Below is a VERY simplified, theoretical, circuit for interfacing a high-voltage flame sensor to an Arduino. Please note, the current through the resistor is in MICROamps, not milliamps.
I've been researching this problem for over 8 years to solve a problem with my furnace that began when my electric company replaced the old mechanical meter with one of those Smart Meters 11 years ago. The controller in my furnace is unusually sensitive to noise on the mains, and the manufacturers recommended replacments have the same problem. (The problem is more of a nuisance than a safety issue, for now.) So, I want to build my own controller using a PLC or an Arduino UNO or MEGA 2560. But, interfacing to the flame sensor has got me absolutely stumped!
I have seen op-amps that would work on circuits with AC voltages below 70 volts, but I have been unable to find one that can handle the 120 VAC that will be present when there is no flame. Please note, I want to keep the manufacturers original flame sensor setup, for safety reasons at the very least. Thermocouples, RTD's, and infrared sensors would have too slow of a response time in the event of a flame failure.
Unfortunately, I'm not enough of an electronics engineer to "fill in the gaps". I hope somebody can help me.
The output does not need to be a relay, it can be a transistor; the logic can be forward or reverse; I can work with (almost) any thing.
Thank you in advance to anybody who can help me with this.
BTW: Filtering out the noise from the Smart Meter is almost impossible because the signals are at such a low frequency (less than 1 kHz). The only effective, but impractical, means I have found is to convert the supply AC to DC and back to AC. I presently have my furnace on a double-conversion UPS, but that has created minor problems of it's own.
Is your house the only one in the mains transformer?
What equipment have you used to identify the frequency of the interference and is it continuous or intermittent? Is the interference on ALL the wires feeding your house?
Since you have a functioning UPS, can you use it to power your test equipment and watch the mains interference signal while you use the house circuit breakers to turn off the mains power?
Paul
Thank you for responding. In order of your questions:
No; I share a transformer with 5 other homes.
I have not used any equipment to identify the frequency of the
interference. I recognized there was an "outside" influence on the
furnace because of the timing patterns of effects...very regular and at
specific times of the day. The interference is intermittent and short
term, usually less than 2 seconds. I can tell when the interference is
affecting my furnace by it's behavior. If it is not in a heating mode at
the time of the interference, what will happen is that all the
controller outputs will turn on simultaneously (the main blower, the
combustion chamber blower, the igniter, and the gas valve). How long
they stay on SEEMS to be related to the length of time the interference
occurs, usually less than 2 seconds, though I have seen this happen for
close to 30 seconds, but very rarely, thank God. The gas valve won't
actually open until the combustion chamber blower gets upto full speed,
which takes a little more than 2 seconds.
I don't know if the interference occurs on all the wires.
I do not own an oscilloscope or similar test equipment.
What I know about the interference from Smart Meters, I learned by
lengthy on line research...they have a reputation for causing problems
for electronic devices down line. My furnace is not the only one
susceptible to the interference; I have a remote controlled ceiling fan
that has occasionally been affected.
My ex company built the circuit boards for thousands of the meters when they were first being tested. No problem was ever reported by the city doing the testing.
The old mechanical meter at our house was replaced three years ago. No interference or stray signals have ever been noticed by anything in my house or shop or by any device in my extensive amateur radio station.
Where I live, the limit to the number of drops from s single transformer is three.
But, if something is making the interference you are experiencing, then the odds are really good they will also effect in a bad way, anything you create using an Arduino.
MY best advice is to contact an electrician and have an RF filter put in prior to the main entrance breaker box.
OR get the proper test equipment and track down the source of the interference. The only fundamental difference between our old mechanical meter and the new one is the mechanical kept track of the current on only ONE side of your 230 volt feed, but the new, electronic meter keeps track of the total current being used.
Paul
Be aware that there may be legal restrictions in modifying the control system if your furnace - certainly illegal in the UK for safety liability issues
I've spent a lot of time designing for EMC in aerospace and automotive products so I have some background. Also I have a oil burner furnace that has similar components that I'm familiar with because I had some issues with dirty oil. However I don't have the mental image of your system that I have of mine.
As a result of reading your post I did a quick search for "Smart meter causing interference" and was surprised at the number of issues people have.
My thoughts:
I find it hard to believe a <1k Hz signal from a low powered device can cause the issues reported on different posts. I do believe a high frequency being repeated at a <1k Hz rate could.
Your test of using an AC -> DC -> AC being successful could be due to isolation or the additional filtering the DC stage has.
Suggestions:
If you have a cheap AM radio, tune it to somewhere in the lower band where there is not station. You can use it to find the source of the interference. I know it is associated with the smart meter but is it coming from the unit or its wires.
Can you describe the smart meter installation? Mine is connected to the grounded copper water line with a wire going to the exterior of our house to a control/radio box.
Is your smart meter grounded? Or is the water pipe plastic.
If you wrapped the Smartmeter in heavy duty aluminum foil does the interference change?
Can I assume power enters your furnace controller where the thermostats enter? i.e. a single point of mains entry into the furnace "circuit"
I own the house the furnace is in; as long as I comply with all NEC and local codes while doing the modifications I'm legal; however, if I sell the house I will have to remove/replace the furnace. I plan to die in this house.
I have tried a number of filtering technologies: feed-thru EMI/RFI filter at the furnace; parallel whole house EMI/RFI & surge protection filtering; ferrite cores on the power line at the furncae and on the service into the house. Based on my online research, it is nearly impossible to filter out the signal interference these smart meters create. My AC to DC to AC success shows that the interference is coming over the power lines, not through the air.
As for the meter installation, it is in a cluster of 6 meters adjacent to a pad mount transformer. I cannot verify that the meter base is properly grounded, however, I do know that the service to my house is properly grounded at an outside service disconnect.
I don't know who made the smart meters being used on my service - might not be the same as yours.
My home, and the five others, are connected to a pad mount transformer; local codes allow upto 8 homes on a single pad mount transformer - in a mobile home park.
BTW, I am an electrician with 20+ years of experience; and as I mentioned in a reply to another post, I've tried a number of filtering technologies, with no success.
None of my other Arduino or PLC projects have been affected by the interference, because they are fully isolated from the AC mains by DC filtering. The controller on my furnace is NOT isolated from the AC mains; the AC is distributed throughout the controller to relays and thyristors and one 24 VAC transformer for the gas valve. My Arduino or PLC setup would fully isolate the controller from the AC mains and control the furnace through external relays.
BlockquoteThen have no other suggestions. My mfg. home has aluminum siding which is fully grounded and makes a wonderful metal sided box! Is yours the same?
Blockquote
No, my house is vinyl sided, and after 25 years of aging, it's gotten a little brittle .
Oh, easy! Just act like you are going to cut the line and put it on you clipboard. A little flag will pop up and say "quote" on it. If you click it you will have a quote.
I realized from your first post that you had done a lot of work and have enough background to accomplish what you have. And I didn't mean to dis you in anyway it's just the only way I have been successful in solving this type of issue.
When it comes to RF susceptibility there are only two paths, conducted and radiated. From the little I was able to glean from my quick google search Conducted seemed to be the more likely.
Have you put EMI filters right at the mains entry to your furnace? You would have to filter both hot and neutral with the EMI filters grounded to your furnace mains entry box. The goal is to create a barrier from the unwanted RF signal.
There is still the question of the entry wires from your thermostats. If you haven't already tried EMI filters at the furnace, I would suggest you give it a go**.
At the furnace and on the mains to the house, hots and neutral and ground. The problem is the filters are all intended for much higher frequencies than these smart meters generate; supposedly they generate interference frequencies less than 1 kHz, which is just too close to the 60 Hz line frequency for effective filtering, short of AC to DC to AC filtering. The double conversion UPS I'm currently using was intended for my entertainment system, computers, and a few other devices that don't like power outages; I had never intended to run my furnace on it, although it handles the extra load quite well (It's a 2200 VA TrippLite Expandable UPS).
In any case, this thread seems to have gotten way off my original question about interfacing a high voltage flame sensor to an Arduino.
Sorry I can't help with the flame sensor. I wouldn't feel comfortable messing with a furnace safety device.
I've performed many many official EMC tests and I find it hard to believe a 1k signal from a low power device could cause such interference. I'm not trying to contradict you but my experience suggests there is more to it.
If you feel like it the AM radio test would be most useful. If it is really 1k the radio will remain silent, however if you can here a sharp increase in noise from the radio then you will know there are other frequencies involved.
If you could post the brand and model of you furnace controller. You've got me curious on what type of commercial controller is so sensitive to external signals.
(I've 'speed read' the above comments, so apologies in advance of either repeating misinterpreting) Could your 1KHz be pulses of Rf with a 1KHz prf?
You said that IR would not be quick enough, how quick do you need a response?
(As an aside, a student friend of mine, who worked for the Fire Research Establishment did his thesis on probablity of flame distribution. IIRC he used a thermocouple and conducting probe to measure temperature and presence of ionising gasses (i.e. flame). He felt it was all made redundent by IR cameras. That was about 40+ years ago)
.