We all know WOTters are the sportiest people in the world, but sometimes they need some extra exercise. This was achieved by fitting an old motor from a kid’s toy on an old rowing machine out of gym, connect to some batteries and we have renewable energy!

Because the motor from the kid’s toy is a three phase permanent magnet motor, it generates a 3 phase Alternating Current (AC) when rotated. This is very useful for high power motors such as in trains or big milling equipment, but not so much for small scale off-gird power. This is why the generator is connected to a ‘full bridge rectifier’, this takes the three phase AC input and converts it to Direct Current (DC). Cheers for AC/DC! The output DC of the rectifier is then connected to a lead-acid battery (also DC) and voilla; we can put muscle power into a battery!

Small off-grid systems often save solar energy in a battery and later convert that to 230 V AC. This can be done here too! So during the day one could use a solar panel and at night human power.

In the pictures above, the generator can be seen on the left. It is connected with gears and chain from a bike. This increases the rotational speed of the generator. Because the generated voltage is proportional to rotational speed, we can now generator higher voltages. On the right picture you can see two batteries in series (24 V). With higher voltages we can handle more power. With this setup, you could easily start a full sized diesel lorry (50 ton) if you row for a while!

Of course not everyone has an off-grid setup or a lorry with empty batteries. The energy in the battery can also be injected into the grid. Below is a picture of the generated power on the screen (every peak is someone pulling on the rower). To the left is a small solar converter that can inject DC into the national grid. This way we are powering an infinitesimal bit of the world!

In the past the Breurram has been built and tested by the WOT. It is a type of hydraulic ram pump, made out of standard components using an airbuble in a piece of flexible tube as expansion ‘tank’.
More about the mechanism behind the Breurram can be found in this youtube video.

Several times the WOT has been asked: What would happen if you scale up the Breurram?
Our earlier demonstration and test set-up uses 1” valves. Following yet another advice request and subsequent visit at the WOT, we decided to test a larger version of our beloved ram pump.
The diameter of the Breurram can, according to the manual, not be larger than it’s supply line.
Because the supply of water from our ferrocement tank/cistern to our testing rig ends in a 2” pipe, a 2” Breurram has been constructed. Only the valves have been purchased, as other materials were present. The cost for these came down to about €95.

Upon first testing it became apparent that the system wouldn’t work reliably without the optional tensioning spring. Next to this problems were found with starting the pump. Therefore a bracket has been placed on the spill valve connected to a bolt sticking out of the pump, allowing the valve to be moved manually at start up. Depending on the setting of the tensioning spring and the pump head the system was able to work without human intervention. The frequency of the pump cycle was far less constant compared to our other, smaller, Breurram. What this is a result of remains a question at this point, but suggested are lower flow speeds and/or quality of components.
Tests are done using a supply head of 184 cm (approximately) and pump heads of 397, 582, 767 and 952 cm. Using a return line and a petrol pump pumping back used water to the supply tank, the supply head was held as constant as possible. So far one test has been completed at a constant setting of the tensioning spring. The yields at different pump heads and several curve fits can be seen in the following figures, as well as some pictures /visualizations of the test/setup.

The plot displays the data points along with an automatic second-order curve fit using the least squares method. It is expected that the flow should approach zero at a pump head approximately ten times the supply head. However, the second-order curve fit does not exhibit this behavior, even when considering the mean and standard deviation. To address this issue, a new curve fit was developed by defining a trial function and making several coefficient guesses.

So far, in order to draw more accurate conclusions several more tests will be conducted, of which the results can be expected in the near future. This will also include a (rough) comparison with the yields of our other, smaller, Breurram. You can read more about hydraulic rams here.

Now that the winter has subsided, we decided it was time to turn on our solar shower again.
The system was turned on during the April working weekend, and two WOTters actually enjoyed a (still very cold) shower, a very welcome surprise after all the mud that was picked up at the other activities 🙃.

At an association night (Wednesday the 10th of April) we noticed two problems with the system.
-The water temperature was rising unexpectedly fast on Saturday and Sunday (for the sun intensity those days) and dropping too fast at nights.
-The pump failed to switch on on Monday and Tuesday. The system thinks it switched on the pump but the return-line temperature was not following the collector temperature. Which also meant the tank-temperature was not increasing on those days.So that Wednesday we checked the water level of the tank and sure enough the tank was almost empty, during dinner we let the tank fill up. And we used the bottom flush of the tank to try to flush some rust out of the system, which might have been blocking the pump.

We were hoping this would work, but during the following days, we noticed that the pump still was not switching on correctly. On some days it was (like Saturday the 12th of April), but on other days it was not. We can however see that the tank now heats up and cools down slower than it did when it was not completely filled.

Here you can find a graph displaying the data that is measured live from the solar shower system. This graph shows the temperatures of various sensors and the time the pump was on for every 10 minutes.
-The pink line shows how long the pump was on, the pump cycles on automatically when the collector temperature is higher than the tank temperature.
-The dark-blue line shows the temperature of the collectors, it quickly rises if the sun is shining.
-The red line shows the temperature of the water flowing out of the collectors and into the tank.
(This means it follows the collector temperature while the pump is on but the temperature quickly drops towards ambient once the pump has shut off)
-The light-blue and yellow line show the actual water temperatures in the isolated storage vat this is the water that is used to heat the shower water through a heat exchanger.

On Monday the 14th of April we flushed more rust out of the pump, this time by actually manually disconnecting the pump and running it with some clean water. This did not seem to work, because on the 18th the pump was jammed. On the 23th a small hole was found in a pipe and had been welded shut, this explains why the tank was as good as empty during the working weekend. The pump was replaced with another one on the 24th and this seemed to have helped, because the pump has been functioning every day up to the day of writing (the 3th of May). Since the 29th of May the days kept getting warmer and on the 1th of May the tank was at a temperature of 45°C and someone had a nice and warm shower 😊.

Since we already installed a number of rope pumps in the Netherlands, we were looking for a safety brake that is reliable and stultifies the appearance of the rope pump not too much. As users of rope pumps know, the wheel and handle swing back firmly when they stop pumping. In developing countries the user gets instructions to use the hook to prevent turning back of the handle. For public places where anyone can use the pump, the blocking hook will not be used and there is a chance of injuring the user.

Therefore, we searched for an automatic device that prevents the reversal movement.

We have already been experimenting with a kind of rattle that inserts in the spokes of the wheel. Tests showed that the forces were too high resulting in bending the rattle pin. Another solution might be a band brake. This is also tested. These solutions do not look very pretty and they are susceptible to vandalism. Also, the braking power is not sufficient and reliable enough. A third type of a one-way brake with a nail into a wedge-shaped space in the bearing bushing gave good results. Better than expected. The video below shows the principle operation, construction, and testing of the one-way brake.