Director,
T.E.(Terry)
Manning,
Schoener 50,
1771 ED
Wieringerwerf,
The
Tel:
0031-227-604128
Homepage:
http://www.flowman.nl
E-mail:
(nameatendofline)@xs4all.nl : bakensverzet
Incorporating
innovative social, financial, economic, local administrative and productive
structures, numerous renewable energy applications, with an important role for
women in poverty alleviation in rural and poor urban environments.
"Money is not
the key that opens the gates of the market but the bolt that bars them"
Gesell, Silvio The
Natural Economic Order
Revised English
edition, Peter Owen, London 1958, page 228
Edition 9:
The following pages are about
checking the installation and water flow. If you are more interested in other
aspects concerning installation, please return to the installation index.
|
PHYSICAL |
Solidity of panel support
structure. |
|
SAFETY |
Safety rope fastened. |
|
Sealing off of bore-hole. |
|
|
|
|
|
ELECTRICAL |
Voltages of controller and motor
matching. |
|
Electric cable to controller
connections. |
|
|
Controller to panel connections. |
|
|
Panel to panel connections. |
|
|
ORIENTATION |
Panels pointing south (North H.)
or north (South H.) |
|
Freedom from obstructions. |
|
|
Array correctly tilted. |
|
|
HYDRAULICS |
Water in bore-hole ? What level ?
Installation depth OK? |
|
Feed pipe without kinks. |
|
|
Feed pipe correctly embedded. |
|
|
Water tank or other recipient
connected. |
|
|
In some cases the valves may
stick after storage.See notes below. |
When the pump is connected directly
(therefore without batteries) to the solar panels, operations under way and the
status of the system can be checked by looking at the four LED lights which can
be seen when the lid of the controller box is removed.
Do not touch any metallic or non metallic part inside the controller.
These four LED lights enable an initial rapid check of operation to be made on
first installation or on re-installation after maintenance.
Only a few things can in practice happen with a pump running directly from PV
panels through a Sunprimer controller. To carry out a full diagnosis, at least
two commercial testers (= 2 voltmeters and 2 ampmeters) are needed as well as
the LED lights.
When all four LED lights are off then EITHER the controller is not connected to
the panels OR the safety devices having discovered damage to the main
transistor such as to prevent normal operation of the controller, have put the
PV panels into short circuit.
THE LED LIGHTS ARE NUMBERED ACCORDING TO CIRCUIT
LOGIC.
The first LED, the third from the top, (shown in the
drawing as no. 1) lights up when a voltage greater than about 10V reaches the
controller from the panels.
The second LED, the second from the top, (shown on the drawing as no. 2) lights
up when the main capacitor is charging. This phase lasts about three minutes
with the Mk I/d and Mk I/e controllers and about 2 minutes with the Mk II
controller. Note that in certain circumstances the capacitor may be impeded by
built in safety mechanisms from charging.
The third LED, the one at the bottom, (shown on the drawing as no. 3) lights up
when the controller sends output voltage greater than about 10V to the pump
motor.
The fourth LED, the one at the top, (shown on the drawing as no. 4) lights up
when the pump motor stops because the external (float) switch is in OFF mode.
If no LED lights come
on, then either the connections between the panels or the battery set and the
controller are such that no current arrives to the controller, or that the
built in safety devices having sensed damage to the main transistor have put
the panels or the battery set into short circuit. In case of the battery set,
the fuse should have burned out. First check the electrical contacts. Then
remove the electrical supply to the controller, where appropriate (battery
installations) substitute the burned out fuse, and reconnect the power supply
to see if the fault is still there. Should operation become normal then there
was a temporary system block due to a false alarm caused by accidental voltage
surges which were not self-correcting. If the defect is still there, the
diagnosis should be completed by placing an ampmeter and a voltmeter between
the panels or the battery set and the controller. If voltage is too low and no
current passes, then there is a fault with the electrical connection or with
the PV panels. If current passes normally from the panels but there is no
voltage (or just a few volts) then either the battery fuse continues to jump or
the PV panels or the battery set are in short circuit mode and there is a
defect in the controller.
If just the first LED,
the third from the top marked on the drawing as no. 1 comes on but not the
second LED (the second from the top marked on the drawing as no. 2), or if both
the first (the third from the top marked on the drawing as no. 1) and the
second LED (the second from the top marked on the drawing as no. 2) light up
regularly but after at least three minutes (Mk I/d and Mk I/e) or two minutes
(Mk II) not the third LED (the third bottom LED marked on the drawing as no.
3), then the voltage of the main capacitor is insufficient for the system to
work normally. This can be due to:
a) Insufficient insolation.
b) PV panels are not suitable.
c) Faulty electrical input connections.
d) Battery set is discharged.
e) Break in the electric cable going to the pump.
f) Brush wear.
g) Controller break down.
To check whether the the capacitor voltage is in order without the need to open
or access the internal part of the controller box, connect a voltmeter to the
ends of the two float switch wires. Since these wires are connected to the
capacitor through resistors for short circuit protection purposes, they can be
safely used to check capacitor voltage.
To check for brush wear and
eventual break in the electric cable, use a "tester" to check the
global circuit resistance. If this is about 10000 Ohm, the brushes are worn
out. If it is nearly infinity, then there is a break in the cable.
To check the other factors, without
removing the first voltmeter, add a second voltmeter to the panel circuit and
an ampmeter on the pump motor circuit.
a) Where four panels are installed,
each pair of panels has to be checked either contemporaneously or one after the
other. If panel voltage is normal but there is no current or a current of just
a few milliamperes and the capacitor voltage does not go up, then the main
capacitor charging and discharging thyristor is broken. If there is a current
of several amperes then an attempt should be made to stop the controller by
putting the two float switch wires in contact with one another BY MEANS OF A
GOOD QUALITY FAST SNAP SWITCH (NEVER NEVER USE YOUR HANDS FOR THIS
OPERATION!!). If current in the pump motor circuit fails to switch off then the
main controller transistor has burned out and at the same time the resistance
of the cable between the controller and the pump motor is too high (see
"Cable resistance" below). If the current switches off then the main
controller transistor is in order and the cable between the controller and the
pump motor offers too much resistance (i.e. it is too small). Should capacitor
voltage rise, check the voltage level reached.
b) Where
four panels are installed, if voltage does not exceed 60V then either there is
something wrong with the panels or there is a fault in the automatic
series/parallel switching mechanism of the controller with the result that the
panels remain in parallel instead of switching to series. This can be confirmed
by measuring the maximum voltage reached by the capacitor and contemporaneously
the voltage reached by each pair of panels to see whether the maximum voltage
of the capacitor is the sum of the voltages given by the two pairs of panels or
whether it is equal to the voltage of just one pair of panels.
If the first LED (the third from
the top marked on the drawing as no. 1) and the second LED (the second from the
top marked on the drawing as no. 2) light up regularly and after about 2 minutes (Mk II controller) the third
LED (the bottom LED top marked on the drawing as no. 3) comes on and then goes
out then the pump has not managed to start, or started but switched off
immediately. This can happen repeatedly on a cyclical basis because:
a) Insolation is too low.
b) The PV array is too small for the application in question.
c) The pump is too deeply submerged under water.
d) The resistance of the cable between the controller and the pump is too high
(= the cable is too small).
e) The load is too high for the pump to handle and/or the cam size used is too
large.
f) Built in safety devices have acted to cut current where current has reached
a value of 4.5 amps for half a second due to blockage of the feed pipe through
freezing or other causes.
g) Very rarely, failure to start due to reduction of current surge caused by
obstruction of the small self-cleaning nipple of the residual pressure release
device situated at the lower end of the feed pipe. Should this occur, just
wobble the wiggle wire potruding from the nipple.
h) It can also happen that a four panel system fail to start because the two
pairs of panels are not equal or insolation on them is not equal.
If the first (the
third from the top marked on the drawing as no. 1), second (the second from the
top marked on the drawing as no. 2) and third LED (s light up properly and
operation appears normal but no water is pumped, then an ampmeter should be
inserted in the pump motor circuit. If current equal to or less than that shown
on the tables is produced when the third LED lights up, and there is no water,
then either the pump is out of the water (= running dry) or it cannot
self-prime, in which case refer to the further notes on priming in this section
of the instructions. If, instead, the third LED lights up but no current
passes, then either the electric cable is cut or the motor brushes require
replacement. This can be counter-tested for a 10000 Ohm resistance as set out
above.
If everything operates as expected
and the pump starts normally after initial connection to the panels or with the
early morning sun but then after the passage of a cloud or for some other
reason fails to start up again, then either:
a)The main thyristor of the capacitor charging device has broken down
b)Resistance between the pump motor and the controller is too high (cable is
too small)
c)Excessive cable resistance coupled with damage to the main controller
transistor.
In such
case, insert a voltmeter between the float switch wires (use a good switch for
the purpose NOT your hands!!) and an ampmeter in the pump motor circuit and
then try to recreate the situation which led to the malfunctioning, e.g. by
turning the panels out of the sun or shadowing them to simulate the passage of
a cloud causing a reduction in insolation requiring the intervention of the
controller.
If, in case of failure to
re-start, voltage is normal but there is either no current or current is just a
few milliamperes, and capacitor voltage does not go up, then the main capacitor
charging and discharging thyristor is defective (=the thyristor manufacturer
has failed to respect his stated manufacturing tolerances) and the thyristor
may need replacement.
If current to the order of a
few amps passes then try to turn the controller off by putting the two float
switch wires in contact (use a good quality switch for this purpose, NOT your
hands!!). If current in the pump motor circuit is not switched off then the
main controller transistor has burnt out AND ALSO the cable resistance is too
high (= the cable is too small)(see "Cable resistance" below). If the
current switches off, then the problem is just the resistance and size of the
electric cable between the pump and the controller.
If capacitor voltage goes up, then refer back to
"Only the first LED lights up" above.
The overall resistance of the cable
between the controller and the pump must be less than 1.5 Ohm. Purely
indicatively this in normal conditions will correspond to cables with a section
of 2.5mm2 for lengths up to 80m, 4mm2 for lengths up to 150m, and 6mm2 for
cables longer than 150m, assuming at all times that the wires are pure copper
and cable joints are few and well executed. Since the resistance of the motor
windings is about 1.4 Ohm, the global resistance of the pump motor circuit
measured at the tips of the wires leaving the controller should be less than
2.9 Ohm. This resistance must be measured very carefully as some commercial
testers are too inaccurate at lower resistances while others using very low
test currents may be disturbed by weak electical forces caused by interference
amongst various materials present in the circuit to be measured.
The resistance of the cable between the controller
and the pump must be low because the controller governs the outlet voltage
towards the pump motor and if the voltage falls below 20V it cuts current
supply to the pump as it feels the pump is about to stop. If the cable resistance
is too high, voltage could because of ohmic effects (V=RxI) remain over 20V
even where the pump has stopped, with the result that the controller cannot do
its work and cut current supply to the motor. Were this to occur for instance
after the passage of a cloud, the system would correct itself only once current
has been cut either autonomously at nightfall or through the operation of built
in safety systems where current reaches 4.5 amps for half a second or through
the operation of the float switch or other external control.
If after a period of normal
operation at parity of insolation during the middle of the day the pump
produces less water than expected, the cause can be:
a)Badly connected panels cables
b)Incorrect seasonal or latitudinal panel orientation
c)Undersizing of PV panel array
d)Incorrect cam size (cam size TOO LARGE) with the result that the pumps stays
in parallel mode instead of passing into series mode.
First check the electrical connections.
Then insert a voltmeter in the pump circuit.
If a pump unit with a four panel system starts later in the morning and stops
earlier in the evening, then the series/parallel switching device of the
controller could be damaged or electrical connections could be faulty. Use a
voltmeter in the pump circuit to check electrical connections. Voltage during
normal high insolation operation will typically be 50-60V. Where insolation is
weaker, voltage is halved to about 30V to ensure operation for instance in the
morning or in the evening or in cloudy weather. If insolation becomes very
weak, then the pump switches off. If, after having put a voltmeter in the pump
motor circuit, there is no evidence of operation at the lower voltage in the
morning or in the evening or when passing clouds are simulated by turning the
panel array gradually away from or back towards the sun, then there is a defect
in the series/parallel switching device.
If the pump never passes into series (48V) mode or
passes into series (48V) mode only for a short period during peak insolation,
then:
a)Check panel orientation
b)Check controller for faults. To do this take pull the pump out of the
borehole and test the system in a low load situation with the pump with an open
outlet (no head) in a bucket. If the pump in this situation does not pass into
series mode then the series/parallel switching device of the controller is
damaged. If instead the series/parallel mechanism works, then the installation
parameters adopted are incorrect.
c)Use a larger array
d)Change pump model (=use a smaller cam size)
If there appears to be
damage to the pump itself it is useful to study the behaviour of the pump at
sight, letting it work outside the borehole in a suitable recipient, with an
ampmeter in the electrical supply circuit and a manometer and a valve on the
outlet to permit the required pressures to be reached. It is not necessary to
use the controller for this purpose, and the pump itself may be connected
directly to the PV panels. Capacity is best measured using the traditional
method of a bucket and a stop watch.
It is good practice to
run the pump for a few minutes in a recipient BEFORE fitting the feed pipe on.
If the valves stick shake the pump. If they continue to stick it may (rarely)
be necessary to apply suction at the outlet. It is good practice to let the
pump run from an appropriate recipient for a few minutes without connecting the
pump to the feed pipe, to facilitate initial priming of the pump. It is not
absolutely necessary to use the controller for this purpose and the pump can be
temporarily connected directly to the pV panels or to a 24V battery set.
If the insolation is
good (say, 800Wm2 or over) water should start flowing at the outlet within
about three minutes (Mk I/d and MK I/e controllers) two minutes (Mk II
controller), depending on the depth of installation and the length (and
therefore internal volume) of the feed pipe. It may take a little longer where
air is trapped inside. If insolation is low (say, from 400Wm2 to 800Wm2) it
will take longer for water to exit at the feed pipe outlet, as the booster must
first kick-start the motor, and pump capacity will be reduced. If insolation is
very low ( say, from 150Wm2 to 400Wm2) some patience will be required, as it
may take the booster some time to start the motor, and pump capacity will be
very low. Once water flow is smooth, the volume of water pumped can be compared
the published flow tables for the power, head and insolation levels in
question. The pump needs a few hours to run in, after which capacity can be as
much as 10% more than after initial installation. Flow testing should therefore
be carried out after the pump has seen a few hours' operation.
Should, after a reasonable amount
of time, usually about 2 minutes (Mk II), no water reach the feed pipe outlet
after start up :
1) Repeat the checks on the check list in this section.
2) Have you turned the power switch on the panel support structure on?
3) Check insolation. If insolation is very low : wait longer, especially if the
weather is overcast (clouds passing), if it is early in the morning or evening,
if it is winter and the sun is low on the horizon, if it is very hazy.
4) Check if motor turning. Feel the feed pipe where it comes out of the
bore-hole. If there are vibrations, the motor is turning. If there are no
vibrations, keep your hand on the feed-pipe for a minute or two more, as the
booster may be storing energy to kick-start the pump.
5) If motor is turning, the valves could be stuck after a storage period. In
this case case, shake the pump by the feed pipe to help expel air from the
system. In rare cases it may be necessary to apply suction at the outlet to
assist the initial priming. Refer to the section "What to do where the
valves stick" above for further information.
Should the valves for any reason continue to "stick" after a period
of lay-off, they may on rare occasions, be caked with deposits and require
cleaning. Refer to the Maintenance Section of the manual for instructions on
how to open the valve groups and clean the "caked" valve rubbers and
parts, or contact your supplier for help.
6)Make sure the self-cleaning nipple (part 308) situated at the top of the
pump, below the non return valve, is not obstructed. Obstruction of the nipple
can make pump start-up more difficult. To clean the nipple, just wobble the
little metal wire protruding from the nipple.
7) Measure actual PV panel output against
insolation. OK ? Make sure, where applicable, the pairs of panels are equal and
exposed to the sun in the same way.
8) If not, check that panels are mounted in series or in parallel or in groups
in series or in parallel, according to your system design. Then re-check panel
readings.
9) If O.K. re-start trouble-shooting cycle as above.
10) If O.K. and no result at 9), check output in amps and voltage at Sunprimer
outlet.
11) If not O.K., check that Sunprimer voltage regulation correct.
12) If readings O.K., then current is either not reaching the motor at all or
insufficient current is reaching the motor.
At this point, proceed according to
the instructions in the maintenance manual, working backwards from the source
of power :
a) The joint(s) in the electrical cable. Check cable type. Check for current
losses. UNDERSIZED CABLES ARE THE MOST COMMON CAUSE OF STARTING DIFFICULTIES.
In case of doubt, try using larger cables eventually together with a smaller
cam.
b) The connections to the pump motor.
c) Signs of water leakage to the
motor.
d) The motor brushes.
e) The motor winding
Note that Solar Spring pumps usually need a few
hours to run in. Performance of the pump when run in may be as much as 10%
higher than that of the pump immediately after first installation.
Start-up
problems in 36V or 48V battery systems without controller can be checked by
using an ampmeter (better still if a voltmeter is also available) placed in the
pump circuit. A standard commercial "tester" is quite suitable for
the purpose.
1.If the pump produces no
water and the motor does not absorb any current, then the electrical contact is
defective or the electric cable is broken.
2.If the pump produces no
water but absorbs the expected amount of current or a bit less, then either the
pump is running dry out of the water or is unable to prime itself. In this case
refer to the notes above.
3.If the pump produces no
water but absorbs more current than expected, or the fuse burns out, then
either the feed pipe is blocked or the pump cannot start because the load is
greater than what it can handle, or the fuse used is too small. If the feed
pipe is not blocked, then the load on the pump must be decreased by taking the
following steps in order:
(a) Increase fuse size by one ampere.
(b) Decrease submersion.
(c) Decrease cable length.
(d) Increase cable size.
(e) Use a pump with a smaller cam.
4. If a pump after a period
of normal operation unexpectedly stops or fails to absorb current when the rest
of the system is in good condition, then either the electric cable is broken or
the motor brushes have worn out. To enable immediate diagnosis of brush wear a
resistance of 10000 Ohm has been placed in the motor in parallel with the
brushes. If the global resistance of the electrical circuit across the ends of
the electric cable is measured with a "tester" before pulling the
pump from the borehole is found to be about 10000 Ohm, we know the fault lies
with the brushes. If the resistance is practically infinite, then the cable is
broken.
5.If a pump after a period
of normal use absorbs the expected amount of current or a little less but
produces no water, then the level of water in the borehole has gone down and
the pump is running dry.
6.If a pump after a period
of normal operation fails to produce water and absorbs a current greater than
that expected, or if the fuse burns out, then the feed pipe is blocked. Very rarely,
in marginal situations, where everything is in order and the feed pipe itself
is not blocked, the start up possibility of the pump may be reduced by the
blocking of the self-cleaning bleeder placed below the foot valve at the bottom
of the feed pipe to reduce pressure in the valve system. In this (very rare)
case the pump needs to be pulled and the little wiggle wire projecting from the
nipple moved around to eliminate the block.
7.If the pump appears to be
damaged, pull it and without remving any part, place it in a recipient so that
its operation can be obersved, inserting an ampmeter in the supply circuit and
a valve on the feed pipe so as to be able to vary system pressure. The best way
of measuring capacity is by using a watch and a bucket. In case of true system
breakdown, call your supplier for help, as maintenance should always be carried
out by qualified persons.
It is STRICTLY
PROHIBITED to test the Solar Spring pump in the laboratory with the Sunprimer
electronics unit. This is because the response times of the Sunprimer unit are
far too fast for normal laboratory testing equipment.The Sunprimer unit can be
field tested using photovoltaic panels together with the Solar Spring pump as
indicated in the accompanying drawing. The Solar Spring pump can be laboratory
tested without the Sunprimer electronics unit following the scheme indicated in
the attached drawing. The equipment required for testing purposes takes into
account the fact that the Solar Spring is a high speed single action piston
pump incorporating certain inertia characteristics. Shock and missmatch in the
system are absorbed by means of elastic elements and flexible feed pipes. Any
testing using methods and/or equipment not expressly approved in writing by the
manufacturer will not be recognised by the manufacturer for guarantee or other
purposes and may well produce seriously false and/or misleading data.
Because the Solar
Spring pistons are single acting, the pump behaves as a monocylindrical pump
with a pulsating flow. The ovoid in A) and the balls in the Hyboost unit B) are
designed to absorb pressure variations and regulate flow. They are enough to
handle flow in a normal installation C) because the length of polyethylene pipe
and the free outlet tend to absorb residual missmatch. The mentioned pressure
and capacity pulsations can, however, create problems where tests are being
carried out using a tank where the pump outlet is not free but is occupied by measuring
and control instruments. To avoid problems in measuring test results from a
tank, refer to drawing D).
Solar Spring pump (1) (WITHOUT
SUNPRIMER ELECTRONICS unless the test rig is directly connected to PV panels!!)
in tank (2) with power supply measured by volt-meter (3) and ampmeter (4). At
least 10-20m of polyethylene feed pipe (5). An expansions tank (6) which serves
to absorb residual pulsations and simulate free outlet as in normal field
installations. Pressure measurement (7). Pressure regulating valve (8).
Flowmeter (9). Current from PV panels should be measured as well as water flow.
Currents and voltages relating to given water flows should be recorded, because
even if they are not immediately useful, they can be used later on for study
purposes to understand the phenomena in play and help with diagnosis of
eventual problems.
The
response time of most cmmonly used instruments such as ampmeters and voltmeters
is too slow to measure the reactions of the Sunprimer electronics accurately.
Menu installation Solar Spring pumps.
Some recommended technologies.
List of attachments to the Model.
Typical list of graphs and
drawings.
List of abbreviations used.
List of key words.
Documents for funding
applications.
