One of the Guys at the radio club has the same sort of problem, event got the feds out with one of their Cat-Detector-Vans and they could find nothing. Turns out it was his unused party-phone line, that dates from the 1940s. He was at the end of the line and before it him it ran though the farm next door, and all its old out buildings. One houses the router and computer for the automated milking management system and the old phone line was picking up some of the RF from that. In the end a simple ferrite bead on the incoming phone line solved it along with a 45 ohm terminator to ground.
If the power line comms are anything like the crappy radios that use the power cord as the antenna, I wonder if you could simply open the units up and decouple the rf from the power line, hooking it instead to a dummy power line only shared with the other TED devices. Alternately, put those devices on their own, minimal-length circuit, and add filters to block the noise from going anywhere else.
Modern gear is rarely 'off'... it goes into 'standby' with a portion of the circuits remaining powered, in order to monitor the software triggering 'on' button when it is cycled. I have a preference for real AC disconnection. I also dislike all the wireless 'stuff' and EMI throwing devices that plague the modern era.. Loss of easy access to non emitting thermionic light sources has been another problem.. LED and CFL bulbs are horrid noise sources. I did find that 'Rugged' series filament light bulbs were not banned..but they are costly and scarce. I now have 2 complete lighting systems in my lab.. one for assembly and general work, and another for measurement time, 100% filament thermionic, that it does not throw EMI-RFI.
Most of my even vintage HP gear is AWFUL for this. Crystal ovens stay on so they can stay up to temp due to the 72 Hour warm up time. Why I like having the RB unit for the stability.
I assume that the TED unit you show is some sort of master station that polls the other units and provides the web interface over ethernet. It causes a loud chirp because it's close to your bench, but the slave stations provide a much smaller chirp because they're further away. Maybe different power but there's no obvious reason why that should be the case. So what if you put the master unit in the utility cupboard with the slavers ? Maybe even screen the utility cupboard with metal foil or mesh. Then, wire it to your ethernet. Perhaps use an optical bridge so there's no conduction of the chirp along the ethernet wiring, and put ferrites on the power cables coming out of the cupboard. Then the transmissions will be made well away from your bench. All you'll lose is those leds on the front. If you have slave pickups distributed around the house this won't be so easy, but you should at least be able to move the worst source away from your bench. I wouldn't bet on the supplier dropping the mains transmission, it will suit a lot more of their customers than ethernet.
This was really cool and informative, thanks! Hypothetical: what if you were to build a lab out in the country somewhere and located it in a building which you disconnected from the power grid, instead powering the building via solar panels and specialized filtered power supplies, with a separate dedicated ground rod for the building? Would you see even lower noise on the spectrum analyzer? Suppose you even put a Faraday cage around the building to filter out as much stray RF as possible. I wonder how clean and sensitive you could get...
I am gearing up to move buildings in the medium term. I am starting to look at what I should do with the power systems to get the quietest and cleanest power. My current thoughts are to use a battery conditioned UPS to feed the entire lab with an isolated ground for the lab but I am not sure yet. Common mode noise as well as all the wireless can be REALLY hard to get rid of. There was a famous instance where NASA though they were hearing signals from space that turned out to be a tech making popcorn in a microwave at 1 AM.
The video shows the chirp and proves which unit was the perpetrator. However, if the video explained how the TED was making the problem, I missed it. Maybe turning on/off the current coil to read the current was causing a capacitively coupled "thump" on the Line leg of the Mains and spreading the thump via the Mains connection to everything? Please explained. If I missed it, I apologize. Thanks for the help. Cheers!
This unit uses power line networking it is using the ground and neutral conductors my current theory is when it is actively communicating it is causing the ground reference for the equipment to "not be ground" temporarily which is messing with the measurements. I tried isolating the unit to a high quality power strip as well as a UPS but it was still present. Through the power panel this data impulse was spreading to everything in the lab. I still want to do the power monitoring and The Energy Detective is getting ready to release a new version of this unit so I am currently in a holding pattern waiting for the new unit to release. I am hoping they us ethernet networking or WI-FI for the communication as we have plenty of that in the lab.
@@ZenwizardStudios Yes, I understood from the video your theory of the problem. And, I understand that it may not have been worthwhile to identify the problem within the TED. However, I would have found the exploration into the TED's problem to be very interesting and educational. So, I think our theories combine... Yours - That the Ground is being momentarily lost... Mine - that the TED's sensor coil is conveying an inductive pulse to the Mains legs, when the TED turns on/off the sensing coil to take a measurement. A pulse would change the Ground reference, as you say. If this were to be the issue with the TED, it would be an issue with a component failure, or the circuit's design (both obviously.) But, I would not think that changing the TED's method of communication (ethernet, or WiFi) would correct the problem, as the "bump" is being injected into the Mains. The cure would seem to be to eliminate the inductive bump. I do recall that current sensing coils include a resistor, for some purpose... I forget what. If it is a bleed-off resistor, maybe it failed, due to the repetitive high current spikes. Offered humbly, by a hobbyist. Thanks for the video and explanation! Cheers.
Welcome! It kind of got named the chirp because you could actually hear it in a signal tracer. It was audible, that was the first time it really caused trouble.
Sometimes tracing out those interference issues are a challenge, Great job.
good video. that 3585 looks freaking pristine
One of the Guys at the radio club has the same sort of problem, event got the feds out with one of their Cat-Detector-Vans and they could find nothing. Turns out it was his unused party-phone line, that dates from the 1940s. He was at the end of the line and before it him it ran though the farm next door, and all its old out buildings. One houses the router and computer for the automated milking management system and the old phone line was picking up some of the RF from that. In the end a simple ferrite bead on the incoming phone line solved it along with a 45 ohm terminator to ground.
If the power line comms are anything like the crappy radios that use the power cord as the antenna, I wonder if you could simply open the units up and decouple the rf from the power line, hooking it instead to a dummy power line only shared with the other TED devices.
Alternately, put those devices on their own, minimal-length circuit, and add filters to block the noise from going anywhere else.
Modern gear is rarely 'off'... it goes into 'standby' with a portion of the circuits remaining powered, in order to monitor the software triggering 'on' button when it is cycled. I have a preference for real AC disconnection. I also dislike all the wireless 'stuff' and EMI throwing devices that plague the modern era.. Loss of easy access to non emitting thermionic light sources has been another problem.. LED and CFL bulbs are horrid noise sources. I did find that 'Rugged' series filament light bulbs were not banned..but they are costly and scarce. I now have 2 complete lighting systems in my lab.. one for assembly and general work, and another for measurement time, 100% filament thermionic, that it does not throw EMI-RFI.
Most of my even vintage HP gear is AWFUL for this. Crystal ovens stay on so they can stay up to temp due to the 72 Hour warm up time. Why I like having the RB unit for the stability.
I assume that the TED unit you show is some sort of master station that polls the other units and provides the web interface over ethernet. It causes a loud chirp because it's close to your bench, but the slave stations provide a much smaller chirp because they're further away. Maybe different power but there's no obvious reason why that should be the case.
So what if you put the master unit in the utility cupboard with the slavers ? Maybe even screen the utility cupboard with metal foil or mesh. Then, wire it to your ethernet. Perhaps use an optical bridge so there's no conduction of the chirp along the ethernet wiring, and put ferrites on the power cables coming out of the cupboard. Then the transmissions will be made well away from your bench. All you'll lose is those leds on the front.
If you have slave pickups distributed around the house this won't be so easy, but you should at least be able to move the worst source away from your bench. I wouldn't bet on the supplier dropping the mains transmission, it will suit a lot more of their customers than ethernet.
This was really cool and informative, thanks! Hypothetical: what if you were to build a lab out in the country somewhere and located it in a building which you disconnected from the power grid, instead powering the building via solar panels and specialized filtered power supplies, with a separate dedicated ground rod for the building? Would you see even lower noise on the spectrum analyzer? Suppose you even put a Faraday cage around the building to filter out as much stray RF as possible. I wonder how clean and sensitive you could get...
I am gearing up to move buildings in the medium term. I am starting to look at what I should do with the power systems to get the quietest and cleanest power. My current thoughts are to use a battery conditioned UPS to feed the entire lab with an isolated ground for the lab but I am not sure yet. Common mode noise as well as all the wireless can be REALLY hard to get rid of. There was a famous instance where NASA though they were hearing signals from space that turned out to be a tech making popcorn in a microwave at 1 AM.
The video shows the chirp and proves which unit was the perpetrator. However, if the video explained how the TED was making the problem, I missed it. Maybe turning on/off the current coil to read the current was causing a capacitively coupled "thump" on the Line leg of the Mains and spreading the thump via the Mains connection to everything? Please explained. If I missed it, I apologize. Thanks for the help. Cheers!
This unit uses power line networking it is using the ground and neutral conductors my current theory is when it is actively communicating it is causing the ground reference for the equipment to "not be ground" temporarily which is messing with the measurements. I tried isolating the unit to a high quality power strip as well as a UPS but it was still present. Through the power panel this data impulse was spreading to everything in the lab.
I still want to do the power monitoring and The Energy Detective is getting ready to release a new version of this unit so I am currently in a holding pattern waiting for the new unit to release. I am hoping they us ethernet networking or WI-FI for the communication as we have plenty of that in the lab.
@@ZenwizardStudios Yes, I understood from the video your theory of the problem. And, I understand that it may not have been worthwhile to identify the problem within the TED. However, I would have found the exploration into the TED's problem to be very interesting and educational. So, I think our theories combine... Yours - That the Ground is being momentarily lost... Mine - that the TED's sensor coil is conveying an inductive pulse to the Mains legs, when the TED turns on/off the sensing coil to take a measurement. A pulse would change the Ground reference, as you say. If this were to be the issue with the TED, it would be an issue with a component failure, or the circuit's design (both obviously.) But, I would not think that changing the TED's method of communication (ethernet, or WiFi) would correct the problem, as the "bump" is being injected into the Mains. The cure would seem to be to eliminate the inductive bump. I do recall that current sensing coils include a resistor, for some purpose... I forget what. If it is a bleed-off resistor, maybe it failed, due to the repetitive high current spikes. Offered humbly, by a hobbyist. Thanks for the video and explanation! Cheers.
I have an iammeter. It is wifi. Also it has open interfaces, like a local mqtt and http as well as a full remote server.
Not a chirp.. Aww
First time watching your channel, 30sec in and I've already subscribed 😅
Welcome! It kind of got named the chirp because you could actually hear it in a signal tracer. It was audible, that was the first time it really caused trouble.
Too much WiFi stuff! What is in good for - conveniece? SMPS same topic - bullshxx.