fixing design flaws in commercial products.

Help Support GroupDIY:

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
21,494
Location
Hickory, MS
I have mixed feeling about this. I enjoy problem solving, but I don't enjoy having to fix new products I recently purchased.

Long story only slightly less long, I have been using a RO water filter for decades because the chlorine(?) in my tap water kills beer years. Fast forward to about 6 months ago... I woke up to a flooded kitchen floor. It took me a while to figure out exactly what caused my leak.

The under sink water filter industry uses quick connect press in plastic tubing fittings. These are generally reliable but I discovered an unrecognized weakness the hard way. Both press in tubing lines going to and coming from my UV disinfection unit stopped holding. 1632848174238.png
The retainer clip on the left is a new one with all four fingers intact, that lock the tubing in place. The two gray clips are from my old UV unit. The one with two finger left just barely held the plastic tubing in, the one with no fingers left didn't hold at all. Upon reflection I connected a few dots like how quickly plastic containers left outside in direct sunlight become brittle and fall apart. I expect I had the perfect storm of using clear plastic tubing in and out of my UV unit. The clear plastic tubing allowed enough UV light to degrade the plastic clips (UV light breaks polymer chains inside plastic). If I has used any other color tubing than clear, my clips would probably still be holding.

1632850794775.png
My overkill solution was to purchase a UV unit with a stainless steel housing. The most desirable feature for me was ability to use proper brass fittings that should laugh at UV light. It turns out that this particular SKU was a cheap Chinese knock off of a successful American product, I didn't realize this at the time. My new drama was that the housing made by the Chinese knockoff factory threaded the bosses with 1/4"-18 straight pipe thread, not the correct tapered NPT (T stands for tapered). My first sign that something was amiss was that the brass fittings refused to screw in all the way, and I could see fine slivers of brass, cut by the un-tapered steel threads in the housing. My next move was to purchase a proper 1/4"-18 NPT tap to test out my theory because I was still speculating about the problem.

Sure enough the NPT tap opened up the threaded boss to interface with commercial off the shelf NPT fittings properly, while trying to force the brass fittings into the steel housing trashed the threads.

I suspect the water filter industry still considers me an outlier with a personal problem, and I concede I have had decades of good experience with quick connect push in fittings, except for this one case of UV exposure. The plastic NPT fittings can be forced into the straight pipe thread and probably not leak if only done once.

For even more overkill I just upgraded the electronic ballast powering my UV lamp to one with a flow switch so it only powers on when water is flowing. The energy savings from cycling off a 6W bulb is not significant, but now my RO water is no longer warm from 24x7 UV exposure. This will surely improve bulb life (I hope***) and eliminate any possibility of UV damage to plastic bits.

JR

*** I love the internets... so much information at our finger tips. Researching UV lamp life, apparently UV lamp output left turned on continuously degrade (loses UV output) over time. Further short cycling on/off frequently is bad for lamp life. The filaments in UV lamps are probably a little like incandescent bulbs stressed by repeated thermal cycles. Reportedly electronic ballasts are kinder to UV lamps than inductor ballasts. So I guess I will see about lamp life. My life is one long scientific experiment.
 

NOON

Well-known member
Joined
Nov 25, 2014
Messages
172
Location
Melbourne, Australia
Do you have fluorescent tube style UV lamps? Frequent on-off switching will kill them fairly quickly as the electrodes get stripped a bit each time it strikes an arc at starting. You can get soft starters that make the process a bit gentler, usually built into electronic ballasts but you can also get soft starters for inductive ballasts.
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
21,494
Location
Hickory, MS
In an ideal world, the dedicated electronic ballast designed to turn the UV lamp on/off from a water flow switch should anticipate a higher number of on/off cycles than typical. This is a relatively new feature for this product category so I won't presume that they anticipated lamp life issues.

From observation I see an orange colored light coming from the original energizing of the filament before it shifts to fluoresce blue. I don't know what a soft start looks like. From another web search, soft start involves heating the filament before striking an arc so maybe it is already soft start.

This is yet another thing I have no expertise about.

I'll see how it goes...

JR
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
21,494
Location
Hickory, MS
I realize that far UVc (short wavelength) LEDs are still unobtanium but regular wavelength UVc LEDs are available so that might be a robust solution for water disinfection. Looking at bare UVc LED chips on Digikey they look pretty expensive ($3 to $10 ea).

JR
 

Latest posts

Top