Hello,
Does the switching ON and OFF of **all ** Air conditioning units result in mains transients?, or is it only old AirCon units that exhibit mains transients when switched ON/OFF?
We ask because our garden lamp products seem to be failing due to mains transients. Our new factory has four new airconditioning units protruding out of the side wall, and so If these are a convenient source of mains transients, then we can put our lamps on the same mains supply, and see if they fail.
I would get the scope out and look on the mains myself, but the boss won’t let me do this kind of investigation unless I provide some proof up front.
The following article, on page 2, states that mains transients to 800V are caused by Air Conditioning Units switching on and off….
http://www.seered.co.uk/sunvic_capacitor_information.pdf 
I have run across this problem before in industrial environments. The usual solution is to move the affected equipment to its own circuit or install extreme power conditioning systems like online UPSes.
Cycling motors, such as air conditioners and elevators, frequently cause EFTs, which are capable of producing spikes up to 1kV. The operation of arc welders and motor starters can cause transients as high as 3kV.http://www.ecmweb.com/contractor/looking-sources-transient-overvoltages
The usual solution is to move the affected equipment to its own circuit or install extreme power conditioning systems like online UPSes.
Thanks, i take it you mean, the equipment other than the aircon unit needs to get power conditioning protective circuitry, to protect it from the aircon unit spikes?
Depending on the source impedance (the impedance of the circuit coming in to the area), there will be more or less 'flicker' caused by high inrush currents. This is a fairly easy calculation given the cable size and length from the transformer and similar information on the circuit feeding the AC unit. It is based on Locked Rotor Amps for the AC unit. At my house, lights visibly dim when the AC system kicks in but it's rather larger unit than a small window cooler. Nothing gets damaged by this momentary drop in voltage. In every environment, 'flicker' is an issue to some extent. Even in large industrial establishments, starting large motors, even using Y-Delta starters, causes some amount of flicker.
Once again, I think you're looking in the wrong place. If it was a problem, it would be a problem for everybody's product. It's your product that fails. You can hardly expect your customers to leave their AC off in order to guarantee survival of a garden light.
A better place to look might be for some reference material related to electrical design for survivability in residential environments. Something that spells out what you need to design for. Then you need to build a test setup that will cause those conditions and test your product.
As to the article touting capacitors, remember that it is written by a capacitor company. Of course their product will solve all the problems of the world!
Semi modern aircon is generally fitted with electronic variable frequency drives to at least soft start the compressors, it is the OLD stuff that tended to star/delta starters on the compressors that could cause all sorts of mess.
If you want to be able to easily test you really want a proper test source so that you can inject common and differential mode pulses onto the supply to your gear on demand and at known pulse shapes and amplitudes, using a nearby noise source will not do this in a repeatable way.
Get the right gear, it will save you more in R&D time and very expensive customer returns then it will cost you, and a local EMC house will probably be quite prepared to advise.
Regards, Dan.
Quite possible. I had a fridge, every time it's compressor stopped, my webcam disconnected from the PC. And a pop sound came from my coffee machine. AC motors are nasty things. I placed one of these surge suppressors on both ends, but that didn't help. Extra ferrite to the camera did. And I got rid of the old fridge.
I have run across this problem before in industrial environments. The usual solution is to move the affected equipment to its own circuit or install extreme power conditioning systems like online UPSes.
Cycling motors, such as air conditioners and elevators, frequently cause EFTs, which are capable of producing spikes up to 1kV. The operation of arc welders and motor starters can cause transients as high as 3kV.
http://www.ecmweb.com/contractor/looking-sources-transient-overvoltages
The real issue is how much energy is in a spike. A sub-millisecond glitch isn't likely to be noticed anywhere in the system. Even if it is, if other equipment is dealing with it, a simple garden light should survive. A simple inrush 'flicker' shouldn't affect anything - by design.
I remember the bad old days of computer facilities where everybody worried about everything. Mainframes were typically fed 400 Hz power to a) allow smaller transformers and b) ensure they had dedicated power. It was costly but it worked well. Today, we have more horsepower in our desktops and they run on nasty old residential power. And they run well.
The reason they run well is that the power supply designers have taken the environment into consideration in the design.
AC motors are nasty things.
The motors are fine, it is the cheap, shonky thermostat that is the nasty thing, combine the worlds cheapest relay or bimetalic strip with an inductive load and then value 'engineer' out the snubber and yea, RFI city, but the motor has little to do with it.
Regards, Dan.
We ask because our garden lamp products seem to be failing due to mains transients.
OMG more crap undesigned products from your company, what for gods sake name does it trade under so I can AVOID ??
OMG more crap undesigned products from your company, what for gods sake name does it trade under so I can AVOID ??
Sounds to me like they're trying to work out why the products are failing, presumably to improve them. I wouldn't look down on that.
A/C and refrigeration compressor motors generate spikes far beyond their inductance.
They are worst source for mains transients I've ever seen, next to lightning.
There is (mechanical) energy storage causing backspin on the rotor when the motor switches off. This (piston) back "kick" I've seen generator-action, and the motor dishes out a many kV spike. I thought induction motors could not generate anything, but until the rotor flux decays, there is a kick.
I would try move the garden lights to another phase, not the one the A/C units are on. Or try extra surge-protection. Ask them if other equipment is failing or giving them problems, as the A/C can disrupt many devices.
treez, I invite you to stop going small and ultra low cost on your products. The extra $0.50 for a bigger mov outweighs any savings now. Product failures out in the field are very costly.
I've seen people use induction motors as generators, I was surprised at first but it does work. You just have to connect some capacitors across the motor and then spin it at the proper speed and it will self excite assuming some residual magnetism. If you load it too heavily the output will suddenly drop to nothing but if you need to MacGyver a generator out of stuff you have laying around it is possible to do.
OMG more crap undesigned products from your company, what for gods sake name does it trade under so I can AVOID ??
Sounds to me like they're trying to work out why the products are failing, presumably to improve them. I wouldn't look down on that.
When you continuously assume the cause is outside your product that is called blame shifting. This individual seems stuck on proving over voltage from an outside source is killing the product. Even if that is true then the answer is their design because it seems these uncontrollable factors are everywhere. In another thread they have diagnosed the failure to something they 'ass u me' is over voltage. I think they need to start with explaining why they believe that and see if there could possibly be other things going on here. Maybe the design is OK but they are assembled with counterfeit parts? Saw a guy get stuck with the SSR's off of sleazebay causing a problem in a product.
Sounds to me like they're trying to work out why the products are failing, presumably to improve them.
That's how I read it.
My suggestion is that relying on the unknown parameters of your air conditioning units to provide transients for testing purposes is the wrong way to approach it. If your units fail on site, then that means the design needs attention - but if they don't fail, it
doesn't mean the design is OK.
The transients you get are going to be unknown in a number of parameters, not limited to when they occur and their magnitude. This is an extremely haphazard way to test. You really need a proper transient generator. You could modify your design to work within your factory locale - but when it goes out into the field, the units may be exposed to more severe electrical conditions.
In other words - you can do it, but I would not rely on the results as confirmation of your design in any way, shape or form.
To test right, you need to spend the money.
Edit: Spelling
OMG more crap undesigned products from your company, what for gods sake name does it trade under so I can AVOID ??
Sounds to me like they're trying to work out why the products are failing, presumably to improve them. I wouldn't look down on that.
Neither would I, but in this case it is worth scanning the subforums for threads recently started by the OP, and reading the first post. For example:
https://www.eevblog.com/forum/beginners/mains-in-uk-can-go-up-to-292vac-for-periods-of-several-minutes/I'm beginning to think that I would avoid any product designed by a company in Southampton that is related to a combination of LEDs, streetlights/lights, WiFi.
I would get the scope out and look on the mains myself, but the boss won’t let me do this kind of investigation unless I provide some proof up front.
Does boss = wife? I really get the impression you are a cottage industry trying to make a buck off of the industry that has sprung up around people growing medicinal plants at home? You've read enough to get a product out the door but are just now learning engineering has lots of quirks you haven't learned yet?
The usual solution is to move the affected equipment to its own circuit or install extreme power conditioning systems like online UPSes.
Thanks, i take it you mean, the equipment other than the aircon unit needs to get power conditioning protective circuitry, to protect it from the aircon unit spikes?
I am sorry if I was not clear. Yes, that is correct. Although if the solution involves a separate circuit, it could be for either. Office environments sometimes run across this with laser printers because the heater for the fuser draws a lot of power to minimize warmup time.
I would get the scope out and look on the mains myself, but the boss won’t let me do this kind of investigation unless I provide some proof up front.
Does boss = wife? I really get the impression you are a cottage industry trying to make a buck off of the industry that has sprung up around people growing medicinal plants at home? You've read enough to get a product out the door but are just now learning engineering has lots of quirks you haven't learned yet?
I have done this before and a safe and simple method is worth mentioning. Just connect the oscilloscope input to the line through a low voltage AC power transformer, set the triggering conditions in normal sweep mode so that only large surges will cause a trigger, and check it periodically.
I have done this before and a safe and simple method is worth mentioning. Just connect the oscilloscope input to the line through a low voltage AC power transformer, set the triggering conditions in normal sweep mode so that only large surges will cause a trigger, and check it periodically.
Thanks, this sounds a good idea.
And i assume the reason that you wouldnt just connect a 100:1 probe to the raw mains is because of possibility of accidentally touching it and getting shocked?
We were also thinking of a microcontroller circuit 'watching' the output of a comparator which 'looks at' a divider after a mains rectifier.
Hello,
Does the switching ON and OFF of **all ** Air conditioning units result in mains transients?, or is it only old AirCon units that exhibit mains transients when switched ON/OFF?
We ask because our garden lamp products seem to be failing due to mains transients. Our new factory has four new airconditioning units protruding out of the side wall, and so If these are a convenient source of mains transients, then we can put our lamps on the same mains supply, and see if they fail.
You really need to tell us what the make and model of the four new air conditioners are to have any chance of answering this accurately.
A reasonable rule of thumb tho is:
Older style air conditioners with constant speed compressors = more transients, disturbances when the compressors start and stop
Modern Inverter driven units = less as the inrush only occurs with the initial power up of the Inverter section. However more chance of issues with harmonics EMC issues (if circuitry is not well designed)