A young start up, ARM Ltd, had a contract to replace the mechanical guts of slot machines with micro-processors and electronics. One of the requirements was that the system should continue to work when an electric arc welder was used very nearby and powered from the same circuit.
Why such a crazy requirement? Seems the slot machine company already had a micro-processor solution but it had a problem. Players learned that lighting a cigarette with a electronic cigarette lighter would often cause the machine to pay out. So players started to do that a lot!
The arc welder test was in lieu of the EMC testing we have today.
Luckily the ARM Ltd solution worked and the company went on to create the ARM processor we know and love today.
Then take the box to earth at the electrical outlet that is in the pic. Verify the ground actual goes somewhere. Mount a plywood board in the back of the box. Then mount my box onto the plywood backing so it is isolated from the metal enclosure. My box has chassis to ground also. The conduit will have a path to ground through the box.
Then I will talk to the AC company to find out any suggestions on how to get the compressor to turn on and off as quickly as possible and see what I see on a Prop scope. I have solved some crazy noise issues on the Prop scope before in the field, so it is better than no scope. They will not tolerate a cigar or cigarette as mentioned, but maybe a candle will have to be the only option unless the AC guys have some ideas.
I have a feeling that since the instances of false triggers or reset is only once or twice a week, that the box and conduit will make an improvement. I will beef up the reset pin with cap/res on the board.
Then I will talk to the AC company to find out any suggestions on how to get the compressor to turn on and off as quickly as possible and see what I see on a Prop scope.
That's why you should try and emulate this on the bench, and 'harden' from there.
Can you also make a re-boot less noticeable ?
"Then I will talk to the AC company to find out any suggestions on how to get the compressor to turn on and off as quickly as possible" - I can almost bet money they won't know anything
IF this problem arises from a "spark" ... then depending on the nature (magnitude) of the spark, metal shielding might not help. Yes, metal shielding will protect against electrical noise, but not magnetic noise that couples into your circuit. With a spark generated from motors, relay contacts, etc. the voltages can be several thousand volts from the collapsing magnetic field in the coil ..... but a collapsing spark GAP itself is slightly different... A spark GAP can generate high voltages, much higher from an Air purifier in the order of hundreds of kV. When you establish a voltage potential across two points that become large enough for a spark to jump across those two points, the spark itself has a relatively LOW resistance value (less than 1k Ohms in many cases.) ... if you treat that spark as a wire carrying current, then a magnetic field is also produced around that "wire" .... Any long runs of parallel wires, loops, transformers in the vicinity will pick that up as if it was a secondary winding to a transformer ( the primary winding being your spark ) and could inject an unwanted signal or voltage spike into a circuit ... <-- Even though you have a shield. The best solution in this case is that you don't run any sensitive electronics near an electrostatic air cleaner or similar device.
BTW) Turning an air compressor ON and OFF faster will likely make the problem worse by generating even higher voltages in the collapsing magnetic field and/or spark gap
If the compressor motor turning on/off is causing the problem and it is an AC motor then turning it on and off at the zero crossing should eliminate the problem.
I suspect that the only odor that might have is caramel. Not so sure about how to fuse the thing to get it started. Buying fuse might require answering a lot of questions in today's world.
++++++++++++
Of course, you might just use a few sticks of incense - pine, sandalwood, or whatever.
I suspect that the only odor that might have is caramel. Not so sure about how to fuse the thing to get it started. Buying fuse might require answering a lot of questions in today's world.
++++++++++++
Of course, you might just use a few sticks of incense - pine, sandalwood, or whatever.
Not sure if it will work for Potassium Nitrate and Sugar but a piece of fine nichrome wire or light bulb filament works for igniting fireworks.
I am reading on the Honeywell info about interference on radios and TVs, and they are referring to the problem as radio frequency interference coming from the air cleaner. They have suggested some filters like this:
I think you might be missing what I am trying to say. The physical spark itself is only part of the equation and depending on the nature of the spark will depend on what sort of problems you may experience.
Type I -- If a spark is caused from a collapsing magnetic field when a circuit is broken such as a motor or relay, the collapsing magnetic field creates a high voltage which can in turn arc across the broken contact(s). This high voltage can usually be snubbed with a capacitor, diode, and/or a capacitor and resistor in parallel with the broken contact(s). ... and with a bit of shielding can usually be solved
Type II -- When a deliberate voltage is generated across two terminals ( usually at a greater distance than a Type I spark gap ) until a spark initiates such as in the picture below (yes, that is my hand and my creation in the photo) and much like what an air purifier would do. In this type II spark, the current through the spark is intense... just like a wire, when you pass current through it it creates a magnetic field around the conductor. In this case, the conductor is literally a spark (typically a few Ohms of resistance). The intense current through this "spark-wire" also creates an intense magnetic field that surpasses the electric field you typically can shield against. This also is like the primary winding in a coil ... with anything in the immediate vicinity that even remotely resembles a conductor acts as a secondary coil. This magnetic coupling is often a silent killer and the culprit of "odd", scratching you head moments of electronic malfunctions .... You can shield all you want with a metal box, but that Faraday cage will only stop electrical noise ... not magnetic noise.
Thanks Beau. I understand the difference as you explain it and that is a good reference. Just pointing out that then info I can find on this Honeywell electronic air clear is referring to radio interference only and solutions, never mentions magnetic.
I don't think Honeywell or any of the other air filter makers are really going to comprehensively document what their air filters might interfere with. They likely mention the radio interference as they might feel the need to comply with FCC rules.
This is a common problem in today's engineering environment -- tons of documentation that is either promotional literature or legal defensive material. But when you have a genuine application concern, you are on your own.
Hopefully, the air filter is not the problem, but it needs to be considered and ruled out. My experience with them is that they spark more and more as they accumulate dirt and age. And since the cleaning can be tedious and the replacement of filters can be expensive, people put up with the sparking for quite awhile before taking action. So if it is indeed the source of the problem, it will worsen over time unless you discover a way to shield.
As I understand it, they only spark when there is a particle passing by between two plates. So the particle is setting the spark in motion as the particle is part of the conduction I assume. It is not sparking non stop. So maybe this will negate the condition Beau talks about where the spark happening without a cause.
"So maybe this will negate the condition Beau talks about where the spark happening without a cause" ... oy!! ...
...The air purifier is a type II as I mentioned above. The "cause" is that the voltage potential across the plates is generated and sustained. Type II plates (or spark gap) are generally further apart then a Type I meaning there will be higher voltages and currents involved. A type I is caused by a magnetic field collapse within a coil and is generally short lived and comparatively lower in voltage than a Type II.
When I talked of shielding circuity from E fields in a metal box I was thinking the kind of sparks you might get from a Van_de_Graaff_generatorhttps://en.wikipedia.org/wiki/Van_de_Graaff_generator. High voltages but minimal currents. A metal shield will protect against capacitive coupling of such E fields.
I think Beau has a point. A much bigger spark, with a lot of energy behind it, will give rise to a big current spike. That current is flowing in a large loop of a circuit. It will be fast rising. The loop driving the spark is an inductor producing a magnetic field that a metal shield will not keep out.
Presumably one has to:
1) Get the circuit as far a way as possible.
2) Minimize the "receive" loop areas in ones circuit to minimize magnetic coupling.
3) Arrange that ones circuit is tolerant of whatever that magnetic coupling does manage to induce.
Not sure what that dust particle has to do with it. A spark is a spark. The cause is the high voltage and stored energy.
Comments
A young start up, ARM Ltd, had a contract to replace the mechanical guts of slot machines with micro-processors and electronics. One of the requirements was that the system should continue to work when an electric arc welder was used very nearby and powered from the same circuit.
Why such a crazy requirement? Seems the slot machine company already had a micro-processor solution but it had a problem. Players learned that lighting a cigarette with a electronic cigarette lighter would often cause the machine to pay out. So players started to do that a lot!
The arc welder test was in lieu of the EMC testing we have today.
Luckily the ARM Ltd solution worked and the company went on to create the ARM processor we know and love today.
http://www.mcmaster.com/#75065k69/=zbj2pv
Then conceal all visible cable with this:
http://www.mcmaster.com/#73145k4/=zbj2t8
Then take the box to earth at the electrical outlet that is in the pic. Verify the ground actual goes somewhere. Mount a plywood board in the back of the box. Then mount my box onto the plywood backing so it is isolated from the metal enclosure. My box has chassis to ground also. The conduit will have a path to ground through the box.
Then I will talk to the AC company to find out any suggestions on how to get the compressor to turn on and off as quickly as possible and see what I see on a Prop scope. I have solved some crazy noise issues on the Prop scope before in the field, so it is better than no scope. They will not tolerate a cigar or cigarette as mentioned, but maybe a candle will have to be the only option unless the AC guys have some ideas.
I have a feeling that since the instances of false triggers or reset is only once or twice a week, that the box and conduit will make an improvement. I will beef up the reset pin with cap/res on the board.
Can you also make a re-boot less noticeable ?
IF this problem arises from a "spark" ... then depending on the nature (magnitude) of the spark, metal shielding might not help. Yes, metal shielding will protect against electrical noise, but not magnetic noise that couples into your circuit. With a spark generated from motors, relay contacts, etc. the voltages can be several thousand volts from the collapsing magnetic field in the coil ..... but a collapsing spark GAP itself is slightly different... A spark GAP can generate high voltages, much higher from an Air purifier in the order of hundreds of kV. When you establish a voltage potential across two points that become large enough for a spark to jump across those two points, the spark itself has a relatively LOW resistance value (less than 1k Ohms in many cases.) ... if you treat that spark as a wire carrying current, then a magnetic field is also produced around that "wire" .... Any long runs of parallel wires, loops, transformers in the vicinity will pick that up as if it was a secondary winding to a transformer ( the primary winding being your spark ) and could inject an unwanted signal or voltage spike into a circuit ... <-- Even though you have a shield. The best solution in this case is that you don't run any sensitive electronics near an electrostatic air cleaner or similar device.
BTW) Turning an air compressor ON and OFF faster will likely make the problem worse by generating even higher voltages in the collapsing magnetic field and/or spark gap
http://www.wikihow.com/Make-a-Smoke-Bomb-from-Potassium-Nitrate-and-Sugar
I suspect that the only odor that might have is caramel. Not so sure about how to fuse the thing to get it started. Buying fuse might require answering a lot of questions in today's world.
++++++++++++
Of course, you might just use a few sticks of incense - pine, sandalwood, or whatever.
Not sure if it will work for Potassium Nitrate and Sugar but a piece of fine nichrome wire or light bulb filament works for igniting fireworks.
http://www.amazon.com/Pashion-Ferrite-Magnetic-Suppressor-Diameter/dp/B01587GZZ8/ref=sr_1_5/175-6069374-5075520?ie=UTF8&qid=1444698924&sr=8-5&keywords=rfi+filters
This does not address the 240VAC compressor which as Beau mentioned could be electromagnetic.
Type I -- If a spark is caused from a collapsing magnetic field when a circuit is broken such as a motor or relay, the collapsing magnetic field creates a high voltage which can in turn arc across the broken contact(s). This high voltage can usually be snubbed with a capacitor, diode, and/or a capacitor and resistor in parallel with the broken contact(s). ... and with a bit of shielding can usually be solved
Type II -- When a deliberate voltage is generated across two terminals ( usually at a greater distance than a Type I spark gap ) until a spark initiates such as in the picture below (yes, that is my hand and my creation in the photo) and much like what an air purifier would do. In this type II spark, the current through the spark is intense... just like a wire, when you pass current through it it creates a magnetic field around the conductor. In this case, the conductor is literally a spark (typically a few Ohms of resistance). The intense current through this "spark-wire" also creates an intense magnetic field that surpasses the electric field you typically can shield against. This also is like the primary winding in a coil ... with anything in the immediate vicinity that even remotely resembles a conductor acts as a secondary coil. This magnetic coupling is often a silent killer and the culprit of "odd", scratching you head moments of electronic malfunctions .... You can shield all you want with a metal box, but that Faraday cage will only stop electrical noise ... not magnetic noise.
http://custommagneticshielding.magneticshield.com/category/mumetal-sheet-and-foil
There are a few products advertised for magnetic shielding.
This is a common problem in today's engineering environment -- tons of documentation that is either promotional literature or legal defensive material. But when you have a genuine application concern, you are on your own.
Hopefully, the air filter is not the problem, but it needs to be considered and ruled out. My experience with them is that they spark more and more as they accumulate dirt and age. And since the cleaning can be tedious and the replacement of filters can be expensive, people put up with the sparking for quite awhile before taking action. So if it is indeed the source of the problem, it will worsen over time unless you discover a way to shield.
...The air purifier is a type II as I mentioned above. The "cause" is that the voltage potential across the plates is generated and sustained. Type II plates (or spark gap) are generally further apart then a Type I meaning there will be higher voltages and currents involved. A type I is caused by a magnetic field collapse within a coil and is generally short lived and comparatively lower in voltage than a Type II.
I think Beau has a point. A much bigger spark, with a lot of energy behind it, will give rise to a big current spike. That current is flowing in a large loop of a circuit. It will be fast rising. The loop driving the spark is an inductor producing a magnetic field that a metal shield will not keep out.
Presumably one has to:
1) Get the circuit as far a way as possible.
2) Minimize the "receive" loop areas in ones circuit to minimize magnetic coupling.
3) Arrange that ones circuit is tolerant of whatever that magnetic coupling does manage to induce.
Not sure what that dust particle has to do with it. A spark is a spark. The cause is the high voltage and stored energy.