Andres
Newly Enlightened
Hi all - I've been lurking around CPF for a while but thought I would sign up and start a thread so that I could get some of the collective experience of CPF to cast a critical eye over my road bike lighting design.
:wave:
Just as a bit of background which hopefully justifies some of my design criteria;
I've been interested in torches every since I was given a Minimag many years ago. Obviously things have come a long way since then, and still have a long way to go!(i.e. a multimode, regulated, 1xAA, focusable, tailstanding, forward clicky light in a nice body!!!) But I digress...
Lately I have been using a MTE 5-Mode 100 lumen rebel, 1AA light zip-tied to my helmet as forward lighting. This has been pretty good for commuting, but obviously there is potential to be a lot better. A lack of output, and non-helmet mount are my main reasons to upgrade.
I understand a helmet mount 'spotlight' is handy (essential?) for MTB, but for road and commuting, I find there is usually enough 'other' lighting to see what's around, but I need a much brighter light to see the big potholes / branches far ahead to give ample time to avoid them. Furthmore the angle of my head changes substially from when I'm rising with my hands in the trops to when sitting back on the top bar. Unfortunately the angle tends to work the 'wrong way' too, with the torch pointing at my front wheel when going faster in the drops, and way of in the distance when peddling slowly and sitting back.
Having seen the output of my (from all reports- very underdriven) Romisen RC-T5 4xCree light up the street very well, I believe I should be able to get more than enough light out of 2 Q5 Cree emitters, driven at well under 1000mA for all but the darkest, fastest sections of road.
The lighting system will be attached to my Learsport 8500 roadie. It's a nice cheapish bike with Ultergra componentry and Mavic wheels. I use it both as a commuter 3-4 days a week and also for longer 'pleasure' rides.
I have decided the lighting should be left on the bike 99% of the time, although the batteries will not. I believe this will allow for a stronger mounting method which will be more resiliant to both vibrtions and potential theft (although rarely will the bike be in a location where theft of the lights could occur).
I will go into more details, hopefully with some CAD images, of the mounting method later, but first I need to make sure I have the electrical side of things correct;
The plan thus far is to have:
3x 18650 Li-ion batteries connected in series, to a bFlex controller, connected to 2x Q5 Crees in series.
From what I've read, the Q5s can need up to almost 4 volts at the higher currents, and taking into account some driver lossses, 3x3.7=11.1V of voltage from the batteries should mean there's always enough juice to prevent the bFlex from having to fall back to direct drive, and also allowing a decent runtime. While very crude, and doesn't take into account the diminishing effeciency at higher current etc, I was able to get over 3 hours from my RC-T5 on high from 2x18650 batteries pushing 4xCrees, so I figure as a very rough estiamte it is safe to assume I will get over 3 hours from the 3x18650 driving 2x Q5s at a sensible level. If anyone thinks that this is unlikely, please say, as a 'safe' 3 hours is my aim for running at arund 750mA. Any longer than that and I'm sure I'll either be at my destination, or happy to take a 5 minute break to switch batteries. Furthermore, that allows for 5, 30 minute commutes, meaning the batteries could be recharged roughly once a week.
The other part I'm looking for a little though on, is switching for the bFlex controller. From reading other threads and the manual, I believe it requires a momentary action switch, but does not need to handle the full power of the lights, as it is only a 'signal' switch? As such, I thought it would be a very elegant solution to use the momentary switches deisgned into the "Flight Deck" Simano shifters on my bike. These are actually designed to control a matching Shimano cycling computer, but from all reports, the computers aren't too great, and regardless, I have a nice Sigma computer I'm quite happy with as it is.
Is there any forseeable issue with this electrical design, or anything significant I have overlooked?
PS Sorry about the essay length, the last two paragraphs are the importnt ones though!
:wave:
Just as a bit of background which hopefully justifies some of my design criteria;
I've been interested in torches every since I was given a Minimag many years ago. Obviously things have come a long way since then, and still have a long way to go!(i.e. a multimode, regulated, 1xAA, focusable, tailstanding, forward clicky light in a nice body!!!) But I digress...
Lately I have been using a MTE 5-Mode 100 lumen rebel, 1AA light zip-tied to my helmet as forward lighting. This has been pretty good for commuting, but obviously there is potential to be a lot better. A lack of output, and non-helmet mount are my main reasons to upgrade.
I understand a helmet mount 'spotlight' is handy (essential?) for MTB, but for road and commuting, I find there is usually enough 'other' lighting to see what's around, but I need a much brighter light to see the big potholes / branches far ahead to give ample time to avoid them. Furthmore the angle of my head changes substially from when I'm rising with my hands in the trops to when sitting back on the top bar. Unfortunately the angle tends to work the 'wrong way' too, with the torch pointing at my front wheel when going faster in the drops, and way of in the distance when peddling slowly and sitting back.
Having seen the output of my (from all reports- very underdriven) Romisen RC-T5 4xCree light up the street very well, I believe I should be able to get more than enough light out of 2 Q5 Cree emitters, driven at well under 1000mA for all but the darkest, fastest sections of road.
The lighting system will be attached to my Learsport 8500 roadie. It's a nice cheapish bike with Ultergra componentry and Mavic wheels. I use it both as a commuter 3-4 days a week and also for longer 'pleasure' rides.
I have decided the lighting should be left on the bike 99% of the time, although the batteries will not. I believe this will allow for a stronger mounting method which will be more resiliant to both vibrtions and potential theft (although rarely will the bike be in a location where theft of the lights could occur).
I will go into more details, hopefully with some CAD images, of the mounting method later, but first I need to make sure I have the electrical side of things correct;
The plan thus far is to have:
3x 18650 Li-ion batteries connected in series, to a bFlex controller, connected to 2x Q5 Crees in series.
From what I've read, the Q5s can need up to almost 4 volts at the higher currents, and taking into account some driver lossses, 3x3.7=11.1V of voltage from the batteries should mean there's always enough juice to prevent the bFlex from having to fall back to direct drive, and also allowing a decent runtime. While very crude, and doesn't take into account the diminishing effeciency at higher current etc, I was able to get over 3 hours from my RC-T5 on high from 2x18650 batteries pushing 4xCrees, so I figure as a very rough estiamte it is safe to assume I will get over 3 hours from the 3x18650 driving 2x Q5s at a sensible level. If anyone thinks that this is unlikely, please say, as a 'safe' 3 hours is my aim for running at arund 750mA. Any longer than that and I'm sure I'll either be at my destination, or happy to take a 5 minute break to switch batteries. Furthermore, that allows for 5, 30 minute commutes, meaning the batteries could be recharged roughly once a week.
The other part I'm looking for a little though on, is switching for the bFlex controller. From reading other threads and the manual, I believe it requires a momentary action switch, but does not need to handle the full power of the lights, as it is only a 'signal' switch? As such, I thought it would be a very elegant solution to use the momentary switches deisgned into the "Flight Deck" Simano shifters on my bike. These are actually designed to control a matching Shimano cycling computer, but from all reports, the computers aren't too great, and regardless, I have a nice Sigma computer I'm quite happy with as it is.
Is there any forseeable issue with this electrical design, or anything significant I have overlooked?
PS Sorry about the essay length, the last two paragraphs are the importnt ones though!
Last edited:
one. 
