Hello!

Here I am working on a system that involves a few electrical components controlling and driving a DC 12v Water pump (max current 6amps). My desire is for a capacitive water level sensor to kick on the pump when a specific level of water is detected, and alternately shut off the pump when the water level is too low. Furthermore, an integrated water flow reed switch shuts off the pump when no flow is registered on a parallel piping system.

My realization is that the inrush current (and operation current) of the pump burns out both the capacitive sensor and the reed flow switch. To combat the operation (and inrush) current of the pump, I have discovered two possible solutions: Inrush Current Limiters (ICL) or MOSFET.

I attempted to calculate the ICL I needed, but apparently did so incorrectly. After integrating an ICL I purchased into the system, it protected the sensor (yay!), but it did not allow enough voltage to pass through to the pump when the sensor was submerged in water.

Questions:
1) Which of these solutions is simplest and least expensive to implement?
2) Are there other solutions that I should investigate?
3) As far as the rating of MOSFETs and ICLs go, would I have to purchase different MOSFETs/ICLs to protect the different individual components (sensor and switch)?
4) And lastly, if I've supplied enough information, which MOSFET/ICL would I need?

Links and tech data of the components I am using:
Pump
Amp draw: 3.6A (max 6A)
Voltage: 12vdc

Capacitive water level sensor
Rated operational current (I_e): Continuous ≤ 50 mA
No-load supply current (I_o): ≤ 10 mA
Voltage: 10-30vdc

Flow Switch
Current: 0.08A
Voltage: 12-130vdc

I am not married to any of the components, however, I am content with the operation of the flojet pump. I hope the dilemma is clear enough - any help would be much appreciated.

Best

 

Without seeing a schematic of your system, it's all a guessing game.

 

As you have learned, neither the sensor nor the switch can control the pump directly. This is because 6 A is much greater than either 0.05 A or 0.08 A. What you want to do can be accomplished by adding to your circuit either a solid state relay (SSR) or a power MOSFET and some resistors and capacitors, *maybe*. maybe, because the description of the inline switch is confusing

ak

 

Inductors are intrinsic ICL components. Take a length of wire ... appropriately sized .. wrap the wire around a piece of PVC pipe, used as a form, and tape together with electrical tape. The number of turns will have to be determined experimentally ... start with 10 or 20.
... Place the wire coil in series with the pump circuit. The only time there will be substantial voltage drop across the coil is when the circuit is energized or de-energized.
\(V_L=L{\frac {di} {dt}}\)
Note ... The inductor strength, L, can be enhanced by inserting a length of iron rod into the coil form ... e.g. a piece of steel rebar ... this may be necessary to obtain sufficient inductance.

 

Can you estimate the size and weight of an inductor to limit a 6 A inrush down to less than 1 A for two seconds?

ak

 

Can you estimate the size and weight of an inductor to limit a 6 A inrush down to less than 1 A for two seconds?

ak

.. that will require some research.

The main difficulty would be getting a practical estimate of the switch time transient.

... Suggest making an experimental trial.

 

It looks like there is a quantity called the surge impedance :
\( Z=sqrt{\frac L C}\)
... Just improve that factor by a fraction.
It may not be necessary to completely compensate for the overall switching transient.

 

jbigej........I'm not sure if I understand your setup. But a relay is a great device for controlling a large current with a small current.

 

If you derate everything by 50%, then he has 25 mA of control current available. Other than a DC SSR, I don't think that is enough to drive a relay with 10 A contacts.

ak

 

... Just ...

I don't think it is quite that simple. The static current is 3.6 A and the transient current is 6 A. To reduce the transient current by 5x or 10x will take a large inductor. Also, you now have a huge inductive kick to deal with at turn-off.

His initial plan to control power to the pump directly with the two components will not work.

ak

 

... can't tell without the actual circuit, but the possibility of extra added capacitance, due to the capacitive sensor, suggests that the transient impedance is less than it would be otherwise. Maybe the sensor is cause for excessive inrush current. So, in a qualitative sense, the thing to do would be to add inductance.
... Not enough details to make any numerical estimates though.

 

... can't tell without the actual circuit, but the possibility of extra added capacitance, due to the capacitive sensor, suggests that the transient impedance is less than it would be otherwise. Maybe the sensor is cause for excessive inrush current. So, in a qualitative sense, the thing to do would be to add inductance.
... Not enough details to make any numerical estimates though.

It's hard to imagine that any of the quirks of inductors or capacitive sensors really matter until we've addressed the fact that the load current is two orders of magnitude higher than what the sensors can handle. Even ignoring the inrush current, you can't run 3.6 amps (3600mA) through something that's only rated for 50mA. No amount of inrush limiting is going to solve the fundamental problem.

Clearly a circuit needs to designed to take the signal from the sensors and use it to control an appropriate switch off some sort. A pre-made DC SSR would be the easiest, but most expensive, solution. A reasonably simple MOSFET-based switching circuit would probably be more cost effective if the OP is ready to do a little soldering and figure out how to enclose the electronics.

EDIT: ...I suppose it's possible I've misunderstood the specs on the sensors. If those specs represent what the sensors themselves consume, but they're capable of switching more current, then that would change things. In that case we'd need the rest of the specs. I think we need part numbers or datasheets, along with a schematic (sketch on a napkin, whatever) of the current wiring scheme.