The SolidLights “1303” 10W and SolidLights “1203D” Headlights

A Review.

Introduction:

This review is in three parts. In the first part, the reasons for purchasing the Solidlights 1303 are discussed as well as my experiences with the light over the first 9 months of use. In the 2^nd part, the newer dynamo driven Solidlights 1203D light is described. In the 3^rd part, photographs and measurements of light output are presented.

Part 1: The Solidlights 1303.

Christmas time 2004, I decided to treat myself to yet some more lights for my bicycle in my quest for decent lighting... After scouring the forums and Internet for reviews it appeared to me that the lights at the cutting edge were a choice between Overvolted Halogen lights, High Intensity Discharge (HID) lights and the new High Power Light Emitting Diode (LED) lights. Top end systems are powered by Lithium Ion rechargeable batteries.

From what I can glean from specifications, reviews and FAQs, the following pros and cons apply:

What I really wanted was a reliable lightweight unit that could give me adequate light most of the time with a high power boost when needed in very dark situations. Much of my cycling is also on the daily commute when about 90 minutes of lighting is needed in the winter months. HID systems give “daylight” but runtimes suffer unless you pay a lot more. A dual headed Halogen system is the most cost effective viable alternative but, again, for long runtimes, the batteries need to be bigger and heavier. In choosing to take the plunge for a high powered LED solution I employed the following reasoning.

• For “town” riding, I needed a light to be “seen by” rather than to light up the road. My current 0.5W (EL300) and 1W (EL500) LED lights have ample brightness for this purpose and a further 1W LED light would be quite adequate.

• For quiet lane riding where dark adaptation is good, I have found my existing 10W halogen more than adequate and reasoned that a 3W LED would provide about the same amount of light with improved runtime.

• A 10W LED should provide enough light for other rare situations (fast and dark) where a large amount of light is needed.

Of the LED lights available at the time, the SolidLights and USE Exposure lights seemed to be the best on offer. The 10W SolidLights 1303 unit uses three 3W Luxeon LEDs while the Exposure uses two 5W Luxeon LEDs. The 3W single LED “1103” SolidLights offering was discounted because it would not provide any real improvement over my existing 10W Halogen for my commuting runs. According to the LumiLeds website, these Luxeon LEDs are quoted to emit between 25 and 40 lumens per Watt. The USE Exposure Enduro light weighs in at 290g and has its battery and charger included in the unit while SolidLights provides one or two external batteries with an external charger. All in all, after deciphering what is meant by the Exposure's power ratings, run times are similar depending on battery choices and power levels. The SolidLights 1303 have a shorter high power runtime but a longer low power runtime. This is apparently due the the tendency for the Exposure unit to drop the power supplied to the LEDs from an initial high of just under 7W down to just over 2W as the battery discharges while the SolidLights 1303 circuitry keeps the power output constant at over 9W regardless of the battery discharge state by using a boost regulator. I had also heard rumours that Exposure lights interfered with heart rate monitors. Emails to Exposure yielded no reply while Chris Jones at SolidLights replied promptly in great detail and was very helpful. After some soul searching cogitation, I closed my eyes, took a deep breath and plumbed for a SolidLights 10W unit with 2 batteries (430g for just under 3 hours high power or 20 hours minimum power runtime) Chris at SolidLights was also very helpful in passing on a 5% discount by allowing payment by direct monetary transfer to cut vendor costs.

The light (Serial number 000023) was delivered in a small box by the Royal Mail. The box was well packed and secure. The instruction manual is of good quality and easy to read and understand. The battery instructions state that you have to connect the charger to the battery before switching on the mains supply – this done, it took about 3 hours to charge fully. The battery is a two cell sealed unit contained in a small very tough bag with a long integral Velcro strap. One side of the bag has a non-slip red rubberised surface which is kind to the bicycle surfaces. The lead from the battery terminates in a good quality miniature three pronged socket with a twist lock outer casing. Mating with the lamp unit is positive and secure. The light unit is a small rectangular aluminium extruded section with machined side panel box with a row of three LEDs and lenses behind a transparent front panel. The central LED has a different (fresnel) lens providing a tighter beam profile for distance penetration whereas the outer two LEDs provide a wider beam. The box looks and feels strong and is manufactured to close tolerances. The side panels, one sporting the single, round, red push switch, are screwed on with four flush posidrive screws. All edges and corners are smoothly rounded. The button is easy to press and activate even while wearing thick gloves. All in all, although the light looks a little odd with its square profile, it is very well made, occupies less room than an EL300 and I have grown to like it.

Installing the light onto the bicycle was a simple matter of attaching a thumb screw quick release mount onto my space grip (handlebar extension) and sliding the light unit onto it. The quick release bracket seems to be of good quality. The battery was easily and positively connected using the locking plug and socket. The small battery pack can be attached to any spare bit of bicycle tubing such as the top tube, headset or stem by wrapping the strip of velcro around. Simple, almost invisible and effective. The battery really is very small and light – about ¼ the size and weight of my existing 4.1Ah NiMH halogen battery. The light is activated and deactivated by holding down the large red button on the side of the light unit for about 2-3 seconds. The power level can be toggled between one of three settings (1W, 3W and 10W) by pressing the button repeatedly. Installation and familiarisation took about 10 minutes.

In the evening I took the bicycle out to give the light its first real life test. In addition to the new light I also had my existing 10W TrailRat halogen, Cateye EL500 and Cateye EL300 for comparison. All were fully charged.

First, I did a “brightness” test. In the dark of the garden, I positioned the bicycle in the pitch dark to point in the direction of the patio doors and looked at the reflection. On the 10W high power setting, the SolidLights outshone all the other lights and was by far the brightest, appearing at least three times as bright as the Halogen. On the medium power (3W) setting, it still outshone the Halogen due to it being a whiter light but was probably about as bright objectively. On the low power (1W) setting, it was less bright than the Halogen but brighter than the EL300 and not much different to the EL500. However, the beam profile of the EL500 is much narrower and visibility “off axis” was much better with the SolidLights. The beam profile of the SolidLights and the Halogen were about the same. On the low power setting (20 hours with 2 batteries) it is, in my opinion, better than both the EL300 and EL500. As a light to be seen by, I cannot fault it on any of the power settings. As can be seen from the photographs of the reflections, the appearance of the 10W halogen most closely matches the 3W setting of the 1303 aside from the colour of the light. The 10W setting is a whole lot brighter.

Next I took the bicycle out on well lit streets. As expected, none of the lights added much to the existing fairly bright street lighting and running the SolidLights 1303 at the 1W setting is more than adequate.

On less well lit streets, the light put out by the SolidLights 1303 in the 3W setting was about equivalent to the 10W Halogen and much better than the EL300 and EL500. And I am satisfied that in these situations, it will perform as well as the 10W halogen with a much longer runtime (about ~4 hours on one battery as opposed to 2.2 hours for the Halogen).

Out on a dark country lane, the 10W halogen and the SolidLights 1303 on the 3W setting were pretty much equivalent in terms of light output. The character of the light was more yellow with the Halogen. Some things were slightly easier to see with the yellow light but others were slightly easier to see with the whiter SolidLights. However, on the high power 10W setting, the SolidLights easily outshone the 10W halogen in both the medium and far distance and I estimate that it puts out at least double the amount of light in subjective visual terms. The difference in light output and visibility is not much short of what I experienced when I went from a 3W halogen to a 10W halogen at the beginning of last winter. In this test, it appears that the actual light output from a current high power LED light is of the order of 2 or 3 times that of a normal halogen of similar power. This concurs with what Chris at SolidLights intimated but is a little less than implied by Exposure in their “blurb”.

Over the next few days, I used the light on my 30 mile commuting runs comparing it with the 10W halogen. Overall, I could not detect any significant difference between the visibility provided by the 10W halogen and the SolidLight on the 3W setting despite the difference in “colour”. Towards the end of a commute, the SolidLight definitely had the edge with the halogen starting to lose brightness due to its battery starting to run down. On the 10W setting, a single battery managed 1h20 as specified and was an absolute pleasure to use on the commute. The amount of light on the road compared favourably with the amount provided by passing cars with dipped headlights. Road signs shone out and cateyes were visible a long distance ahead. Peripheral illumination was good. On the dark, potholed downhill sections of Common Lane and Quiet Lane, I was able to maintain almost the same speed as I would in daylight. In the face of oncoming cars, it was possible to see the road illuminated by the light at all times as they passed – not possible with the 10W halogen. On the dark winding section of the commute, oncoming cars dipped their headlights before coming into view around corners – something that never happened with the 10W halogen. Runtime with a single battery is just right for the 30 mile commute using 3W setting for about 60% of the distance and 10W for the dark and winding sections. My overall impression is that although this is not a proper “daylight” system, it comes close when running at 10W.
The 1303 unit therefore became my main lighting unit for commuting and rides for the bulk of 2005 and I used it for all my Winter commutes and for a night-time section of a 300k Audax. As a commuter light requiring only 2 hours of runtime at 3W with intermittent 10W use, it works very well. “3W” runtime is a little shorter than expected; just over 3 hours with a single battery before switching to 1W “reserve” I suspect, given the light output measured, that it is running at closer to 4W than 3W. The Audax ride section lasted about 2 hours and was run at 3W providing satisfactory lighting with a single battery.
One other quirk I have noticed is that, if left connected, the 1303 will slowly discharge the battery. I would therefore advise leaving the light disconnected from the battery when not in use for any extended period. However, according to an October 2005 information release from SolidLights, the latest version of electronics has been altered to minimise current drain when the light is switched off but still connected.

Part 2: The 1203D (serial number 000029) arrives...

In late September 2005 I happened to acquire a pair of Schmidt SON28 dynamo hubs and had them built into wheels for my commuter and Audax bicycles. Around this time, Solidlights also released their new 1203D dynamo driven light using twin 3W luxeons at a cost of £140. Not able to resist temptation, I bought one!

This also arrived promptly and was easy to fit and install using the same mounting system as the 1303. I now have both the 1303 and the 1203D mounted upside down on a spacer bar off the front handlebar. The construction of the 1203D is almost identical to the 1303 except that it is 2/3 the width and instead of having a rear power lead socket built into the unit, there is a short flylead with and in-line socket instead. I suspect this is to allow mounting on a fork crown mount without the socket getting in the way.

Initial impressions of the 1203D in operation were good – not as bright as the 1303 at 10W but quite acceptable. The beam profile is almost identical to the 1303. The amount of light produced depends on the speed at which the dynamo is turning. If the light has been in use, there is a standlight of about 500mW brightness available for about 5 minutes. Once moving at less that 2 mph, the light emits bright flashes on a single LED. Between 2 and about 8mph (12kph) the single LED lights up to give steady light of increasing intensity. Above 8mph the second LED starts to light up reaching full brightness at about 16mph (25kph)
I also noticed that, when starting from “cold”, the second LED takes a couple of minutes to come on or comes on at a higher speed – I suspect this is due to the standlight reservoir being charged and thus consuming the excess power.
Activation of the light requires a simple press of the button on the side. A second press will put the light into flashing mode and a third press will switch it off. Unless the standlight reserve is charged, the wheel must be turning to switch the light – if the standlight is working it will allow switching. In flashing mode, a single LED is used.
Road tests revealed subjective brightness of the 1203 to be equivalent to my 10W halogen at speeds between 5mph and 8mph (8-12kph) and better then the halogen above 8mph (>12kph).

Brightness tests show the light to be very visible and at least as good as the 1303 on 3W. The light from the 1203D has a slightly pinkish hue – I suspect these must be a slightly later generation LEDs compared to those in my 1303 – indeed, the LumiLeds web site specification sheet now list these LEDS to have a 100,000 hour 70% brightness life compared with 20,000 hours listed previously when I last looked when researching my original 1303 purchase.
I also did a daylight test of the light on flashing mode – It prompted a bus driver coming the other way to wind down his window and comment that the flashing mode was extremely visible!

Part 3: The pretty pictures and the measurements.

To give an idea of how all these lights perform in “real life”, herewith some photographs of the illumination given by my various lights taken in pretty much pitch darkness. Exposure factors for all the photographs are identical and the background reference light is constant. The photographs clearly show that the SolidLights 1303 at the 3W setting is as good as or even better than my 10W halogen with a matching beam pattern. The 1023D at “8mph” with 1 LED lit to maximum brightness also has gives similar illumination to the 10W halogen. The first “bollard” is at 10m, the second at 16m, the third at 22m, the wall in the distance is at 26m and the fence is at 28m (90ft) With the better beam profile, the SolidLights at 1W is superior to the EL500 running the same power. The relative intensities of the various lights as shown do match real appearances although, with dark adapted eyes, actual visibility is slightly better than implied by the photographs, especially at the lower intensity levels.

The final photograph in the series below reveals the full lighting potential of my current headlight set-up using both 1203D and 1303 running at full tilt – very pleasing!

Finally, I did some exposure and pattern measurements in a similar manner to that described by Phil White in Arrivee 86 (Winter 2004, p48) My LED units were measured in turn. They were placed at a fixed point 3 metres from a blank featureless wall . The beam pattern was measured on the surface and beam angle calculated. The incident light was then photographed and the exposure values measured. The same set of measurements were taken in each case with the only variable being the light used. In essence, light output for all the lights matches the power input although I suspect the SolidLights 1303 at 1W slightly outperforms the Cateye EL500 because it gives a much broader beam. Of the three LED lights, I am most impressed by the SolidLights for light quality and quantity coupled with a more useful beam profile. The 1203D and 1303 beam profiles are almost identical.

Herewith the Beam patterns for the various LED lights:

The measured and calculated beam areas, beam angles and light exposure levels recorded are tabulated below. Measuring the beam width and areas of the EL300 and EL500 is relatively straightforward due to them having quite sharp falloff characteristics. However, the smooth profile and wider beam of the Solidlights units makes measurement more difficult. The values quoted are my “best guess” from repeated measurements. The Solidlights beam profile does not change with the different power level settings. The “Exposure Value” quoted is a relative number calculated from the camera exposure factors for the total light falling on a fixed surface area at a 3m fixed distance. About 95% of the SolidLights total coverage was used as the reference measurement area.

In summary:

The SolidLights 1303 has fulfilled my expectations. I think it will provide a lightweight, good light. The burn time with one battery is just about right for my daily 30 mile commute in the winter and two batteries should be just right for overnight summer Audax rides by using the 1W and 3W settings judiciously. The 10W runtime is too short for sustained Audax use unless the night time section is short, it is used very sparingly or spare batteries are carried. The ease of switching between power levels, even with heavily gloved hands, is a bonus. The light is expensive (£280 inc for 2 batteries) but build quality is good and, as far as I have been able to determine, you get what you are promised. The 1303 with a single battery running at 3W is an equivalent light to my 10W halogen but weighs about half as much (295g vs 570g) and gives almost double the runtime (~4hrs vs 2h20). All in all, 3.3 times the “performance” for about 2.5 times the cost. The Solidlights at 1W clearly outperforms the Cateye EL500 in all respects. Ideally, I would have liked a longer high power runtime but that would mean heavier and more expensive batteries. With the LEDs rated to last at least 20 years before any performance loss at my current usage of about 10 hours per week, I think I will probably be using them for a long time hence; they may even outlast me! It is also likely that with improvements in manufacturing methods and materials that both LED efficiency and battery capacities will continue to improve significantly in the medium term. At 10W the SolidLights 1303 provides good road visibility but not equal to an HID “daylight” system. To get “daylight” for the runtimes I want, I would have to pay a lot more and I still would not get the very long low power runtimes.... The SolidLights supplied batteries are so small and light (133g) that it would be no hardship to carry a couple of spares if even longer runtimes are needed although at £39 each they are not cheap! However, 2500 mAH 7.2V Li Ion battery packs are available in the US for $23 and prices will drop as demand increases. There are also some concerns about the longevity of the LiIon technology with batteries losing capacity significantly after about 2 years of use.

The addition of a 1203D has much improved the lighting system. It can be used on its own as a lighting system to provide ample light for almost any ride but being able to use the 1303 in addition to provide that extra daylight when needed make the combination a ideal lighting system giving good light with unlimited runtime off the dynamo while having a bit of semi-daylight and a backup available when needed. The extra drag due to the hub dynamo (Schmidt), is barely noticeable.

Postscripts:

May 2005: I had a failure of the mounting shoe on the 1303 – hammering my aluminium boneshaker over 2600 miles of Sheffield potholes on the 30 mile daily commute has taken its toll! My EL300 light also failed at the same attachment point. However, an eMail to Chris at SolidLights resulted in three replacement attachment shoes in the post a couple of days later; the Cateye, on the other hand, cannot be repaired and had to be binned.

21^st October 2005: Today it was the turn of the 1203D attachment shoe to disintegrate! What fierce potholes we have here in Sheffield... Chris Jones was kind enough to send me a replacement by first class post the next day. It appears that the screw holding the shoe on had been tightened too vigorously during manufacture – I am reliably informed that assembly line instructions have been altered to prevent this from happening again.

14^th November 2005: More trubble on't front, it seems! On arrival at the end of my morning commute today, I noticed that the 1203D was no longer lit – I usually switch on the light on leaving home in the dark dawn and arrive with the sun so I did not notice when the light failed toward the end of the commute. Once I stopped and the dynamo wheel stopped turning the light seemed to partially “reset” and produced some light when turning the wheel again. However, the unit would not switch to “off” or to “flash”, only one LED lit, the standlight did not charge and low speed flashing persisted to a higher speed. At home, I tested the light with a DC source : it would not light more than a single LED; neither would the standlight charge. It seems therefore that part of the light electronics failed. This was not a catastrophic failure – it still produced 3W LED power for the 15 miles home. An Email to Chris this evening brought an instant reply promising a replacement in the post as soon as he was able to send it. Mine is apparently the very first light he has had in which the electronics have malfunctioned and I have heard of no other failures or malfunctions on the on-line fora I frequent so I must assume it is an unlucky one-off. I have packaged up the old one to return for for forensics...

22^nd November 2005: 1203D received back from SolidLights with the following comments from Chris Jones:

“The fault was simple component failure - just one of those things, unfortunately. There is a tiny 3.3V regulator in there which provides power for the microcontroller and various other control circuits, and it had ceased to regulate. This in turn had damaged the microcontroller which is what was causing the odd behaviour.
With those two components replaced everything is fine now: I ran your light on my test rig at 50km/h for 25 minutes (stressful conditions for any dynamo light!) and encountered no problems. The infuriating thing about the fault, apart from it happening at all, was that I spent a considerable amount of time during the design stage evaluating different types of 3.3V regulator to find out which was the most tolerant of the wide voltage variations the dynamo light sees, and had not had one fail in service without external provocation!
I also took the opportunity to upgrade the case screws and sealing to the latest specification. The 1203D is now supplied with stainless screws and o-ring seals to make it better able to withstand being left out in the rain when switched off.”

The is light now back on the bike and functioning normally. Having used my old BM Oval on the Dynamo in the meanwhile makes me appreciate it even more!

May 2006: I have now used the 1203D for 30 miles every day over the entire winter until now. I would not be without it. It is fully waterproof, looks the same as the day I bought it and has functioned faultlessly since it was returned to me with the new regulator. I usually exceed 35mph over about 2 miles on the way to work every day as well. Otherwise, I have nothing more to report! I get on the bicycle and pedal. During the winter the 1203D has been "on" all the time - I did not ever bother to switch it on and off. I only switched the 1303 on on the very dark and winding bits of the commute when the fancy took me. Both lights (and mounting brackets) survived a tumble due to broken forks unscathed.

September 2006: An email arrived from Chris Jones whilst I was on leave advising of a free upgrade to the 1203D and 1203DR series lights, apparently to correct occasional failures when run at high speeds. 1203D duly sent off the unit for the upgrade and received it back within 3 days....
The 1303 mounting bracket (the bit that fits to the bike, not the mounting shoe) has also failed this month – this is the first time this component has failed after about 10,000 miles of Sheffield potholes so it owes me nothing. No other light/mounting bracket I have owned has survived this distance of all season commuting yet; even my old BM Oval bolted to the crown head disintegrated!
So far, therefore, the weak points of these lighting systems have proved to be:
1. Light mounting shoe – easy and relatively cheap to replace.
2. 1203D regulator failure – Now fixed by the manufacturer with upgraded components.
3. Monting bracket failure after 10,000 miles. Also relatively inexpensive.
I have pre-empted the next mounting system failure by stocking up with spares and adding a zip tie around each light (As shown in the photograph in “Part 2” above)to securely fix the lights to their bracket. Easily cut and replaced if the lights need to be moved or replaced...

November 2006: Battery life. With the advent of a new season of winter commuting, I have been testing the battery packs and their capacity. Lithium Ion batteries are said to have a life of about 2 years (75% capacity) This seems to be correct – the 1303 batteries are now a few weeks short of 2 years old and are now giving me 55-60 minutes at full power on a single pack which works out at just about 75% of the runtime when new. Since I mainly use the light on its 3 Watt setting (in conjunction with the 1203D) with rare intermittent 10W use, it is still lasting the course for my 2 hours a day commute so I won't be replacing the battery yet. However, I suspect that by the beginning of next winter, a new battery will definitely be needed....

May 2008: Another winter of commuting and rides done and dusted and the lights are still working flawlessly.... No further breakages or problems. A expected, the 1303 Li-Ion battery capacity has dropped a bit more and I doubt it will survive next winter; it only just had enough capacity to cope with the commutes these last 6 months. I am well pleased with it lasting 4 winters of commuting, though.... The 1203D is still working flawlessly and is remains quite adequate for my needs. I have seen several reports of newer LED lights with better output over the past couple of years but have not yet felt the need to splash out for a replacement or "upgrade". Maybe next year....

October 2009: Another winter has started and lighting systems reviewed. In September this year, I decided to take up Chris Jones' offer (for £70, it must be said) of upgrading the 1203 light to the new XB2 version with more modern Seol P4 bin 'U' LEDs. I sent off the light and received it back a week later completely refurbished with the new LEDs in place. (And a new mounting plate – the previous having failed again....) The new light is about 2.5 times brighter than the old version with a slightly sharper beam pattern. It now gives me pretty much the same light output and beam pattern as the combination of 1303 and 1203D used to give previously. The battery on the 1303 is now no longer good enough to last a full commute and the light has been retired until I make a new battery for it. The 1203D is quite sufficient on its own. The operation of the 1203D has also changed – both LEDs now come on at the same time. I suspect that, with brighter LEDS, the low speed performance improvement of using a single LED below 10kph is negated...
Further news, I have also purchased a Bush & Muller IQ Cyo Senso+ and a Supernova E3 Triple in the last year. These are Pretty Good Lights! The Cyo has a rectangular, more confined beam pattern but puts it squarely (!) on the road with good dazzle suppression to oncoming traffic. The visibility is slightly better than the SolidLights, mainly due to concentrating the available light output to the road. Added to this is the automatic activation on falling ambient light levels. As a fit and forget light, the B&M IQ Cyo would be my recommendation for ease of use, light levels and price. The Supernova E3 triple is Very Bright^TM. The beam pattern is almost the same as the 1203D but more than twice as bright (quoted 550 Lumens as opposed to ~220 Lumens for the SolidLights XB2). For general commuting, this is just a little too unfriendly... The dazzle factor due to the amount of light escaping above the horizontal, even when the light is tilted down 15 degrees as it should be for maximum road illumination, is a little too much. I am considering making a little reflector to modify the beam pattern.... I suspect it will be used mainly for Randoneuring and quiet night rides....

October, 2011: Final update to this article..... Solidlights have stopped making and shipping their range of lights. The era has ended. The "XB2" continues to give good service. It now occupies space on my "summer" commuter which is also used in the spring and autumn and is likely to remain there. Small, unobtrusive and efficient. The Winter commuter now uses the IQ Cyo on the front with a matching rear light. This gives a more concentrated beam pattern and thus slightly more light on the road. It is my preferred dynamo light in the wet and in winter.

© Pieter Meiring, 2004,2005,2006,2008,2009,2011.

Last Updated: 18:20, 2^nd October, 2011)

Appendix 1: Solidlights Specifications

Appendix 2. USE Exposure - comments and power measurements

Quoting Chris Jones:

“The (USE) Exposure unit uses a pair of 5W Luxeon V Portable emitters wired in parallel driven by an unusual regulator design. The LEDs are connected almost straight across the battery (two Li-Ion 18650 cells - nominally 7.4V 2150mAh, although I haven't verified this) with only a 0.22 ohm resistor (used almost certainly for current sensing) and a switching MOSFET in series with them. Note that there is no inductor involved - this isn't a conventional buck or boost regulator. The regulation is done digitally using a microcontroller, and results in perceptible flicker in the light output, as well as variation in current consumption: the ammeter on a bench supply used to power the unit wanders up and down by about 10%.

The regulator design is more like a PWM (Pulse Width Modulation) halogen regulator than a conventional LED driver, so making meaningful measurements involved a lot of work with an oscilloscope and a spreadsheet. I can provide more details of the measurement method if anyone's interested.

The results are interesting. The power delivered to the pair of LEDs varies from 6.6W with fully charged batteries down to 2.3W just before the unit shuts itself off.

The voltage measured is the battery voltage under load: the highest value, 7.50V, was determined by running the unit from a freshly-charged battery. The odd jumps up and down in power at certain voltages were repeatable - presumably a consequence of the regulator's algorithm.

Looking at the results, it's clear to me, at least, that the surprisingly long run-time for the claimed power is mostly due to the fact that the LEDs are only being driven at just over half power for most of the battery's discharge curve.

The full table of results is here:”

Appendix 3: Chris Jones' comments on the SolidLights 1303 design:

Regarding interference with other devices: “Solidlights are designed using 'best practice' as regards RF emissions, with awareness of the strict regulations now in place. Indeed, the design sacrifices some efficiency to reduce the potential for RF emissions. They have been tested with a wide variety of devices - everything from a long-wave radio to a GPS receiver - and found to cause minimal interference. My GPS receiver (a Garmin Geko) works fine even when placed on top of a 1303 light, and my own Polar heart rate monitor works when given a few inches of space from the light, so whilst it's impossible to test every situation, I'm reasonably confident that any interference problems could be overcome.”

Regarding SolidLights power delivery to the LEDs: “A Solidlights 1303 under normal conditions will deliver more than 9W to the LEDs regardless of the state of the batteries - the actual output power is deliberately temperature-dependent, and reduces as the internal temperature rises to protect the LEDS. The 1303 will never actually cut off due to overheating: the case temperature stabilises at around 70 degrees Celsius with no airflow and an ambient temperature of 25 degrees.”

Regarding LED longevity: “Longevity with LEDs is largely a matter of keeping their temperature down, and to that end the thermal contact between the LEDs and the case in a 1303 is designed to be good.”

Regarding SolidLights power supply and regulation: “The regulator in the 1303 is a boost regulator, which means that its output voltage must always be higher than its input. In this case, the output voltage is 3 LEDs in series, which is just under 11V. If the input (battery) voltage rises above that, the regulator loses control. To prevent damage, the microcontroller automatically shuts the LEDs down if it detects an input voltage which is too high. The working voltage range of the 1303 is really 6.0V to 11.5V. I can build a custom version of the 1303 with a buck regulator (as used in the 1103) which will work from 12V to 16V, for people whose batteries are in that range. The reason there isn't one model covering both ranges is simply efficiency: a regulator which can cope with the output being lower or higher than the input is hard to make with more than about 70% efficiency in practice, which is too great a compromise for me.”

Appendix 4: Power supply characteristics of Cateye EL300 and EL500 LED lights.

In comparing lights for this article I spent some time trying to find out the quoted power consumptions for the Cateye LED lights that I use but drew a blank. As a result, I decided to investigate the power consumption characteristics myself. Examining the units, there is no regulation of the supply voltage to the LEDs in these lights; the batteries appear to be connected directly to the LEDs via a switch and current limiting series resistor. The value of the resistor was measured at 16 ohms on the EL300 and 12 ohms on my EL500. Measurements revealed that the current supplied to the units is voltage dependent with an almost linear increase in current with voltage above 4V. The power supplied to the LEDs thus varies with the square of the supply voltage.

With the EL300, supplied power varies from 170mW at 4V to 450mW at 4.8V up to 1W at 6V.
With the EL500, power varies from 340mW at 4V to 700mW at 4.8V up to 1.3W at 6V.

The 2500mAH NiMH batteries used in the EL500 during my tests provide 5.3V fully charged giving 0.9W.
The 1800mAH NiMH batteries used in the EL300 during my tests provide 5.1V fully charged giving 0.5W.

Measured power input to the EL300 and EL500 units:

Given the lower voltage output of most rechargeable batteries (1.15-1.35V per cell) and the relatively rapid voltage falloff in Alkaline batteries with discharge, I suspect this is why these lights give lower than advertised performance. However, the linear decline of supplied current with voltage, saving the discharging battery as its voltage drops, does explain why they continue to emit light, albeit reduced intensity light, for such a very long time.

Access Count: