Why don't lamps have dimmers?

[n?π]

Off on a tangent from Stu's thread.(http://ukcaving.com/board/index.php?topic=11296.msg145738;topicseen#new)

Why do modern lamps have set brightness levels rather than continuous dimming?

With bulb lamps its simple, 2 bulbs, 2 brightness levels. To have more levels mean more bulbs hence more faff and a bigger lamp.

Modern lamps tend to have more levels (4 or so) from a "camp" up to "megasuperultrabright".

As i understand it, the dimming is done by driving the led with a plused constant current and altering the ratio of on pulse to off pulse creating the effect of brighter or dimmer light. This pulsing is done by electronics often a microcontroler. (I hope this is correct?)

So why aren't people using a variable resistor as an in put to the microcontroler to give a continuously variable brightness level? this would allow you to tune your light to your needs (you wouldn't turn your lamp up to full for a passage that needs a bit over medium) and save power?

I can see for a reto fit (Bisun/Retro 2) fitting one in would be awkward and that push buttons are easier to water proof (little general i think?) but a lot of new lamps (scurrion, petzl ultra, sten) use a custom made rotory input that could surely drive a variable resistor as easily as a switch.

Is it more tricky to engineer? Do people not like a variable light source for some reason?

Just wondered.

Cheers

n thing n

 

[cap n chris]

Love to see this helmet-mounted....

or something a bit more techie?....

Clearly it would need to be retro-fitted to something apt, like an Oldham or stinky conversion, just for the pukka "steam punk" characteristics.

 

[Elaine]

Sounds like a good plan, though although I know nothing at all about anything in the whole world I have noticed that the dimmer switches I have owned have not lasted as long as any 'normal' switches.

 

[Rob]

I have made one before, for photography underground. Had a waterproof rotary resistor which set the output via a Fatman LED Driver (http://www.taskled.com/techfatman.html - now discontinued).

This is great for photography, where you can set the brightness to exactly what you need for the shot. However it was quite fiddly to accurately adjust, I think if it was a headlight you would spend quite a lot of time setting it to a level which is nice to cave on, then turn it up to full to see a big chamber, then again spend quite a lot of time setting it to a lavel which is nice to cave on! Nice and simpe though...

 

[shortscotsman]

One advantage of set levels is knowing the burn time.  On the powerfull lamps there is often only a couple of hours on the highest setting so you have to use this sparingly.  With a dimmer you could be running higher than you think.

 

[Addy]

A single button/switch can be water/crud proofed fairly easily. A rotary control to give 'infinite' variability is far more difficult to integrate whilst maintaining a suitable level of water/crud proofing and sufficiently rugged.
It's not so easy to come up with an intuitive method of controlling up/down with a single button. Adding a second button is an option but more complication and expense.
It's easier to 'step up' and/or 'step down' using long or short pushes. Theoretically it's possible to have loads more steps is you're prepared to push/hold the button until it's just right.
The argument about duration is a good one.
The lamps I've made have 3 or 5 levels but I've also made lamps with infinitely variable controls and whilst ergonomically more intuitive, are not really necessary - you set a level and leave it there until you need it brighter or dimmer...

 

[SamT]

awwww - I was expecting a punch line 

(too many christmas cracker jokes in our house.)

 

[potholer]

As i understand it, the dimming is done by driving the led with a plused constant current and altering the ratio of on pulse to off pulse creating the effect of brighter or dimmer light. This pulsing is done by electronics often a microcontroler. (I hope this is correct?)

The chances are that most multilevel lights will use a microcontroller (though the Sten doesn't(or at least, didn't)), and whatever actual power control method is used, it would be possible to have more or less continuously variable output as far as the microcontroller is concerned.

...this would allow you to tune your light to your needs (you wouldn't turn your lamp up to full for a passage that needs a bit over medium) and save power?

Personally, I make multilevel lights and set the levels at a factor of 3 apart since I experimented and didn't see much point having them any closer than that.

Though it's a personal thing, I don't see a particular passage as needing a precise amount of light. The first proper LED light I made myself was about 10x dimmer per output level than the ones I make now, and it was perfectly adequate for caving with, (and rather better than what most people seemed to be using) so even though I prefer the latest model, the idea of a passage demanding a precise lighting level isn't one I'd go for.
Also, the amount of light I use will depend how frugal I'm being - in domestic trips where I'm unlikely to flatten my first rechargeable pack even going flat-out, I'd quite possibly use much more light for a given kind of passage than if I was on expedition trying to make a few flatpacks last me a month.

I'd definitely echo the points made about power consumption predictability - I do like to have an idea how long a given level might last, and to be able to provide that information to other people. Only having worst-case and best-case figures with no idea of intermediates would be definitely less good from my point of view.
Given that the human eye is pretty bad at judging brightness levels, unless there were some kind of engineered click-stops in a rotary switch, people wouldn't know what output/consumption they were at. And if there were click-stops and they used them, they'd negate the point of having variable power.

There is the odd variable-power headlamp and flashlight around, but the fact that they're so rare does rather suggest there isn't a huge untapped market crying out for them.

 

[n?π]

ah, good points I knew I was overlooking something :-[

It would seem variable brightness lamps aren't around as: you have no idea how long they will last and nobody really wants/needs one.

cheers folks.

 

[potholer]

It was a good question -  I think it's really a case of there seeming to be few people wanting one loudly enough to get someone to invest the time and effort into making one, and the people who have made one-offs not having been bowled over by demands to go into production.

On the runtime side, many users might well not be much concerned, since for many light/user combinations, a lot of trips are unlikely to flatten a battery pack.
However, the people who actually make lights are probably also the kind of people who do have an interest in runtimes.

There are a few headlights/flashlights with continuously variable output, but they do seem to be very much niche products in a fairly large market.
Caving lights are a small market, and there may just not be space for a niche product.

 

[TheBitterEnd]

Just a thought, should any lamp makers be reading this (and relating to the original thread about blinding people with really bright lights)-  has anyone tried to develop an "auto dim" facility so that if you look at another caver the lamp automatically drops down to a low setting because it detects light from their lamp?

 

[Stu]

Just a thought, should any lamp makers be reading this (and relating to the original thread about blinding people with really bright lights)- 

Just to clarify, my thread wasn't really about blinding people with super lights. It was about at what point do the returns of big lights start falling. Light is good!

 

[footleg]

Since no-one else mentioned it, the use of a variable resistor to limit brightness also has the disadvantage that the lamp would be less efficient. Any resistor consumes power, which is why LED lights are more commonly dimmed by flashing them on and off very fast. That way the battery lasts longer for the same apparent brightness level.

I think the LEDs are also less efficient when run below their full power, or at least do not dim in brightness in a linear relationship to the input voltage. So unlike an incandescent traditional bulb, a simple dimmer circuit does not work so well. (I am not an expert on these last points so corrections welcome!)

 

[potholer]

Even when rapid on-off pulsing (PWM) is used for power control, there's usually something limiting the current in the 'on' periods, whether it's a resistor or electronic current limiter, or in the case of some cheap AAA lights, the ability of alkaline AAAs to provide current.

As for LEDs, I think there was a time when some were more efficient at higher power, but that isn't the case for recent bright white ones (pretty much anything from Luxeons onward).
If an LED was driven constantly at a given current, it would give more light (though these days maybe only marginally more) than one driven at twice the current for 50% of the time.

While efficiency does keep increasing as power reduces almost to zero, the output colour can (or at least did) tend to shift at the lowest powers, so there was a minimum level below which it was recommended to use PWM to keep the colours fairly constant (though whether that would actually be important for caving is debatable).
In practical terms, actually running a pure constant current drive at very low currents can be troublesome anyway, and it can be easier using PWM for reasons of electronic simplicity even if colour shifts aren't an issue.

 

[hoehlenforscher]

My LED Lenser P7R headtorch has a variable dimmer on it. Mind you the cable has now broken for the 3rd time in a year so wouldn't advise anyone getting one!

 

[Peter Burgess]

For me, reliability wins hands down over functionality when it comes to something as important as a caving lamp. My Duo hasn't let me down once in over three years, although there have been a few times when it would have been good to have a brighter more focussed beam. When I recall all the faffing about trying to repair and adapt Oldhams and NiCad conversions in the past, it is quite reassuring not to have to do this any more. That's not to say a dimmer isn't a good idea, but it is one more thing to go wrong, and it would have to be virtually faultless to make it worth having, were it to be introduced. And for a caving lamp it would need to be failsafe, so if the dimmer started to play up the lamp would default to working rather than not working!

 

[Addy]

Since no-one else mentioned it, the use of a variable resistor to limit brightness also has the disadvantage that the lamp would be less efficient. Any resistor consumes power, which is why LED lights are more commonly dimmed by flashing them on and off very fast. That way the battery lasts longer for the same apparent brightness level.
.....

My most recent, albeit experimental lamp has a Cree LED and uses a single Li-Ion cell (18650 type). I've used a linear regulator as the Li-Ion couple voltage is close to the LED voltage and with a low VceSAT transistor, quite an efficient and linearly variable lamp can be made. I've not really noticed the colour shift at low currents but it is the case that PWM if often used to maintain full current and therefore true colour, even at low duty cycles.

There are some very expensive lamps available and some very good lamps or 'innards that are quite affordable, particularly if you're willing to use an Oldham type headset...

 

[potholer]

I've not really noticed the colour shift at low currents.

I can't say that I actually noticed any colour difference when I switched from linear to PWM drive for low power settings, though in those days LED colour was so variable from one to another that slight differences in one light probably wouldn't have stood out much anyway, especially at lighting levels where colour vision wasn't exactly running on all cylinders.

I only made the change since it helped reduce flickering which the linear regulator had issues with at very low output, and fortunately, I managed to do it without actually changing anything other than a bit of programming, which was nice.

 

[footleg]

For me, reliability wins hands down over functionality when it comes to something as important as a caving lamp. My Duo hasn't let me down once in over three years, although there have been a few times when it would have been good to have a brighter more focussed beam.

As a fellow Duo fan, I can strongly recommend the Bisun Duo conversion if you have not tried one yet. I upgraded my 14 LED Duo which was using a 1 watt LED screw in replacement bulb in place of the halogen bulb, and I have been really impressed with this unit in use over the past 18 months. The spot beam is excellent compared to anything I have had in a Duo previously, and the great thing is that the 14 LED unit is left in place and works just as before. So no worries about reliability. The Bisun conversion just slots into the case (you remove the obsolete plastic reflector to make space for it, and just screws into the halogen bulb holder for power. So it operates off the existing switch. Took only a few minutes to install. Battery life with the Bisun spot on appears similar to the life I get from the 14 LED unit. So you can choose between spot and flood (the 14 LED unit) as you prefer during trips. I have found myself often preferring the Bisun Spot for general caving rather than the 14 LED unit.

Here's a link. http://www.bisun.co.uk/solo.html

 

[Peter Burgess]

Thanks Footleg. It's been on my "get a round tuit" list for when all the December bills are eventually paid off!