Ranco etc burned out?

[spyderrobotics]

Arcing does not create an imbalance. But again, is it a GFCI or an AFCI? AFCI would assume arcing is a failed extension cord - cut off power.

Semiconductor switching would eliminate arcing using techniques such as switching on zero crossing (so as to not create excessive heat). Also necessary are numbers for that heater and for switch rating. Switches that cut off more than 1 amp will arc. But a switch is designed to minimize that arc and to not cause contact damage. Snubbing will not eliminate the arc - but will significantly reduce destructive effects.

If a GFCI trips simultaneously with arcing, then current is going somewhere else - where it should not be going. GFCI would be tripping as designed; reporting a potential human safety issue.

Line filters are expensive? What is expensive - a number required?

https://www.google.com/url?sa=t&rct=...68rtmX327KQOjg

Semiconductor switching of course is best solution but has its own problems with heat output and is not something that can be added to correct the problem with the thermostat they already have.

Usually arcing occurs when the contacts are broken so that would be good info to know. If it trips during release (when the power is turned off to the heat).

Your link to line filters is for components and not a plug in solution. I was thinking more towards line conditioners that would be a plug in solution and may have built in line filters and such.

[westom]

Semiconductor switching of course is best solution but has its own problems with heat output and is not something that can be added to correct the problem with the thermostat they already have. ...
Your link to line filters is for components and not a plug in solution. I was thinking more towards line conditioners that would be a plug in solution and may have built in line filters and such.

If semiconductor heat is so problematic, then dimmer switches would even be hotter. Semiconductor switching is problematic when a switch is not properly designed. For example, if a switch uses a transistor, then heat can be problematic. Other semiconductor solutions (as found in dimmer switches) exist.

For example, that thermostat can drive a box containing a better switch (ie relay); that does not arc. A thermostat that conducts well less than 1 amp (to power that box) should not arc.

Retail price of filters is provided. Manufacturers box same filters (purchased at wholesale prices) and add a plug and receptacle. Then it sells at profit for a similar price. Demonstrated are filters (and costs) found inside filtered appliances and inside plug-in filters. Or build one. So again, a question. What number defines expensive?

Other numbers. What is an electrical specification for that thermostat switch? What numbers are for a heater. Better answers exist when numbers are provided. No numbers often results in answers that are only speculation or hearsay. And may even explain a previously discussed 'near fire'.

[spyderrobotics]

If semiconductor heat is so problematic, then dimmer switches would even be hotter. Semiconductor switching is problematic when a switch is not properly designed. For example, if a switch uses a transistor, then heat can be problematic. Other semiconductor solutions (as found in dimmer switches) exist.

Dimmers use triacs same as proportional thermostats but dimmers are most commonly 600 watts and lower. They are also rarely at full load. They sink heat to the metal frame of the switch in most cases. There is small resistance across a triac and hence heat. It is problematic and has to be dealt with via heatsink and airflow. The higher the wattage the more heat but not linear. I have extensive knowledge and testing in this area. Mosfets have lower on resistance but then in most circuits AC is converted to DC to use the mosfet and then you have the voltage drop/resistance across the diode bridge which then also creates alot of heat. No free rides there. Ideally a trigger circuit using a double mosfet in reverse to each other would be the best solution but it is costly. And when I say costly I mean more than what can be done in other methods that are acceptable. Yes, a $5-$10 additional cost doesn't seem like much but when you add additional labor, complexity, markups, and so forth there is always a balance to be had. So I guess you have to be clear whether you are talking about a one off solution to solve a individuals issue or something that is a production unit.

[westom]

So I guess you have to be clear whether you are talking about a one off solution to solve a individuals issue or something that is a production unit.

This is obviously a 'one off solution'. Since other heaters do not have this problem. Semiconductors remain a simple solution (BTW, heat is greatest when operating a dimmer at half intensity. Dimmers must works continuously in a worst state and still not create too much heat. They do). Other solutions have also been mentioned. All must be replied to.

A most critical questions remain ignored. Is it a GFCI or AFCI. Nothing useful can be posted without answering all questions - especially this one. Also not provided were numbers. Semiconductors (ie thyristors) work successfully (not too hot) for same reasons why dimmer switches (at highest heat setting ie at halfway dimmed) also produce so little heat. Details are irrelevant because every question was not answers - including all most important ones.

[spyderrobotics]

BTW, heat is greatest when operating a dimmer at half intensity. Dimmers must works continuously in a worst state and still not create too much heat. They do).

This is not true. Only a resistive based rheostat generates more heat during dimmed function because it burns the excess current off through resistance. In solid state devices like triacs or mosfets they only pass through current when the device is in the on state and therefore at half power they only produce half the heat (only on for part of the half AC cycle). Rheostats are practically extinct today. Even cheap dimmers use this method. Which often is why dimmers are not rated for ceiling fans (motors in general). Because without a snubber on the triac they can self trigger and not turn off at zero crossing. We actually use a more expensive triac with built in snubber on our units which allows them to work with misting pumps and such.

[westom]

In solid state devices like triacs or mosfets they only pass through current when the device is in the on state and therefore at half power they only produce half the heat (only on for part of the half AC cycle).

When does a dimmer switch using triacs create highest heat? When at the half dimmed position. How much heat? Depends on how it is designed. Semiconductors are routinely used to switch power - even hundreds of thousands of volt transmission lines. AC is converted to DC from Washington to California, all lines incoming to Quebec, and the connection from NJ to Long Island. Then more semiconductors convert that DC back to AC - without excessive heat. But again, semiconductors are problematic when one does not implement additional design concepts. Hundreds of megawatts controlled and delivered by semiconductors that do not overheat.

When does a computer's power supply (that is constantly switching) create greatest heat? Typically when operating at half power. Heating is minimized at full power - when semiconductors are selected and properly designed to perform switching.

Denials only exist because you could not make semiconductors and snubber circuits work. Others have used those well proven solutions successfully. Which one is best for the OP? He must first provide necessary information. Otherwise only solutions based in speculation (ie a protector) will exist.

[spyderrobotics]

When does a dimmer switch using triacs create highest heat? When at the half dimmed position. How much heat? Depends on how it is designed. Semiconductors are routinely used to switch power - even hundreds of thousands of volt transmission lines. AC is converted to DC from Washington to California, all lines incoming to Quebec, and the connection from NJ to Long Island. Then more semiconductors convert that DC back to AC - without excessive heat. But again, semiconductors are problematic when one does not implement additional design concepts. Hundreds of megawatts controlled and delivered by semiconductors that do not overheat.

When does a computer's power supply (that is constantly switching) create greatest heat? Typically when operating at half power. Heating is minimized at full power - when semiconductors are selected and properly designed to perform switching.

Denials only exist because you could not make semiconductors and snubber circuits work. Others have used those well proven solutions successfully. Which one is best for the OP? He must first provide necessary information. Otherwise only solutions based in speculation (ie a protector) will exist.

I don't think you understand. "When does a dimmer switch using triacs create highest heat? When at the half dimmed position." This is not true. In a half dimmed position on a solid state dimmer is is only passing current half the time and therefore generating half the heat. During the ac cycle the voltage goes from 0 to 120V and back to zero in a half cycle. Triac based control waits for zero crossing and then delays and then triggers in that half cycle and stays on until back to zero volts. The longer the delay the less power. If you wait half the cycle then you are not passing any current and not sinking any current during that half phase therefor not generating any heat.

"semiconductors are problematic when one does not implement additional design concepts" This is true.

"When does a computer's power supply (that is constantly switching) create greatest heat? Typically when operating at half power." This is wildly false. They do have better efficiency at certain loadings (usually more than half loaded) but they definitely do not create more heat at half power. This is also why many switching power supplies do not turn on the fan until load goes up. You are confusing linear power supplies with switching power supplies.

"Denials only exist because you could not make semiconductors and snubber circuits work." I certainly have proven to make semiconductors work just fine with 10+ years of building and designing thermostats which have snubbers built in. I also had snubbers working in my relay experiments but in order to see much improvement it required a large cap and due to size restrictions was not the best option. There is a difference between successful and not practical for a circuit.

And with that I have thermostats to build.