Power Conditioners and RCDs

[jamesdesilva]

Hi everyone,

I'm having an issue with two power conditioners and my mains RCD in my home.

I have two identical power conditioners, both of which work properly when used individually. They are housed in two seperate racks, with different audio stuff in each. When I power up the first, everything is OK, and so I switch on the outlets on the back to power up the other units in the rack. I then power up the second, and all still seems OK, but... the moment I switch on the outlets on the second power conditioner, my mains RCD in the house trips out. Both units work individually as explained before, and it doesn't matter in which order I power up the outlets, still the same outcome, a tripped RCD.

I wonder if anyone has had this problem before, and wonder what I might be able to do to cure it???

For now I have removed one of the power conditioners and am running direct from the mains.

 

For reference:

Rack 1 Contains: Thomann VM-100 Power Conditioner, Phonic Helix Board 18 Firewire (1st Gen), 3 x Behringer Autocom Pro-XL MDX1600 Compressors

Rack 2 Contains: Thomann VM-100 Power Conditioner, 2 x Behringer EP2500 Power Amp, Alto X23 Active Crossover

[ceecrb1]

sounds like the power conditioner needs a service...

 

I'm sure others who are more local can give suggestions of repair places?

[Simon Lewis]

James,

 

It looks as if the leakage current that each conditioner introduces is (in addition to any other leakage current in your house) is sufficient to trip the RCD. It's also possible that the house RCD is faulty. We don't do D0m35t1c stuff here, so we can't help with that side, but someone with a suitable portable appliance test meter could let you know if the power conditioners themselves were causing a problem.

 

Why do you need them anyway? Is your local mains giving problems, or are they a convenient means of providing the necessary mains sockets?

 

 

Simon

[Andrew C]

As you say that these are audio racks, we'll skip the domestic bit for now.

 

My guess at to the cause of your problem is this:-

 

You have a PC somewhere leaking a few milli-amps

The fridge leaks a few milli-amps

The first power conditioner you plug in leaks a few milli-amps...

...etc

 

You are now up to 25mA

 

You plug the second one in - 29.9999999mA

 

Switch on the first bit of kit - CLICK - it all goes dark

[djrichie_t]

It seems to me that there is alot of interference on your mains supply, which when you plug in a power conditioner, is causing some of it to go to earth. Two power conditioners will dump twice the current to earth as one, which means that switching any one on will be below the 30mA threshold, however switch the 2nd one on and the current to earth will go above this, and trip the rcd.

 

Power conditioners work in many different ways, and often with a miniscule current leaking to earth however depending on the mains interference, spikes etc... this can cause harmonic probs which will throw the rcd.

 

First thing I would do is unplug everything in the house, switch off the mcb's and have only the two racks plugged in. See if they power up fine when you switch them both on.

 

a) If they do switch on ok, then slowly start plugging everything in the house back in until the breaker trips, and thats what is adding to the problem. Could be a kettle left plugged in, or a dodgy immersion heater element. It could also be something like a dodgy fluorescent light fitting causing surges from the choke to go back into the mains, and therefore when the surpressors try and clamp it, there will be too much current to earth and the rcd trips.

 

b)If however, when you switch both racks on the breaker trips straight away then its probably an issue with a very 'noisy' (unclean) mains supply which is causing enough current to leak back from the power conditioners to throw the rcd. I would then consult the electricity board who will send out an engineer to test the mains supply and see how much interference there is. I've witnessed times when I have put a scope on different mains supplies, and its definately not a pure sine wave. Even moreso if your house is at the end of a street (less if its a more modern PME system)

 

Also it might be worth checking just to see if the rcd is not faulty (or the wrong one, it should be 30mA but I have seen times when idiots have put in a 10mA one)

 

Have a chat with your neighbours too and see if they have had similar problems with nuisance tripping.

 

I have installed mains supressors directly between the electric meter and consumer unit, the zymax ZM1000 is a good choice, click here for details.

 

HTH

 

Richie.

[Ike]

Power conditioners at this level are nothing clever, they simply consist of a MOV or number of them to "short out" any high voltage surges and a low pass filter to reduce the amount of harmonics and other noise at higher frequencies than mains. OK the fancy ones contain an LED or neon to show you when one of the MOVs has died and the really fancy ones such as those made by Zymax have a buzzer and some electronics to tell you when one is going to die but the principle is exactly the same.

 

 

The number and severity of surges your equipment will be subject to is hugely dependent on local conditions however it's fair to say that frequency decreases with an increase in severity. A typical property might have something in the region of 1000 1kV surges in a year and only 1 at 6kV. Incidentally surges over 6kV in domestic and commercial buildings are very unlikely as at this voltage a standard 13A socket acts as a surge suppressor with the transient simply tracking between the terminals. The european requirements for surge immunity of equipment are very flexible allowing the designer to decide what a pass actually is and don't cover anything greater than 1kV. Most equipment is however designed to much higher standards so as to reduce warranty costs and bad public relations concerning equipment failure. While MOVs do cause some earth leakage particularly during surges and it is possible for the switching of reactive components in the second rack to cause such surges it hugely unlikely that this will cause any problems with RCDs compared to the relatively high leakage caused by the low-pass filter.

 

 

The idea here is that any interference likely to affect the function of equipment is going to be higher than 50Hz or the local mains frequency. These could be harmonics which are integer multiples of the fundamental (mains) frequency or other interference from sources such as nearby TV or radio transmitters, noisy switches etc. Unfortunately for filter designers the worst interference is normally that closest to the mains frequency which is also the hardest to filter out without also filtering out some of the mains and thus causing earth leakage. Matters are made substantially worse when you cascade filters for example by using one on the input to a rack and then inbuilt ones in individual pieces of equipment. The filters stop acting as individual filters and start interacting causing even higher earth leakage than expected, you also run the risk of them beginning to self resonate and making your problems a million times worse. It's fair to say that even a perfectly well designed and working filter on a perfectly clean supply can leak a significant amount to earth, cascaded filters will leak a lot more.

 

 

RCDs are not accurate, they are designed to trip at less than there rated current rather than more than it in the interest of safety. It's safe to say they should not trip at half there rated current and indeed this is one of the tests required by most periodic inspection reports however an RCD that trips at anything more than this is usually perfectly acceptable. Assuming it is a 30mA RCD 15mA of leakage is not a lot depending on what you are powering from it, if it's a split consumer unit with all your household sockets on a single RCD then it certainly seems reasonable that it would trip.

 

 

Personally I'm not a fan of the recent craze of fitting filters to everything, it's important to remember that filters do not absorb interference but simply move the problem from one place to another. If you have a problem and fitting a filter "solves" it then good for you but I'm far from convinced they should be fitted just in case. MOV surge suppression probably does save at least a small amount of equipment however it's worth bearing in mind that decent equipment will already feature it and cheep plug top surge suppression such as those from pound shops and the like will be 99% as effective as even the most expensive Zymax models. If you are going to invest in one try and get one with a neon or LED fail light and replace it when the lights comes on, leaving a dead MOV in a circuit for prolonged periods is a good way to cause a fire.

[adam2]

As others have sugested, it is probable that the leakage current of either of the power conditioners is less than the RCD trip current, but that the total leakage current of the two is sufficient to trip the RCD.

Without straying too far into forbidden d0m3estic territory, you may wish to have an electrician install a couple of dedicated sockets for your equipment, not on the main RCD.

[SteveAATW]

Just adding to what Ike said, have you installed the power conditioners because there is an actual need to or because you thought it was best to? 9 times out of 10 they give no discernible benefits other than giving you a pretty readout of the mains voltage.

 

Also you shouldn't be running power amps through them, they are really designed for small signal protection only.

 

Ultimately the cheap ones (sub £1000's) do little more than what any EMC compatible appliance already does. I've put small signal audio gear in some fairly hostile places and never needed a power conditioner, most problems are best solved at signal level or by routing out the root cause.

[djrichie_t]

I came up with a completely fool-proof solution for some1 that had a problem needing pure clean power.

 

Very simple, and even better than a normal UPS system that you find on computers...

 

basically you get (or get made) a heavy duty 24v charger, onto a bank of batteries, then get a true sine wave inverter and plug your equipment into that

 

Not the cheapest but definately the cleanest and most secure solution.

[Ike]

...until you plug something into it.

 

Whenever someone genuinely needs a clean supply I think it's safe to assume they also need it to have a very low impedance, if it didn't as soon as they plugged their kit in you'd more than likely end up with all the rubbish you were trying to remove in the first place. The output impedance of "true sine wave inverters" is never great, even the really expensive ones don't have anything low enough to be genuinely called clean supplies.

 

The one area where clean supplies are a requirement is in electro-magnetic compatibility (EMC) testing. The supplies used here tend to consist of a power amp (exactly the same as the noise boys use) fed with a 50hz sine wave and negative feedback to further reduce the output impedance. There are a few manufacturers who have streamlined the design a bit, removed unnecessary features and put everything in a nice white box but the principle is the same.

[djrichie_t]

You can get decent sine inverters with a low output impedance, these are heavier as they use a proper step up transformer rather than switched mode circuitry. Basically its a row of mosfets fed straight off the batteries, fed with a sine wave and feeding into the transformer. Bit old-school but definately nicer in my opinion. Just the weight can be quite hefty.

[Ike]

Well you learn something new every day!

 

It's still a safe bet that the transformers impedance would be too high at higher frequencies for a lot of test and measurement purposes but I guess you could counteract that by adding capacitors, still seems a bit of a long winded approach though.

[SteveAATW]

You can get decent sine inverters with a low output impedance, these are heavier as they use a proper step up transformer rather than switched mode circuitry. Basically its a row of mosfets fed straight off the batteries, fed with a sine wave and feeding into the transformer. Bit old-school but definately nicer in my opinion. Just the weight can be quite hefty.

 

In essence that is just a power amp though - it just uses a DC rail from the batteries rather than rectified mains. The impedance can be a problem - batteries, especially heavy duty deep discharge capable lead acid ones take a finite amount of time to react to load demand changes. Add to that the reactance of the transformer and you can have a noticeable lag in step response. The harder you work them the worse that lag gets, you'll also have a voltage variation on your load as the batteries terminal voltage will decrease with load.

 

Also if the batteries are online and fed by a rectifier from the mains, problem transients can flow straight through them - it's sometimes easier for them to go to your output circuit than into the battery. Its arguable that the performance is sometimes no better than just rectifying the mains to DC and using that as the DC bus for the power amp, plus by using batteries your efficiency takes quite a big hit and life cycle can be an issue. By using some clever capacitor arrangements on the DC bus you can present a dammed clean DC supply with a very low impedance, and no banks of batteries to worry about.

 

There are some new lithium based batteries which don't have as many problems in regard to reaction speed and terminal voltage but then you bring in a whole load of other costs to monitor the cells, not to mention the cost per kw probably being 20-30 times that of lead-acid.

[Dmills]

Synchronous motor, big flywheel, alternator, and some metalwork! I have actually seen this sort of plant in use.

 

Actually, I quite like old school ferrorresonant transformers as a source of smallish amounts of clean power.

They are frequency sensitive however and are inherently current limiting so are probably better employed on

single small critical loads then as a general fix.

 

I did once end up using a MA5000VZ and signal generator to run an American tonewheel organ that turned up with a

band unexpectedly and NEEDED 60hz, hey, it worked....

 

Regards, Dan.

[djrichie_t]

I did once end up using a MA5000VZ and signal generator to run an American tonewheel organ that turned up with a

band unexpectedly and NEEDED 60hz, hey, it worked....

 

Excellent, love it