0
\$\begingroup\$

I'm working on a small water fountain that's powered by solar panels, and I would love some help identifying a root cause for a water pump dying. This is my build so far:

  • Water pump

  • Solar panels

  • I have 4x 6V-100mA solar panels in parallel, so I can generate 6V at 400mA. Under full sun, the panels's voltage is around ~7V and under this voltage, the pump draws 240mA (it is rated for 220mA)

  • I get 3h of direct sunlight every day so the pump will turn on, and turn off 3h later

My first run was great, but the pump died after 3 days of usage. I identified several possible causes:

  • running at 17% above the rated voltage (I didn't expect this was going to be such a problem as I'm used to 1.25 safety rating)
  • low quality pump (I found a very similar model on Amazon that recommended to use them for experimentation only)
  • Turning on and off every day, beeing worse than running continuously
  • [not the case] dry run burn out (the pump never worked under dry conditions

I've ordered a better pump that should be more suitable for this usage, but the size is starting to be an issue. Any thoughts on what's the most likely reason for the pump dying in the first place? (if there is hope that such fountain can run on small pumps like that one, it is great as it can fit many other projects)

Bonus question: as the light increases slowly, the pump starts drawing high current while not running as it hasn't reached its starting conditions. How much of an issue is that?

\$\endgroup\$
5
  • 3
    \$\begingroup\$ That "getting power but not enough to turn" situation sounds bad. Perhaps you should try to build something which only powers it when it can really run, the catch is that the panel voltage will drop a lot when the pump comes on. You may be able to model what is a sufficient, or rig up something where you periodically try. If you switch the motor off for an instant and it is spinning, you should see the back EMF. If it's not spinning, wait a few minutes before trying again. Or you could build in some sort of measurement test load. And MCU control could also let you modulate at high sun. \$\endgroup\$ Aug 4 '20 at 19:35
  • 1
    \$\begingroup\$ Something got stuck in the impeller but any of the reasons listed are possibilities. \$\endgroup\$
    – Andy aka
    Aug 4 '20 at 19:35
  • 1
    \$\begingroup\$ I think you have the "safety rating" upside down. The rated voltage should be 1.25 times the maximum actual voltage applied, not the other way around. \$\endgroup\$ Aug 4 '20 at 22:53
  • \$\begingroup\$ @Chris Stratton, thanks for the answer. Do you think a Zenner diode would do the trick? From what I've tested, the voltage is either a few mV or >5V (clouds are very rare) \$\endgroup\$ Aug 5 '20 at 15:43
  • \$\begingroup\$ No, a zener diode won't help by itself, it could maybe a very crude reference to a comparator. From your description, this may not be an issue at all. \$\endgroup\$ Aug 5 '20 at 16:18
2
\$\begingroup\$

The Bonus question is almost certainly the culprit. Under that condition, it is essentially the same as when the motor is stalled, or locked rotor condition. When a motor is not spinning it does not generate any back EMF and so will act as a short. The only current limit is in the DC resistance of the wire and/or the current limit of you source.

This condition creates a lot of heat, and can/will melt the insulation on the wire (even magnet wire insulation). The other thing it can do is heat up the bearing or bushing for the rotor, which once that deforms also kills the motor.

A solution to the above, would be to use some sort of circuit as @Chris Stratton is describing in the comments of the OP.

\$\endgroup\$
1
  • \$\begingroup\$ I think that the pump is probably using brushed motor and is acting as a short circuit. If he can find a pump with a BLDC motor his issue will be resolved. \$\endgroup\$
    – JaySabir
    Aug 5 '20 at 4:26
1
\$\begingroup\$

Filtering water input to prevent input of objects that will not easily pass through the pump "is a very good idea".

In my experience it's very unusual for low wattage pump motors to die in normal use.
Many AC mains powered pumps use a magnetically coupled rotor and the motor can spin even when the rotor is trapped. DC versions are more liable to be direct coupled centrifugal - which yours appears to be.

You could try measuring pump current at various voltages dry / with water / with stalled rotor to get a feel for fault conditions. I would expect most pumps to run at reducing power and current as available energy reduces.
A PV panel is more like an insolation (sunlight) varied current source than a variable voltage source - so the motor will load the panel to whatever voltage matches the power when the PV panel can provide a given current level.

It would be quite easy to provide a "hiccup" protection scheme that turned off power for say 10 seconds if some limit was exceeded (too low voltage or excessive current or ...).
We can provide protection circuit details if wanted - BUT odds are, a half decent input filter is all that is needed.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks for the detailed answer Russell. I checked the broken pump and I could turn the impeller with no resistance. Agree the filter is important. \$\endgroup\$ Aug 5 '20 at 15:36
  • \$\begingroup\$ @JulienAltieri If you connect the two motor wires together and then try to manually spin the motor, is it really difficult? If not, then the wires or internal fuse have most likely opened. \$\endgroup\$
    – Aaron
    Aug 5 '20 at 16:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.