Creating An Activity Meter From A Pulse Width

It's the output pin (9) of the 556, which I believed generated
the pulse width (on that side of my board I have three pots
for centering the servo, and on the other side of the timer I
have the servo connection), but it doesn't make sense to me
that the servo is only getting that small of a voltage signal
to drive it. The side of the timer that has the servo connected
to it (I believe) generates the refresh (48Hz), but according to
my schematic attached in my first post (which lists two 555s),
that would make it backwards on my circuit board. I do get
the refresh signal on that side of the timer though. Soo, I'm
guessing it really is the PW generator side of the circuit that
is putting out that mv signal. Either way, when I make the
magnet stay still in the coil, that V signal stays extremely
constant, fluctuating up to 0.1mv, which in a scale of 1.0mv,
would be acceptable. So, I would bet the signal (where it is
I'm getting it at), is VERY stable. The other output pin (where
I get the oscillating signal), has a much higher V output, also.
But again, the signal I'm getting, from whatever part of the
circuit it is, changes with servo motion, remains stable, and
does not adversely affect the servo's operation when a draw
from my meter is applied.

hi mhs,
I will read thru your last post/edit.
The attached image is the way it could be done, this way the amp is 'ac' coupled so a slow drift on the 180mV will not be a problem. I have not finalised the design at this time, I plan to to do a hardware version check.

I am assuming the 'tremor' frequency will be higher than 1 cycle per second.?

I think we will have a problem using the output of the 556.??

I was not aware you are online at this moment.
I have an attachment I am sending which is my
specific circuit, so watch for it in the next few moments...

here you go...It is the signal sent to the servo, pin 9...

madhippiescientist said:

here you go...It is the signal sent to the servo, pin 9...

Click to expand...

hi,
Copied that, OK.
EDIT:

Simulated your circuit using LTspice, regarding the 180mV +/- 1mV measured on pin 9 of the 556, how did you measure the signal.
If you used an ordinary voltmeter that would explain the reading you saw.

The servo expects a changing mark/space ratio, usually its a 20mSec frame rate with a ~2mS m/s deviation.
Look at the simulation plot and you will see this deviation, I have had to use a massive inductor value in order to get a reasonable plot.

Based on this test I would say the method for displaying the difference in servo feed signal is a non starter.

Ignore the right side circuit on the image.
The LM3914 circuit that we discussed is working on the bench with a simulated 180mV +/- 1mV signal... this is of course not usable with the 556.

Let me know what you decide.

Sorry, but you've lost me. I measured the signal with a DMM.
I do understand how the servo works ("cycling" a signal width).
From there, I don't know what you mean by the bench, or when
I looked at your circuit, it seems to only show (possible)
manipulation of my servo's circuit (I didn't compare your values
to mine). Why would it matter if it's two 555s or a 556? I just
need to amplify the signal that changes when the servo moves,
and zero the meter to ignore the expected continuous output.
Sooo, I don't know what you're saying by "let me know what
you decide" ???

madhippiescientist said:

Sorry, but you've lost me. I measured the signal with a DMM.
I do understand how the servo works ("cycling" a signal width).
From there, I don't know what you mean by the bench, or when
I looked at your circuit, it seems to only show (possible)
manipulation of my servo's circuit (I didn't compare your values
to mine). Why would it matter if it's two 555s or a 556? I just
need to amplify the signal that changes when the servo moves,
and zero the meter to ignore the expected continuous output.
Sooo, I don't know what you're saying by "let me know what
you decide" ???

Click to expand...

hi,
I am saying that the +180mV +/-1mV is not usable in the way we talked about.
The 180mV read by your DVM is not a steady DC signal with the +/-1mV sitting on top.
As you know you the DVM is reading the average value of the servo drive square wave.

Regarding the 'bench' test, if you recall yesterday I said I had a circuit in simulation but I wanted to test the actual circuit by building it, which I did and it runs OK.
But thats with a steady 180mV DC signal and +/-1mV on top, which of course isn't what we are getting from the output from the 556 to the servo, so we can't go that route.
Using a pair of 555's to replace the 556 is not a problem.

Can you measure the voltage signal across the coil when the pendulum is moving as expected.??
It may be possible to pick off that signal, amplify it and drive the LM3914 that way.
You will need a scope to get true coil voltage.

Fun project. I think you need a couple of more specs. before you can get to far. Let's say the pot is turned as quicky as possible from center to 50% of maximum. Should all the lights light because of the speed of the change or 50% of the lights to represent the amount of the change? In either case how long should they stay on until the 50% rotation becomes the new "normal".

Activity

Eric,
Here is a circuit I have used to convert pulse to voltage. Is it what you are looking for?

hi ronv,
Thanks for the circuit, I do have a different version.
I have attached the LTS asc of the circuit showing the 556, which 'mhs' posted, thought it would be helpful for you to try your pulse to Volts circuit.
I have changed a couple of the astable components in an attempt to make it a 20mS frame.
Ignore the OPA on the rightside of the LTS, its a hangover from when I thought we had a 180mVdc with a +/-1mV superimposed.

Thanks, We may have lost "MHS". Would you mind sharing your 556 model with me?
Thanks

PS I can't uzip files from this site for some reason so txt files if you can?

ronv said:

Thanks, We may have lost "MHS". Would you mind sharing your 556 model with me?
Thanks

PS I can't uzip files from this site for some reason so txt files if you can?

Click to expand...

hi ronv,
Download this free IZARC zip/unzip program, seems to cover most zipped files.
**broken link removed**

If you cant download, contact Electromaster.

Added ronv1.zip 556 files

I had tried IZARC earlier with the same results. I turned it in maybe he can fix me up.

ronv said:

Thanks, We may have lost "MHS". Would you mind sharing your 556 model with me?
Thanks

PS I can't uzip files from this site for some reason so txt files if you can?

Click to expand...

hi Ron,
MHS has no web access for a couple of weeks...

ronv said:

Eric,
Here is a circuit I have used to convert pulse to voltage. Is it what you are looking for?

Click to expand...

hi Ron,
This is a version that would do the job.
I have edited the 556 timer to a 20mS frame and a mark of 1mS to 2mS to suit a standard servo..

Activity

ericgibbs said:

hi Ron,
This is a version that would do the job.
I have edited the 556 timer to a 20mS frame and a mark of 1mS to 2mS to suit a standard servo..

Click to expand...

As you can see I haven't figured out how to plug your 556 into LTSpice yet, but here is another cost reduction since it is really only a visual indicator and not a servo. I'm still trying to figure out if he wants it to return to zero (AC coupled) and just show a change. How do you read it?

PS 7 ZIP works for my ZIP problem.

ronv said:

As you can see I haven't figured out how to plug your 556 into LTSpice yet, but here is another cost reduction since it is really only a visual indicator and not a servo. I'm still trying to figure out if he wants it to return to zero (AC coupled) and just show a change. How do you read it?

PS 7 ZIP works for my ZIP problem.

Click to expand...

hi Ron,
He is driving an XY plotter with a stepper and servo, recording Earth tremors/quakes and he wants the LM3914 output to start the plotter and also indicate on the LM3914 LED's the magnitude of of the tremor.

I read it a couple more times. Seems like this is the right approach. See if we agree on the following:
1-The signal will always be AC it can't go to 2 ms and stay there. Therefore AC coupling so the LEDS return to center is not required.
2-If you had 2 sets of LEDS with 1.5 MS at the center you could make the LM3914's so that under vibration they would both appear to light more LEDS in both directions.
3-The earth can't shake at over 50HZ.
After babbling it seems like you have it. What do you think?

hi.
If you look at post #24 circuit that the OP is using, I am not satisfied with some of the component values/timings also the description on how the circuit works.
What do you think.?

Hello Guys, I AM back (you haven't lost me, per se).

It doesn't matter if the meter converts a V or a PW
to display. What I'm aiming for is to have the meter
max out just as the servo starts to move, kinda like
a heads up for me to walk up to the servo as activity
begins so I can watch it take place first hand. The
circuit description (of the schematic driving my servo)
indicates it idles at 1.5ms PW, and when there is (so
to say) no motion of the pendulum, the servo centers
itself somewhere nearby this width. I hope this answers
some of the questions I saw that arose while I was away,
and if there is actually a circuit already posted, or another
perspective for which you have inquiry, please let me know
and I will respond. Otherwise, I haven't factioned much of
the few pages posted while I was gone...