Stamp tachometer problems

[steamer]

--There's a site up with an excellent description of how to build a
tach with a Basic Stamp, at:
http://www.floridaame.org/Plans/Kamran%20Nili/rpm/tachometer.html
I built this circuit the other day, then had an expert pal make sure
I hadn't done anything wrong (left out one wire!).
Once all was set to rights I tried the three programs to see what
would happen.
First one: got some readings on the debug screen when waving a
magnet back
and forth near the sensor: a good sign, yes?
So then I loaded the second program and got the proper responses
from the display, so all was well there.
After loading the third program things started to get weird; i.e.
we were unable to get consistent readings. What I did is tape a neodymium
magnet to the edge of a little 2" sanding disk, chuck it in a small drill
and power it up near the sensor. It seemed to do *something* but not what
I expected; i.e. sometimes I'd get a reading after I removed the rotating
magnet,
but not while it was spinning. I thought maybe the neodymium magnet
was too strong and switched to a weaker one but results were just as
weird.
I'm wondering: might there be a problem with the "flywheel"
being too small in diameter? Don't think this could be the problem but one
never knows.
There were also times when the numbers displayed were almost
random; i.e. readings being much higher than the rate at which we waved
magnets near sensor. It was all a bit baffling.
We're sort of in the dark here, so any clues folks can pass along
would
be greatly appreciated. Thanks,


--
"Steamboat Ed" Haas : Currently broke and
Hacking the Trailing Edge! : looking for a job...
www.nmpproducts.com
---Decks a-wash in a sea of words---

 

[Tim Wescott]

On Tue, 22 Dec 2009 22:36:01 +0000, steamer wrote:

--There's a site up with an excellent description of how to build a tach
with a Basic Stamp, at:
http://www.floridaame.org/Plans/Kamran%20Nili/rpm/tachometer.html
I built this circuit the other day, then had an expert pal make
sure
I hadn't done anything wrong (left out one wire!).
Once all was set to rights I tried the three programs to see what
would happen.
First one: got some readings on the debug screen when waving a
magnet back
and forth near the sensor: a good sign, yes?
So then I loaded the second program and got the proper responses
from the display, so all was well there.
After loading the third program things started to get weird; i.e.
we were unable to get consistent readings. What I did is tape a
neodymium magnet to the edge of a little 2" sanding disk, chuck it in a
small drill and power it up near the sensor. It seemed to do *something*
but not what I expected; i.e. sometimes I'd get a reading after I
removed the rotating magnet,
but not while it was spinning. I thought maybe the neodymium magnet was
too strong and switched to a weaker one but results were just as weird.
I'm wondering: might there be a problem with the "flywheel"
being too small in diameter? Don't think this could be the problem but
one never knows.
There were also times when the numbers displayed were almost
random; i.e. readings being much higher than the rate at which we waved
magnets near sensor. It was all a bit baffling.
We're sort of in the dark here, so any clues folks can pass along
would
be greatly appreciated. Thanks,

Did you try it slow? Did you try the 1st and 2nd program with the wheel
on the drill? If you have an oscilloscope, you should look at the output
from the pickup -- that'll tell you if it's clean.

You're getting a program for free, and you're hoping to gain more value
from it than you paid, it could just not work right.

--
www.wescottdesign.com

 

[ehsjr]

steamer wrote:

--There's a site up with an excellent description of how to build a
tach with a Basic Stamp, at:
http://www.floridaame.org/Plans/Kamran%20Nili/rpm/tachometer.html
I built this circuit the other day, then had an expert pal make sure
I hadn't done anything wrong (left out one wire!).
Once all was set to rights I tried the three programs to see what
would happen.
First one: got some readings on the debug screen when waving a
magnet back
and forth near the sensor: a good sign, yes?

Yes, a good sign. Did you get the proper number displayed
that matched the number of times you waved the magnet
back and forth?

So then I loaded the second program and got the proper responses
from the display, so all was well there.

8888 and then 2845, as per the site you mentioned?

After loading the third program things started to get weird; i.e.
we were unable to get consistent readings. What I did is tape a neodymium
magnet to the edge of a little 2" sanding disk, chuck it in a small drill
and power it up near the sensor. It seemed to do *something* but not what
I expected; i.e. sometimes I'd get a reading after I removed the rotating
magnet,
but not while it was spinning. I thought maybe the neodymium magnet
was too strong and switched to a weaker one but results were just as
weird.

The motor may be affecting the sensor, or broadcasting
electrical noise into the circuit. Try a different
setup that does not use that drill - or any electrical
motor, for that matter. Also, be sure to run your
circuit from a good 9V battery - NOT a power supply.

I'm wondering: might there be a problem with the "flywheel"
being too small in diameter? Don't think this could be the problem but one
never knows.

No. Flywheel diameter has nothing to do with it. Spacing
between the magnet and the hall sensor is a factor, as is
RPM at which the flywheel is moving.

There were also times when the numbers displayed were almost
random; i.e. readings being much higher than the rate at which we waved
magnets near sensor. It was all a bit baffling.

That makes it sound like the results from program one were
not correct. You need to be able to count the number of
waves correctly:
"At this time you test the program moving the magnet close to the Hall
sensor. In the debug screen will see how many times you moved the magnet
back and fort per sec, that number than is multiplied by 60 and gives
you pluses per minute."

If you can't get the correct results here, then you won't get
the correct results with program 3. I'd suspect a wiring error.

Ed

We're sort of in the dark here, so any clues folks can pass along
would
be greatly appreciated. Thanks,

 

[steamer]

Tim Wescott <tim@seemywebsite.com> wrote:

Did you try it slow?
--Yes; max rpm maybe 400.

Did you try the 1st and 2nd program with the wheel on the drill?
--Yes
If you have an oscilloscope, you should look at the output
from the pickup -- that'll tell you if it's clean.
--Good idea; will try that next.

You're getting a program for free, and you're hoping to gain more value
from it than you paid, it could just not work right.
--Heh, yeah; still I live in hope that some day I'll be able to get

reliable numbers for my steam engines..

--
"Steamboat Ed" Haas : Currently broke and
Hacking the Trailing Edge! : looking for a job...
www.nmpproducts.com
---Decks a-wash in a sea of words---

 

[steamer]

ehsjr <ehsjr@nospamverizon.net> wrote:

Yes, a good sign. Did you get the proper number displayed
that matched the number of times you waved the magnet
back and forth?
--Not really; chalked it up to inconsistent speed of magnet, but...

So then I loaded the second program and got the proper responses
from the display, so all was well there.

8888 and then 2845, as per the site you mentioned?
--Yes, just so.

After loading the third program things started to get weird; i.e.
we were unable to get consistent readings. What I did is tape a neodymium
magnet to the edge of a little 2" sanding disk, chuck it in a small drill
and power it up near the sensor. It seemed to do *something* but not what
I expected; i.e. sometimes I'd get a reading after I removed the rotating
magnet,
but not while it was spinning. I thought maybe the neodymium magnet
was too strong and switched to a weaker one but results were just as
weird.

The motor may be affecting the sensor, or broadcasting
electrical noise into the circuit. Try a different
setup that does not use that drill - or any electrical
motor, for that matter. Also, be sure to run your
circuit from a good 9V battery - NOT a power supply.
--Hmmm; well I'll point out that the drill was on a flex shaft so

motor was well removed.
--Re: battery vs wall wart: will try that too; thanks!

I'm wondering: might there be a problem with the "flywheel"
being too small in diameter? Don't think this could be the problem but one
never knows.

No. Flywheel diameter has nothing to do with it. Spacing
between the magnet and the hall sensor is a factor, as is
RPM at which the flywheel is moving.

There were also times when the numbers displayed were almost
random; i.e. readings being much higher than the rate at which we waved
magnets near sensor. It was all a bit baffling.

That makes it sound like the results from program one were
not correct. You need to be able to count the number of
waves correctly:
"At this time you test the program moving the magnet close to the Hall
sensor. In the debug screen will see how many times you moved the magnet
back and fort per sec, that number than is multiplied by 60 and gives
you pluses per minute."

If you can't get the correct results here, then you won't get
the correct results with program 3. I'd suspect a wiring error.

Ed

We're sort of in the dark here, so any clues folks can pass along
would
be greatly appreciated. Thanks,

--
"Steamboat Ed" Haas : Currently broke and
Hacking the Trailing Edge! : looking for a job...
www.nmpproducts.com
---Decks a-wash in a sea of words---

 

[Bob Masta]

On 22 Dec 2009 22:36:01 GMT, steamer
<steamer@sonic.net> wrote:

--There's a site up with an excellent description of how to build a
tach with a Basic Stamp, at:
http://www.floridaame.org/Plans/Kamran%20Nili/rpm/tachometer.html
I built this circuit the other day, then had an expert pal make sure
I hadn't done anything wrong (left out one wire!).
Once all was set to rights I tried the three programs to see what
would happen.
First one: got some readings on the debug screen when waving a
magnet back
and forth near the sensor: a good sign, yes?
So then I loaded the second program and got the proper responses
from the display, so all was well there.
After loading the third program things started to get weird; i.e.
we were unable to get consistent readings. What I did is tape a neodymium
magnet to the edge of a little 2" sanding disk, chuck it in a small drill
and power it up near the sensor. It seemed to do *something* but not what
I expected; i.e. sometimes I'd get a reading after I removed the rotating
magnet,
but not while it was spinning. I thought maybe the neodymium magnet
was too strong and switched to a weaker one but results were just as
weird.
I'm wondering: might there be a problem with the "flywheel"
being too small in diameter? Don't think this could be the problem but one
never knows.
There were also times when the numbers displayed were almost
random; i.e. readings being much higher than the rate at which we waved
magnets near sensor. It was all a bit baffling.
We're sort of in the dark here, so any clues folks can pass along
would
be greatly appreciated. Thanks,

If you want to test your sensor setup on an
independent tachometer, you can download my
Daqarta app that uses your sound card for input.
One advantage of this is that you can see the
input waveform and play with the triggering, which
may give clues to your Basic Stamp problem.

If the discussion below isn't adequate, note that
you can right-click any control to get Help about
it. (Or ask me!)

Click on the thin button below 'Input' on the
toolbar to open the Input control dialog, and make
sure Line In is set. You can connect your sensor
to a standard sound card audio cable cable using
mini-gator clips. The tip is Left, which is
easiest to grab, and the sleeve is ground. Toggle
Input on and you should see the raw waveform from
your sensor. Adjust Input Level (more-positive
values are more sensitive) for an unclipped
waveform.

You can play with the Trigger controls to get a
stable display if the default isn't acceptable.
(I usually prefer Normal mode instead of the
default Auto.)

Now go to the Options menu and hit Frequency
Counter (or use CTRL+F). Click on RPM and you
should be in business. Note that the Frequency
Counter window borders can be dragged to resize
it, and the readout will grow or shrink to match.
Depending on your display, you may need to make it
wider the first time you use it, to see the
controls at the far right of the counter. The Cyl
control works with RPM for use with engines, but
it's also for reading RPM from gear teeth. You
need to set it to *twice* the number of teeth to
get a direct RPM reading.

This setup should give you an independent view of
the signal into the Basic Stamp. Daqarta's
30-day/30-session free trial should be enough to
debug your Stamp problem. Enjoy!

Best regards,


Bob Masta

DAQARTA v5.00
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Scope, Spectrum, Spectrogram, Sound Level Meter
Frequency Counter, FREE Signal Generator
Pitch Track, Pitch-to-MIDI
DaqMusic - FREE MUSIC, Forever!
(Some assembly required)
Science (and fun!) with your sound card!