Waterproofing the U/V electric assembly of a pond filter.

[Chainhurstman]

Three of my older pond filters have an integrated U/V assembly in the filter lid, which is raised well above the level of the water in the filter and therefore never intended to get wet (at least, all the while that the lid pipework itself remains sound). One, unfortunately, overflowed, so water entered the lid and the electric assembly tracked down to earth. It has remained out of use ever since, but I decided to fire it up today to see if it had recovered from its involuntary immersion. It appears to have done so, and the RCD has so far not tripped. However, I am intending to leave this filter when (if !) we sell the house, and I would like to do my best to ensure that the purchasers do not experience any problems with it. As the unit has already tracked down to earth once, I would like to improve its impermeability if I can, and I was considering using one of the modern "nano" products such as Nanoprotech Electric. I would therefore be interested to know if any forum member has experience of this or of similar products, and if so, whether they would recommend it for the purpose I have in mind. Also, might there be any benefit in using instead Nanotech Pro Electric which is considerably more expensive. I note, by the way, that both products shew as "out of stock", which does not bode well ...

 

[Midwest]

I have a small water feature, with a few fish in. To my cost I used the Hozelock Bioforce range. End up replacing them every few years, cos' they fill up with water/condensation. The purchaser probably won't care about your (beloved) pond, so I wouldn't spend too much money trying to resolve the issue. Mines getting ripped out when I try to sell.

 

[Chainhurstman]

Ah yes, the Bioforce -- that is one of the "all-in-one" pressurised units, if I remember rightly. I very nearly made the mistake of buying one of those, just because it looked so neat and elegant, but fortunately I could not afford it So now the pond is filtered with an Oase Biotec 10.1 (which I will be taking with me) and I plan to leave a Green Genie 24000 (the one with the compromised electrics) and a Green2Clean 12000. Assuming, that is, that the buyers want to continue with the pond.

 

[LeeH]

2 x 150ml Bottles of Magic Gel | City Electrical Factors - http://www.cef.co.uk/catalogue/products/4319079-2-x-150ml-bottles-of-magic-gel?q=Gel

This may be suitable.

 

[Chainhurstman]

2 x 150ml Bottles of Magic Gel -- This may be suitable.

That looks good (fillings degree of protection IP68). Thank you, Lee.

 

[Marvo]

Maybe Maplin PCB Lacquer | Maplin - http://www.maplin.co.uk/p/maplin-pcb-lacquer-n67an

Not sure if it's UV-C stable though.

 

[telectrix]

just paint the pcb with yacht varnish.

 

[Chainhurstman]

Thank you for those suggestions. In practice it is not just a PCB, as the U/V tube is physically separate and joined to the PCB with flying leads and electrical end caps, which is why I was initially thinking in terms of a spray-on insulating medium. But Marvo's point is something I hadn't considered, and clearly should have done. If my chosen insulating medium isn't UVC-stable, the whole thing will clearly be a waste of time. More research needed !

 

[Chainhurstman]

Well, I spoke too soon; it started raining here about an hour ago, and a few minutes the ring RCD tripped; I reset, and it tripped again after a couple of minutes; I removed the U/V from the circuit, reset a second time, and it has not tripped again. Clearly it needs urgent protection against water ingress. What is slightly odd is that it is upstream RCD that tripped (the one for the ring involved) rather than the downstream spur RCD. Ah well ...

 

[Marvo]

...If my chosen insulating medium isn't UVC-stable, the whole thing will clearly be a waste of time. More research needed !

UV-C is highly corrosive to many materials, even materials that could be normally be considered UV stable which usually means they're okay in occasional sunshine. These a big difference between a UV-C light source and partial sunlight. The high ozone concentrations produced by UV-C tubes can also be another issue....

Looking at a picture of this thing there's no controls, just a ballast, starter and tube. I'd be tempted just to buy a replacement electronic ballast and dry out the tube with a hairdryer. There's not much else can cause problems.

 

[Chainhurstman]

Thank you for your suggestion, Marvo. Yes, I think that you are right -- as there is no guarantee that the purchaser will have any experience at all of what to do if the U/V trips the RCD and all power to the pond is thereby cut off, I should probably play safe and replace the whole electrical assembly. It's not too expensive (circa £40 inc. p&p) but money is rather tight at the moment and I was hoping to be able to waterproof it for somewhat less ...

 

[Octopus]

What is slightly odd is that it is upstream RCD that tripped (the one for the ring involved) rather than the downstream spur RCD. Ah well ...

Which is what normally happens - and why having 2 RCD's of the same value, in series, is pointless...

 

[Chainhurstman]

True, but required by regulations, is it not ? The upstream one protected a domestic ring, the downstream one protected an outside spur.

 

[Octopus]

True, but required by regulations, is it not ? The upstream one protected a domestic ring, the downstream one protected an outside spur.

It doesn't work like that. RCD at fuse board covers circuits downstream

 

[Chainhurstman]

OK, thank you for the clarification.

 

[Chainhurstman]

Managed to locate an exact ballast replacement at only €9.48 inc. p&p so almost certainly cheaper than any waterproofing solution (so long as it is the ballast that is tracking down to earth rather than some other component). Of slight concern, though, is that Tridonic (the manufacturers of the ballast unit) recommend a 4uF capacitor in parallel for PF correction, but the unit as supplied by Lotus has only a 0,047uF. What should I infer from this ?

 

[Marvo]

Probably that the Tridonic ballast is wire-wound and highly inductive whereas the new replacement ballast may be electronic and less inductive. Any PFC capacitor will be directly across the supply so make sure it's an 'X' type.

 

[LeeH]

Of slight concern, though, is that Tridonic (the manufacturers of the ballast unit) recommend a 4uF capacitor in parallel for PF correction, but the unit as supplied by Lotus has only a 0,047uF. What should I infer from this ?

Someone doesn't know their Farads from their elbow?

 

[Chainhurstman]

They are identical model numbers, Marvo, so I (rather naïvely, perhaps) assumed that they would be identical physically. Looking at the construction of the old one, I would conjecture that it is wire-wound (physically heavy, what appear to be two separate inductors sharing a common core and perhaps some integrated electronics that cannot be seen), and this is the one that has only a 0,047μF capacitor across line and neutral where they enter the assembly. When the replacement arrives, I shall report back as to whether it is indeed identical or not ... The spec. sheet is online, and the exact model is an EC 30 C502K.

Lee : Slightly worrying, if you are right !

 

[Marvo]

Ahh, I didn't realise it was a like-for-like replacement in which case any PF correction should be the same. I wouldn't worry too much about it, for warranty purposes just install the new ballast together with whatever capacitor is supplied with it.