美国工程师的高性价比太阳能车道照明灯项目

我的太阳能车道照明灯是一个又棒、又便宜的科研项目——我最近安装了一些现在已经常见的太阳能电池供电的车道照明灯，(“安装”也许显得太正式，“在路边插了一些车道照明灯”也许更准确)。在当地每一个机车基本路段大约花费3美元，他们在短期和长期的工作情况是很值得期待的。

他们采用一个单一的、用户可访问的、易于更换的AA型镍镉(NiCd)充电电池，这一点使其显得特别便利。(讽刺的是，我不得不返回一个单位，这是一个有趣的失效模式失败：它不能熄灭照明LED，而是整天都有亮着，因此无法完全地自行充电。)

由于这里是夏天，这些灯都能够获得很多自由光子，所以在运行了7个小时之后，即日落的时候，这些灯将会自动开启。但是，我也在考虑初冬的情况，我将会对他们的运行情况进行比较观察。

都获得这些自由光子地段及运行约7个小时后，他们又在自己的夕阳。但我也想到了冬天的前面，我会看他们如何在比较执行。我发现由于季节变化可能影响他们工作的几个因素：

*最明显的，将有光照时间缩短，和太阳的低角度引起的光线减弱

*由于天气寒冷，气温降下降到零度或者以下，这将影响电池的蓄电和充电的效率

*太阳能电池本身可能不能有效的将光子转换成电流

*电池内部因素，当温度下降到零度或者以下，照明级别装置开关电流也会受到影响

*在设计上，镍镉电池将有一个每日充电/放电周期，电池的充/放电次数将是多少？这也是个问题。

虽然我不能建立一个正式的数据采集站来收集太阳照射角度平均值、周围环境温度、黄昏之后的精确时间，以及其它参数。但是我还是将继续每天(包括夜晚)做一些非正式的情况记录。显然地，这是一个充满热情的、初级的科学家或工程师在低成本的情况下，根据实验来获取一些数据收集和分析的参与过程。

但我的理想不止于此。我的下一个步可能会申请拨款，正式研究这些太阳能灯，也许分析深入比较研究不同的机型。我需要一个温度可控制的环境舱，包括太阳能相当于照明和校准控制/测量设备。如果我真的雄心勃勃，我甚至可以做一些全生命周期环境成本，因为那些灯可能只有几年的使用寿命——电池、太阳能电池、充电电流或者塑料外壳中的一样或几样将会退化或彻底的失效。

毕竟，这个课题拥有了所有的“热门关键词”：太阳能供电，可再充电，绿色，无废气排放，以及类似的词汇。所以，我该去哪里申请？

点击进入第二页：美国工程师们对该项目的看法

{pagination}

Duane Benson：多年前我也曾买了两盏太阳能灯。我将它们当作挂灯，而不是固定在地面。但是同样的东西，在当时需要花费大约20美元。其中一盏灯用过了两个冬天；而另一盏灯只用了两个月。

原本我以为质量会随着时间而得到改进，但是一些现象告诉我随着价格的不断下降，质量也不可能得到提升。

我的经验跟Duane Benson先生的很像，需要一个更为复杂的充电电路、运动传感器和大容量电池，才能照亮我的车库大门。那些可充电电池设备工作了3年。

这些设备的花费将是替代而不是改进。这就是现代市场和早前市场的区别，这也是一个非常可怜的经济发展规律。

……

点击进入参考原文：My solar-powered driveway lights are a nice, cheap, science project, by Bill Schweber

翻译/增值：Ana Hu

《电子工程专辑》网站版权所有，谢绝转载

{pagination}

My solar-powered driveway lights are a nice, cheap, science project

by Bill Schweber

I recently installed some of those now-common solar-powered driveway-marker lights ("installed" is perhaps too dramatic; "stuck them in the ground" is more accurate). At around $3 each at the local Home Depot, hey, it's worth seeing how well they work short- and longer-term (see here).

They use a single, user-accessible, easily replaceable AA-size nickel-cadmium (NiCd) rechargeable battery, which is especially convenient. (Ironically, I had to return one unit which failed with an interesting failure mode: it would not turn the illumination LED off, but instead stayed on all day, and so was unable to fully charge itself.)

Since it is summer here, these lights are getting lots of those free photons and running for about seven hours after they turn themselves on at sunset. But I am also thinking about the winter ahead, and I'll be watching how they perform in comparison. I see several possible factors working against them as we shift seasons:

　 * Most obvious, there will be fewer hours of sunlight, and those rays will be weaker due to the sun's lower angle

* It will be colder, getting down to–and likely below–freezing, which will affect the battery efficiency at storing and returning charge

* The solar cells themselves may not be as efficient in converting photons to current

* Their internal, light-level-based turn-on/turn-off circuit also may be affected by the temperature, as it drops down 0°C or lower

* By design, the NiCd cells will have a daily charge/discharge cycle; how many of these cycles can the batteries tolerate?

While I won’t be setting up a formal data logging station to collect data on mean solar exposure, ambient temperature, after-dark "on" time, and other parameters, I'll keep informal notes on the daily (and nightly) situation. Clearly, this is a experimental set-up that an aspiring, junior scientist or engineer could get involved with, and gain some data-collection and analysis experience at low cost

But I am thinking much bigger. My next step may be to apply for a grant to formally study these solar-powered lamps, and perhaps analyze different models for an in-depth comparative study as well. I'll need a temperature-controlled environmental chamber, including solar-equivalent illumination (think of all those watts!) and calibrated control and measurement equipment. If I get really ambitious, I could even do some total life-cycle environmental costing, since these lights probably have a viable life of only a few years –one or more among the battery, solar cell, charging circuit, or plastic enclosure will likely deteriorate or fail outright.

After all, this topic has all the "hot buttons": solar-powered, rechargeable, green, no emissions, and similar. So, where do I apply?

Comments：

Duane Benson: I bought a couple of these a number of years ago. They were hanging lights rather than stick-in-the-ground lights, but otherwise about the same thing. They cost about $20.00 at the time. One lasted through two winters. The other lasted about two months before it's on-time dropped below about 15 minutes.

I'd like to think that the quality has improved in the intervening years, but something tells me that along with the drop in price, they are unlikely to have much of an increase in quality.

My experience is much like Mr. Benson's, but involved a unit with a more sophisticated charger circuit, motion sensor and larger battery, to illuminate my garage door. The device lasted about 3yrs, which in AZ is good for a rechargeable battery.

Ongoing spending for these devices will be for replacement, not improvement. This is what differentiates the modern market from previous markets, and is a pretty poor formula for economic development.

Dr.G.S.Darbari, India

The first thing I did to improve the boost converter was to throw away the cheap axial lead 100uH inductor and replace it with a good SMT low ESR 10uH inductor. This boosted the light output substantially. I was able to add two more LEDs in parallel and the current draw from the batter was only 13mA compared to 10mA with the only inductor and one LED.

The next step was to throw away the cheap aluminum electrolyitc capacitor and replace it with a good 1uF ceramic capacitor. This capacitor was across the LEDs and served to bypass the high frequency current spikes from the boost.

I next added a 10uF ceramic capacitor at the input of the boost converter. While this did not result in an improvement in current or brightness, it did drop the low end of the boost converter input voltage range by about 50mV.

The other part of my project is to improve the runtime of the lights since I live in the Seattle area where we don't get as much sun as other parts of the country. I plan to design a small, efficient converter to boost the voltage from a pair of small polycrystalline cells to charge the battery.

I figure that by the time I am done with this project, I will have taken a $4 piece of junk and turned it into a $20 gadget that only a true nerd could love.