How to create a water proof 12 v strip light

[ians]

This seems to be a bigger challenge than I expected ... much like wifi controlled strip lights ... a huge market for these 2 ... why because people want to control their strip lights using the app they use to control other lights and stuff around the house ... there are product available ... however trying to find one that works and links to various apps ... thats the challenge ... this is more about weather proof 12 V strip in particular.

Why 12 V weather proof strip lights ... some of the projects I am busy with a the moment are in areas classified as zone 0/1 or 2.

You can buy 24 volt IP rated strip lights which are IP rated ... but not 12 v (that I am aware of)

So I am thinking its time to create a environment which will seal against the ingress of water.

This is what I think would be compliant.

First solder a 5 m long tail onto the strip ...

fit the end caps ... Pour a thin layer of epoxy resin along the bottom of the aluminum u channel ... then peal and stick the the strip to the epoxy ... then pour another thin layer over the LED strip.

You now have a completely water tight strip light.

My concern is the heat generated by the LED strip and what effect it will have on the strip over a period of time.

You are probably wondering why I am going to so much trouble ... firstly because of the location ... they are now being exposed to water.

the lights look amazing and lastly it is a challenge (you know how I love a challenge )

To make it a little more difficult the room is only 2.4 m and a concrete slab with slab boxes.

So what is the big deal ... no junction boxes in zone 0/1/2 ... so 2.4m is 100 mm too low ... the volt drop of the 12 v system ( we could go with thicker wires) and the 12V safety transformer will have to be mounted in the junction box in the middle of the bathroom ... about 2.6 m from the shower.

It is all going to look stunning ... but more important safe.

[ians]

The other option would be to fit a clear heat shrink tube over the entire length and then just fill with silicon.

Any suggestions?

[Justloadit]

Provided the heat of the LEDs does not exceed 70°C they will last many hours.
Your challenge is to try one out and have a temperature probe in the center to tell you what the temperature is after 4 hours.
Dropping of the voltage slightly will also reduce the heat build up, just remember that heat is based on square nature curve, so a small drop in current has quite a bit of influence, and the drop in lumanance may not be noticeable.

[GCE]

I have been into Province lighting LED assemble plant where they use a resin to fill the channel to an IP spec for exterior lighting on buildings.
It is a clear resin done in a controlled environment. They make up according to dimensions supplied
If it was me , I would approach them to make up the units - At least there would be warranties etc
They have the experience in that particular field vs you having to pay all sorts of school fees

[adrianh]

I was involved with swimming pool led lights many years ago. We used to make AimFlow lights in the following way:

One circular PCB full of 5mm leds. Another circular pcb board below with a bridge rectifier, voltage regulator etc. The two boards were put into the white or clear Aimflow ball. The ball was then filled with resin covering the pcbs but the tops of the leds were allowed to stick out. If one covers the entire led with resin then the light is diffused too much.

I still have lots of them and their parts lying around.

We made 7 and 9" lights as well.

The whole business of encasing the leds in resin does have problems - the resin tends to degrade and become coated with muck over time. We also had to fit a gland for the wire because the water would seep in along the boundary between the wire and the resin (probably due to different shrinkage rates)

We also did color changing lights with little pic controllers - Although we never developed the ability to synchronize multiple lights I came up with a way to do it after we left the market. One or two big companies pushed all the little guys out so we never took it any further.

[ians]

The reason I am thinking it might actually be better to use a re enterable gel because of the flexibility.

[ians]

The challenge got real ... so I decided to call a friend.

There is a concrete slab above the bathroom ... just to make the challenge a little more difficult ... fortunately I did the conduit installation so there were additional conduits installed for just in case.

The power will be fed from the switch outside the bathroom to relay in the roof for the pump and to a junction box in the middle of the bathroom ... where a 12 safety supply will be installed ... from the safety supply to the RGB controller ... from the controller fed with 3 m of 1.5 mm x 4 core cabtyre to a junction box above the shower where it will spit 2.5 and 2.5 m down to the sides ... instead of using an aluminum U channel ... a 16x16 pvc channel will be used ... sealed against the ingress of water.

This is how any project should be designed ... you read the SANS regulations and take all the points into consideration and create the safest method to achieve the goal.

What have we learnt from all of this ... I believe and it was confirmed that the reason you cant find a 12 V IP rated strip light unless you sell a kidney and even then you would need to take into consideration the amount heat generated by a 12 VDC strip (14W) which will cause discolouring of the resin/epoxy used over a period of time.

This could be why you can only get 24 V and higher IP rated strip lights.

Installing a 24/230 IP rated strip light around the inside lip of a pool is not legal and just because the lighting supplier or customer tells you it is doent make it right.

The reason we spend years training to become sparkies is so that we know better ... and that people is going to be what the judge will point out if you are ever put in the witness box.

Make sure you are a member of the CYA club ... dont worry about the DOL and ECA and ECB and all that crap.

[ians]

Just when I thought I had figued it all out ... try find a 230/12 VDC constant voltage driver with an isolation transformer.

swimming poo lights are 12 VAC ... so it easy to find a 230/12 VAC isolation transformer ... now I need to figue out how to make the 12 VAC side DC and constant voltage.

A question ... why in the sans regs do they refer to low voltages AC 12 VAC and 50 VDC.

If you look at it like that it means you could use a 24-50 VDC light in zone 0 ?

I tried phoning around and spoke to a few lighting suppliers ... it made me realize how little people know about all this type of stuff.

Please correct me if I am wrong ...but an IP rated transformer doesnt make it a isolation transformer.

It seems that most people feel that you can throw a 24 VDC light into a pond ... so long as it has an IP 67 rated driver ?

[GCE]

Just when I thought I had figued it all out ... try find a 230/12 VDC constant voltage driver with an isolation transformer.

swimming poo lights are 12 VAC ... so it easy to find a 230/12 VAC isolation transformer ... now I need to figue out how to make the 12 VAC side DC and constant voltage.

If the driver is marked as SELV then you can use it as long as it is outside the zone 0

Have a look at the previous thread you had going

https://www.theforumsa.co.za/forums/showthread.php/27660-Water-features-ponds-and-strip-lights

[Justloadit]

Contact Mantech, they have a bunch of Mean Well Isolated constant current LED supplies..
If you check the data sheet, it shows that it is isolated.

They also have it in 24V and 36V DC.
They used to have a better range in the past.

[ians]

Contact Mantech, they have a bunch of Mean Well Isolated constant current LED supplies..
If you check the data sheet, it shows that it is isolated.

They also have it in 24V and 36V DC.
They used to have a better range in the past.

Something I noticed in the data sheet was reference to a class 2 power unit.

Fixed appliances
(e.g. luminaires,
underfloor
heating, mirror
heating, towel
heaters )

zone 1 and 2 refers to B2 - denotes that class II appliances shall be used;

some interesting information with regards to appliance class types

https://www.inspec-bv.com/what-are-the-different-class-types-for-appliances-class-i-appliances-class-ii-appliances-class-iii-appliances

For those who don't know ... this is the symbol for an isolating transformer

https://www.iso.org/obp/ui#iec:grs:60417:5947

[ians]

I tried opening the spec sheets for other popular brands ... not much to share compared to the mean well drivers.

When you open the mean well technical data sheet ...

All the info you want to see ...

Certifications

symbols for double insulated

Isolating transformer

SELV

type class 1 or 2

even dimming options.

[Justloadit]

Just a point of interest with respect to Class 3 or isolating transformers, that I have observed.
Assuming that the using an isolating transformer makes the secondary circuit safe can be a grave error, which is highly dependent on the manner in which the isolated circuit is being used.
This is the reason that CTs and isolating transformers have one terminal earthed.

If you are interesting in understanding the reason for this then read on

I have noticed this more and more in modern day circuits due to the higher frequencies that are used in electronic power supplies.
Effectively there is an electric connection between the primary and secondary, albeit it extremely low. Simply imagine that the primary of the transformer is a capacitor plate, the secondary as the second plate. Any material between the plates will act as a dielectric. Effectively the transformer is a very low value capacitor.

A description of a capacitor is as follows :-
A capacitor is a device that stores electrical energy in an electric field. It is a passive electronic component with two terminals.

A description of a dielectric material is as follows :-
The capacitance of a set of charged parallel plates is increased by the insertion of a dielectric material. The capacitance is inversely proportional to the electric field between the plates, and the presence of the dielectric reduces the effective electric field.

The value of the capacitor is as follows :-
Capacitive reactance is measured in ohms of reactance like resistance, and depends on the frequency of the applied voltage and the value of the capacitor. The symbol for reactance is X. To specify a specific type of reactance, a subscript is used.

Reactance of a capacitor is calculated by using the formulae 1 / (2 x Pi x Frequency x capacitor value)
The component which is interesting in the formulae is the part - the 'Frequency', the higher the value, the smaller the 'Reactance' or resistance to the charging voltage.

What is interesting to note, that when we think frequency , we always think this nice slow rise of time to the crest of the wave, which is what we experience with our 50Hz mains sine wave, this is how we measure the frequency, from zero to the crest is 25% of our frequency. This time is what charges the dielectric material. Depending on the material, it may charge well or it may charge poorly, non the less it does charge. It then discharges on the down cycle towards zero, and charges in the reverse direction to the negative peak and then discharge to zero on the following last quarter cycle.

But what happens when there is a surge or lightning strike in the vicinity of our circuit?

The rise time is almost instantaneous, and have been measure to rise from a zero value to a few thousand volts in a few hundred nanoseconds, (that is 10 to the power of minus 9 of a second, or 0.000000001 of a second). From the previous description of the capacitor charging in the first quarter, one can then calculate the frequency being in being in the Mega Hertz range. Applying that value into the reactance of the capacitor formulae, it immediately changes the reactance value to very low resistive value. This is where the danger occurs. Usually lightning or surges do come in thousands of volts, the secondary circuit of our isolating transformer could charge at a high voltage if you happen to be touching the equipment and ground. Similar to static electricity spark you build up when walking on a carpet and then touching some one else or a door handle.

With the higher and higher frequencies being used in SMPS (Switched Mode Power Supplies), the transformers are getting smaller, which means that the Primary and Secondary windings are getting closer to each other, effectively reducing the distance between the windings, increasing the dielectric value, which reduces the 'resistance' between the windings. To counter this and to maintain the "galvanic" isolation of the circuits, manufacturers place a small capacitor on the one side of the secondary circuit to earth.

I am sure many of you have experienced the mild shock when you touch a lap top and a table top PC chassis plate when connecting cables. This is part of the charged dielecric which has nowhere to go until a you create a path for it.

[ians]

It is possible to get a double insulated (not that you need double insulation on the transformer) ... isolating transformer with the some form a safety rating ... which can produce enough power to supply 5 m of 12VDC strip light.

Make sure you take note 12 VDC will have a volt drop which must be taken in to consideration if longer than 5 M ... especially if the power supply is mounted away from the strip.

From the little research I did ... I can only assume the reason it is difficult to find a IP rated 12 VDC strip light ... is due to the high current and thus ... heat generated ... hence the 24 VDC strip being the option offered by all the lighting suppliers.

Was it worth all the time to get it right ... this is what I really enjoy about my job ... looking back at the finished result makes it all worth it.

Going back to a site I completed 12 years ago ... and everything is still like I left it ... all the DB's are labelled ... the wires are labelled ... as built sketches and wiring diagrams ... made it so simple for me to identify circuits ... everything is neat and tidy.

Take pride in your work ... it feels good when people walk into a building and the first question they ask the customer is who did all the cable trays and cabling .

[Justloadit]

You can get double insulated transformers, in fact you can get safety transformers where the primary and secondary are wound on separate parts of the bobbin, so if the primary wire heats up and fails and melts, there is no condition in which it could short to the secondary.

This works for 50Hz transformers, but does not bode well with high frequency transformers.

Isolating transformers are wound, primary first, then a copper sheet is placed over the primary, and carefully not shorting out around the transformer, or else it could work as a single turn secondary, and then connected to the earth. Then the secondary wire is wound over the copper sheet. Under this construction, under no condition, could the primary ever short to the secondary.

The problem with the 12V AC transformers, depending on the manufacturer, the 12V secondary output voltage can vary by more than 25% when loaded and unloaded. Cheap transformers usually suffer in this way. An proper design transformer is expensive, as more copper is used to ensure a constant 12V output over the operating range.

Now the question is why is it a problem if the voltage varies?
Well LEDs work with D.C. voltage.
DC voltage rectified from the mains is voltage times 1.414 to give you the peak of the sine wave.
So 12V AC will yield 16.97V peaks to the LED circuit.
So lets give it a 25% over voltage, then 15V will yield 21.22V, that the LEDs will now have to contend with.
Manufacturers may or may not accommodate for this, but it costs money, and knowing that customers are not prepared to pay for product, opt out to go the cheap way. They use a current limiting resistor to attempt to control the voltage. The problem is that resistors have a fixed value, and the current increases with the voltage, following Ohms Law, Current = Voltage / Resistance. So looking at our LED circuit, depending on the transformer and the load, the LED will see higher or lower currents depending on the circuit design. The problem is that Power follows is a square function, so that 25% over current makes a huge difference to the LED, it heats up very quickly. The secret is to maintain a straight line luminescence on the LED over a range of voltages, and keep with in the LED power rating. Another problem is that Red LEDs use far less current than Green and Yellow, so this adds to the mix of varying voltages, especiall6y when doing dimming.

Invariably cheap transformers are used with current limiting resistance LEDs, and we have a recipe for failure.

Another way to get some control, is to use a DC SMPS power supply, which has a limited voltage output control, where the user can lift the voltage slightly to accommodate the voltage drop.

Better control would be to use AC to supply the strips, and in each strip there is a SMPS to control a fixed length strip. This would reduce the amount of heat at the transformer, PCB tracks as each strip would dissipate its heat generated, there by improving performance. The heat is shared by every strip. But alas price is what customers want, so cheap solutions are employed, and we live with the consequences.