Searching \ for '[OT] - Input signal conditioning for Magenetic Pic' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/ios.htm?key=input
Search entire site for: '- Input signal conditioning for Magenetic Pic'.

Exact match. Not showing close matches.
PICList Thread
'[OT] - Input signal conditioning for Magenetic Pic'
2007\03\12@044110 by David

flavicon
face
I am wanting to read Engine RPM from a large water pump motor. The motor
is fitted with a 550mm diameter, 118 teeth flywheel
that a Magnetic pickup can be connected in close proximity.

The Range of engine RPM is 500 to 2000.

The manufacturer's datasheet shows a graph with an output voltage going
up as high as 100V p-p when loaded with 100k Ohms.

Does anyone have suggestions regarding suitable circuitry to condition
this signal to be read by a PIC micro counter input ?

My initial thoughts were to isolate the input using opto coupler and
then run into a comparator with hysteresis before the PIC but
I wonder if it is better to put a load across the input then AC couple
the input signal, reduce the level and filter it first as an analogue
signal to remove noise and then into the comparator ?

What would be the most used technique ? Anyone possibly have an example
circuit they would be willing to share ?

Thanks in advance.


Regards

David

2007\03\12@074154 by Gerhard Fiedler

picon face
David wrote:

> I am wanting to read Engine RPM from a large water pump motor. The motor
> is fitted with a 550mm diameter, 118 teeth flywheel
> that a Magnetic pickup can be connected in close proximity.
>
> The Range of engine RPM is 500 to 2000.
>
> The manufacturer's datasheet shows a graph with an output voltage going
> up as high as 100V p-p when loaded with 100k Ohms.
>
> Does anyone have suggestions regarding suitable circuitry to condition
> this signal to be read by a PIC micro counter input ?

I'm not sure how exactly your sensor looks like, but you can look for
"variable reluctance" sensor chips. I've used the (non-RoHS predecessor of
the) ON Semi NCV1124. If you only need one or two and can get them as
samples, that's fine. Also if you need a lot. Otherwise, it seems to be
hard to get by right now in smaller quantities currently. National has also
one (LM1815) that works a bit differently.

> What would be the most used technique ? Anyone possibly have an example
> circuit they would be willing to share ?

If you want to do it yourself, the inside circuits of these chips can give
you some ideas.

Gerhard

2007\03\12@081754 by Peter P.

picon face
The voltage is frequency dependent so using a RC lowpass filter after its output
will equalize the amplitude regardless of frequency. You are looking at about
1000 to 4000 Hz output (assuming one pulse per tooth). This is easily equalized
with a RC low pass. Then one uses a divider if necessary and then one uses a
hysteresis element as you suggested or a specialized chip, f.ex. from National.
With good conditioning you can probably feed the signal directly to a Schmitt
input on a PIC. E.g.:

              ^     ^
              |     |
             33k   0.1u
     c1       |     |
in o--||--22k--*-----*---o pic
     0.1u     |     |
             33k   0.1u  
              |     |
              v     v

The symmetrical input helps with startup and transients. Add Schottky clamps if
necessary. This circuit will divide a 150Vef transducer output (at 2kHz) to 6Vpp
at the pic input. Adjust the 22k resistor for other dividers. C1 must withstand
400 Volts and the 22k must also withstand high voltage.

Peter P.


2007\03\12@123208 by David VanHorn

picon face
On 3/12/07, David <spam_OUTdhuismanTakeThisOuTspambigpond.net.au> wrote:
>
> I am wanting to read Engine RPM from a large water pump motor. The motor
> is fitted with a 550mm diameter, 118 teeth flywheel
> that a Magnetic pickup can be connected in close proximity.
>
> The Range of engine RPM is 500 to 2000.
>
> The manufacturer's datasheet shows a graph with an output voltage going
> up as high as 100V p-p when loaded with 100k Ohms.



Loading inductive pickups with resistance is somewhat problematic.
The waveshape will vary with frequency, and there are resonant points that
will be interesting.

What I have done, is to load the pickup with a virtual short.
An op-amp configured as a current to voltage converter.
In this case, I might put a pair of 1N4148 diodes across the pickup as well,
in case the output current is higher than the opamp can match.

Loading it this way, you've damped down any of the effects I've mentioned
above, limited by the resistance of the wire in the pickup. The Q of any
resonant points has been ruined, and you'll get a much more stable waveform
over changes in speed.

2007\03\12@152950 by Orin Eman

picon face
On 3/12/07, David <.....dhuismanKILLspamspam@spam@bigpond.net.au> wrote:
{Quote hidden}

My car computer uses .001 uF across the sensor, 15k resistor to a
zener to clip the signal, followed by a 2:1 voltage divider using 15k
resistors, 470 pF across the lower leg and finally a schmitt trigger.

Cable to the sensor is shielded with one side of the sensor and shield
grounded at the computer.

I've picked up the schmitt trigger output and connected to a PIC input
for a data acquisition system for this computer, so the circuit does
work for a PIC input.

Much more fun is the circuit for the once per revolution timing pin...

Orin.

2007\03\12@153653 by Recon

picon face
I know the building is a lot of the fun,  But is youwant one already
built they make one for about $30 USD  that sells at small engine
shops.   Just wrap the wire around the spark wire and mount the read out
with the double stick tape provided.  got one on my riding tractor been
working over 2 years now.

RECON


David wrote:

{Quote hidden}

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...