Strange 24/12V system behaviour with dimmable lighting

[alexhibbert]

Hi all, I wonder if an electrics expert could help.

I have designed and installed my boat's electric system, which mostly supplies power for engine start, lighting, external cameras, and ducting fans. Everything so far has worked first time (probably more luck than judgement!). Until now.

The batteries supply 24V, to be used by 24V hardware, or regulated down to 12V for those items that need it.

I have three sets of LED strip lights - at 12V. For the first two, I step the voltage down to 12V (a Victron unit each) and then each have a dimmer switch. They work fine - the LEDs are getting 12.3V at the top end, and smoothly dim until cutout at 5V.

The problem is that the third, although as far as I can see wired the same, won't dim. The switch appears to have no effect at all, and even when the dimmer says OFF, the full 12V are passing though. I thought the dimmer might be busted, but I swapped in one of the others and the same happens. That trial dimmer continued to work fine when put back into its own circuit, so the dimmers aren't being actively broken by an action of mine.

I wondered if it was only this sort of dimmer. As I was installing a ducting fan with an inline dimmer switch (not the same model or brand as the LED one), the same thing. All the connections are made carefully, but the full voltage passes through and ignores the dimmer. These are 24V fans and the dimmer is rated to both 12 and 24V.

Any ideas why the dimmers are being 'ignored'? I imagined if I'd busted them with a mistake or a short that they would stop working entirely?

 

[jwfrary]

Same LED strip? They won’t all dim.

 

[neilf39]

What make and models of dimmers and what make and model if LEDs and fans? Some LEDs can't be dimmed and some motors won't respond to a dimmer. When you say the voltage is not going down is that measured or the fact the motor / LED is unaffected? Did the dimmer that did not work on the LED work on the other LEDs?

 

[alexhibbert]

What make and models of dimmers and what make and model if LEDs and fans? Some LEDs can't be dimmed and some motors won't respond to a dimmer. When you say the voltage is not going down is that measured or the fact the motor / LED is unaffected? Did the dimmer that did not work on the LED work on the other LEDs?

The LED strips are dimmable, and two out of three dim fine. The fan motor runs more slowly if I offer up a 6V power source, so I can only assume it's dimmable too.

I've measured the voltages both in and out of the dimmers, so it's not just based on the fact the lights are constant.

That final test is a combo I haven't tried, but since the reverse test supports that the dimmer isn't broken.....

For the smaller dimmer for the fan, I've taken it home and connected it to a 12V source (from an AC adapter). 12V measured in, and 12V measured out, regardless of the dimming setting, even when clicked to OFF.

 

[Alex_Blackwood]

I am not up to speed on dimmers and don't know what you have , but are they DC dimmers or possibly ac? Don't really know why they would work on two sections of lighting and not the other if all are identical units.

 

[alexhibbert]

I am not up to speed on dimmers and don't know what you have , but are they DC dimmers or possibly ac? Don't really know why they would work on two sections of lighting and not the other if all are identical units.

All DC, correct voltage ranges, and yep, I'm confused too!

 

[alexhibbert]

LEDs are similar to these: Premium LED Tape by Tagra®, Cool White, 12w p/m (5 Metre Reel) and dimmers Single Colour/Single Zone Inline Dial Dimmer, 12/24v, 8A. Just to reiterate that this combo is proven to work on two out of three setups.

The fans are Rule inline blowers (24V versions) and the dimmer is a standard 12/24V 5A inline dimmer with a click to off setting. Also to say again that the blower does indeed run slower at a fixed lower voltage, so should be 'dimmable'.

So, this can't be about wrong equipment, and I'm looking to see if anyone knows of any known reasons why dimmers might be 'bypassed', either through a malfunction or common wiring mistake etc.

 

[Sandyman]

Do hope those external cameras are working well. Wouldnt want to be caught out at sea in a Force 8 without them.
Can LED Lights be Dimmed? All About LED Dimming Problems - LampHQ

 

[alexhibbert]

Do hope those external cameras are working well. Wouldnt want to be caught out at sea in a Force 8 without them.
Can LED Lights be Dimmed? All About LED Dimming Problems - LampHQ

Thanks for the link. None of those appear to be it. The components are of the correct type and rating, and two other identical set ups work fine.

Re: the cameras. Well the guy in Poundland said they were waterproof and for the outdoors, so.

 

[VicS]

So, this can't be about wrong equipment, and I'm looking to see if anyone knows of any known reasons why dimmers might be 'bypassed', either through a malfunction or common wiring mistake etc.

A wiring mistake seems the most likely explanation.

 

[philwebb]

Well, I’ve not looked at DC dimmers, but AC (mains) dimmers don’t reduce the peak voltage, they clip off part of the wave form, so that it’s not a complete sine wave any more. The effect of this is to reduce the available power, so that a lamp is dimmed. A lot of motors are not happy with the odd wave form. I wonder if the DC dimmers, instead of reducing the voltage, switch it on and off rapidly so that the amount of power is reduced. This would dim an LED as it would be rapidly switched on and off, which your eye would see as being less bright. It wouldn’t necessarily have the same effect on a motor. Depending on your test meter response time, the instrument would probably still show the full 12volts.
To see what is happening you would have to look at the output from the dimmer on an oscilloscope. If you don’t know anyone who has one, I could look forat it for you if you post it to me.

 

[alexhibbert]

Well, I’ve not looked at DC dimmers, but AC (mains) dimmers don’t reduce the peak voltage, they clip off part of the wave form, so that it’s not a complete sine wave any more. The effect of this is to reduce the available power, so that a lamp is dimmed. A lot of motors are not happy with the odd wave form. I wonder if the DC dimmers, instead of reducing the voltage, switch it on and off rapidly so that the amount of power is reduced. This would dim an LED as it would be rapidly switched on and off, which your eye would see as being less bright. It wouldn’t necessarily have the same effect on a motor. Depending on your test meter response time, the instrument would probably still show the full 12volts.
To see what is happening you would have to look at the output from the dimmer on an oscilloscope. If you don’t know anyone who has one, I could look forat it for you if you post it to me.

Thanks for your carefully considered advice here. I completely agree about the problem if it were AC - but the entire set up is DC, which as far as I'm aware doesn't use such a complex dimming method. The voltages on the lights that do work correctly are varying between 12V at full power and down to 5V where they cut out.

I'm trying to work out a way I've wired them wrong, or have somehow physically bypassed the dimmer so the current is flowing via another route, but these circuits really aren't too complex, so it's hard to see a place where I've gone wrong.

I've tried to post a picture on here showing the smaller dimmer showing this bizarre behaviour here at home where I'm set up a little rig - but I get an error as I try to post. Will try again.

 

[alexhibbert]

I'm not sure if this picture attaches ok. It shows the smaller dimmer (12/24V 5A rated) with a 12V power source. The output is being measured, and regardless of the dimmer setting (off, halfway, or full) it's just the same 12V.

Ok I can't embed an image here (perhaps I've not been a member long enough?) so here it is: https://static1.squarespace.com/sta...6c67718e510041c349/1613145966358/IMG_6117.jpg

 

[philwebb]

Hi Alex, What happens if you connect it to a 12volt car bulb (filament type) ? That may tell you something. It could be that the voltage stays the same but the current is controlled.

 

[alexhibbert]

Hi Alex, What happens if you connect it to a 12volt car bulb (filament type) ? That may tell you something. It could be that the voltage stays the same but the current is controlled.

Thanks - have ordered one. I'm not entirely sure what it will do apart from take the LEDs out of the equation, as the odd, unchanging 12V outputs of the dimmers can be measured before even connecting in the lights. Anyhow, it's £3 so can't hurt.

ps the image link for the test rig I have here is now in the post above.

 

[alexhibbert]

Ok I seem to have make some progress.

The LED 8A dimmers are voltage reducing ones, and so it's likely one of them is faulty. Or at least possible. It doesn't explain why another working dimmer had the same effect on the third LED strip - perhaps that's faulty too.

The fan 5A dimmer is actually a more modern design PWM dimmer which does (as I've now learned) not reduce voltage, but instead cuts off the current at a high frequency, and so reduces the power getting through. It explains the measurements that didn't make sense, and how the motor wasn't affected by the dimmer. So I'm ordering a voltage reducing dimmer/motor controller for this, even though they are considered less efficient and sophisticated, and hoping for a result.

 

[TernVI]

Basic 12V LED strips generally have 3 leds and a resistor in series, repeated along a the strip.
An LED will have a voltage/current characteristic where current increases steeply around say 3.2V. Maybe double the current for an extra 100mV.
So in the simple set up, current is controlled by the resistor and the voltage and life is simple.

More modern/efficient/sophisticated is to have a conversion circuit so the LED current is constant regardless of volt input. You have a chip controlling the current and converting the power. So it works over a wide voltage range and gives a constant output of light.
Next step is a chip that can be controlled somehow. Either sending it data to set the current, or as in mains 'dimmable LED bulbs' the phase chopping of a thyristor dimmer is decoded by the chip to decide on the regulated current.

If you want to do simple dimming either lok for old-skool LED,LED,LED, Resistor strips or get a strip with its own compatible controller.

There are variations on these themes including those xmas light strings that flash, multi-coloured and all that tat. Too clever for its own good, and mostly just irritating.

Like wise things like fan motors can contain controller circuits which make it work the same over a wide voltage range. Nothing is simple it's not 1985 any more!

 

[PaulRainbow]

Sounds complicated and expensive just to dim some LEDs.

Why not do it by remote ?

Mi Light Touch Panel Wall Controller DC12V 24V For 5050 RGB RGBW LED strip Lamp | eBay

or

FUT007 Dual White CCT Remote Control for MiLight Dual White CCT LED lighting | eBay

Or both. Either of them will control four zones. Just choose the remotes and the controllers to suit your LEDs. Versions available for plain white, dual white or RGB.

Edit: If you want to use 12v LEDS, just fit one 24v to 12v DC-DC converter and connect that to all of the LEDs. no need for multiple converters.

 

[alexhibbert]

Sounds complicated and expensive just to dim some LEDs.

Why not do it by remote ?

Mi Light Touch Panel Wall Controller DC12V 24V For 5050 RGB RGBW LED strip Lamp | eBay

or

FUT007 Dual White CCT Remote Control for MiLight Dual White CCT LED lighting | eBay

Or both. Either of them will control four zones. Just choose the remotes and the controllers to suit your LEDs. Versions available for plain white, dual white or RGB.

Edit: If you want to use 12v LEDS, just fit one 24v to 12v DC-DC converter and connect that to all of the LEDs. no need for multiple converters.

Long story short - no remote ones due to RF interference with other gear, and in a normal world dials and knobs tend to break less!

And having multiple units due to redundancy and the boat being expedition spec, where we can't just pop back for spares if something breaks. Having a couple of 10A step downs is about the same price as a 20A step down.

 

[chrisclin]

Thanks - have ordered one. I'm not entirely sure what it will do apart from take the LEDs out of the equation, as the odd, unchanging 12V outputs of the dimmers can be measured before even connecting in the lights. Anyhow, it's £3 so can't hurt.

ps the image link for the test rig I have here is now in the post above.

I don't know if it's at all relevant but I had a problem with downlighters at home when I replaced halogen bulbs with LED. They continued to glow dimly even when turned off. The manufacturer of the dimmer switch assured me it worked with led but suggested I needed to add some resistance to the circuit. I replaced one of the LEDs with a halogen and everything was sorted. They tried to explain what the problem was but I couldn't make sense of it and they agreed it was complex.
Is it possible that the 2 working circuits have more resistance than the other - even just a longer wire?