BasicMicro - Forums

Very quick history
Trying to tap into the signal line coming from the speed sender from my street rod tranny going to the speedo. Using the pulse count to shift tranny and lock up torque converter.

Anyways ...

the pulses are nothing more than 12v being turned on and off once per rev.

With the speedo disconnected .. my fluke sees the pulses as 6v (averaged I guess) and
the atom counts the pulses nicely.

BUT !!!!

With the speedo CONNECTED .. tapping into the line .. my fluke only sees 1v and is
not enough to trigger the atom.

soooo ... I have two choices .. I can either amplify the pulses with a transistor or
opamp ...

or ..

Is it possible for me to feed the pulses directly to the atom .. and maybe even process
them .. by speeding them up or slowing them down .. and then .. passing them
trough onto the speedo ??

thanks
mike

I'm surprised that whatever is driving your tach signal is so whimpy!

Without knowing more about your setup it's hard to say how to boost your signal without loading it down and disturbing something else downstream.

One thing I can say is that you should run your tach signal (once it's boosted) through a flip-flop. Trying to count the duration of your pulses can be quite innacurate.

If you employ the flip-flop it will provide a high signal for one complete rev, with microsecond accuracy, and a low signal for one complete rev. The Atom Count command works very nicely in this scenario.

I did this with a rotor tachometer on my helicopter.

You didn't say, but I assume your "fluke" is a Digital Volt Meter (DVM), not an Oscilloscope.

Measuring pulses with a DVM is very iffy. A DVM is designed to read 60 Hz AC accurately, and DC accurately. Pulses, it tends to "average", so that who knows what the number actually means.

The DVM has very high "input impedance", which means it should affect the signal very little. The BS2 also has very high "input impedance" when a pin is set for an input -- like 10 Megohms.

Now if your "fluke" IS an oscilloscope, I too am surprised that the pulse amplitude is around 1 volt. You COULD add an 8-pin op-amp to your BS2 to amplify that signal to something the BS2 could read. I agree with the flip-flop idea, too.

Allanlane5's suggestion for an op amp is a good one, but I'm thinking a comparator like the LM311 might be the best way to go. We use this in a design at work to track the output of a thermistor. The idea is to feed the signal to an input, which allows you to 'snoop' the tranny output without breaking the circuit. This is a 8 pin device, DIP or SOIC, which can be had at DigiKey for 50 cents in single quantities. Which is a pain when you realize you'll pay a $5 handling fee and lots more for shipping, but... you DO want this to work, right?

This part will work with voltages (inputs, outputs and rails) up to 30 Volts. You can think of this as a voltage translator. It has an open-collector output, so can be used to switch dissimilar voltages between the input and output. Depending on how you set up the inputs (inverting or non-inverting) you can have a high out when the input is low, vice versa, or follow the input. In your case, connect the LM311's Vcc to the car's supply and the pullup resistor on its output to the 5V driving the Atom. When the higher voltage of the tranny pulse goes from ground to 12-13V, the comparator moves between ground and 5V. You can use a voltage divider on one input of the comparator to determine when it "fires", or switches state. Using a potentiometer, you can dial this down until it fires reliably without dipping down into the noise (it's a car - OF COURSE there's electrical noise!) and jerking around the count.

As for pin assignments, go to DigiKey and use their excellent site to find and download the data sheets for this. There are TONS of sites discussing comparator usage.

And have fun doing it!
kenjj

Ok .... I have reading you guys suggestions and reading and reading data sheets.

I am thinking that the comparator may the way to go also.

>kenjj wrote
>When the higher voltage of the tranny pulse goes from ground to 12-13V, the comparator >moves between ground and 5V. You can use a voltage divider on one input of the >comparator to determine when it "fires", or switches state

When I read the voltage with the speedo disconnected I DO get 12-13v .. but .. when I
have the speedo in the circuit .. so that I am "tapping" into the signal .. it drops to
like 2v ... (this really surprised me)

anyways ...THAT is my problem - finding something that a 2v square wave will trigger

but ... Ill go get a 339 or 311 and maybe a 741 and start breadboarding away.

oh .. and I did buy a flip-flop to see if that cleans up the whole mess ...

anyways .. thanks so much for the basic elctronics help ..

mike

That's the beauty of the comparator. This particular part can resolve signal differences of a few thousand MILLIVOLTS! Two Volts is a sledge hammer to this thing! Put the circuit in the car and before connecting the tranny signal adjust the pot so the input pin reads 1V, even down to half a Volt. Every time the signal swings back and forth through this level the comparator changes the output level. You just may find a setting this low picks up noise, so may have to adjust things a little higher. At least until the noise stops having an effect. You could put a resistor-capacitor filter at the signal input to filter out most of the noise.

Later!
kenjj