Constant Current incan modes. Why not?!!?

[bykfixer]

Once upon a time this was a fun thread.
Maybe Kestrel will come back the fork in it so the poor dead horse can rest in peace.

 

[chillinn]

Once upon a time this was a fun thread.
Maybe Kestrel will come back the fork in it so the poor dead horse can rest in peace.

Personally, I'm not accepting the empty explanations for dening the possiblity of PWM-less moded incan. I feel like I have provided the proof of concept of a way it could work already, through HKJ's suggestion concerning lowering the voltage of Li-ion secondary cells to the voltage level of Lithium primary cells. I have a feeling there's probably a half-dozen ways to limit voltage no one has mentioned. Accepting that a lamp driven at a lower voltage than it is specified for is not going to be lit idealy is a given. I don't see it as much of a problem if what you get for the sacrifice of ideal color temperature is quadrupal runtime on lower modes. I kind of wish the negative nellies would move on, as there's only so many ways to express the personal opinion "I don't think you can do that," before it starts sounding like "I don't want you doing that."

 

[ssanasisredna]

Deleted

 

[bykfixer]

When they come out with an incan with pwm I can imagine using one while walking in a thunderstorm.
The song "singing in the rain"- the disco version will be playing in my head.

As my progress continues I'll chuckle as everyone in my beam will be flailing about in epileptic siezures.

I see what you are getting at ssandre, but it seems pwm is not something that would be on a list of wants by the author of this thread.
It'd be like the time Richard Petty built a race car then refused to drive it. Why? He replied "man that thing'll kill ya".

 

[ssanasisredna]

When they come out with an incan with pwm I can imagine using one while walking in a thunderstorm.
The song "singing in the rain"- the disco version will be playing in my head.

As my progress continues I'll chuckle as everyone in my beam will be flailing about in epileptic siezures.

I see what you are getting at ssandre, but it seems pwm is not something that would be on a list of wants by the author of this thread.
It'd be like the time Richard Petty built a race car then refused to drive it. Why? He replied "man that thing'll kill ya".

That's quite a wild imagination.

 

[Kestrel]

Once upon a time this was a fun thread.
Maybe Kestrel will come back the fork in it so the poor dead horse can rest in peace.

Lol, not against PWM discussion per se; in fact it may be a primary aspect of this thread topic - the two regulated incan iterations that I know of both utilize pwm.

But I posted earlier because the discussion was starting to get personal - the old standby still works: "attack the post, not the poster".

That's quite a wild imagination.

Posts like this, come to mind. :ironic:

 

[bykfixer]

the two regulated incan interations that I know of both utilize pwm.

:

So pwm does not mean automatic premature death sentence to the light bulb?
Cool!!!

My morning commute involves tree lined interstate travel at sunrise so the constant flicker of sunrise at 60+mph is no big deal. Nor is working around massive multi colored disco light flashers on vehicles. But I do understand PWM bothers a whole lot of people. It is an awesome invention. But it seems many manufacturers have figured out how to throttle back LED's without dramatic tint changes and being an arm chair quarterback kinda figure somebody could do it to the old edison bulb.
Perhaps my krypton bulb'd table lamp is flickering away without me noticing? Hmmm, maybe that's why my wife and one of the kids keep getting headaches?

Thanks for keeping this open sir.

 

[fyrstormer]

Thanks fyrstormer, first part feels like another death blow, but an example would still help.

I have a Lumens Factory 9V 320lm incandescent drop-in. Lumens Factory says it will run for ~40 minutes on a pair of RCR123 batteries. RCR123s are rated at .75 amp-hours when new, so if they can be drained in 40 minutes by an incan bulb then the bulb is consuming an average of 1.125 amps. That doesn't seem like much, but remember that it's running without regulation, so it's actually pulling several amps when the batteries are fully-charged, dropping steadily downward as the batteries drain. The peak amperage is probably closer to 3 amps. That still isn't a significant amount of amperage for short bursts, but to run for more than a couple minutes the current-regulating electronics would need heat-dissipation to avoid overheating. The drop-in occupies the entire head of the light, so there would need to be a second compartment for the current-regulating electronics to sit in, and adequate heat dissipation would be difficult to provide with the electronics sitting right next to a chamber occupied by a white-hot piece of tungsten that's heating the entire body of the light. They would probably be better-positioned in the tailcap. The incan light that I'm using for my example is fitted with a 4-mode PWM tailcap that I pulled from an old Jetbeam LED light, and it's about a half-inch longer than a normal P60-size tailcap to make room for the PWM circuit board and the mode-selector button. I honestly don't know how much larger it would need to be to contain current-regulating electronics that could handle 3 amps for a sustained period of time; the only light I have that burns that much power is a P60-compatible light with an Oveready 1000lm drop-in, and that uses PWM in the megahertz range to regulate power to its 3-LED array. Even using PWM the drop-in is still as big as a normal P60 drop-in; at a guess, current-regulating electronics that could do the same job would probably take up the same amount of space as the entire drop-in does, including the 3-LED array and the TIR optics that focus the LEDs. So for a current-regulated P60-compatible incan light, you'd probably be looking at a light big enough to hold two P60-size drop-ins instead of one, with separate chambers at each end of the battery compartment. I guess that's not a problem if you prefer large flashlights, but I prefer ones that easily clip into my pocket with room to spare.

Second part there... concerning this business about seeing PWM. Whether it is seen is irrelevant. I cannot see 17.5kHz PWM. Using the Lightsaver tailcap (for incan), I generally cannot see the PWM in the lower modes. This is kind of a red herring bit of evidence trying to support the idea that PWM isn't bad. You can't see CO2, so it's not killing you. You can't see that cavity, so it must not be causing you any pain.

Consider that incandescent lightbulbs in your house all "flicker" at 120Hz, because the alternating current power supply oscillates at 60Hz. Power goes one direction through the filament for 1/120th of a second, then stops, then goes back the other way for 1/120th of a second. If you can't see the flickering from incandescent lightbulbs operating on what is effectively 120Hz PWM, then you definitely can't see the flickering from 17,500Hz PWM. But again, you CAN hear 17,500Hz, because that is within the range of human hearing, and that may be causing you discomfort. I have a couple flashlights with audible high-frequency PWM and they are definitely more irritating to use for long periods of time.

 

[fyrstormer]

So pwm does not mean automatic premature death sentence to the light bulb?
Cool!!!

No, certainly not. If it did, household lightbulbs wouldn't be able to run on alternating current.

It seems many manufacturers have figured out how to throttle back LED's without dramatic tint changes and being an arm chair quarterback kinda figure somebody could do it to the old edison bulb.

Not possible, unfortunately. LEDs and incans generate light in fundamentally different ways. An LED generates light the same way a radio antenna generates radio waves; it is effectively broadcasting electromagnetic radiation on a frequency that is pre-tuned by the molecular composition of the emitter. As such, it can be cycled on and off extremely quickly to reduce its power consumption without changing the frequency of the EM radiation it emits when it's powered-on, and the rest of the color spectrum is filled-in by a phosphor layer on top of the emitter. An incandescent bulb, on the other hand, emits light in a wide range of frequencies at once, and the range of frequencies is determined moment-to-moment by the temperature of the filament. ANYTHING that reduces the power consumption of the filament will also reduce its temperature, and thus also change the spectrum of light it emits.

 

[bykfixer]

^^ Good info! Thanks!

My mind contemplates a 3 or 4x 123 cell length light with a 2 cell operating system. Based on the above posts that is. At the tail end the cooling system for the regulator taking up room one cell does in a multi cell light.
And perhaps a 2 stage switch to power a pair of bulbs. Like those triples etc both could be lit for high, or one be lit for low via the switching system. That way lesser output would stay pretty similar in tint.

I have a 1aa incan with a 2aa (#222) bulb in it to provide a dimmer, more orange output. It's about the size of a Solitaire, but brighter with the magnifier tip'd #112.
In complete darkness it is surprisingly effective at my end and surprisingly unseen by someone else. I did mirror testing using a beam profile to see how it worked.
With my hand cupped around the bezel I can see surprisingly far while sneaking up on a would-be villain inside my home. And would be villain, pre-occupied with stealing the silverware would not see me until it was too late. "Flash-bang"

I say that to point out that if regulation to reduce output changes the tint to create a better stealth mode... I'm ok with that.

 

[Need a Light?]

So, could be completely off topic, as I read a bit, but also skimmed a bit.

I have an old 6D Incan 'Dyna-Lite' made in OH. (2 actually). It has a slider switch, that when you start sliding, turns the bulb juuuust barely on, and sliding it all the way up gives full brightness, with everything in between. Basically a dimmer switch.

Granted, you lose some color temp as you go dimmer. But why is this the only light I know if that does this?

Is it just resistance in the switch, as you slide it it has a wider contact patch maybe? I wonder if it still pulls the same current when it's dim?

It's my only dimmable Incan and I love the light, but didn't know where to post about the dimming, this thread seemed reasonably appropriate.

 

[chillinn]

...so there would need to be a second compartment for the current-regulating electronics to sit in, and adequate heat dissipation would be difficult to provide with the electronics sitting right next to a chamber occupied by a white-hot piece of tungsten that's heating the entire body of the light. They would probably be better-positioned in the tailcap....

Unlike with LED, nearly all of the heat from the tungsten filament is shooting right out the front with the light it provides. Only a fraction of the heat from the filament is heating your entire flashlight, and most of that heat is likely coming from your cells as they provide voltage and current.

I haven't seen one first hand, and though it is an LED tower, CPF member Tana is currently offering moded LED tower SingLED for E series (at least), and somehow has stuffed his driver into the little drop-in. I have no idea if an incan driver could be made small enough to fit that size housing, but it is at least a proof of concept for a spot for an incan driver. If SingLED provides enough heat sinking for LED, where the heat is mostly NOT shooting out the front, then a similar housing should provide enough heat sinking for an incan driver. I wouldn't mind the driver in the tailcap, but as has been addressed in this thread already, it would be more ideal for the driver to be in the head than the tailcap (for the reasons, see the above posts). I have no preference, personally, just repeating the info already addressed.

Consider that incandescent lightbulbs in your house all "flicker" at 120Hz, because the alternating current power supply oscillates at 60Hz. Power goes one direction through the filament for 1/120th of a second, then stops, then goes back the other way for 1/120th of a second. If you can't see the flickering from incandescent lightbulbs operating on what is effectively 120Hz PWM, then you definitely can't see the flickering from 17,500Hz PWM. But again, you CAN hear 17,500Hz, because that is within the range of human hearing, and that may be causing you discomfort. I have a couple flashlights with audible high-frequency PWM and they are definitely more irritating to use for long periods of time.

This second point has also been addressed. The flickering in fixed incan lighting is not a fair comparison to PWM, as the effects are not at all similar. Fixed lighting oscillating at 120Hz is not causing any issues. The flicker is minuscule, and the change in brightness is smooth and tightly centered. PWM drops the voltage to zero, then back up to full. It has been said that PWM in incan is undetectable because the filament doesn't have time to cool enough to drop the light output. But it drops it enough in some implementations to be detectable if you look carefully (like the Lightsaver Miser), but detecting it isn't the issue. The effect it has, whether it is detectable or not, is the issue, and the effect, which is not immediate but cumulative over the exposure, is that it causes pain in some, including me.

Again, whether PWM can be seen is a red herring, and not any sort of proof that it can't be having any painful effect. Lots of things that can't be seen can hurt you, please don't make me start listing the counter-examples again, there are so many, and I have listed them too many times, annoying the owner of the site, among others.

I'm going to carefully say that it is possible that a good implementation of something like PWM (voltage pulsing) in incan may be acceptable (but might never be acceptable with LED), if it can at least achieve what fixed incan lighting achieves with its 120Hz cycles. If the rise and fall of the brightness is so fast and tight, light doesn't dim further than what can only be detected with extremely slowed down video of incan fixed lighting, then the effect should be the same, i.e. no effect.

However, as it has already been shown (in CPF threads of the past since the availability of aftermarket PWM tailcaps such as Lightsaver Miser) that the claim that PWM in incan maintains the tint and prevents yellowing at lower modes is simply false. You lower the voltage to a lamp designed for a higher voltage, and the light is going to get warmer and yellower as the voltage drops and the light dims. Bearing this in mind, PWM doesn't live up to it's claim (for incan; PWM maintaining tint in LED is another story), I honestly can't see the point of even having it. If it isn't needed, because it doesn't provide any advantage in tint or battery capacity over a constant current solution (really, a constant voltage solution), then it is superfluous, i.e. unnecessarily adding complexity to a system without benefit.

The incan constant current (i.e. constant voltage) solution I am requesting since my initial post should be a simpler system, and more efficient, giving better runtimes. All things being equal, the simpler solution is best, with the caveat and understanding that the lower modes provided by lower voltage will not be powering the lamp ideally, causing tint shift. Tint shift is a small price to pay for avoiding a migraine trigger.

I have an old 6D Incan 'Dyna-Lite' made in OH. (2 actually). It has a slider switch, that when you start sliding, turns the bulb juuuust barely on, and sliding it all the way up gives full brightness, with everything in between. Basically a dimmer switch. ... Is it just resistance in the switch, as you slide it it has a wider contact patch maybe? I wonder if it still pulls the same current when it's dim

That sounds like a potentiometer, as described above in other posts. Other suggestions to achieve the same effect include QTC or electron quantum tunneling, such as that used by Peak in their lights. I have also suggested that perhaps a Zener diode or avalanche diode could be implemented to lower voltage, sparing a drop in current. Unfortunately for me, I my electrics kung fu is weak to non-existent, and I am at the mercy of clever, knowledgable, experienced, kind, and patient CPF members to come along and explain how some of these ideas are valid, some are not, and the reasons why, and how moded incan can be implemented practically without PWM, revealing the path ahead for us. I hope someone else can address your post better.

 

[Minimoog]

I'm not an expert in circuit design but I have a few comments as I built around 5 lights with a PWM regulator fitted to regulate the brightness. Firstly it really is good to regulate incan lights for more reasons than to just keep the beam brigher for longer. I have found that regulting lights:

Almost eliminates flicker from slightly ressitive battery contacts
Makes bulb last longer as it does not get an inrush of current nor overvoltage
Can be set very accurately with a Fluke to run bulbs as bright as possible
Batteries last longer as less load - the more batteries the better!

Prior to setting up the regulator I do testing on the bench using a regulated bench PSU. I can run bulbs through burn cycles and find out how hard I can drive them before lifetime suffers. Using a regulator I find that I can go way above the rated voltage with little penalty - as opposed to using batteries direct that will blow bulbs quite readily if not rested after recharge etc. For instance in one of my lights I use Gibraltar 1.25 Volt bulbs (big globe, single strand filament) and with a regulator I can get them up to 1.85 Volts without a problem. This gives a lovely silvery light. Same for more powerful bulbs - run them over rated with rarely any problem.

As to flicker, the filament does not flash like an LED due to heat in the filament - same as removing metal from a fire - stays glowing for a good time. As to the buzz I house all my regulators inside the light and seal with thick cell foam - or you could use epoxy and leave just the adjuster free.