Reactive power compensation for a bicycle hub dynamo USB charger

@davee The bike people must do a better job, I can quickly find battery operated lights, some using AA or AAA and they all last from a low of 24 hrs to a high of 80 hrs. I don't recall how long my lights were good for when I had a bike, but it was a long time for my use.

Remember he at least initially said NO batteries, AFAIK he was just wondering if the common 'dynamo' could be improved to do ????? maybe more lights?

Later he seemed to accept batteries would be needed, but until he can update his requirements it's a bit of a guessing game.

I wasn't trying to be that humourous when I said he wanted a 'doomsday prepper' solution. I can understand that taken to its ultimate end, batteries will not last so we need some way of creating electricity (for what?) on the fly.

Maybe that is the correct question to ask him, why the need for electricity at all?

Travelling at night in a doomsday environment would be ill advised and showing a light even more so. That is why firepits and caves are preferred for cooking in days when the night was the time of the hunters.

Hi Ron @zander,

   I know nothing of bike lights, post about 1970, excepting when I see one passing me in the street. As for headlights (for people's heads, not cars), there are plenty of expensive LED ones that others have got, which do indeed run for longer, but only offer a tiny fraction of the light. The one I measured is cheap, but apparently offering a similar light output to those I have seen bike riders using as they pass by. Whilst I am not suggesting the light I have is the 'most efficient' or 'best', I suspect it provides a rough idea of how much power is needed to provide a fair beam to light a dark road, etc. 

Personally, I am not really concerned as to whether his project is worthwhile, etc, or not. He posed a technical question and I have done my best, to answer it. Beyond that, is his challenge, not mine. I wish him well, no more.

Best wishes, Dave

@davee The better LED lights have selectable power output with of course corresponding higher battery drain. The one I had for my bike was good for 10 days on low power. Low power meant I could see 50 ft. As you know doubling the power and battery drainage does NOT double the light output and does not double the 'seeing' distance. I think it's a log10 relationship, so to double the seeing requires 10x the current drain????

In any case low power and 10 days use was great for me. They were pricey, but not nearly as much as the rear lights with radar that increased the brightness and flashing frequency as a following car got closer. They were EXPENSIVE!

Hi Ron @zander,

   My simple headlight has basic, 1 step PWM dimming to about half power ... I haven't measured it, but it effectively strobes, as there are no inductive/capacitive components to 'smooth' the current, and can be seen when it's raining. I am not sure about absolute maximum distance, as for my purposes, the road bends are usually the distance limit, but watching for much closer, but hard to see, trip hazards, etc., is my self-interest. In most cases, the light reflected from the road/pavement is far brighter than provided by street lighting alone. 

As I say, my last memory of actually using bike lights was more than half a century ago, but my memory of the lights then, even with fresh batteries, were like "glow worms", compared to the headlight.

--------------

However, I am not judging this project's merit, merely trying to suggest a technical approach to a question.

Best wishes, Dave

@davee For sure lights today are much better. When I still had the truck, lots of folks complained about the brightness of the lights. They were regulation, but since the lights were much higher from the road I guess even on low they blinded them.

As far as the OP's project, I am curious, but at the moment I am still stymied as to what the heck he is trying to build.

Hi Ron @zander,

  Drivers being momentarily blinded by LED lights from vehicles is a subject that the UK government is just waking up to, (after several years ...), especially as many vehicles are 'SUVs', which puts their lights a few inches higher than many other cars.

The 'human headlight' I was discussing is obviously much smaller and lower power, but when you have a group of people using them, they all have to be careful not to shine in each others' faces, as it is a very unpleasant experience.

Best wishes, Dave

I see i had putted too much context in my question. In fact i am asking here for an efficient way to compensate the reactive power for a low power and frequency changing alternator. Not for a complete solution for whatever my purpose is.

Despite of i want to use it in my bicycle, you can imagine other types of low power generators which could benefit from a good idea, like very small hydro or wind power turbins. Especially small wind turbins would have exactly the same problem. But in general, a good technical solution will always find its application somewhere.

So coming back to the topic. Does someone know a low-ON-resistance AC switch, which is not a relay and not a 1.5-2V drop triac ?

@j526 NO

@zander do you have experience in AC power optimizing applications and that means NO, SUCH CERTAINLY NOT EXIST or does it mean NO, AT LEAST I HAVE NO IDEA

@j526 NO I have no ideas on this topic.

Why?

I used to be an industrial electrician working on very high voltage equipment and several thousand HP MG sets. Before that I did QA testing at Westinghouse on similar equipment. I know how many engineer hours were spent so have zero confidence I can outsmart them although my uncle and I did on one occasion outsmart the UK engineers and fixed one of the first constant current devices.

I do NOT however have any experience with the smaller machines you want to optimize but can appreciate that the small size may not attract the same degree of engineer hours so there may be an opportunity. I do encourage you though as many people around the world can potentially benefit from improved AC generation such as wind and water powered devices common in the 3rd world use alternators (AC)

I wish you much luck, keep us informed.

@davee Over here in the colonies, we do a few things to alleviate that. The first is I have numerous cameras on my car that 'sees' oncoming cars and dips the lights. The higher speed highways are divided so the problem is almost 100% eliminated. When not divided, many major higher (80km) speed roadways are 4 or more lanes wide so if both parties are obeying the law and travelling in their rightmost lane the problem is again largely eliminated. For those left lane hogs, they deserve it.

The remaining (majority) roadways with lower speeds (mostly 50Km) are less dangerous due to the lower speed. 

I have to mention, there is technique involved as well. Do NOT stare at the oncoming lights, in fact look away so only your peripheral vision is involved as it is much better at detecting danger and is not nearly as affected by the brightness. I forget the name of this phenomena, but we also taught that in astronomy classes when looking through the eyepiece of a telescope.

If you are letting politicians determine what to do, you will end up far worse off. The only conclusion they will come up with is to reduce the power of the lights. Unfortunatly that will end up in more deaths from drivers unable to see danger ahead. 

Do the math, 100kph (62mph) is about 100 feet per second. The average driver takes 1 second to move his/her foot from the gas pedal to the brake. Total reaction time is 2.5 seconds. Now we are  2,500 feet further down the road. Since legal high beams are only good out to about 400 feet I think you can see we have a problem. Fortunatly most 100kph highways are lit so that helps, but take a country road at 50kph/30mph and that is 44 feet per second so you go 1,250 ft. Now that big dog, moose, deer, cow, farmer is hamburger and you and your family are either dead or injured.

This is why I exchanged the factory headlights for illegal brighter lights, and for that country road added rally style driving and fog lights.

I didn't need to do that with the truck as it had lights that were good enough for my driving conditions. My new car also has none and I have almost hot several moose in the past so I might add some lights but I am seldom out at all nowadays let alone at night so maybe yes ,maybe no.

@davee One quick follow up. One of the main reasons I like to drive with adaptive cruise turned on is that not only can it likely 'see' better than me, but its reaction time is nearly instantaneous thus reducing the stopping distance by hundreds or even over a thousand feet.

Posted by: @davee

↑

Sorry, I haven't really thought my answer through, as it isn't something I was personally interested in doing, but a Google based suggestion from Wikipedia, might be interest you, if you haven't already seen it.

https://en.wikipedia.org/wiki/Active_rectification which includes:

-- attachment is not available --

 

I have done some simulation staff and it seems that this approach is an absolutely no-go if there is a frequency changing generator on the input side. The output is very sensitive for frequency changes, a compensation capacitor in series is anyway required and it can not be of constant capacitance in frequency range of 4-200 Hz which we are talking about. It would require very very capacitors and very very switches to work, and a heavy transformer of course.

Hi @j526,

  I am not surprised you decided it was impractical ... I thought it looked a very tough, and possibly expensive, project, but I pointed it out as a 'technical' approach that 'theoretically' met some of the requirements.

Sometimes people take on 'impossible' challenges and find ways of overcoming the difficulties, but of course, for every success, there will be a long list of failures. The tricky bit is spotting the future 'success', without also working through the failures.

I hope you found the search interesting, and maybe even find it useful, sometime in the future in a different context.

Best wishes, Dave

@davee Not surprised.