Help with Counting frequency of tones

lfreeze

I have been struggling through the Counter Modules and Circuit Applications section
of the propeller education kit lab series (most of it way beyond my skill level).
I have been able to assemble a counter
program (attached as COUNTER1) Which counts the cycles of tones generated by
the propeller. the tones are generated in a separate COG and are measured on an
asynch basis. I take enough snippets of the tone to get a consistent number
for each frequency.

I thought I would take this a step further and generate tones external from the
propeller and measure them using a smaller propeller program (attached as COUNTER2).
I used a BS2 and the freqout command, to generate the tones. I used the same
frequencies and intervals, but got very different unusable
results. I don't have a scope, so I could only verify the tones by listening to them.
They sound okay, but the measurements are not. I made a direct connection from the PIN on
the BS2 through a 1K resistor to the Propeller pin.

I don't know if I have created a bad program which will only work with propeller generated
tones, or I have to build an interface from the stamp2 to the propeller.
Any help or guidance would be appreciated. Here are some examples of the results I am
getting with both of the programs. Please excuse the length of this question, but I wasnt
sure how else to ask it.

Propeller measuremet of 1K, 2K, and 3K Propeller generated tones

---

Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 1010
Frequency = 1010
Frequency = 1020
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 2020
Frequency = 2020
Frequency = 2020
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 2240
Frequency = 3030
Frequency = 3030
Frequency = 2720
Frequency = 0
Frequency = 0
Frequency = 0

Propeller measurement of 1K, 2K, and 3K Freqout tones from STAMP

---

Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 48640
Frequency = 48690
Frequency = 25850
Frequency = 0
Frequency = 0
Frequency = 47350
Frequency = 47750
Frequency = 47720
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 47690
Frequency = 47710
Frequency = 47740
Frequency = 0
Frequency = 0
Frequency = 13880

Thanks in advance for reading this and for any help you can provide.

Larry

Phil Pilgrim (PhiPi)

The numbers you're seeing from the BASIC Stamp are probably accurate. The Stamp's FREQOUT statement does not put out a square wave of the selected frequency. Rather, it outputs a square wave that, when low-pass filtered, produces an approximate sine wave of the chosen frequency. To accomplish this, there are many extra transistions in the square wave. You could try low-pass filtering the Stamp output and measuring the result. I'll bet you come a lot closer to what you expect.

-Phil

lfreeze

Thanks Phil, that solves the mystery. I· had thought that freqout produced a good sine wave as I had used it

to create DTMF tones and it worked okay.· ·I will take your suggestion and try the low pass filter. I will

also find another source to generate the tones as a further benchmark for testing..

Larry

deSilva

@Ifreeze: Your measurements are rather unprecise... The reason is that you you perform many instructions other than the WAITCNT in line 50.

A substantial improvement of accuracy can be achieved when you set PHSA to zero just before the WAITCNT and read it immediately after the WAITCNT.

Subtracting 2300 is not a good idea for many reasons...
WAITCNT should take around 800 ticks extra only (ca. 10 µs @80 MHz)

lfreeze

I inserted the low pass filter that Phil suggested and
It certainly improved the results (see below). I built a trial
and error filter using devices I had on hand. I still think working with
analog signal levels and frequencies is close to witchcraft, but I can't argue with
success. The RC filter component combination
that produced the best results were an a 1K ohm
resistor with .1MFD capacitor. I also changed the program to include
the suggestions DeSilva made. I also increased the reps by reducing the
waitcnt. (revised program attached)

Frequency = 21020
Frequency = 21420
Frequency = 21690
Frequency = 21130
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 22200
Frequency = 33530
Frequency = 32910
Frequency = 32470
Frequency = 27870
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 143470
Frequency = 146090
Frequency = 142430
Frequency = 144220
Frequency = 6450
Frequency = 0
Frequency = 0
Frequency = 0
Frequency = 0

Thank you both for your time and help.
Larry