Category: Projects

“Off-grid eCharger”

The aim of this ‘off-grid’ solar energy project was to build a charger for an eBike so that its battery could be charged (when not in use) from power collected from the sun via a ‘fixed’ off-grid installation, rather than relying on mains energy. This can also be used for other battery charging and small backup supply purposes.

Four main elements were put together to create the solar charging ‘system’:

Solar Panel

This fundamental part performs the key task of converting sunlight into electrical energy. There is a wide range of panel types to choose between. Firstly, sizes include the very small, through 50W, 60W, 80W and up to 100W. Thereafter, systems are typically built from connecting multiple panels together.

Then, there are four main types of solar array:

a) Poly Crystalline – cheaper to manufacture, sensitive to high temperature, less efficient with a shorter lifespan
b) Mono Crystalline – more expensive, more efficient, with a longer lifespanc)
c) Thin-Film Amorphous (A-Si) – flexible, lower cost – easier to produce, shorter lifespan, lower power
d) Concentrated PV Cell (CVP) – very high performance, solar tracking and cooling required, very expensive.

Finally, there is the choice of construction type:

i) Rigid – cheapest and most solid, heavy and longest lasting
ii) Flexible – PET – light-weight and bendy, ideal for vehicles and temporary installations
iii) Flexible – ETFE – longer lasting, marine grade, can be walked on.

For this project, the aim was to use something lightweight that could be easily attached to an existing small building structure by means of the re-enforced ringlets in its corners (not all panels have these, so choosing had to be careful), rather than a permanent roof-top installation, and so the Flexible PET type was indicated.

After some consideration of likely ‘sun-hours’ – estimates tend to indicate an average of 2-3, depending on the season and weather – and also cost, a 50W 12V Mono Crystalline Flexible PET panel was selected. This size could prove to be an under-estimation for the task, but this can be supplemented later as necessary since its possible to join panels in parallel to increase current (or series, to increase voltage). That will require ‘T’ or ‘Y’ joining pieces for the MC-4 solar cable connectors, and potentially in-line fuses to protect separate ‘strings’ of connected panels (on their own they don’t need the fuses since their short-circuit current rating is sufficient).

If a panel becomes partly shaded, its whole output diminishes. Hence the advantage of having separate panels is that they may contribute more energy individually that using one large single panel. Another aspect is that it is likely that the cost of solar panels will reduce over time, and so adding to the system over time makes sense, rather than trying to ‘future proof’ the power requirement.

Charge Controller

The task performed by this element is to ‘condition’ the output of the solar capture, converting an unregulated supply of around 18V (for a single panel) to a more stable 12V, the energy from which can be directly used or stored in the battery. The use of a controller avoids the risk of over-charging of the battery or issues with varying power, and will shut off its output to the loads when the supply is diminished and the battery is discharged. There are many models, which vary in power rating and functionality, some with simple LED displays and others showing a range of information, and may include sense terminals for long cable runs, temperature measurement for adjustment due to the weather and equalisation functionality for spreading charge across multi-battery systems.

The controllers fall into two main types:

a) PWM – Pulse Width Modulation

b) MPPT – Maximum Power Point Tracking

The latter of these is much more efficient, likely to convert more of the captured solar energy into stored power. However, these tend to be at least 3 times more expensive, often much more.

A modestly–priced (reasonably-functioned) PWM unit was selected for this system, with a 20A rating (more than enough for the initial project requirements). The reasoning for this product selection is this: with a small system, should the captured energy be not enough, then purchasing additional panel(s) to increased the wattage is more cost-effective in terms of energy produced, compared with the gain from the more efficient controller. It only becomes worth purchasing the MPPT version when much more energy is being generated (and therefore ’lost’ with a less efficient charge controller).The controller model selected is capable of both 12V and 24V operation, which allows for the increase the voltage rating of the system if desired in the future.

Battery

The traditional choice for this type of storage application is the ‘Lead-Acid Cell’, though it needs to be of a Deep-Cycle ‘Leisure’ variety, which is intended to be frequently discharged and recharged, rather than the ‘starting’ classification which is used in a vehicle for short-bursts of energy to get a motor running, after which its mainly in stand-by and therefore expected to be 100% charged most of the time.

Apart from this distinction, there are two main types of product:

a) ‘Flooded’ Wet-Cell or Valve Regulated Lead-Acid (VRLA)

b) Absorbed Glass Mat (AGM) or Gelled Electrolyte (Gel)

Of these, the Wet-Cell traditionally required periodic ‘maintenance’, topping up with fluid, though VRLA sealed versions are now available, and tend to ‘vent’ gas and can’t be tipped over due to the fluid inside. The AGM and Gel types however are available maintenance-free, don’t ‘vent’ gases and don’t have fluid that can spill out.

The AGM-type was selected for this project as, whilst being a bit more expensive, offered the convenience of a product shouldn’t require special attention; these are less sensitive to over-charging and can stand for 30 days totally discharged without harm.

There are a number of measurements of the capacity and capability of batteries:

• Reserve Capacity (RC) is how long the battery can deliver a constant current of 25A at 80F
• AmpHour (AH / C20) represents how much charge is stored. It is the energy delivered continuously in 20 hours at 80F without falling below 10.5V
• Cold Cranking Amps (CCA) is a measure of the ability to start an engine cold. It is the number of Amps delivered at 0F for 30sec by a 12V battery whilst maintaining at least 7.2 Volts

A capacity of 110AmH was chosen for this project.  Whilst ‘the more the merrier’ in terms of storage is certainly the case, this capacity should be sufficient for the basic storage needs.  It is advisable however to allow for discharge of only 50% of the stated rating, whilst also bearing in mind the possibility of a few days without sun.

It is possible to add batteries in Parallel to create more storage, or in Series to increase the voltage of the system. For a modest system set-up, 12V is sufficient and convenient for the purpose. Adding another battery to double the system voltage to 24V would have a side-benefit of requiring lower gauge of wiring.

Inverter

In order to convert the 12V DC power generated from the Solar Panel and stored in the battery, to the 240V AC required by the Load, the Li-Ion battery charger adapter, an inverter is needed. This wouldn’t be the case if a suitable 12V charger-adapter could be sourced, which would avoid the inefficient ‘double-conversion’ of 12V DC into 240V AC and back into 42V DC again. The inverter does however provide the flexibility for powering other equipment requiring an AC ‘mains’ supply and therefore is ideal for a range of charging and backup tasks.

Apart from the selecting the correct voltage, which is normally either 12V or 24V (12V required to fit with this system set-up), there are two main choices:

a) Modified Sine-Wave

b) Pure Sine-Wave

The first type is the low-cost product, and available in a wide range of power-ratings, so if high-power is required then that tends to be the preferred choice. The draw-back is that it is less-efficient, creating a ‘blocky’ AC waveform which can upset sensitive equipment. The second type produces a smooth sinusoidal AC waveform equivalent to the standard AC ‘mains’ supply, and therefore is suitable for all equipment types, though at higher cost and usually lower-power rating.

Given the performance uncertainty and potential for harm to some equipment that the MSW product might bring, and the modest power needed, it seemed sensible to select the PSW variant. A 300W product was chosen, partly limited by cost, but also since it is recommended to ‘match’ the power rating to the consumption requirements; in practice the inverter size should be around 3-7 times the power needed by the load appliance. A much higher-than needed inverter is ultimately less efficient for smaller loads, consuming more power than necessary.

Additional parts required were cabling and mounting accessories, plus some optional ‘add-ons’ in the form of other battery chargers and lights to make full use of the generated power:

System accessories

• An extension cable to connect the solar panel to the solar controller. The panel itself comes with about 1m of ‘tails’, fitted with MC-4 connectors which are standard for solar equipment. A ‘male’ connector provides the positive supply, and a ‘female’ connector offers the return negative connection. Two cables are therefore required, with the appropriate MC-4 connectors at one end, and unterminated at the other to connect to the solar controller. A ‘top tip’ is to obtain a single fully-terminated extension cable of twice the required length, and cut the cable half-way. For this project, a 25m extension cable was sourced, which provided 12.5m as a sufficient length for both positive and negative connections once cut in half. Thickness of the cable is an important consideration, so that the rating is sufficient for the power of the system and the loss due to the length of the connection. For this installation, a 4mm2 (12AWG) cable was selected, suitable for up to around 15m and 23A, meeting the needs.
• Battery terminal leads
• Midi and Blade Fuses and holders for fusing of the output from the battery supply (40A) and protecting the input to the charger controller (15A). The inverter came with a 35A mini-blade input fuse.
• Lugs and wire for the Solar supply and DC power connections
• Fixing screws for the solar panel, and cable clips
• 12V Lamp with remote and motion sensor, providing a light powered by the solar energy for the indoor work area
• AA / AAA / C / D / PP3 battery charger, for supplementary battery charging requirements, fed from the 12V output from the charge controller (avoiding unnecessary 240V conversion)

The resultant assembly captures energy from the sun via the solar panel, ‘conditioned’ by the controller (to avoid over-charging or issues with varying power) and stored in the battery. Naturally, this can take place throughout the day when the sun shines, without the need for a load to be present. Then when required, the load can be supplied from the stored energy.

The inverter is utilised to supply a replacement AC feed to the specialised Li-ion charger required for the load battery, which can therefore be used as an ‘off-grid’ power source for other uses in place of a regular ‘mains’ supply.

Charging additional Li-ion or Ni-Cad batteries for other equipment can be also be facilitated using the 12V output to supply an appropriate battery charger, and as well as powering 12V LED lights for illumination of the work space.

Hopefully this project is of interest and of use to anyone contemplating establishing a solar energy system, with the aim of harvesting energy from the sun, a ‘free’ resource available to us all!

@YellowsBestLtd our mission is in “Keeping Customers Operational”. We’re always keen to enhance our range of #business services, widen our #enterprise infrastructure support  and expand our mix of #sustainable solutions we offer for supply and maintenance of new and legacy #technologies and products for our customers.

Please help us understand what would be of interest to you by getting in touch to discuss your management services or solutions requirements, whether you’re implementing new systems or maintaining existing infrastructure to serve your operational business needs. We look forward to hearing from you.

I am pleased to announce ‘finally’ the completion of this construction project which dates back to 1991, when it was first commenced and then ‘shelved’ for a quarter of a century. Returning to it, the most amazing thing was the intact condition of the mechanical build and electronic components, as well as the paper notes and Silvine Originals exercise book (thank goodness, what chance would an electronic ‘soft’ copy had it existed be readable now?) which had been preserved perfectly despite various moves and long-term storage.

It’s based on a design by John Becker published in Practical Electronics magazine, which appeared in three parts from November & December 1988 to January 1989. The magazine title subsequently merged with Everyday Electronics to form Everyday Practical Electronics (EPE) (and then later absorbed Electronics Today International) which survives today.

The project itself comprises the build of three main electronic elements, forming the three parts of the published design: PSU, Time-Base and Y-Drive. Coupled with these is the mechanical housing to enclose all the elements, and not least is the inclusion of a Cathode Ray Tube (CRT), which provides the traditional display device for the instrument.

Mechanical Housing

This element of the project needed to ‘come first’ as it forms a fundamental part of the construction, not only containing the PCBs of electronics but also serving as the mounting frame for the necessary controls, switches, inputs and outputs, including the display screen formed from the CRT tube. Naturally finishing this stage, including the wiring of all parts, had to wait for the completion of the PCBs and so was the ‘last part’ as well.

A suitable 19” rack style box was chosen, providing more than sufficient enclosing space, the actual constraining factor was the necessary size of the front panel. The template provided from the magazine notes was photocopied and enlarged several times from the published artwork to produce a ‘full size’ design which was then stuck to the front panel of the housing. This then enabled all the holes for the parts to be marked and drilled, including the large circular hole (formed by drilling a series of small holes in a circle) for the CRT face, which was then smoothed and lined with a split insulator from cable. The final touch to the front panel was the display screen, which was again photocopied from the magazine template, this time onto clear plastic and later pin-mounted over the CRT face hole.

Two of the PCBs (Time-Base and Y-drive, once built) were later attached to the rear of the front panel via their panel-mounting control switches, and the PSU PCB was fixed to the floor of the enclosure, as was the mains transformer. To the rear of the enclosure a mains socket, fuse and on/off switch were mounted. The CRT was mounted on a bracket composed of pipe clips and packaging tray and positioned so that the face would meet the front panel hole for the display.

PSU PCB

The complete power supply is formed from a PCB plus transformer to convert standard 240V AC mains supply into the +/- 5V low voltage DC supplies required by the rest of the project, and +HT 250V and –HT 350V DC supplies for the CRT display as well as a 6.3V AC supply for the CRT heater. Also produced is a 2.5V (peak) 50Hz reference signal, and focus and brilliance control voltages for the CRT.

The Mains input consists of ‘kettle lead’ socket, fuse and on/off ‘neon’ switch which were wired to the primary of the transformer, which provides secondary windings of 1x 250Vac (for the +/- HT) and 2x 6.3Vac (for the +/- 5V supplies and CRT heater voltages).

A purpose-made PCB contains all the other PSU components, including rectifier diodes, smoothing capacitors, voltage regulators, resistors and variable resistors (for adjustment of the +HT and 50Hz reference output voltages) and drive transistors, producing the HT and stabilised 5V supplies.

The construction and testing of this stage was the most personally challenging element of the project given the presence of high AC voltage and even higher DC; the +/- HT combination required to drive the CRT display amounts to 600V! Hence extreme care was made building and testing, and probing the voltages was done with much due respect and caution!

The main ‘trouble-shooting’ required at this project stage was the diagnosis of a failed drive transistor (BF259); hardly surprising since this component is responsible for the production of the +HT 250V supply, and care was needed to avoid a short circuit whilst building and testing.

Once the PSU itself was working, four ‘temporary’ 1M ohm potentiometers were wired to the CRT base to control the X and Y deflection plates. This enabled the testing of the Tube supply circuit, allowing manual up/down and left/right movement of the ‘spot’, and well as the two permanent 1M potentiometers providing the focus and brilliance controls.

Time-Base PCB

The most complicated part of the electronic design is contained with the Time-Base PCB, providing the horizontal sweep oscillator creating the left-right ‘motion’ of the beams. For the subsequent ‘fly-back’ of the X-axis beam, a saw-tooth waveform generator is employed. A collection of circuit stages make up the various elements of the required Time-Base functionality.

The Ramp-Generator circuit consists of six capacitors selected by a switch to produce a range of ramp rates, with a variable resistor providing intermediate rate variation. As the selected capacitor discharges, the first-stage comparator IC output goes high, tripping the second-stage comparator IC output high. As the capacitor charges, the first stage comparator IC output then go low, but the second-stage comparator IC initially is held high by a diode, until a transistor goes off, tripping the comparator low and starting the repeat of the sequence.

The Sync-Retriggering circuit controls the Ramp Generator, since capacitor discharging only occurs if the diodes in this stage are grounded, which occurs when a clock pulse from an internal Pulse IC or external source causes the 4013 (dual D-type flip-flop) IC output to go low. Hence the ramp is synchronised. Switches select between +/- and external/internal trigger sources, with a variable resistor setting the correct trigger level. The external sync is decoupled and limited by capacitors, resistors and diodes to the +/- 5v power levels.

The Inhibitor circuit ensures that each X-trace sweep begins at the same position. At the end of the trace flyback with the ramp generator comparator IC output going low, the output of the inverting Schmit IC goes high. A positive pulse across a capacitor resets the Sync-Retriggering 4013 IC, inhibiting the Ramp Generator operation. The Ramp starts again after the next clock pulse from the Sync-Retriggering Pulse IC.

The Auto-Retriggering circuit provides the X-trace sweep in the absence of an external signal for sync. The ramp of the X-trace restarts via a diode when the external/internal switch is closed without the need for an external signal; with the switch open, external sync only occurs. A switch selects one of six capacitors which charges once a 4013 IC output goes high until the inverting Schmit IC trips, causing its output to go low and enabling a discharge path for the capacitor via a diode. Thus the switch-selected capacitor provides sync until pulse occurs.

The CRT circuit provides the required boost to drive the 7cm display, since 21V per cm on for one plates and 37V per cm on other plates results in 147V and 259V needed; +HT of 250V is utilised. The output voltage from the Ramp Generator first-stage comparator IC is fed via a resistor and variable resistor, setting the full ramp range output, to the base of one half of a push-pull transistor pair, which provides the HT voltage to one CRT X-plate; the other is supplied by the other transistor controlled by another resistor and variable resistor, varying the screen trace position..

The Flyback-Blanking circuit ensures the beam is not displayed on the right-left ‘return journey’ by providing -50V to the CRT grid, with respect to the cathode. When the Ramp-Generator second-stage comparator IC output goes high, a BF259 transistor goes on, and a negative pulse occurs across a capacitor, with a diode limiting the positive going pulse edge.

Trouble-shooting this project stage was concerned with obtaining a working oscillator, which was achieved once some overflow solder was removed from the transistors and the -5V supply connection restored. The X-plate connections to the temporary potentiometers were removed to allow the Time-Base to drive the CRT display, with the panel mounted potentiometers centering the trace and adjusting the width across the screen.

Y-Amps/Drive PCB

Two key elements form the final PCB. The first are the two identical amplifier stages for the two Y-inputs, the components for which are duplicates. These are then combined using a ‘beam-split’ function fed from the Time-Base second-stage comparator IC to form the dual traces, with each channel then driving in turn the Y-plates of the CRT.

The Input signals to each of the Y-amps are switched for AC decoupling or DC pass-through, with diodes limiting the inputs to the +/- 5V power lines. A switch selects between three resistors to provide an attenuation stage of normal, one tenth and one hundredth; a non-inverting Op-Amp IC provides a gain stage and buffer, with a further three resistors offering selection between gains of 10, 50 and 100.

For both Y-amp channels, the output is brought to a switch to select either for the Time-Base sync. Following the Op-Amp IC, panel-mounted potentiometers control the signal level through a 4066 IC to drive one half of a transistor pair (similar to the push-pull transistor pair of the Time-Base PCB) for the vertical drive of the CRT Y-plates. The offset position is adjusted by two more panel-mounted potentiometers, allowing the Y-position of each trace to be independently positioned vertically.

The Beam-Split circuit takes the output from the Time-Base second-stage comparator IC to the input of a 4013 flip-flop IC, providing twin outputs of opposing logic levels. Each time the input goes high, the pulse triggers the 4013 IC into its next state. This feeds the control pins of gates which when open allow analogue signals to pass unattenuated. The gate outputs are commoner and fed to the Y-position to other half of the transistor pair driving the CRT Y-plates.A panel-mounted switch controls the displays modes, allowing dual or single Y1/Y2 by forcing one of the gates closed, or alternatively divert in the Y2 amp input to the X-plates in place of the sweep generator, for the display of ‘Lissajous’ figures.

Trouble-shooting this project stage discovered an error in the orientation of the diodes which were pulling the +/-5V supplies together. This was as a result of the use of an analogue multimeter displaying the opposite polarity result when switched to the “resistance” function, with the black lead positive (+) and the red lead negative (-) because it is easier to manufacture it that way!

The Y-plate connections to the temporary potentiometers were removed to allow the Y-amps to drive the CRT display, with the panel mounted potentiometers providing movement up/down the screen.

Control Wiring

Once complete, the Time-Base and Y-Drive PCBs were mounted onto the back of the enclosure front panel via their PCB / panel-mounting control switches, and all potentiometers, switches and sockets wired to provide the necessary controls for X and Y adjustments.

Trouble-shooting this final part found nothing from the 50Hz reference output, which was traced to a loose connection on the socket. Once properly connected, this output was used to verify and adjust both traces, demonstrating the oscilloscope project to now be fully functional!

CRT and final thoughts

The most specialised component of the project is the CRT, which of course was once common-place in consumer TVs, now almost completely replaced by LEDs and Plasma displays. The age of this project is therefore naturally reflected in the use of a CRT, but even then it was quite a special component, being of modest size and ideally suiting the instrument. Indeed it was the availability of this device which enabled Practical Electronics to provide the design; prior to that only surplus display tubes just after the war made such projects feasible, which became increasingly hard to find..

The almost certain lack of such components now means a modern design would need a completely fresh approach, making this project now very special indeed. I doubt that there could be very many others in existence, not including commercially produced products. Hence it is therefore particularly satisfying to have completed and made working this (probably) unique project challenge!

Photos (including views of the components inside the enclosure and of screen traces) of the completion project are on our Facebook page.