but no success so called out installer. He couldn’t see anything wrong but managed to reset the trip. HP OK after.
(10th Jan) Tripped some time between 06:40 and 08:10 during post-setback heat up (so heat pump at high duty - clue, I thought). Managed to reset the trip after 3-4 tries. Operated OK after that.
(Today 14th Jan, 10:30). The house lights flickered off-on briefly and the heat pump stopped (surge?). I was able to reset it again (2-3 tries), and it’s been OK since. This time I was actually monitoring the heat pump with SNET-Pro2, and will post the resulting spreadsheet when I have tidied it up.
The only other thing of note is that I’ve had 4 defrosts today between 07:00 and 10:00, and I’ve never seen more than 1 per day on this HP: (You can see the trip at 10:35, and it took me 10 minutes to reset the trip - and remember to reset the roomstat - to get things going again.)
I presume you can’t tell whether it is tripping on earth leakage or overcurrent? Is there a lot of condensation anywhere, which might give you an earth fault?
I haven’t seen anything like this with my Samsung, but I replaced the Cheil ELCB with an MK RCD. I still have the Cheil one somewhere and I can happily send that to you if your installer wants to try a replacement test.
The ELCB is looking for an imbalance between current in the Live and Neutral, so anything that causes a leakage to Earth above the trip level could be the problem, condensation on PCBs and wiring, if the immersion is wired in the same circuit and is old then these often develop leakage paths.
It might just be failing. An electrician will have a piece of test kit that can determine if the device is OK.
I have seen RCDs trip on transients. It’s usually associated with kit that contains EMC filters where there is a path to Earth by design. The Samsung outdoor and indoor units will almost certainly have such filters for susceptibility and emission mitigation.
As far as I can tell from the Korean data sheet, it’s an ELCB, not an RCBO, so an RCD in UK parlance.
The data sampling rate isn’t great, and may have skipped the moment in question, but at 10:35 when the trip occurred, I see nothing to suggest overcurrent (the starting currents for the preceding defrosts were much higher). Also, the controller, although in an unheated garage, is fairly well ventilated (I keep the cover off nowadays, so I can get to the trip reset easily…).
As @toadhall suggests also, I had been starting to suspect that the Cheil was faulty, but at £120 or so a pop I wasn’t going to rush out to buy a replacement. I’ll see what Samsung Helpdesk say, but at the mo I’m not a happy bunny…
At least I know it’s not that - I isolated the da*n thing years ago.
My partner (a sometime electrical technician) asks whether the MK RCD is a simple drop-in replacement (same 70mm CC screws)? A drop-in MK would certainly be cheaper than finding (and trusting) a local electrician to test the Cheil…
Unfortunately I was only logging at 30s intervals, but some time between 10:35:48 and 10:36:18 the HP went into “CompDown” mode and threw some interesting Error Codes at me. The 466 implicates inverter DC Link over/undervoltage (the others are just comms errors), so I’m wondering if I’m the hapless victim of grid voltage swings? (I know we were nudging 260V last summer, due partly to my own PV panels trying to squirt output into an undersized local system…)
[Edit: A cursory glance at the DC Link data (column U) blows the above theory out of the water, so the search for an answer goes on…]
No, unfortunately not. The stuff meant for the UK is all designed to sit on the DIN rail found in Consumer Units, so no mounting screws holes. Mine is in a separate Consumer Unit dedicated to the heat pump. If you have access to an electrically competent person (so to speak) it would be relatively straightforward to screw a DIN rail offcut in place of the Cheil ELCB and clip a standard UK RCD onto it. If you decide to do this, make sure it is at least a Type A RCD, and of course your warranty would be probably be invalidated.
The first thing I noticed were these earth wires. Might be nothing, I’m not even sure if this would cause the issue you’re seeing (I’m not an electrician), but it’s probably worth checking there isn’t a loose connection here. Sounds like your partner is switched on enough to have noticed this and would be able to tell if it’s an issue, but it’s worth a try.
Thanks @jakeymd1.
Those earth wires might look dodgy but they are firmly attached.
And yes I do have the old PCB (Samsung checked it out), but I’m reluctant to get it put back except as a last resort (I’d want my installer to do it, and that would cost me…).
The Helpdesk has had me go through all the components on the WRC Self-Test screen, and I found no bad actors, so the puzzle remains.
Having replaced the Honeywell T3R roomstat with an EPH CP4v2 (both wireless), the problem seems to have gone away. I got the clue from the MIM trip that occurred as I loosened the T3R receiver cover, and suspect that the (ageing) receiver had developed a fault in the MIM output circuit.
.
