Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Flow sensor has two no flow states

Status
Not open for further replies.

xray

Electrical
Mar 3, 2001
73
I am inquiring about some ideas on how to handle a turbine pickup flow sensor, cheap model about $12. The sensor has three states: 1. NO FLOW - gnd, 2. NO FLOW - +V, 3. FLOW - variable frequency square wave. The signal is low pass filtered and drives the gate of an nmos fet indicating flow or not.

Problem: the flow state and the +V no flow state drive the gate of the nmos fet to on. Any ideas on filtering this signal short of an MCU?

Much appreciated

Cy Drollinger
 
Replies continue below

Recommended for you

Wouldn't one be looking at the frequency of the output?
 
Yes, but Im only interested in flow not how much. So frequency indicates flow and yields a dc value turning on a fet, but so does one of the off states +V, when the turbine stops at the pickup.
 
So you need a circuit (or a commercial module) called a zero rotation detector, or shaft stop detector, or words like that. My point is, it's been done.

It's been done with 555s, of course, but you may find that a small PIC is cheaper than the resistors you need for any analog solution.

Mike Halloran
Pembroke Pines, FL, USA
 
The 555 circuit is called a 'Missing Pulse Detector' in the 555 cook books. As soon as the applicable transition is late (as it would be with either stuck-high or stuck-low), it triggers the output.

 
Yes, that will work. Other solutions require more sweat equity.


Mike Halloran
Pembroke Pines, FL, USA
 
Well, Im working on two circuits driven by the flow sensor: peak detect with final value at about 9V driving 100k resistor to the drain of FET_1 and in parallel a voltage divider consuming very little current driving the gate of the same FET_1.
The peak detect at approximately 9V at the drain of FET_1 drives the gate of FET_2 (flow on signal)while the voltage divider at the gate of FET_1 is not enough to turn FET_1 on.
When the sensor stops at a 12V output also seen by the drain of FET_1 a the voltage divider Vgs large enough to turn FET_1 on dumping the current of 12V through a 100k resistor.

Cy Drollinger
 
Why not write a spec for what you are actually trying to do, and hire an experienced EE to execute it?



Mike Halloran
Pembroke Pines, FL, USA
 
Sometimes one needs to tear-up what was badly implemented, and simply start again fresh. Interfacing from your choice of 'missing pulse detection' options (several options listed above) to the gates of FETs is relatively straightforward.
 
It could be done with just an op amp. Switch is tied to a pull up resistor, pulsing signal feeds through a capacitor and rectified by two diodes. A resistor and capacitor provide smoothing and time constant. op-amp connected as a comparator. Simple enough for a neophyte to diagnose an issue.

Maybe it's just me but I cringe at the thought of using a 555.
 
Maybe it could be done with just an op amp, but we still don't know what 'it' actually is.


Mike Halloran
Pembroke Pines, FL, USA
 
All is well, all you all are great, you all have fine suggestions. Have a great day!
CY
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor