HELP:Low-pass filter on frequency counter

[caius]

Hi all,
I have to made some measurements with a frequency counter (model VC3165) on
low frequencies (up to 60 Hz) .I know that in case of low frequency
measurements a low-pass filter is needed in order to obtain stable and
precise readings.
My frequency counter has only an AC/DC button (to be used with a low-pass
filter as said in the service manual) but it doesn't have an integrated
low-pass filter (many counters have it).

I built some simple RC filter ( using a resisitor and a capacitor) with
various cut-off frequncy ( I tried 15 Khz and 50 Khz) but I alway obtain
floating and not precise readings on low frequencies.
How can I solve this problem?Maybe using a low-pass filter like these(very
expensive IMHO)?:

http://cgi.ebay.com/AGILENT-HP-10856A-LOW-PASS-FILTER-KIT-5KHz-50KHz-500KHz-/120668081898?pt=LH_DefaultDomain_0&hash=item1c1860ceea

or building a proper one (but as said before simple RC filters didn't work)?

 

[William Sommerwerck]

A simple RC filter gives only 6dB/8ve rolloff. You probably need something
much sharper, which would require active filters.

 

[caius]

"William Sommerwerck" <grizzledgeezer@comcast.net> ha scritto nel messaggio
news:ihcj1s$hfc$1@news.eternal-september.org...

A simple RC filter gives only 6dB/8ve rolloff. You probably need something
much sharper, which would require active filters.

Ok, thanks, I imagined this.Is it difficult to build an active low-pass
filter (with a 50KHz cut-off frequency)?

 

[William Sommerwerck]

Ok, thanks, I imagined this. Is it difficult to build an
active low-pass filter (with a 50KHz cut-off frequency)?

It's easy to design one. But... with such a high cut-off, you'll op amps
with a very high gain-bandwidth product to get a stable filter (ie, one
whose characteristics don't change much with variations in component
values).

 

[Fred McKenzie]

In article <4d39cb36$0$2048$4fafbaef@reader1.news.tin.it>,
"caius" <briccus@yahoo.it> wrote:

I have to made some measurements with a frequency counter (model VC3165) on
low frequencies (up to 60 Hz) .I know that in case of low frequency
measurements a low-pass filter is needed in order to obtain stable and
precise readings.

Caius-

Your problem may not be related to high frequencies unless you are
measuring low level, high impedance signals in the presence of high
frequency electromagnetic fields.

One possibility is that you are over-driving the input to the counter.
Using an attenuator may help.

Another possibility is that the signal actually varies in frequency over
time.

Fred

 

[Fabio]

"Fred McKenzie" <fmmck@aol.com> ha scritto nel messaggio
news:fmmck-7261FF.15494021012011@5ad64b5e.bb.sky.com...

In article <4d39cb36$0$2048$4fafbaef@reader1.news.tin.it>,
"caius" <briccus@yahoo.it> wrote:
One possibility is that you are over-driving the input to the counter.
Using an attenuator may help.

My counter has a X20 button attenuaor but form manjla I didn't understand if
it on channel A or B
Anyway if I press down the reading become stable on low frequecies>

Another possibility is that the signal actually varies in frequency over
time.

I think not.Because , for example, now I'm measuring the pin5 of the EL4583
(an IC sync separator) , this pin provides vertical sync output.As video
input I'm using a video composite sync, so I should obtain a stable reading
of 60Hz on pin 5 of EL4583.The reading is floating ( goes up and down from
60 to 63 Hz) instaed.

One possibility is that you are over-driving the input to the counter.
Using an attenuator may help.

My counter has a X20 button attenuator but from manual I didn't understand
if it's on channel A or B
Anyway if I press down the button the reading become stable on PIN5 of the
EL4583, may it's as you said?.This is the manual:

http://www.emclr.co.uk/productdownloads/prod_45/VC3165%20User%20Manual.pdf

Anyway a friend of mine uses another counter which has a 50KHz low-pass
filter and measuring the PIN5 of the EL4583 the reading is stable with the
low-pass filter switched on:

http://www.tti-test.com/products-tti/pdf-brochure/prec-pfm1300-2p.pdf

From this manual:

"For Input A, a low pass filter with a cut-off frequency of 50kHz is user
selectable to ensure stable readings at low frequencies"

So, I'm bit confused..do I need a low-pass filter or an attenuator to obtain
stable readings on low frequencies using my counter?

 

[caius]

Sorry for the previous post sent from 'Fabio', by mistake I used another
account form another computer but I'm always 'Caius' the author of the post

 

[Michael A. Terrell]

caius wrote:

Hi all,
I have to made some measurements with a frequency counter (model VC3165) on
low frequencies (up to 60 Hz) .I know that in case of low frequency
measurements a low-pass filter is needed in order to obtain stable and
precise readings.
My frequency counter has only an AC/DC button (to be used with a low-pass
filter as said in the service manual) but it doesn't have an integrated
low-pass filter (many counters have it).

I built some simple RC filter ( using a resisitor and a capacitor) with
various cut-off frequncy ( I tried 15 Khz and 50 Khz) but I alway obtain
floating and not precise readings on low frequencies.
How can I solve this problem?Maybe using a low-pass filter like these(very
expensive IMHO)?:

http://cgi.ebay.com/AGILENT-HP-10856A-LOW-PASS-FILTER-KIT-5KHz-50KHz-500KHz-/120668081898?pt=LH_DefaultDomain_0&hash=item1c1860ceea

or building a proper one (but as said before simple RC filters didn't work)?

Time period measurement is more accurate at low frequencies, as it
reduces the timing errors. You measure the time period, then calculate
the reciprocal.


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

 

[Phil Allison]

"caius"

I have to made some measurements with a frequency counter (model VC3165)
on low frequencies (up to 60 Hz) .I know that in case of low frequency
measurements a low-pass filter is needed in order to obtain stable and
precise readings.
My frequency counter has only an AC/DC button (to be used with a low-pass
filter as said in the service manual) but it doesn't have an integrated
low-pass filter (many counters have it).

I built some simple RC filter ( using a resistor and a capacitor) with
various cut-off frequency ( I tried 15 Khz and 50 Khz) but I alway obtain
floating and not precise readings on low frequencies.

** Errr - as you only want to measure 60Hz, why not have a LP filter at say
100Hz ??

A 10kohm resistor and a 0.15uF film cap ought to work - plus keep the input
level down to something the frequency counter likes.



...... Phil

 

[caius]

"Michael A. Terrell" <mike.terrell@earthlink.net> ha scritto nel messaggio
news:ZbSdnTlf2LOflKfQnZ2dnUVZ_sydnZ2d@earthlink.com...

Time period measurement is more accurate at low frequencies, as it
reduces the timing errors. You measure the time period, then calculate
the reciprocal.

Yes, I have this period function on my counter but I didn't understand how
to use it, if you could help me.

 

[Dave M]

Fabio wrote:

"Fred McKenzie" <fmmck@aol.com> ha scritto nel messaggio
news:fmmck-7261FF.15494021012011@5ad64b5e.bb.sky.com...
In article <4d39cb36$0$2048$4fafbaef@reader1.news.tin.it>,
"caius" <briccus@yahoo.it> wrote:
One possibility is that you are over-driving the input to the
counter. Using an attenuator may help.

My counter has a X20 button attenuaor but form manjla I didn't
understand if it on channel A or B
Anyway if I press down the reading become stable on low frequecies
Another possibility is that the signal actually varies in frequency
over time.

I think not.Because , for example, now I'm measuring the pin5 of the
EL4583 (an IC sync separator) , this pin provides vertical sync
output.As video input I'm using a video composite sync, so I should
obtain a stable reading of 60Hz on pin 5 of EL4583.The reading is
floating ( goes up and down from 60 to 63 Hz) instaed.

One possibility is that you are over-driving the input to the
counter. Using an attenuator may help.

My counter has a X20 button attenuator but from manual I didn't
understand if it's on channel A or B
Anyway if I press down the button the reading become stable on PIN5
of the EL4583, may it's as you said?.This is the manual:

http://www.emclr.co.uk/productdownloads/prod_45/VC3165%20User%20Manual.pdf

Anyway a friend of mine uses another counter which has a 50KHz
low-pass filter and measuring the PIN5 of the EL4583 the reading is
stable with the low-pass filter switched on:

http://www.tti-test.com/products-tti/pdf-brochure/prec-pfm1300-2p.pdf

From this manual:

"For Input A, a low pass filter with a cut-off frequency of 50kHz is
user selectable to ensure stable readings at low frequencies"

So, I'm bit confused..do I need a low-pass filter or an attenuator to
obtain stable readings on low frequencies using my counter?

From the spec sheet that you posted on your counter, the X20 button is not a
filter; it's an attenuator. The description says that channel B has an
attenuator (X1,X20), & AC/DC coupling for a low-pass filter. It attenuates
the signal from the B input (selected by the function switch) by a factor of
20. According to the table at the top of the page, in order to measure
frequencies below 100Hz, you should be using DC coupling. AC coupling is
really a high-pass filter, which, from the table, attenuates frequencies
below 100Hz.
Which input are you using to measure your 60Hz sync signal? If you're tring
to use the B input, of the A input with the Function set to the 2-50MHz
range, then it's no wonder that you're getting erratic readings. It's a
wonder that you're getting any reading at all.

From the specs on the channel A input, you should be OK to measure a high
level signal (up to 30VP-P. The fact that your reading is varying from
60-63 HZ, that would indicate that your signal is mixing, or being
overwhelmed by mains coupling. Try isolating the counter's ground from the
power line. That brings another question... Where are you connecting the
counter's signal ground? Are you relying on the power line ground for the
connection? If so, that's a very bad idea. Use a scope probe or shielded
cable, and connect the probe's shield to the signal ground of the circuit
that you're testing.
I really doubt that a low-pass filter is going to help you here, unless the
signal is not clean, that is, has a lot of noise riding on it.
How clean is the power supply that is powering the circuit under test? If
it has a lot of ripple, that could explain your erratic readings. Make sure
the Vcc rail is clean from ripple and noise.
--
David
dgminala at mediacombb dot net

 

[Michael A. Terrell]

caius wrote:

"Michael A. Terrell" <mike.terrell@earthlink.net> ha scritto nel messaggio
news:ZbSdnTlf2LOflKfQnZ2dnUVZ_sydnZ2d@earthlink.com...
Time period measurement is more accurate at low frequencies, as it
reduces the timing errors. You measure the time period, then calculate
the reciprocal.

Yes, I have this period function on my counter but I didn't understand how
to use it, if you could help me.

Download this manual. It is for a '60s HP 5245L, but it has good
desriptions on how varous functions work, and how to use them. Period
measurment starts on page 22. Then ask any questions you still have and
I'll try to help you. It is a scan of an old printed manaul, so I can't
copy the text for you. It is a 16.1 MB file of the 1973 version of that
manual.


<http://cp.literature.agilent.com/litweb/pdf/05245-9044.pdf>


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

 

[caius]

"Phil Allison" <phil_a@tpg.com.au> ha scritto nel messaggio
news:8pug4kFs56U1@mid.individual.net...

** Errr - as you only want to measure 60Hz, why not have a LP filter at
say
100Hz ??

A 10kohm resistor and a 0.15uF film cap ought to work -

Ok, but where exactly i have to put this low-pass filter?In the circuit to
be measured (before the input) or between the counter and the probe?The
circuit is this:
http://img838.imageshack.us/img838/8206/snap1nx.jpg

R3 and C2 in this circuit should be a simple RC circuit, right?

plus keep the input
level down to something the frequency counter likes.

Did you mean the input impedance?According to the manual of my counter,
Channel A has 1Mohm of input impedance.How can measure it on the circuit?

 

[Phil Allison]

"caius"

"Phil Allison"
** Errr - as you only want to measure 60Hz, why not have a LP filter at
say 100Hz ??

A 10kohm resistor and a 0.15uF film cap ought to work -

Ok, but where exactly i have to put this low-pass filter?In the circuit to
be measured (before the input) or between the counter and the probe?

** Your are obviously a babe in the woods.

The RC filter goes at the *input* to the counter.

A counter is a dumb animal that tries to count whatever you feed it - so
if you expect to see a stable reading then make sure you are feeding it a
stable SINGLE frequency.

Any signal that contains harmonics of the fundamental or hum or noise is bad
news.

Look at the signal on a scope.

You have one of them - right ??

plus keep the input
level down to something the frequency counter likes.

Did you mean the input impedance?

** So you have no idea of the difference between " input level" and
"input impedance" ??

The woods is a dangerous place for you - pal.


...... Phil

 

[caius]

"Dave M" <dgminala4444@mediacombb.net> ha scritto nel messaggio
news:wOOdnXELXJK6i6fQnZ2dnUVZ_u2dnZ2d@giganews.com...

From the spec sheet that you posted on your counter, the X20 button is not
a filter; it's an attenuator. The description says that channel B has an
attenuator (X1,X20), & AC/DC coupling for a low-pass filter. It
attenuates the signal from the B input (selected by the function switch)
by a factor of 20. According to the table at the top of the page, in
order to measure frequencies below 100Hz, you should be using DC coupling.
AC coupling is really a high-pass filter, which, from the table,
attenuates frequencies below 100Hz.
Which input are you using to measure your 60Hz sync signal? If you're
tring to use the B input, of the A input with the Function set to the
2-50MHz range, then it's no wonder that you're getting erratic readings.
It's a wonder that you're getting any reading at all.

According to the table, for measuring 60Hz I'm using the Channel A (from
0.01Hz to 50 MHz) with step 3 and DC coupling pressed

From the specs on the channel A input, you should be OK to measure a high
level signal (up to 30VP-P. The fact that your reading is varying from
60-63 HZ, that would indicate that your signal is mixing, or being
overwhelmed by mains coupling. Try isolating the counter's ground from
the power line. That brings another question... Where are you connecting
the counter's signal ground? Are you relying on the power line ground for
the connection? If so, that's a very bad idea. Use a scope probe or
shielded cable, and connect the probe's shield to the signal ground of the
circuit that you're testing.

Yes, I'm connecting the counter ground alligator clip to the ground of the
power supply (from which the circuit gets power and ground also).So signal
ground of the circuit is the same of the power supply.Is it wrong?

I really doubt that a low-pass filter is going to help you here, unless
the signal is not clean, that is, has a lot of noise riding on it.
How clean is the power supply that is powering the circuit under test? If
it has a lot of ripple, that could explain your erratic readings. Make
sure the Vcc rail is clean from ripple and noise.

Excuse me the ignorance but how can I verify this?I have no scilloscope...

--
David
dgminala at mediacombb dot net

 

[Dave M]

caius wrote:

"Dave M" <dgminala4444@mediacombb.net> ha scritto nel messaggio
news:wOOdnXELXJK6i6fQnZ2dnUVZ_u2dnZ2d@giganews.com...

From the spec sheet that you posted on your counter, the X20 button
is not a filter; it's an attenuator. The description says that
channel B has an attenuator (X1,X20), & AC/DC coupling for a
low-pass filter. It attenuates the signal from the B input
(selected by the function switch) by a factor of 20. According to
the table at the top of the page, in order to measure frequencies
below 100Hz, you should be using DC coupling. AC coupling is really
a high-pass filter, which, from the table, attenuates frequencies
below 100Hz. Which input are you using to measure your 60Hz sync signal?
If
you're tring to use the B input, of the A input with the Function
set to the 2-50MHz range, then it's no wonder that you're getting
erratic readings. It's a wonder that you're getting any reading at
all.


According to the table, for measuring 60Hz I'm using the Channel A
(from 0.01Hz to 50 MHz) with step 3 and DC coupling pressed

From the specs on the channel A input, you should be OK to measure a
high level signal (up to 30VP-P. The fact that your reading is
varying from 60-63 HZ, that would indicate that your signal is
mixing, or being overwhelmed by mains coupling. Try isolating the
counter's ground from the power line. That brings another
question... Where are you connecting the counter's signal ground? Are
you relying on the power line ground for the connection? If so,
that's a very bad idea. Use a scope probe or shielded cable, and
connect the probe's shield to the signal ground of the circuit that
you're testing.

Yes, I'm connecting the counter ground alligator clip to the ground
of the power supply (from which the circuit gets power and ground
also).So signal ground of the circuit is the same of the power
supply.Is it wrong?
I really doubt that a low-pass filter is going to help you here,
unless the signal is not clean, that is, has a lot of noise riding
on it. How clean is the power supply that is powering the circuit under
test? If it has a lot of ripple, that could explain your erratic
readings. Make sure the Vcc rail is clean from ripple and noise.

Excuse me the ignorance but how can I verify this?I have no
scilloscope...

Then you're shooting in the dark. In order to effectively troubleshoot this
type of problem, you need the right equipment. At this point, I'd suggest
that you Google for a PC sound card scope, There are a couple freebies on
the internet that you can download. Bear in mind that they're very basic,
uncalibrated, and low frequency (audio) response. An example is at
http://www.tech-systems-labs.com/osc251.zip.
Do you have any friends, neighbors or relatives that can help?
Your responses to other postings indicate that you're a novice in the world
of electronics. You really need to learn more of the basics in order to
understand the intracies of dealing with measuring and evaluating complex
signals using basic test equipment.
Google the net for tutorials on basics of circuit theory, using basic test
equipment, evaluating waveforms, etc.

--
David
dgminala at mediacombb dot net

 

[David Nebenzahl]

On 1/21/2011 3:30 PM caius spake thus:

"Phil Allison" <phil_a@tpg.com.au> ha scritto nel messaggio
news:8pug4kFs56U1@mid.individual.net...

plus keep the input level down to something the frequency counter
likes.

Did you mean the input impedance?According to the manual of my counter,
Channel A has 1Mohm of input impedance.How can measure it on the circuit?

Input level = input *voltage*. Forget about impedance (for the time being).

And beware Phil: he can be dangerous (or at least appear to be) if he's
off his meds and you somehow offend him ...


--
Comment on quaint Usenet customs, from Usenet:

To me, the *plonk...* reminds me of the old man at the public hearing
who stands to make his point, then removes his hearing aid as a sign
that he is not going to hear any rebuttals.

 

[caius]

"Phil Allison" <phil_a@tpg.com.au> ha scritto nel messaggio
news:8pug4kFs56U1@mid.individual.net...

** Errr - as you only want to measure 60Hz, why not have a LP filter at
say
100Hz ??

A 10kohm resistor and a 0.15uF film cap ought to work - .

I added a RC filter with a a 10kohm resistor and a 0.15uF film (cut-off
frequency at 106.2 Hz) but now I get no readings at all.

plus keep the input

level down to something the frequency counter likes

The sensivity of my counter is:

"AC" 100Hz-50MHz<120mVrms

"DC" 0.01Hz-1Hz?500mVp-p,

1Hz-100Hz?80mVrms.

How can I know the amplitude or strenght) of the signal I'm measuring anf if
it's out of the sensivity of the counter?I know only that the signal is at
+4.78 Volt



This is the manual:

http://www.emclr.co.uk/productdownloads/prod_45/VC3165%20User%20Manual.pdf