What Do You Think of This Circuit?

[SUNRAY]

I'll repeat my earlier warning.

Usually the manufacturers instructions will show the acceptable connexions, if they don't show a bypass switch, it will be for a reason and I would not try it as I have encountered failure for this reason.

I know it doesn't make sense but learning by experience overrules guessing.

 

[SUNRAY]

Would they make a PIR which could be damaged in such a way when wiring similar to that is common-place?

YES

 

[333rocky333]

YES

Sometimes the instructions show you can Parallel up more than 1 device , if so then would that in turn verify there safe

 

[SUNRAY]

That may stop the RCD from working, have you tried it? and did the RCD still work? and what type of RCD was fitted? I have read conflicting reports on type AC RCD's and DC current. The test rig I saw used had pure DC, and clearly that is different to half wave rectified DC, they also used a three phase RCD and fed the DC through the unused phase connection, so not sure if they really stop working or if the test rig was designed to worry those who watched the video.

I have wired a farm house with DC lighting as it was on a generator, and did use a diode matrix so any bedroom light worked hall light so one knew if bedroom light left on, but no real need in the UK.

As yes have found PIR units which did not work correctly if live fed to both sides, it did not harm them, just stopped them working, the main thing is where there is no neutral to switch and it needs a slight volt drop across the unit to power it. Same applies to all electronic light switches.

1/2 wave rectified is still close enough to AC, it may modify the performance of the RCD but theres no reason for it to stop it functioning.

Items like this do exactly the same thing, albeit to operate a relay, and I have used them to do the same thing with AC commercial lighting circuits
https://www.foxtamcontrols.co.uk/assets/img/117-YWRICD-YWA9-YWA15-11WA8.pdf

 

[SUNRAY]

Sometimes the instructions show you can Parallel up more than 1 device , if so then would that in turn verify there safe

Yes this is a better way of describing my 'bypass switch' statement.

 

[ericmark]

1/2 wave rectified is still close enough to AC, it may modify the performance of the RCD but theres no reason for it to stop it functioning.

Items like this do exactly the same thing, albeit to operate a relay, and I have used them to do the same thing with AC commercial lighting circuits
https://www.foxtamcontrols.co.uk/assets/img/117-YWRICD-YWA9-YWA15-11WA8.pdf

This has come up before see here and we were directed to

this video and it is due to what was talked about in that thread which makes me raise the question about RCD protection if using half wave rectification. Since many countries have banned type AC RCD's in many countries this would not be a problem, and many items now stipulate type A RCD must be used, think Bosch central heating is one of the items.

If the thread had not been started by @securespark I would not have muddied the water, most DIY people would not even know there are different types of RCD, however @securespark has shown over the years he does know what he is doing, however in my own house I ended up with the wrong RBCO I read what it said on the box, it said type B however latter realised it was curve B type AC I had got fitted.

In the thread there is also a University report on type AC RCD's which seems to reverse what the u-tube video shows, so I am really not sure how bad the DC really is, I note there are many types of RCD, AC, A, B, F etc, if type AC could do all, question must be why are the others made?

 

[securespark]

Sometimes the instructions show you can Parallel up more than 1 device , if so then would that in turn verify there safe

I have very often used two PIRs to switch the same loads and never had an issue!
Similarly I have fitted override switches with (seemingly) no ill-effects.

 

[SUNRAY]

I have very often used two PIRs to switch the same loads and never had an issue!
Similarly I have fitted override switches with (seemingly) no ill-effects.

Yes and so have I and probably most others in this thread. But I have also encountered those which cannot be used that way which is why I said to look at MIs.

 

[securespark]

Will do that, thanks.

 

[aptsys]

In the thread there is also a University report on type AC RCD's which seems to reverse what the u-tube video shows, so I am really not sure how bad the DC really is, I note there are many types of RCD, AC, A, B, F etc, if type AC could do all, question must be why are the others made?

Any DC bias on the AC line is a very unusual circumstance caused by a fault - A properly working piece of equipment should not be injecting DC current.

People seem to associate SMPSUs with this, particularly car chargers for some reason, but they would not get approvals if this was the case. The better power supplies have power factor correction which basically makes it impossible to occur anyway.

 

[bernardgreen]

Descriptions of different types of RCD HERE


Waveforms of currents at the sensor due to faults to Earth are shown HERE

 

[plugwash]

People seem to associate SMPSUs with this, particularly car chargers for some reason, but they would not get approvals if this was the case.

It would only take a single open-circuit diode for pretty much any SMPSU to create large DC currents in the line and neutral and this may well not be noticeable to the user.

In terms of earth faults, if there is an earth fault after the rectifier but before the isolation/voltage conversion then that earth fault will be pulsating DC.

As for why car chargers are a bigger concern than most SMPSUs I can think of a few reasons.

1. Cars are typically metal bodied and are outside the equipotential zone. This creates a conundrum when earthing them, connecting them to a TN-C-S earth is considered undesirable because they are outside the equipotential zone but a TT earth means you are totally reliant on a RCD for earth fault protection.
2. Car chargers are much higher power than any other SMPSU likely to be found in most domestic environments.
3. Cars are a pretty hostile environment for electronics.

The better power supplies have power factor correction which basically makes it impossible to occur anyway.

No. There are basically two types of power factor correction for SMPSUs. "passive PFC" puts an inductor in series to slow the changes in input current, active PFC adds a special boost converter between the rectifier and the main input capacitors to allow the current waveform to be controlled electronically.

Neither of them will stop DC currents from happening under reasonably foreseeable fault conditions.

 

[aptsys]

Waveforms of currents at the sensor due to faults to Earth are shown HERE

Those diagrams aren't very representative of real world situations once in circuit and in an appliance. Number 8 is plain incorrect. Yes you can recreate those, the likelihood is vanishingly low except on actually damaged equipment. General purpose power supplies will only be susceptible on the HV rail, which exists only on the primary side.
Many of those faults would result in overcurrent protection operating.

 

[aptsys]

It would only take a single open-circuit diode for pretty much any SMPSU to create large DC currents in the line and neutral and this may well not be noticeable to the user.

How?

No. There are basically two types of power factor correction for SMPSUs. "passive PFC" puts an inductor in series to slow the changes in input current, active PFC adds a special boost converter between the rectifier and the main input capacitors to allow the current waveform to be controlled electronically.

Neither of them will stop DC currents from happening under reasonably foreseeable fault conditions.

Which fault are you thinking of specifically?

The current is drawn over the entire AC waveform with an active PFC rather than just a pulse to charge the HV capacitor. You don't see the same DC biased current waveforms.

Sure you can create situations, but there's a whole multitude of things you can do to create situations that are hazardous but low in probability in real life. Certainly when we design SMPSUs we have to take into account the risks, but on the risk matrix they end up acceptable because of the mitigations in place. The simple schematics drawn in those examples are fantasy compared to what is actually implemented to meet the standards.

 

[plugwash]

In a SMPSU typically the first components are protection. After that is typically a bridge rectifier*.

In normal operation the bridge rectifiers behavior is symmetrical, so there are no significant DC components in the current draw from the supply.

But if one of the diodes in the bridge is open-circuit then no current can flow during one of the two half-cycles. This will mean that all the power the SMPSU needs will be drawn during the other half-cycle and the input current will have a substantial DC component. Depending on how much margin the design has this may or may not be noticeable to the user.

\* Some older SMPSUs when configured for 120V input use a voltage-doubling rectifier configuration, but I have never heard of a PSU doing that for 240V input.