SMPS drawing 5mA DC from the mains?

[cupoftea]

Hi,
We are doing a trailing edge controlled heat controller for a shower.
The zero cross detector doesn't give exactly the same (symetrical) pulse timings for
positive and negative half cycles, and so the current pulse is slightly different between positive and negative
half cycles. As such, there is more than 5mA DC in the mains due to use of our equipment.

But to be honest, any 1kW+ offline SMPS is fed by a diode bridge etc, and inevitably the current draw
is more than 5mA different for positive and negative half cycles, and no one complains about that?

So can we overlook and ignore our >5mA DC drawn from the mains?

I mean, is the >5mA DC law actually enforced anywhere?....if it was then every cheap Triac lighting dimmer
in the world would have been banned by now as they all give non symmetrical
current pulses between positive and negative mains half cycles.

I mean, check the mains on a scope....the flat topping shape is not exactly the same for pos and neg half cycles,
so obviously there is DC in the mains anyway.

Problems caused:
I understand that the biggest problem (of DC in the mains) is saturation of distribution transformers?
This "corrosion of copper pipes" thing, is a myth? Theoretically it would happen if there was DC in the mains, but not in reality?

Equalising SMPS:
Supposing the ">5mA DC" thing is enforced, then is it acceptable to add in an equalising SMPS which ensures that the overall DC drawn by any converter is
overall <5mA DC?
...And if such equalising converter is allowable, then what period must it operate over?....eg supposing it just about ensures that over any 1 second interval, the DC is zero (so the DC may well have been >5mA for eg, a particular half second period).

3 phase mains:
What if you have a 130W offline SMPS that is supplied by a star_3_phase supply with a 3 diode rectifier? (ie each phase will then have DC in it, since its three single phase rectifiers, one on each phase)

 

[aliarifat794]

The primary concern with DC in the mains is the saturation of distribution transformers. When a transformer core is exposed to DC, it can become magnetically biased, leading to overheating, reduced efficiency, and potentially even damage.

 

[cupoftea]

Thanks, so if you just want 130W supply and you have a three phase star supply...then you cant do a 3 diode rectifier and use all 3 phases....but instead, due to this, you must only use one phase and with a full wave rectifier?
...ok, i agree that then means you dont draw DC, but then you are drawing imbalance from the mains (because you are only loading one phase).....and that has more potential problems than dc in the mains?

 

[betwixt]

We are doing a trailing edge controlled heat controller for a shower.

Do we assume from this you are using a triac as the control element?
Although 5mA is probably insignificant in the grand scheme of things, where is the difference in zero crossing coming from. I can understand a triac having slightly different voltage drop in each direction so there may be an overall difference in current it passes but I can't see why that would cause the ZC point to be asymetrical.

Brian.

 

[cupoftea]

Thanks, most ZCD circuits give an asymetrical zero cross...unless they are the really simple, inefficient type using just a double diode and resistor across the mains.
The reason they give asymetry is because most ZCD circuits are made so that they dissipate little power over 90-265vac...and when the cctry is added to achieve this, you end up with the asymetry. None of the low power ones are symetrical.

 

[KlausST]

Hi,

5mA is ignorable for a distribution transformer, especially on the secondary side.

***
Do you know these automatic coffe machines (Italian style). They usually use vibration pumps to generate this around 6 bar of water pressure.
They all work with a diode in series. They only draw current on one half wave. This is much more DC than your 5mA. And they seem to be legal.

Klaus

 

[BradtheRad]

Here in the US we have 120VAC, two live wires, phase inverted. Jointly these two each take their share of house outlets. 240VAC is available by putting the appliance between the 120VAC wires. As I've heard it, electricians are asked to 'roughly' balance usage between those 120V live wires, by giving them equal shares in the breaker box. Whether AC-specific or DC-specific, It seems inevitable we'd have some amount of imbalance hour-by-hour, minute-by-minute.

 

[cupoftea]

Thanks, so if your 3kW, 3 phase, star-supplied equipment needs also a 130W offline SMPS...what do you do....?...
1...Use a 3 phase, 3 diode rectifier and use that as input to the 130W SMPS?
2...Just use 2 phases in a full wave diode bridge as input to the 130W SMPS?

(no PFC is needed as overall its a 3kW, 3 phase product and they dont need PFC)

(1) puts DC in the mains and (2) doesnt.
But (2) puts more ripple current in the post rectifier smoothing capacitor.

 

[Easy peasy]

5mA DC ave is a commonly applied standard to keep earth stake corrosion to a minimum, a bridge rectifier usually ensures this - exceptions are for unbalanced synchronous loads

 

[D.A.(Tony)Stewart]

Hi,

5mA is ignorable for a distribution transformer, especially on the secondary side.

***
Do you know these automatic coffe machines (Italian style). They usually use vibration pumps to generate this around 6 bar of water pressure.
They all work with a diode in series. They only draw current on one half wave. This is much more DC than your 5mA. And they seem to be legal.

Klaus

Geomagnetically Induced Currents (GIC) topic and magnetotelluric effects have been a hot topic since Quebec Hydro observed the effects of solar flares in ‘91.

Over 100 Amps had been recorded which cause severe damage to 1GV HVDC lines and a widespread east coast network blackout.

It would be interesting to know which networks now use DC blocking mega caps to ensure network stability and if all the morning espresso machines with their diode half-wave DC solenoid h2o pumps mimic a GIC every morning

the former might obfuscape the latter effects which would integrate slowly to create asymmetric harmonic distortion from saturation, and loud audible buzz at fo repetition rates that might sound like all the espresso pumps in the neighbourhoods.

I wonder if/how the core fluxes slowly balance themselves to avoid saturation by soft limiting.

 

[KlausST]

@Tony:

You say "hydro" ...
you mean a hydro power plant?
If so, I expect distribution transformers to be inbetween power plant and load. In this case the load_introduced_DC can not hit the power plant, since the transformer blocks DC.

Klaus

 

[D.A.(Tony)Stewart]

@Tony:

You say "hydro" ...
you mean a hydro power plant?
If so, I expect distribution transformers to be inbetween power plant and load. In this case the load_introduced_DC can not hit the power plant, since the transformer blocks DC.

Klaus

Quebec Hydro manage all electricity from the end to end. They are world leaders in research and network control . Thus 5mA DC is a non-issue. Yet there are many large DC and extreme LF disturbances.

If one reads their R&D or similar papers on causes and mitigation they can discover more details and avoid jumping to false conclusions.

Harmonic distortion is one of side effects from DC saturation on one peak which does propagate thru the network.