. You want a peak turn on.  Better option is to just use a "motor starter" as commonly used on larger machines.

This is correct, the zero-crossing switching is the worst possible thing to do for a motor. The current in the stator starts at zero and integrates voltage over a whole half-cycle, the flux follows, the iron saturates sometime after the voltage peak and there will be an enormous current spike as the motor inductance almost disappears. Switching on at the peak means that the current only increases over a quarter-cycle, the saturation flux is never reached and no current spike occurs.

I'm not aware of any SSRs that will do voltage-peak switching but maybe they exist or a capacitively derived supply for the control part as TerminalJack505 suggests could trick one into working that way. Or just use a chunkier mechanical relay or SSR.

[NiHaoMike]

Use a contactor with a 120V coil. Connect a hair dryer or something in series with the saw motor and use the contactor to bypass it. Connect the coil across the saw motor so that it pulls in once the motor is up to a good fraction of its rated speed.

[calexanian]

Sounds like what you really need is a soft starter.. They make motors, switches, and breakers happy. Particularly for woodworking tools that have no extra load at startup.

[turbo!]

Use a contactor with a 120V coil. Connect a hair dryer or something in series with the saw motor and use the contactor to bypass it. Connect the coil across the saw motor so that it pulls in once the motor is up to a good fraction of its rated speed.

Why that arbitrary voltage instead of other arbitrary voltages like 100, 200, 400, 415, 575 etc?

[NiHaoMike]

It should have the same rated voltage as the mains. In the US, it's usually 120V. On 240V mains, use a 240V contactor.