Low IR Velux windows

[mattg4321]

Anyone had this before?

Tested a large new build the other day to investigate tripping RCD. All circuits IR test ok except loft ring final. 0.58Mohm

First thoughts, this seems a little high to be tripping an RCD every week or so, but it's possible it's getting worse at certain times or conditions so it's my best guess at the moment. RCD ramp tested fine.

My real question is has anyone found low IR on Velux windows before - the type that electrically open and electrically operated blinds. There are around 6 in this loft and one had IR of 0.58Mohm and at least 2 more were under 4Mohm. These are all tested with L&N connected together to earth. The units are less than a year old.

Other possibility is poor joints getting damp where the t&e joins the flex from the velux. This joint has been buried though so who knows. I was testing from the SFCU's

 

[snowhead]

I'd say it's more likely to be condensation in the joints to the flexes which are probably in the cold part of the roof.

 

[Dave OCD]

From memory the flexes that come pre-wired to electric Velux windows are only two core as the motor unit is double insulated ? Could be wrong though.

 

[mattg4321]

They’re definitely not double insulated as I’ve fitted some on a recent job.

I suspect it could be the joints too. Velux is generally quality gear, just wanted to see what others thought.

When I connect them up I insist on them being accessible - usually means on a blank plate being visible if there isn’t a loft - but these have been made inaccessible by the idiots who wired the place.

 

[Dave OCD]

Ah fair enough, must be my dodgy memory. If the windows are positioned in a vaulted ceiling for example I'll connect with a proper MF box with a SFCU fitted at an accessible position.

 

[Lucien Nunes]

Did you try an insulation test at 250V? Do they have electronic controls or just PSC motors with an SPCO switch to control? It is just possible that they have VDRs in the control box that are conducting at 500V. Clearly at 0.58 megs the IR is at least 58 times higher than would trip a typical RCD at 24mA, and that's including the IR of the neutral which doesn't normally contribute to leakage. So whilst a possible indicator of a leakage fault, I agree it's rather open ended.

 

[mattg4321]

I did test at 250v with same results. I always try this in case of SPD extension leads. Although these are usually around 0.33Mohm.

Windows are electronically controlled via app on phone and wireless battery powered wall switch.

I thought it odd how all the readings where different on each window, yet they’re all identical units.

 

[Wilko]

I’m sure I’m still living in the Cold War mentally, but they were wired by builder who has made the wiring inaccessible and they’re playing up ...

 

[mattg4321]

These people have plenty of money. They don’t want the original electricians back as they’ve had more than one problem with them (also wired telephone points in alarm cable amongst other things). They just want me to sort the tripping problem out.