Technical issues

Leakage around seal for pulley in one piece

Depending on the type of pulley of the alternator, there are two versions for the seal between the turbine part and the alternator compartment:

When I demonstrated a charger with a pulley in one piece, I noticed that there was some leakage water at the bottom of the alternator compartment. It did not pose a real danger as it comes through the seal as a liquid and not as a fine mist of water droplets that can be sucked into the alternator: One could even drill a hole through the separation sheet at it s lowest point to get rid of it. For that demonstration, I used a pump that could only produce about 2 m head and maybe this leakage only occurs at very low head. I had tested this charger before with heads up to 12 m and then I did not notice any leakage problem, but at those tests, water was splashing around everywhere and I might have missed that some water did come through the seal.

The most likely explanation for this leakage problem I can come up with, is as follows: Probably the water can come through because there is a jet of water leaking away between the runner and the nozzle. This jet hits the separation sheet just outside of the pulley and apparently, some water is forced inwards through the narrow clearance between the pulley and this sheet.

If the above explanation is correct, then a solution would be to fit an anti-splashing ring of 0.5 mm galvanised iron sheet, inner diameter = 6 mm, outer diameter = 100 mm between the runner and the pulley (see bottom left corner of the figure below). This ring should prevent the water jet from splashing against the separation sheet at high speed. I haven' t tested this solution yet, please let me know whether it works.

Seal construction for a pulley in one piece, with-anti-splashing ring
This figure is a detail of new version of figure 4.12 of the building manual: The seal and the way to fix the runner on the shaft (click on it to get the complete figure).
In the printed version of the Firefly building manual, this new anti-splashing ring was not drawn. The drawing in the downloadable version is correct.

Note: When printed from within a browser, most likely the scale is wrong. To get it printed on scale, copy this drawing to your computer, open it with a graphical processing programme and print it at 150 Dots Per Inch.

index page

Mechanical regulators

During the mission to CLSU-ANEC and BSU-ANEC, it revealed that the mechanical regulators they used, were rather unreliable when fitted in a firefly switchboard. Probably, this is because tiny sparks between the voltage regulator relay contact points, makes those contacts stick together a little. Apparently the field current of max. 3 A is enough to act as a welding current, producing very tiny welds between those points. Those tiny welds are strong enough to disturb the balance between magnetic force generated by the relay coil, and the spring force. And it is this balance on which the proper functioning of the mechanical regulator is based. The effect is that the firefly charger overcharges batteries: Once the battery becomes charged and voltage surpasses 14.7 V, the regulator does not reduce field current properly, but continues to provide full field current. Consequenty the battery is charged further and battery voltage rises to well over 15 V.

If this sparking effect happens in the firefly charger, one would wonder why such mechanical regulators do function in a car:

This problem can not be solved by readjusting the regulator to a lower voltage. There is no way to predict how much extra force is needed to pull loose sticking contacts and quite likely, this will differ all the time. So trying to readjust the regulator to a lower voltage could make that some batteries are not charged fully while others would still be overcharged.

Fitting the regulator in the alternator compartment is not a good solution either. Of course, the charger will produce some vibrations but when the runner built and fitted neatly, it will be so well-balanced that the vibrations might not be strong enough to make the contacts come loose. It might seem simple to change the circuit diagramme of fig. 4.25 in such a way that the regulator, the field current lamps and switches can be fitted in the alternator compartment. But moving the points over which the voltage regulator senses the voltage, does have an effect on charging characteristics: If for instance the regulator is connected at the alternator end of the alternator cable, the resistance of this cable and the fuse will make that the regulator senses a higher voltage, and consequently will reduce charging current already at a lower battery voltage, see also par. So if one wants to do this, the best way is to leave the circuit as it is and fit extra wires from the switchboard to the regulator.

Possible solutions are:

If this doesn't help: Look for an electronic regulator.

index page

Electronic switchboard

Compared to the standard switchboard, this electronic version has the following advantages:

  1. It is more rugged. There are no moving parts and as long as it is protected against water, it will work accurately. Also with respect to electrical overload situations, the circuit is well-protected.
  2. It is more user-frendly: There are just two `on-off' switches, to connect the charger to either one of the two batteries that can be charged at the same time. A LED bar indicator shows state of charge of the battery that is being charged and once this battery is fully charged, this is also easily visible on the indicator. The standard switchboard had an indicator with a voltage and current range. For finding out whether a battery was fully charged, both voltage and current had to be read. Finding the actual state of charge of a battery while it was being charged was even more complicated , as one had to compare readings with graphs about the charging process.
  3. The voltage regulator part of it can be adjusted easily for a higher voltage than what is common in cars (most car alternators are non-adjusted and then special tricks are needed to set them to a higher voltage).
  4. With the voltage regulator part, there is a field current adjustment. This has the same function as the field current lamps of the standard switchboard, but it doesn't have the energy loss of those lamps. Therefor a charger with an electronic switchboard will be slightly more efficient.
  5. It can be used for charging two batteries in one go, see below with `discharge current LED'. Each battery connection has its own indicator circuit.

But: It is more difficult to build than the standard switchboard or the charge indicators. It should be fitted in a well-sealed box or casted the electronic circuit casted in epoxy resin to protect it against water. I think it is a kind of gadget: Nice to have, it functions well, but it is not essential. It could be an interesting option if large numbers of firefly chargers were produced and series of this electronic switchboard could be produced in one go. As far as I know, this is not the case and I decided not to promote its use.

The electronic switchboard works as follows (see also drawing of front plate and treshold voltages):


Treshold voltages for the indicator part. The same series of LED's show battery state of charge for both `initial mode' (when a battery that has just been connected and is not being charged yet) and `while charging' mode.

Electronic switchboard made by mr Jaap Koppejan. It took him a couple of days to build and it worked fine.





If you are interested to build one: Contact me and I will send the building manual with circuit diagrammes, PCB design, parts list etc.

index page

Adjustable nozzle

The second prototype I built in Holland, had a adjustable nozzle that had a hinge in runner side (see building manual for names of parts). This way, the end part of runner side could rotate towards the bent side and a less thick water jet would be produced. To avoid that too much water leaked away at the edges of this moveable part, I mounted a piece of rubber over the fixed part of runner side, the hinge and the moveable part of the runner side. Later on, I realized that a thick piece of rubber could serve as a hinge so a simpler construction was possible. I haven't build this type of nozzle yet.

Some notes:

Generally, the standard nozzle will be the best choice. Then blocking timbers can be used in case flow has to be reduced. This is simpler to build, it does reduce peak forces on blades and it can have a slightly higher efficiency at very low flow. Such an adjustable nozzle could be useful for:

index page