Solar Based Up Convertor Handles Remote Sensors

Transcription

1 Solar Based Up Convertor Handles Remote Sensors This project is an up-convertor with a mix of functions. This gadget essentially converts a low voltage source to a higher voltage source. Usually you can accomplish this without much fanfare by taking advantage of one of several switching regulators. This up-converter function isn t particularly exciting unless it is built around an application that does something special such as combining inexpensive solar and wireless technology. While in a Walmart, I noticed some inexpensive solar yard lights. They were the units with the smaller solar cell formats, shown at right in the photo below. These were about 1.25" square. These produce about 1.5 to 2.3 volts at the solar sell output in direct light. I eventually discovered that cell power really is more about how much current is generated. Yard lights with the bigger cell formats were also on sale. The larger ones produce almost 3.5V at the cell outputs in direct light and provided a bit more power. Solar powered yard lights So next, while evaluating a new wireless application for a variety of projects, I happened to be strolling thru Harbor Freight. There I found a remote wireless IR proximity monitor. These units use a 9-volt transmit module and a complete, battery-based receiver display unit. Solar-based power is a solution which has become a popular idea very green, very cool, and even possibly very cheap. Wireless projects have also become popular, I thought that combining solar and wireless would be a winning combination. While I found my wireless tools at Harbor Freight, these items can be found in almost every hardware store. I picked this IR sensor system because it was a really good application for this project, it would have a reasonable wide appeal and it was on sale for $8.99. The up-convertor

2 circuit doesn t actually care what the load is. It is the up-converter function I wanted. I intended to use it again and again whenever needed an up-converter for a project. I have included a load analysis below. This should walk you through the process which identifies the panel/cell loading for this project and also give you a bit of insight on battery needs based on your own battery-based projects. Battery technology has been improving for some time, with 1200 milliamp hour (mah), even 2400mah, AA cells, 400mah 9-volt batteries and so on. Many of the improvements are in the areas of battery-power management. That is, in the regulation and control of the battery's discharge and re-charging efficiencies. The transmitter module comes with a 400mah rechargeable Cad battery. The mah ratings are also a bit misleading. A 100mah battery suggests that a load may draw 100 milliamps, (ma) for one hour. At that time, the voltage will be reduced to 0.7 of the fully charge/rated voltage. Test methods and loads are established to provide a comparison of output power loading versus the voltage. With guidelines to define trickle charging and deep re-charging profiles, applications may not be able to operate at 70 percent rated voltage. The 10ma discharge rates used to create the mah rating may also not represent your application s load. Deeper discharge rates may result in non-linear battery longevity, given specifics of chemistry and heating. Battery performance is (or should be) well documented by the manufacturers, but maybe not clearly given the many variables of a particular application. Referring to the schematic, this project brakes down to three parts. The first is the solar related parts and wiring, the second the up-convertor and the 3rd, the actual application. Since the wireless system is already battery based, a solar powered, battery-based application may seem redundant. Maybe it is, but considering the bigger picture it is still both a good example and has the advantage that it will operate in relative, power, autonomy.

3 Schematic Recognizing that both the battery storage and solar source needed to be characterized for a solarbased project, I used the simple variable current load shown below to measure the no-load voltage and then determine what current can be drawn in direct sunlight. This subjective criterion provides at least a hint as to how much load could be supported by the solar cell. While not a perfect analysis, this, at a minimum, allows you to evaluate the relative power from one panel or cell to another. In direct light, measure the cell output voltage with the SPDT to position one, Vpeak. Calculate 0.7 of this value, record as Vload. With the NO button depressed (closed), adjust the pot until the voltage read on the meter is this Vload value. Switch the SPDT to position two and now measure the voltage. Divide that voltage by the resistance of R1 (10 ohms). This is the current you might expect and the product of Vload.* This current is the power you might expect to be available from the panel or cell.

4 Now instead of 0.7* Vpeak, repeat for the voltage which corresponds to the minimum needed to charge the battery. For the application shown here this is ~ volts. The power from this is might be what you expect to be available to the application. To begin, we really need to due an analysis of what our load will be. Beginning the analysis, there are a few rules. First off, each convertor has a conversion efficiency based on a number of operating parameters. Each component also has an operating current, which is the current drawn at the components power pin that is needed by the part to provide its function. Discrete and other parts (resistors, capacitors, and diodes etc.) will also consume power based on the voltage and current characteristics. Power is determined by the voltage applied on the part, in volts, times the current it consumes in amps. Power is measured in watts. For most applications we can more conveniently refer milliamps, (ma), millivolts, (mv) and milliwatts, (mw), where the milli prefix applies a scale of 1/1000. The operating current is typically referred to as the no-load or quiescent current and it may be specified as at no load current. Remember that the part will be loaded, so if you look closely to the part spec you will find a graph or spec which may have to be interpreted to get to the operational current. Other important power characteristics and efficiency includes temperature, input to output voltage difference, load and even layout. Consequently, for most of us, the best we can do with any analysis is a SWAG (Super Wild-Ass Guess) as opposed to the commonplace WAG (Wild-Ass Guess). An analysis begins with identifying the power modes found furthest away from the source. These are distinct operating states which will have distinct power requirements. For this application, the IR transmitter intends to use a 9-volt battery, which we are replacing. There are at least two power modes identified in the transmitter, one of which operates continuously, powering of the IR source, background mode and a second, upon a detection of an IR target, when transmitting. Using some simple measuring techniques, I found the nominal 9-volt battery drew 90 microamps (ua) when in background mode. During a detected signal this increased to 5.6ma for ~0.8 second during transmit. Referring to the transmitter source this represents a 9-volts*.09ma and 9-volts*5.6ma or 0.81mw and 51mw power modes. For now, we can back this load through the doubler and then take that load through the Maxim part. Because, at least the transmit power mode does not operate continuously we will also have to account for the ratio of on to off time, duty cycle, associated with each mode. Referring to the MicroChip, TC1044 data sheet, the doubler will have a 90 percent efficiency during the loaded, 5.6ma mode and a 75 to 80 percent (0.78) efficiency at the low current mode of the transmitter. The doubler no-load operating power is ~0.6ma, adding a 5*0.6ma or 3mw burden. If we assume the transmitter mode is activated N times per hour and accommodating the duty cycles for both modes, the power load seen at the Maxim, MAX856 part is: (1-N*.8/3600) * 0.81mw/0.9+ N*.8/3600 * 51mw/ mw For instance if N=20 then this is: 0.9mw mw +3mw =~4.2mw

5 At the Maxim 5 volt output this is 4. 2mw/5v or 0.84ma The Maxim part using a Schotkey diode, D1, at this load, is 75 to 80 percent (78 percent) efficient. With the Maxim chip power, discrete part leakage and other power losses, this portion of the hardware will need ~1.2milliamp (ma) from the battery source. The output voltage is fixed and defined by the to ground strapping at pin 2, setting the output to 5 volts, at pin 6 of the MAX856. I replaced the 400mah that came with the stick with two, 1200mah AA batteries. I had found the convertors were more efficient at a higher input voltage and the load, still being fixed, resulted in a lower than proportional current requirement. These batteries, in series, produced 2.6 volts, fully charged, 2.4V nominal. Using the batteries at 2.6V, the power seen by battery is then: 4.2mw/ v*1.2ma or 6.4mw. This corresponds to 6.4mw/2.6v or 2.45ma/hour load. Recall that the 1200mah batteries are based on the battery voltage falling to 70 percent of 2.4 nominal voltage on the batteries in series and that the convertor/hardware is fine with that. This means you could expect 1200/3.3ma or 363 hours of operation, without any recharge. This equates to a very long, dark night or a few cloudy days. There are also rechargeable 2000mah, AA batteries available. Next, I modified the solar yard stick. This particular unit included enough smarts to charge the battery in the daytime, detect dusk and manage the drive to its LED light. Not relying exclusively on light and requiring battery power, even periodically and after dark, I found that it was best to merely cut away the units control boards. These were the larger cells, shown below. They produce about 3.5V (Vpeak) in direct sunlight. There are more than several light stick diameters and lens configurations. I had fortunately selected one, not just on sale, but which had a relatively large lens area. This worked out to an advantage when mounting my assembly and will be a consideration should you follow the general instructions described below. Exact battery life depends on the several variables mentioned. You should note that the number of transmissions per hour had little effect on load. While this power mode is higher it represents a fractional portion of time. The actual, averaged measured load through the meter from the batteries was 2.7ma. Note that the MAX856 is a switcher and the in-rush (start-up) and peak current requirements will be higher then these filtered/average measurements. That is why, for this application and load, a battery is required to fill the relatively short-term peak current demands. Temperature extremes as well as discharging and recharging also takes a toll on the batteries, so please never assume that anything lasts forever. I am pretty confident that this application will provide a nice gadget and in fact has many opportunities that would benefit from solar supplemented power conversion

6 As purchased solar light stick I cut through the existing battery holder shown and provided space for a 2 cell battery container. Unsoldering the solar cell leads, extending them a bit, I routed them thru the cut out made through this battery holder. For applications which would rather maintain the single AA as the battery, the regulator design is changed to substitute the LM285Z-2.5, in the schematic with a LM285Z-1.2. This may be for applications that require periodic power or perhaps those that operate on 3.3volts. The dual AA holder then extends through the secured base into the plastic lenses portion which holds the regulator and convertor hardware. Modified for 2 AA batteries. Regulator portion of hardware mounted separately

7 The second portion of the hardware is the up-convertor. The Maxim MAX856 up converts the batteries 2.6volts to 5volts, while the MicroChip TC1044 doubles the 5 volts to ~9 volts. Boost convertor and doubler Battery, regulator and convertor hardware Finally the connection to the transmitter routes either externally or through the yard light s base. The transmitter battery compartment is weather sealed and the wiring to make this connection should, assuming an exterior application also maintains this sealed philosophy. As only a small diameter is needed to route the two power wires, through both the plastic lenses works and battery lid to the transmitter, these were easily sealed with an external grade silicon caulk. Mounted stick and transmitter assembly My application looked like the example shown in the photo below, however there are several other alternative mounting choices that could have been used. The module s manual suggests for best coverage a 3-foot, elevated, mount for the transmitter. So mounting the transmitter/sensor slightly upward makes sense. For driveway, garage or yard gate applications, options could also include mounting the solar part in one place and the transmitter module in another. The included purchased receive module includes audio and visual indicators. Powered with 3 C cells, there is also the option of a wall based power source, I am sure sold separately. I have started the next gadget project and you may see this solar front end, convertor/doubler hardware again. For you professional power/solar designers, if any, I hope I haven t too terrible misrepresented or simplified the subject.