Snubber needed for triac driving heating element?

[Pimpom]

On 4/20/2020 8:56 PM, Ricky C wrote:

On Monday, April 20, 2020 at 6:11:41 AM UTC-4, Pimpom wrote:
On 4/20/2020 11:05 AM, Ricky C wrote:
On Sunday, April 19, 2020 at 10:41:32 AM UTC-4, Pimpom wrote:
This is about a triac switching a heater on and off. The heating
element is mostly resistive with a small incidental inductive
component due to the coiled structure. The switching is a simple
on/off operation without phase control. Power is about 2kW. The
interval between switchings range from a few minutes to an hour.

Is a snubber needed or desirable for this? Please consider the
case for random firing and for zero-crossing.

What type of device is controlling the on/off signal?

A simple digital timer switching an MOC3023 triac driver,
probably a 4060 rather than a uC for such a simple job. The triac
is a BTA41.

So the timer is a black box with some sort of output? I guess no option to add the zero cross detection in that.


It probably wouldn't be too hard to add zero crossing detection. But why bother?

Exactly. I have random couplers in stock but not zero-crossing ones.

You could just buy a contactor (relay). Simple and effective. They even have 240 volt relays with 12 volt drive. Maybe even 5 volt drive.


At 2 kW I take it this is a 240 volt heater? I'm guessing a UK tea kettle auto turn on for morning
brew?

A convection (not microwave) oven for my daughter,

So 240 volts? Here in the US you can get about 1.4 kW from a 15 amp 120 volt outlet or 19 kW from a 20 amp outlet. Most kitchen stuff is 240 volts. Are you US or UK?


a psychologist
who sometimes dabbles in baking and is now really into it during
the lockdown. The clockwork timer on her oven is behaving
erratically. Indian 230V.

Why a timer??? Every oven I've ever owned has had a timer on it (seems obligatory somehow) but I have not once in 50 years used it. Can't she just set a kitchen timer and turn it off herself?

Where does she live? I'd like to go over for fresh baked bread once this is over. Or even just cookies. Mmmmm...

I've run out of not food exactly, but variety. So I should be going out today to shop. I am so sick of eating my own cooking. I believe I can get some carry out to bring home. I need to find some good restaurants near the store.

I'll respond to the points you made here instead of interleaving
them which I often find inconvenient to read.

Yes, it's essentially a blackbox with power in, timed power out.
That doesn't preclude zero-crossing detection but it's not really
necessary.

Re getting a suitable relay/contactor: ordering parts is not easy
where I live even in normal times.

US or UK? Neither. I'm in one of the most remote corners of India
which has little in common with the rest of the country. We use
230V/50Hz mains.

So you never use the timer on your oven? No wonder you're sick of
your own cooking.

Where does my daughter live? You're out of luck - see above .
So far she's produced cookies, pizza, cakes, doughnuts, ice cream
and some experimental recipes without names. Yesterday it was
roast chicken. Yummm. My wife's a good cook too, especially with
local dishes.

BTW, my family - two engineers, one architect, the daughter, one
daughter-in-law and one granddaughter - all live in the same
house. This is not at all unusual in our society. The married
ones will eventually move out but there's no hurry.

 

[Ricky C]

On Monday, April 20, 2020 at 1:09:47 PM UTC-4, Pimpom wrote:

On 4/20/2020 8:56 PM, Ricky C wrote:
On Monday, April 20, 2020 at 6:11:41 AM UTC-4, Pimpom wrote:
On 4/20/2020 11:05 AM, Ricky C wrote:
On Sunday, April 19, 2020 at 10:41:32 AM UTC-4, Pimpom wrote:
This is about a triac switching a heater on and off. The heating
element is mostly resistive with a small incidental inductive
component due to the coiled structure. The switching is a simple
on/off operation without phase control. Power is about 2kW. The
interval between switchings range from a few minutes to an hour.

Is a snubber needed or desirable for this? Please consider the
case for random firing and for zero-crossing.

What type of device is controlling the on/off signal?

A simple digital timer switching an MOC3023 triac driver,
probably a 4060 rather than a uC for such a simple job. The triac
is a BTA41.

So the timer is a black box with some sort of output? I guess no option to add the zero cross detection in that.


It probably wouldn't be too hard to add zero crossing detection. But why bother?

Exactly. I have random couplers in stock but not zero-crossing ones.

You could just buy a contactor (relay). Simple and effective. They even have 240 volt relays with 12 volt drive. Maybe even 5 volt drive.


At 2 kW I take it this is a 240 volt heater? I'm guessing a UK tea kettle auto turn on for morning
brew?

A convection (not microwave) oven for my daughter,

So 240 volts? Here in the US you can get about 1.4 kW from a 15 amp 120 volt outlet or 19 kW from a 20 amp outlet. Most kitchen stuff is 240 volts. Are you US or UK?


a psychologist
who sometimes dabbles in baking and is now really into it during
the lockdown. The clockwork timer on her oven is behaving
erratically. Indian 230V.

Why a timer??? Every oven I've ever owned has had a timer on it (seems obligatory somehow) but I have not once in 50 years used it. Can't she just set a kitchen timer and turn it off herself?

Where does she live? I'd like to go over for fresh baked bread once this is over. Or even just cookies. Mmmmm...

I've run out of not food exactly, but variety. So I should be going out today to shop. I am so sick of eating my own cooking. I believe I can get some carry out to bring home. I need to find some good restaurants near the store.


I'll respond to the points you made here instead of interleaving
them which I often find inconvenient to read.

Yes, it's essentially a blackbox with power in, timed power out.
That doesn't preclude zero-crossing detection but it's not really
necessary.

Re getting a suitable relay/contactor: ordering parts is not easy
where I live even in normal times.

US or UK? Neither. I'm in one of the most remote corners of India
which has little in common with the rest of the country. We use
230V/50Hz mains.

So you never use the timer on your oven? No wonder you're sick of
your own cooking.

Where does my daughter live? You're out of luck - see above .
So far she's produced cookies, pizza, cakes, doughnuts, ice cream
and some experimental recipes without names. Yesterday it was
roast chicken. Yummm. My wife's a good cook too, especially with
local dishes.

BTW, my family - two engineers, one architect, the daughter, one
daughter-in-law and one granddaughter - all live in the same
house. This is not at all unusual in our society. The married
ones will eventually move out but there's no hurry.

Sounds great! I only said UK because I was playing the odds, lots of limeys here.

I wish I was in a family setting. I spend half a week here alone with five cats and half a week in Maryland with friends and no cats. Not sure which one is preferred, but too much of either is not good.

I can cook ok, but I haven't been cooking for months. I tend to get depressed in the winter and my energy levels drop through the floor. Maybe as the weather warms up I'll be more motivated. The longer days are already helping. Still not cooking yet though.

BTW, I've got a great shrimp recipe for the barbecue grill. It's a bit of a pain to mix the ingredients, but well worth it.

I don't have a timer on my charcoal grill either.

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

 

[Phil Allison]

Chris Jones wrote:

-------------------

Do not trust the TRIAC to stay off where it performs a safety function,
though. A mains spike during the night might melt the TRIAC into a short
circuit,

** Not very likely.

Triacs *self protect*when over voltaged by turning fully on.

Supply spikes simply trigger them on and they will commutate off at the next zero crossing.

I think SCRs do the same.

..... Phil

 

[Chris Jones]

On 21/04/2020 01:34, Ricky C wrote:

On Monday, April 20, 2020 at 9:52:40 AM UTC-4, Chris Jones wrote:
On 20/04/2020 20:11, Pimpom wrote:
On 4/20/2020 11:05 AM, Ricky C wrote:
On Sunday, April 19, 2020 at 10:41:32 AM UTC-4, Pimpom wrote:
This is about a triac switching a heater on and off. The heating
element is mostly resistive with a small incidental inductive
component due to the coiled structure. The switching is a simple
on/off operation without phase control. Power is about 2kW. The
interval between switchings range from a few minutes to an hour.

Is a snubber needed or desirable for this? Please consider the
case for random firing and for zero-crossing.

What type of device is controlling the on/off signal?

A simple digital timer switching an MOC3023 triac driver, probably a
4060 rather than a uC for such a simple job. The triac is a BTA41.

It probably wouldn't be too hard to add zero crossing detection.  But
why bother?

Exactly. I have random couplers in stock but not zero-crossing ones.

At 2 kW I take it this is a 240 volt heater?  I'm guessing a UK tea
kettle auto turn on for morning brew?

A convection (not microwave) oven for my daughter, a psychologist who
sometimes dabbles in baking and is now really into it during the
lockdown. The clockwork timer on her oven is behaving erratically.
Indian 230V.

Do not trust the TRIAC to stay off where it performs a safety function,
though. A mains spike during the night might melt the TRIAC into a short
circuit, and you could have the same experience as these people, albeit
at a more reasonable price:


https://www.theverge.com/2019/8/14/20802774/june-smart-oven-remote-preheat-update-user-error

Strange. “It’s a really wonderful feature to be able to remotely preheat your oven" Really? WTF?


Whilst relay contacts can weld together and stay on when they shouldn't,
still, I would have more trust in a relay than a TRIAC to stay off once
the user (or a supervisory circuit) has confirmed that it is already off.

Even if a triac were to trigger on from a power line spike, it doesn't stay on. It turns off at the next zero crossing. No? It would have to be a scramble in the controlling circuit to make it come on for more than an instant which is what the Verge article is about. That can happen even with a relay.

I agree, if the triac is not damaged. If somehow it gets melted, it will
quite likely stay on. I don't know the exact mechanism by which they get
melted, but I believe that it does happen, at least in consumer-grade
light dimmers.

 

[Chris Jones]

On 21/04/2020 09:14, Phil Allison wrote:

Chris Jones wrote:

-------------------


Do not trust the TRIAC to stay off where it performs a safety function,
though. A mains spike during the night might melt the TRIAC into a short
circuit,


** Not very likely.

Triacs *self protect*when over voltaged by turning fully on.

Supply spikes simply trigger them on and they will commutate off at the next zero crossing.

I think SCRs do the same.

.... Phil

I have seen numerous domestic light dimmers fail in the fully-on
condition. I suspect the failures could have resulted from huge currents
due to lightning, though arcing inside tungsten filament light bulbs as
they burn out is also a possibility. As you would be well aware, after
the silicon melts, it tends to conduct, regardless of the original
function of the device. If you have a different explanation for these
failures I would be interested to hear it. I never bothered to dissect
the failed units.

 

On Tuesday, 21 April 2020 15:26:50 UTC+1, Chris Jones wrote:

On 21/04/2020 01:34, Ricky C wrote:
On Monday, April 20, 2020 at 9:52:40 AM UTC-4, Chris Jones wrote:
On 20/04/2020 20:11, Pimpom wrote:
On 4/20/2020 11:05 AM, Ricky C wrote:
On Sunday, April 19, 2020 at 10:41:32 AM UTC-4, Pimpom wrote:
This is about a triac switching a heater on and off. The heating
element is mostly resistive with a small incidental inductive
component due to the coiled structure. The switching is a simple
on/off operation without phase control. Power is about 2kW. The
interval between switchings range from a few minutes to an hour.

Is a snubber needed or desirable for this? Please consider the
case for random firing and for zero-crossing.

What type of device is controlling the on/off signal?

A simple digital timer switching an MOC3023 triac driver, probably a
4060 rather than a uC for such a simple job. The triac is a BTA41.

It probably wouldn't be too hard to add zero crossing detection.  But
why bother?

Exactly. I have random couplers in stock but not zero-crossing ones.

At 2 kW I take it this is a 240 volt heater?  I'm guessing a UK tea
kettle auto turn on for morning brew?

A convection (not microwave) oven for my daughter, a psychologist who
sometimes dabbles in baking and is now really into it during the
lockdown. The clockwork timer on her oven is behaving erratically.
Indian 230V.

Do not trust the TRIAC to stay off where it performs a safety function,
though. A mains spike during the night might melt the TRIAC into a short
circuit, and you could have the same experience as these people, albeit
at a more reasonable price:


https://www.theverge.com/2019/8/14/20802774/june-smart-oven-remote-preheat-update-user-error

Strange. “It’s a really wonderful feature to be able to remotely preheat your oven" Really? WTF?


Whilst relay contacts can weld together and stay on when they shouldn't,
still, I would have more trust in a relay than a TRIAC to stay off once
the user (or a supervisory circuit) has confirmed that it is already off.

Even if a triac were to trigger on from a power line spike, it doesn't stay on. It turns off at the next zero crossing. No? It would have to be a scramble in the controlling circuit to make it come on for more than an instant which is what the Verge article is about. That can happen even with a relay.


I agree, if the triac is not damaged. If somehow it gets melted, it will
quite likely stay on. I don't know the exact mechanism by which they get
melted, but I believe that it does happen, at least in consumer-grade
light dimmers.

Yes, I have seen it happen a couple of times with touch-controlled
dimmers on table lamps when the lamp failed and there was an internal
arc in the bulb. The triac became a short circuit (and the 32A circuit
breaker on the ring final circuit tripped).

John

 

[Phil Allison]

Chris Jones wrote:

-----------------

Do not trust the TRIAC to stay off where it performs a safety function,
though. A mains spike during the night might melt the TRIAC into a short
circuit,


** Not very likely.

Triacs *self protect*when over voltaged by turning fully on.

Supply spikes simply trigger them on and they will commutate off at the next zero crossing.

I think SCRs do the same.




I have seen numerous domestic light dimmers fail in the fully-on
condition.

** The triacs are often puny and/or have no heatsinking.

Uses can overload them with a few 100W bulbs.

But they do not fail short due to supply spikes.

They trigger on due to limited dv/dt capacity if the spike is fast or sheer voltage.



...... Phil

 

[Phil Hobbs]

On 2020-04-21 17:01, Phil Allison wrote:

Chris Jones wrote:

-----------------


Do not trust the TRIAC to stay off where it performs a safety function,
though. A mains spike during the night might melt the TRIAC into a short
circuit,


** Not very likely.

Triacs *self protect*when over voltaged by turning fully on.

Supply spikes simply trigger them on and they will commutate off at the next zero crossing.

I think SCRs do the same.




I have seen numerous domestic light dimmers fail in the fully-on
condition.

** The triacs are often puny and/or have no heatsinking.

Uses can overload them with a few 100W bulbs.

But they do not fail short due to supply spikes.

They trigger on due to limited dv/dt capacity if the spike is fast or sheer voltage.

With a heavy, noninductive load and very high dV/dt, they can certainly
fail short-circuit. The conducting area moves out from the contact, so
the die area is effectively much smaller at short times.

Of course mains wiring isn't great at transmitting very sharp spikes, so
it's probably safe enough in real life.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Phil Allison]

Phil Hobbs wrote drivel again:

============================

Triacs *self protect*when over voltaged by turning fully on.

Supply spikes simply trigger them on and they will commutate off at the next zero crossing.

I have seen numerous domestic light dimmers fail in the fully-on
condition.

** The triacs are often puny and/or have no heatsinking.

Uses can overload them with a few 100W bulbs.

But they do not fail short due to supply spikes.

They trigger on due to limited dv/dt capacity if the spike is fast or sheer voltage.


With a heavy, noninductive load and very high dV/dt, they can certainly
fail short-circuit.

** No more likely than with normal triggering.

Critical di/dt failure is rare.

Of course mains wiring isn't great at transmitting very sharp spikes, so
it's probably safe enough in real life.

** Correct, plus nearly all light dimmers have an inductor in series with the load.



..... Phil