Confused about Frequency Counters

[Jeff Liebermann]

On Sat, 27 May 2017 19:24:00 -0400, Michael Black <et472@ncf.ca>
wrote:

I've certainly thought about buying one of those cheap ones, hoping I
could bypass the prescaler, though I suspect another issue, the prescaler
isn't a decade counter. Back when Heathkit came out with a frequency
counter, circa 1971, the prescalers were decade counters, and things got
better as they improved, and the frequency counters had higher limits.
But that sort of IC seems out of fashion now, so the prescalers are meant
for other things, and offer a binary division, so bypassing it in the
counter (and maybe adding an input stage) means the clock for the counter
is "wrong".
Michael

http://www.ebay.com/itm/Blue-RF-Signal-Frequency-Counter-Cymometer-Tester-0-1-60MHz-20MHz-2400MHZ/172396798620

To get from 2400 MHz to 60 MHz requires dividing by 40. However, that
doesn't seem to be how this one works.

Checking the prescaler chip from the photos,
<http://img.yunqudao.com/UploadFolder/4f2543ab-bfc8-48f0-9aff-c49e51612b75/Default/40_6.jpg>
I find a Fujitsu MB501L prescaler:
<http://pdf1.alldatasheet.com/datasheet-pdf/view/61659/FUJITSU/MB501L/+4155-UwSHHTTVdRhCtYT+/datasheet.pdf>
That's a 64/65 or 128/129 prescaler for a dual modulous synthesizer. I
would guess(tm) that the eBay counter would use divide by 64 to get:
2400 / 64 = 37.5 MHz
which is too low to utilize the full 60 MHz counter range.

However, there's a problem. The MB501L prescaler is only rated to
1.1GHz. How they claim 2.4GHz will remain a mystery. If I had this
counter (and a clean workbench), I would probe it to see what they're
really doing. I couldn't find a schematic.

There are some handheld counters that count to 60Mhz, and use a divide
by 40 prescaler, which results in the proper frequency ranges:
2400 / 40 = 60 Mhz
However, I can't find an example right now. The 40 is achieved with a
divide by 4 followed by divide by 10, which I guess qualifies as a
decade prescaler of sorts. As I recall, it was usually done with 2
ECL chips, which sucked plenty of power and were not cheap.

Here's another way to use a prescaler:
<http://www.startek-usa.com/FREQ%20CTRS.htm>
The frequency ranges are:
50MHz, 800MHz, and 2800MHz
which correspond to:
/1 /16 and possibly /64
Sorry, but no schematic or block diagram found.



--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Sun, 28 May 2017 16:13:52 +1000, Clifford Heath
<no.spam@please.net> wrote:

On 28/05/17 14:25, Jeff Liebermann wrote:
On Sun, 28 May 2017 08:09:04 +1000, Clifford Heath
no.spam@please.net> wrote:
On 28/05/17 04:35, Jeff Liebermann wrote:
On Sat, 27 May 2017 12:55:02 -0400, oldschool@tubes.com wrote:
Look for something that has a built in prescaler. Something like
this:
http://www.ebay.com/itm/Blue-RF-Signal-Frequency-Counter-Cymometer-Tester-0-1-60MHz-20MHz-2400MHZ/172396798620
I have two of those. I bought the second because I thought I'd
broken the first, it performed so badly. They're as bad as each
other.
Thanks. I was thinking of buying some of those. I should have known
as much of the low cost "modules" that I've purchased seem to have
deficiencies as a result of crude design or cost cutting exercises.

It was developed by a good hobbyist who posted everything online.

I couldn't find any such project. I suspect that it might have gone
the same way as the various M328 component test meters being sold
online. The original project was open source. It was then
commercialized by various vendors with wildly varying pricing. Much
of the stuff I've found was early versions of the board and firmware.
Meanwhile, the project has done on to add features and improve the
firmware, but the online stuff seems stuck with early revisions. This
link includes some history:
<http://www.instructables.com/id/AVR-Transistor-Tester/>
I can't seem to find the original development site, which was in
Germany.

I think that full schematics of slightly earlier versions are
available online. He uses a dual-gate MOSFET before the prescaler
and before the main counter, with the inputs paralleled. I think
that affects the sensitivity (though I don't have measurements)
so for my 2nd module, I cut a track to separate the input paths.

If the amplifier is used to simply produce a square wave out of
whatever it fed into the input, low gain might be a big problem. So
will noise around 0v which is why a "threshold" adjustment is usually
supplied. Getting such a simple amplifier to work from 0.1MHz to
2.4GHz is unlikely, which might explain the lack of sensitivity.

I might wind up adding an independent input amplifier with AGC,
or even a pot to adjust the 2nd gate bias on the MOSFETs for a
manual gain control. A little difficult though, as parts of the
circuit are underneath the LED displays, so I'd need to remove
those.

May I suggest that you remove the input amp and setup something that
give the prescaler a 50 ohm input. Then, design a broadband RF
amplifier that has a chance of working over the frequency range.
Something similar to a CATV or OTA TV/FM amplifier might be suitable.
However, don't worry about getting a flat frequency response. Just
take whatever you can get that produces enough drive to make the
MB501L prescaler happy. A collection of communications freq range
bandpass filters would be nice to prevent triggering on out of band
junk.

>The main counter is a PIC.

I'm not PICky.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

On Sun, 28 May 2017 09:57:35 -0400, Ralph Mowery
<rmowery28146@earthlink.net> wrote:

YOu really need to buy one of the used service monitors. They can be
had for around $ 1000. You get a counter signal generator and many more
things.

I'll pass.....
I'd first have to rob a bank anyhow, in order to pay for it, and I think
the bank clerks would just laugh when I pointed a soldering gun at
them... (Or maybe a glue gun, or a caulking gun)

 

[Jeff Liebermann]

On Sun, 28 May 2017 09:12:06 -0500, Foxs Mercantile <jdangus@att.net>
wrote:

On 5/28/2017 8:57 AM, Ralph Mowery wrote:

You really need to buy one of the used service monitors.

Like this for example:
http://www.ebay.com/itm/272688535932

Overpriced. Look at the prices of the sold listings:
<http://www.ebay.com/sch/i.html?_nkw=Motorola%20R2001&LH_Complete=1&LH_Sold=1>

I have two R2001D boxes parked in my palatial office waiting for the
mythical "spare time" needed to fix them. One has most of the red
LED's on the right bashed in by someone dropping something heavy on
the front panel. The other has a very weak CRT display, which is
probably an HV power supply problem. Both have lock problems, which
means it's time for a tantalum transplant. They've been sitting there
for about 3 years. If I wait long enough, maybe the owner will forget
I have them.

I have a few other service monitors. There are three SSI/Wavetek 3000
series service monitors in this photo, plus one more I recently
acquired. Typical cost was $300/ea:
<http://802.11junk.com/jeffl/pics/home/slides/BL-shop5.html>
In the middle left, is an IFR-1500 service monitor (with an
intermittent power supply). I paid $1500:
<http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html>
There are several cell phone specific service monitors hidden in
various corners.

Yep, service monitors are a good thing to have an use, especially in
the field or on mountain top radio site. It has everything that you
might need to work on radios including a counter. All have TCXO or
OCXO reference oscillators for accuracy. At home, I have a home made
GPSDO for even more accuracy.

However, there's a catch. All the stuff in the photos is from the
1980's which means that components are starting to fail. It's a
continuous battle to keep these things running and usable. If you
decide to invest in an older service monitor, be prepared to
occasionally dive in and do some repairs.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

On Sun, 28 May 2017 10:19:43 -0500, "Dave M" <dgminala@mediacombb.net>
wrote:

I probably got more use from the Freq counter I used in the 70s (which
was borrowed). Back then I was doing a lot with CB radios and that
counter would check the CB channel output for accuracy. But I dont do
much with CBs anymore, since no one uses them now.

Can't blame you a bit for not jumping on those "kits" mentioned previously.
Often more trouble than they're worth when you consider all the other stuff
you have to buy and then all the work to assemble and make work.
The C&C 150 seems like a pretty good deal for you. Certainly in your price
range, and appears to be a decent entry level counter. You can get a manual
from the manufacturer' web site
(http://www.cncinst.co.kr/english/bbs/board.php?bo_table=data_eng_manual&wr_id=13);
(Registration required, but nothing out of reason, like credit card numbers,
etc.)

You'll probably find out, if you research "reciprocal counters" (of which,
this is one) that they offer much better resolution than other "normal"
counters, especially at low audio frequencies. This is a good thing, since
you can select a shorter gate time for the measurement than normal counters.
If you want to measure an audio tone of, say 123.4 Hz, you'd need to select
a gate time of 10 seconds to get the last digit to display. With a
reciprocal counter, you can select a gate time of 1 second, or even 0.1
second, and see all the digits the counter can display. It actually
measures the period of a signal, and a microcomputer inside the counter does
a bit of math to calculate and display the frequency with all the digits the
counter is capable of displaying.

Good luck with your choice,
Dave M

I am now the owner of a C&C 150. For the price, and considering the
specifications, I just took a chance in the dark (literally), since I
drove to a local WIFI at midnight and from my car, I bought it. I had a
feeling it would be sold if I waited until today.

I did find one of those discussion groups on the web, in which they were
discussing it. Some guy bought one (on ebay) for around the same price I
paid, and he was pleased with it, but did not know how to use a lot of
the features and controls. Someone in that discussion posted a URL for
the manual. (Not the same one you posted).
This one: http://clayphillipsracecars.com/other/150-166.pdf
(I bookmarked the URL).

This is the manual that's upside down, which I posted about....

(Apparently, others like this have been sold on Ebay recently). The
discussion occured about a month ago). (There actually was another one
being sold on Ebay, but for about $20 more, with shipping).

All I got to do now, is wait for it to arrive....
Till then, I'll read the manual.
(and wont have to stand on my head to do it).

Thanks for the help!

 

[Jeff Liebermann]

On Sun, 28 May 2017 13:43:26 -0400, oldschool@tubes.com wrote:

http://clayphillipsracecars.com/other/150-166.pdf
This is the manual that's upside down, which I posted about....

Congrats on getting the counter. I downloaded the manual and found
that *ALL* of the pages were inverted. No problem. Fire up
PDF-Xchange and click on the "Rotate CW" button at the top of the
screen. That should fix all the pages. Then do a:
File -> Save As
to replace the original with something readable.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Clifford Heath]

On 29/05/17 02:43, Jeff Liebermann wrote:

On Sun, 28 May 2017 16:13:52 +1000, Clifford Heath
no.spam@please.net> wrote:

On 28/05/17 14:25, Jeff Liebermann wrote:
On Sun, 28 May 2017 08:09:04 +1000, Clifford Heath
no.spam@please.net> wrote:
On 28/05/17 04:35, Jeff Liebermann wrote:
On Sat, 27 May 2017 12:55:02 -0400, oldschool@tubes.com wrote:
Look for something that has a built in prescaler. Something like
this:
http://www.ebay.com/itm/Blue-RF-Signal-Frequency-Counter-Cymometer-Tester-0-1-60MHz-20MHz-2400MHZ/172396798620
I have two of those. I bought the second because I thought I'd
broken the first, it performed so badly. They're as bad as each
other.
Thanks. I was thinking of buying some of those. I should have known
as much of the low cost "modules" that I've purchased seem to have
deficiencies as a result of crude design or cost cutting exercises.

It was developed by a good hobbyist who posted everything online.

I couldn't find any such project. I suspect that it might have gone
the same way as the various M328 component test meters being sold
online. The original project was open source. It was then
commercialized by various vendors with wildly varying pricing.

Yes, that seems to describe it.

Google for Sanjian Studio (which is on the PCB) and you'll find
an English translation of the manual. That had assorted URLs to
www.hellocq.net where the many stages of this project's development
was discussed. You have to log in to see the schematics. I subscribed,
but my account seems to have now expired (I didn't get the spam that
I expected from this subscription). So I only have the assorted files
I downloaded, see here:
<https://www.dropbox.com/sh/z91pqvcpg470tuy/AAAIRnVfRDqrbrJwA3yBVQVPa?dl=0>

I think that full schematics of slightly earlier versions are
available online. He uses a dual-gate MOSFET before the prescaler
and before the main counter, with the inputs paralleled. I think
that affects the sensitivity (though I don't have measurements)
so for my 2nd module, I cut a track to separate the input paths.

If the amplifier is used to simply produce a square wave out of
whatever it fed into the input, low gain might be a big problem. So
will noise around 0v which is why a "threshold" adjustment is usually
supplied.

Yes. High gain is a problem too, causing spurious transitions.

Something similar to a CATV or OTA TV/FM amplifier might be suitable.
However, don't worry about getting a flat frequency response. Just
take whatever you can get that produces enough drive to make the
MB501L prescaler happy.

Yes, except it's meant to be (and mine are) and MB506.

A collection of communications freq range
bandpass filters would be nice to prevent triggering on out of band
junk.

Good idea.

>> The main counter is a PIC.

Earlier schematics show two 4-bit counters before the PIC,
which by itself does not have a 60MHz counter.

Clifford Heath.

 

[Clifford Heath]

On 29/05/17 02:16, Jeff Liebermann wrote:

On Sat, 27 May 2017 19:24:00 -0400, Michael Black <et472@ncf.ca
wrote:

I've certainly thought about buying one of those cheap ones, hoping I
could bypass the prescaler, though I suspect another issue, the prescaler
isn't a decade counter. Back when Heathkit came out with a frequency
counter, circa 1971, the prescalers were decade counters, and things got
better as they improved, and the frequency counters had higher limits.
But that sort of IC seems out of fashion now, so the prescalers are meant
for other things, and offer a binary division, so bypassing it in the
counter (and maybe adding an input stage) means the clock for the counter
is "wrong".
Michael

http://www.ebay.com/itm/Blue-RF-Signal-Frequency-Counter-Cymometer-Tester-0-1-60MHz-20MHz-2400MHZ/172396798620

To get from 2400 MHz to 60 MHz requires dividing by 40. However, that
doesn't seem to be how this one works.

Checking the prescaler chip from the photos,
http://img.yunqudao.com/UploadFolder/4f2543ab-bfc8-48f0-9aff-c49e51612b75/Default/40_6.jpg
I find a Fujitsu MB501L prescaler:
http://pdf1.alldatasheet.com/datasheet-pdf/view/61659/FUJITSU/MB501L/+4155-UwSHHTTVdRhCtYT+/datasheet.pdf
That's a 64/65 or 128/129 prescaler for a dual modulous synthesizer. I
would guess(tm) that the eBay counter would use divide by 64 to get:
2400 / 64 = 37.5 MHz
which is too low to utilize the full 60 MHz counter range.

However, there's a problem. The MB501L prescaler is only rated to
1.1GHz. How they claim 2.4GHz will remain a mystery.

I assume the MB501 was 2c cheaper than the MB506 that the project
was designed with. Find one of the many versions that actually use
the MB506 instead.

 

[rickman]

oldschool@tubes.com wrote on 5/28/2017 1:43 PM:

On Sun, 28 May 2017 10:19:43 -0500, "Dave M" <dgminala@mediacombb.net
wrote:

I probably got more use from the Freq counter I used in the 70s (which
was borrowed). Back then I was doing a lot with CB radios and that
counter would check the CB channel output for accuracy. But I dont do
much with CBs anymore, since no one uses them now.

Can't blame you a bit for not jumping on those "kits" mentioned previously.
Often more trouble than they're worth when you consider all the other stuff
you have to buy and then all the work to assemble and make work.
The C&C 150 seems like a pretty good deal for you. Certainly in your price
range, and appears to be a decent entry level counter. You can get a manual
from the manufacturer' web site
(http://www.cncinst.co.kr/english/bbs/board.php?bo_table=data_eng_manual&wr_id=13);
(Registration required, but nothing out of reason, like credit card numbers,
etc.)

You'll probably find out, if you research "reciprocal counters" (of which,
this is one) that they offer much better resolution than other "normal"
counters, especially at low audio frequencies. This is a good thing, since
you can select a shorter gate time for the measurement than normal counters.
If you want to measure an audio tone of, say 123.4 Hz, you'd need to select
a gate time of 10 seconds to get the last digit to display. With a
reciprocal counter, you can select a gate time of 1 second, or even 0.1
second, and see all the digits the counter can display. It actually
measures the period of a signal, and a microcomputer inside the counter does
a bit of math to calculate and display the frequency with all the digits the
counter is capable of displaying.

Good luck with your choice,
Dave M


I am now the owner of a C&C 150. For the price, and considering the
specifications, I just took a chance in the dark (literally), since I
drove to a local WIFI at midnight and from my car, I bought it. I had a
feeling it would be sold if I waited until today.

I did find one of those discussion groups on the web, in which they were
discussing it. Some guy bought one (on ebay) for around the same price I
paid, and he was pleased with it, but did not know how to use a lot of
the features and controls. Someone in that discussion posted a URL for
the manual. (Not the same one you posted).
This one: http://clayphillipsracecars.com/other/150-166.pdf
(I bookmarked the URL).

This is the manual that's upside down, which I posted about....

Just curious, is the manual a series of images of pages or is the text
selectable?

I received a PDF document of an old PDP-11 listing from someone who wanted
help typing it in. I realized when I clicked my cursor over the text it
would select even though it was clearly created from images. Seems some
software in the path (possibly my reader) was doing optical character
recognition on the document. Most of it came through ok, but once in a
while the slightly out of adjustment printer characters would be misread
like a 9 for a 0, or a 0 for an O. Still, it saved a lot of time.

Anyone else see scanned documents showing selectable text?

--

Rick C

 

[Jeff Liebermann]

On Mon, 29 May 2017 08:45:50 +1000, Clifford Heath
<no.spam@please.net> wrote:

I downloaded, see here:
https://www.dropbox.com/sh/z91pqvcpg470tuy/AAAIRnVfRDqrbrJwA3yBVQVPa?dl=0

Got it and thanks. Nice the way the download is packaged as a single
ZIP file. The docs are in Chinese. I haven't tried Google translate
on it yet. The new schematic is very difficult to read the text. I'm
also having problems decoding the "forum description". It's readable
imported into MS Word as Unicode-8. The photos of the PCB seem to be
the old design, which lack the extra divider chips. This is going to
be a challenge.

Something similar to a CATV or OTA TV/FM amplifier might be suitable.
However, don't worry about getting a flat frequency response. Just
take whatever you can get that produces enough drive to make the
MB501L prescaler happy.

Yes, except it's meant to be (and mine are) and MB506.

It also says MB506 on both the old and new schematics.
<http://www.qsl.net/n9zia/900/mb506.pdf>
That's odd because it's a divide by 128 or 256 that goes up to 1.6GHz.
At 2400MHz, divide by 128 yields 18.75MHz. I originally thought that
it would need to use all of the 60MHz counter frequency range, but now
I'm not certain. The PIC used is apparently slow, and won't go that
fast. 18.75MHz seems about right for the PIC16F628a where the data
sheet says it quits at 20MHz.
<http://ww1.microchip.com/downloads/en/devicedoc/40044f.pdf>
The ForumDescription.txt file says that it's divide by 64 on the first
page, but then claims that the low channel goes to 75MHz, and later
claims that the prescaler is an MB501. Kinda looks like the
ForumDescription.txt file is a mixture of the old and new designs.

Looking at the new schematic, I see that one bipolar front end
transistor was replaced by a dual gate mosfet. However, it doesn't
look like the gates were tied together on the schematic. The original
schematic is tiny, but after enlargement, I don't see a dot where the
wires cross.

Also, if you follow the signal path through the prescaler chip on both
the old and new schematics, the higher frequency range input goes
through the prescaler, into the DG MOSFET, and then to the PIC
counter. If there is a sensitivity problem, it would only be on the
lower frequency range input, which goes to the DG MOSFET directly.

Earlier schematics show two 4-bit counters before the PIC,
which by itself does not have a 60MHz counter.

Probably correct, but I can't tell what those are. Also, I think you
have it backwards. The older schematic shows no dividers, while the
new schematic shows what I guess are dividers.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Mon, 29 May 2017 00:08:29 -0400, rickman <gnuarm@gmail.com> wrote:

I received a PDF document of an old PDP-11 listing from someone who wanted
help typing it in. I realized when I clicked my cursor over the text it
would select even though it was clearly created from images. Seems some
software in the path (possibly my reader) was doing optical character
recognition on the document. Most of it came through ok, but once in a
while the slightly out of adjustment printer characters would be misread
like a 9 for a 0, or a 0 for an O. Still, it saved a lot of time.

Anyone else see scanned documents showing selectable text?

Searchable text is a standard PDF feature, even with bitmapped text.
PDF-Xchange has built in OCR (optical character recognition) that will
read through the graphical text, do its best to convert it to ASCII
text, and save the combined file. After that, you can use the search,
select, edit, functions:
<https://www.youtube.com/watch?v=CWtHOsIKaKw>
<https://www.tracker-software.com/knowledgebase/351-How-do-I-OCR-a-document>
The free version will do all that except edit and save the resulting
text. For that, you need the registered version.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Sun, 28 May 2017 21:22:33 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

On Mon, 29 May 2017 08:45:50 +1000, Clifford Heath
no.spam@please.net> wrote:

I downloaded, see here:
https://www.dropbox.com/sh/z91pqvcpg470tuy/AAAIRnVfRDqrbrJwA3yBVQVPa?dl=0

I missed the translated user manual (in English). It shows a
sensitivity graph for the high frequency range with the following
comment from Pg 12:
It is noted that the UHF channel allows measurements up to about
450 MHz. This path comprises a divide by 64 stage claimed to be
able to operate to 2.4GHz according to the published specifications.
It is therefore surprising that the sensitivity fell as quickly
as it did.
Looking at the schematic and layout, my never humble opinion is that
the designer didn't know anything about RF design and layout.

The schematic shown on Pg 13 is quite different from either the old or
new versions of the design that I previously mentioned. Instead of
the extra divide by 4 packages, it has a 2nd DG MOSFET in front of the
PIC counter. It also lists the prescaler as an MB506 which is divide
by 128/256, not 64. It also shows that the gates of the DG MOSFET are
NOT tied together.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[rickman]

Jeff Liebermann wrote on 5/29/2017 12:36 AM:

On Mon, 29 May 2017 00:08:29 -0400, rickman <gnuarm@gmail.com> wrote:


I received a PDF document of an old PDP-11 listing from someone who wanted
help typing it in. I realized when I clicked my cursor over the text it
would select even though it was clearly created from images. Seems some
software in the path (possibly my reader) was doing optical character
recognition on the document. Most of it came through ok, but once in a
while the slightly out of adjustment printer characters would be misread
like a 9 for a 0, or a 0 for an O. Still, it saved a lot of time.

Anyone else see scanned documents showing selectable text?

Searchable text is a standard PDF feature, even with bitmapped text.
PDF-Xchange has built in OCR (optical character recognition) that will
read through the graphical text, do its best to convert it to ASCII
text, and save the combined file. After that, you can use the search,
select, edit, functions:
https://www.youtube.com/watch?v=CWtHOsIKaKw
https://www.tracker-software.com/knowledgebase/351-How-do-I-OCR-a-document
The free version will do all that except edit and save the resulting
text. For that, you need the registered version.

I'm not sure what "standard" means. I was viewing a document full of imaged
text the other day and none of the permissions were set to preclude
anything. Yet I couldn't select any text as it had not been OCR'd.

I assume the OCR has to be done at capture time. Are you saying a reader
will convert images to text?

--

Rick C

 

[Clifford Heath]

On 29/05/17 14:51, Jeff Liebermann wrote:

On Sun, 28 May 2017 21:22:33 -0700, Jeff Liebermann <jeffl@cruzio.com
wrote:
On Mon, 29 May 2017 08:45:50 +1000, Clifford Heath
no.spam@please.net> wrote:
I downloaded, see here:
https://www.dropbox.com/sh/z91pqvcpg470tuy/AAAIRnVfRDqrbrJwA3yBVQVPa?dl=0

I missed the translated user manual (in English). It shows a
sensitivity graph for the high frequency range with the following
comment from Pg 12:
It is noted that the UHF channel allows measurements up to about
450 MHz. This path comprises a divide by 64 stage claimed to be
able to operate to 2.4GHz according to the published specifications.
It is therefore surprising that the sensitivity fell as quickly
as it did.
Looking at the schematic and layout, my never humble opinion is that
the designer didn't know anything about RF design and layout.

That was my conclusion also, and that (plus the earlier published
versions) is why I said "advanced hobbyist". Not even very advanced,
certainly not RF-experienced

I don't know what frequency the PIC counter input is capable of,
but I know that the AVR counter is clocked; so you can only count
at half the CPU clock frequency. Bah, humbug.

The schematic shown on Pg 13 is quite different from either the old or
new versions of the design that I previously mentioned. Instead of
the extra divide by 4 packages, it has a 2nd DG MOSFET in front of the
PIC counter. It also lists the prescaler as an MB506 which is divide
by 128/256, not 64.

Both my units have MB506.

It also shows that the gates of the DG MOSFET are
NOT tied together.

The units I have have both pairs of protection diodes, and the inputs
are joined only at the connector. I cut the trace and soldered a bit
of co-ax onto the prescaler input capacitor.

I don't have a good RF source (yet - currently building, see
https://github.com/cjheath/AD9851LCD) so I can't evaluate the
sensitivity.

Clifford Heath.

 

[Jeff Liebermann]

On Mon, 29 May 2017 17:29:44 +1000, Clifford Heath
<no.spam@please.net> wrote:

I don't know what frequency the PIC counter input is capable of,
but I know that the AVR counter is clocked; so you can only count
at half the CPU clock frequency. Bah, humbug.

The clock crystal is 4MHz on the old version. I can't read the
numbers on the schematic of the new version. That doesn't look very
promising for measuring 60MHz inputs or even with /4 at 15MHZ.

It also shows that the gates of the DG MOSFET are
NOT tied together.

The units I have have both pairs of protection diodes, and the inputs
are joined only at the connector. I cut the trace and soldered a bit
of co-ax onto the prescaler input capacitor.

Oh swell. So the PCB wiring might not follow the schematic. I
suppose it doesn't matter since the DG MOSFET seems to be badly biased
anyway.

I was having nightmares last night from thinking about this counter.
Maybe I should give up while I'm still sane?

I don't have a good RF source (yet - currently building, see
https://github.com/cjheath/AD9851LCD) so I can't evaluate the
sensitivity.

Before you reinvent the wheel, there are AD9851 based DDS generators
available on eBay.
<http://www.ebay.com/sch/i.html?_nkw=dds+generator+ad9851>
along with the associated LCD display:
<http://www.ebay.com/itm/PIC16f-Controller-for-the-AD9851-DDS-Signal-Generator-Module-/182593721953>
However, those only go up to about 70MHz and the output looks
distorted above 30MHz. If you're going to test the counter all the
way to its rated maximum frequency (2.4GHz), you're going to need a
better generator. DDS has benefits for a function generator and
arbitrary waveform generator, but is limited to lower frequencies.

This looks interesting (and tempting):
<http://www.ebay.com/itm/ADF4350-v4-0-137-5MHZ-4-4GHZ-OLED-display-Signal-generator-RF-signal-source-12v-/262688224985>
137.5MHz to 4.4GHz signal generator in 10KHz steps. Looks ok to about
1GHz, but drops in output and increases in sidebands at higher
frequencies. Looks like the same board, but in a shielded box:
<http://www.ebay.com/itm/137-5MHZ-to-4400MHZ-Signal-generator-frequency-generator-RF-signal-source-dc-12v-/271838837908>
Or maybe this thing:
<http://www.ebay.com/itm/0-5Mhz-470Mhz-RF-Signal-Generator-Meter-Tester-For-FM-Radio-walkie-talkie-debug-/172598060649>
Or maybe something computah controlled via USB:
<http://www.ebay.com/itm/RF-Signal-Generator-35MHz-to-4-4GHz-via-USB-16dBm-Plus-Features-2000-units-sold-/201929990411>
Or maybe a real RF generator from HP, TEK, Fluke or others that can
actually be calibrated and trusted. This is the cheapest HP I could
find:
<http://www.ebay.com/itm/Hewlett-Packard-hp-8656A-Signal-Generator-1-990MHz-rf-signal-generator-04-/252950700229>
I have an HP 8656A but prefer to use an HP 8540B. Top right:
<http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html>








--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Mon, 29 May 2017 02:04:11 -0400, rickman <gnuarm@gmail.com> wrote:

Jeff Liebermann wrote on 5/29/2017 12:36 AM:
On Mon, 29 May 2017 00:08:29 -0400, rickman <gnuarm@gmail.com> wrote:


I received a PDF document of an old PDP-11 listing from someone who wanted
help typing it in. I realized when I clicked my cursor over the text it
would select even though it was clearly created from images. Seems some
software in the path (possibly my reader) was doing optical character
recognition on the document. Most of it came through ok, but once in a
while the slightly out of adjustment printer characters would be misread
like a 9 for a 0, or a 0 for an O. Still, it saved a lot of time.

Anyone else see scanned documents showing selectable text?

Searchable text is a standard PDF feature, even with bitmapped text.
PDF-Xchange has built in OCR (optical character recognition) that will
read through the graphical text, do its best to convert it to ASCII
text, and save the combined file. After that, you can use the search,
select, edit, functions:
https://www.youtube.com/watch?v=CWtHOsIKaKw
https://www.tracker-software.com/knowledgebase/351-How-do-I-OCR-a-document
The free version will do all that except edit and save the resulting
text. For that, you need the registered version.

I'm not sure what "standard" means.

Bad choice of words. I meant that the PDF standard:
<https://en.wikipedia.org/wiki/PDF/A>
includes searchable text as part of the standard. I'm too lazy to
look up the chapter and verse.

I was viewing a document full of imaged
text the other day and none of the permissions were set to preclude
anything. Yet I couldn't select any text as it had not been OCR'd.

Yep. If you scan text as a bit map image, and save it in PDF format,
it cannot be text searched. You have to feed it to an OCR program,
which is capable of attaching the OCR text to the PDF, save it, and
then you can search.

>I assume the OCR has to be done at capture time.

No. It can be done at any time with any reasonable document. I
usually make some effort to realign the text and improve the contrast
to make it easier (and faster) for the OCR program to do it's thing.

>Are you saying a reader will convert images to text?

If the images look like readable ASCII characters, yes. I don't think
size makes much difference, but I haven't done much experimentation
into how badly I can butcher the text and the OCR will still work. I
also haven't tried to edit the text after reading to correct OCR
errors.

Maybe a demo will help. Note that the initial scan and file saves
were done in Irfanview, while the OCR and subsequent saves were done
in PDF-Xchange:

Original document scanned to JPG using Irfanview 4.44:
<http://802.11junk.com/jeffl/OCR%20Demo/JPG.jpg>
This is not searchable.

Same document saved to PDF using Irfanview 4.44:
<http://802.11junk.com/jeffl/OCR%20Demo/PDF-no-OCR.pdf>
This is also NOT searchable.

Same document in PDF-Xchange 6.0 build 322.4 after OCR:
<http://802.11junk.com/jeffl/OCR%20Demo/PDF-after-OCR.pdf>
This one can be searched.

PDF-Xchange screen grab showing a typical search result:
<http://802.11junk.com/jeffl/OCR%20Demo/PDF-Xchange-screen.jpg>

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Mon, 29 May 2017 11:00:12 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

I also haven't tried to edit the text after reading to correct OCR
errors.

Here's how to edit OCR errors using Adobe Acrobat:
<http://blogs.adobe.com/acrolaw/?s=ocr+and+image+layer>
I'm still trying to figure it out using PDF-Xchange Editor.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Clifford Heath]

On 30/05/17 03:16, Jeff Liebermann wrote:

On Mon, 29 May 2017 17:29:44 +1000, Clifford Heath
no.spam@please.net> wrote:
I don't know what frequency the PIC counter input is capable of,
but I know that the AVR counter is clocked; so you can only count
at half the CPU clock frequency. Bah, humbug.

The clock crystal is 4MHz on the old version. I can't read the
numbers on the schematic of the new version.

But I don't think the PIC has a clocked counter anyhow.
I assume that the AVR does it to (sometimes) avoid the
need for a low-pass filter.

The xtal on mine is marked "SCK451C" and "TC,A.426",
whatever that means. It was about 15ppm slow, but seemed
quite stable, based on measurements taken with an HP5386A.

Quite a few people have patched in a TCXO to these units.

It also shows that the gates of the DG MOSFET are
NOT tied together.

I don't understand why they used a DG MOSFET, nor why, since they
did use one, they didn't use the upper gate for gain control.
It seems they're feeding the signal into the upper gate, so
won't get the best bandwidth from the cascode behaviour.

The units I have have both pairs of protection diodes, and the inputs
are joined only at the connector. I cut the trace and soldered a bit
of co-ax onto the prescaler input capacitor.

Oh swell. So the PCB wiring might not follow the schematic.

I haven't found the schematic of the current-manufacture.

I was having nightmares last night from thinking about this counter.
Maybe I should give up while I'm still sane?

I think it's fixable, perhaps with an additional front-end.
It would still be easier and cheaper than building from scratch.

I don't have a good RF source (yet - currently building, see
https://github.com/cjheath/AD9851LCD) so I can't evaluate the
sensitivity.

Before you reinvent the wheel, there are AD9851 based DDS generators
available on eBay.

Who do you think designed those? People like me I have a bit
of that Jedi "build your own light sabre" thing going on.

Plus there's no accessible used test equipment market here in
Australia. Whenever nice gear comes up at bargain prices,
merchants buy it up and slap a stupid price on it.

> <http://www.ebay.com/sch/i.html?_nkw=dds+generator+ad9851>

That's exactly what I'm using for development. They have all
copied a flaw in the output filter design, leading to very low
output at higher frequencies. Some impedance problem, it's not
designed to drive 50ohms. I'll add a buffer.

along with the associated LCD display:
http://www.ebay.com/itm/PIC16f-Controller-for-the-AD9851-DDS-Signal-Generator-Module-/182593721953

I loathe and detest both PICs and those 16x2 displays.
I'm building one with 320x240 colour touch screen.

The Arduino also has TTL-level RS232, so add a $2 USB module
and you have USB control.

However, those only go up to about 70MHz and the output looks
distorted above 30MHz. If you're going to test the counter all the
way to its rated maximum frequency (2.4GHz), you're going to need a
better generator. DDS has benefits for a function generator and
arbitrary waveform generator, but is limited to lower frequencies.

I expect to incorporate an ADF4351 also, and possibly two AD9851's,
to give quadrature (but still cheaper than AD9854 or whatever
the multi-channel DDS chip is).

E.g.
<https://www.aliexpress.com/item/35Mhz-to-4-4GHz-4400mhz-PLL-RF-Signal-Source-Frequency-Synthesizer-ADF4351-Development-Board/32757566484.html>

The ADF351's have the same problem as most of those VCO synthesisers,
that they won't sweep cleanly. Changing the frequency makes them
jump wildly about until they stabilise again.

This looks interesting (and tempting):
http://www.ebay.com/itm/ADF4350-v4-0-137-5MHZ-4-4GHZ-OLED-display-Signal-generator-RF-signal-source-12v-/262688224985
137.5MHz to 4.4GHz signal generator in 10KHz steps.

The Arduino clone and TFT Touchscreen LCD cost me $AU14 all up.
<https://www.aliexpress.com/item/2-4-SPI-Serial-TFT-LCD-Touch-Panel-240x320-Dots-5V-3-3V-Module-ILI9341-Driver/32665656357.html>
Add the $30 ADF4351, a $17 $AD9851, and USB and you have a nice bundle
for half what the above costs.

Or maybe a real RF generator from HP, TEK, Fluke or others that can
actually be calibrated and trusted. This is the cheapest HP I could
find:
http://www.ebay.com/itm/Hewlett-Packard-hp-8656A-Signal-Generator-1-990MHz-rf-signal-generator-04-/252950700229

"Does not ship to Australia"

I have an HP 8656A but prefer to use an HP 8540B. Top right:
http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html

I have to give this HP5386A back, but not in a hurry - my friend also
has mountains of test equipment. He worked in sat-comms, so has contacts
who call him before dealers get there - but he loves to hoard it all

Clifford Heath.

 

[Clifford Heath]

On 29/05/17 03:43, oldschool@tubes.com wrote:

This one: http://clayphillipsracecars.com/other/150-166.pdf
This is the manual that's upside down, which I posted about....
....
Till then, I'll read the manual.
(and wont have to stand on my head to do it).

Other people have also given you a method, but I thought it was worth
mentioning that the standard OSX (Mac) Preview app does PDF natively,
and it's trivial to invert, shuffle, delete, etc. Just show the page
thumbnails on the left, click on one, type Command-A to select all
pages, and hit Command-L twice to rotate. Command-S saves the rotated
file.

You can also select and delete individual pages, re-order pages, and
even drag pages in from another document. It's nice, and it's standard.

Clifford Heath.

 

[rickman]

Jeff Liebermann wrote on 5/29/2017 2:00 PM:

On Mon, 29 May 2017 02:04:11 -0400, rickman <gnuarm@gmail.com> wrote:

Jeff Liebermann wrote on 5/29/2017 12:36 AM:
On Mon, 29 May 2017 00:08:29 -0400, rickman <gnuarm@gmail.com> wrote:


I received a PDF document of an old PDP-11 listing from someone who wanted
help typing it in. I realized when I clicked my cursor over the text it
would select even though it was clearly created from images. Seems some
software in the path (possibly my reader) was doing optical character
recognition on the document. Most of it came through ok, but once in a
while the slightly out of adjustment printer characters would be misread
like a 9 for a 0, or a 0 for an O. Still, it saved a lot of time.

Anyone else see scanned documents showing selectable text?

Searchable text is a standard PDF feature, even with bitmapped text.
PDF-Xchange has built in OCR (optical character recognition) that will
read through the graphical text, do its best to convert it to ASCII
text, and save the combined file. After that, you can use the search,
select, edit, functions:
https://www.youtube.com/watch?v=CWtHOsIKaKw
https://www.tracker-software.com/knowledgebase/351-How-do-I-OCR-a-document
The free version will do all that except edit and save the resulting
text. For that, you need the registered version.

I'm not sure what "standard" means.

Bad choice of words. I meant that the PDF standard:
https://en.wikipedia.org/wiki/PDF/A
includes searchable text as part of the standard. I'm too lazy to
look up the chapter and verse.

I was viewing a document full of imaged
text the other day and none of the permissions were set to preclude
anything. Yet I couldn't select any text as it had not been OCR'd.

Yep. If you scan text as a bit map image, and save it in PDF format,
it cannot be text searched. You have to feed it to an OCR program,
which is capable of attaching the OCR text to the PDF, save it, and
then you can search.

I assume the OCR has to be done at capture time.

No. It can be done at any time with any reasonable document. I
usually make some effort to realign the text and improve the contrast
to make it easier (and faster) for the OCR program to do it's thing.

It is so easy to be misunderstood. I'm talking about the text showing up in
the PDF document. I receive d a document that was clearly a scanned image
in a PDF file. But the text was selectable and copyable. The two options
are the image was scanned and OCR when the PDF was made, or the PDF viewer
had OCR scanning built in. Since I couldn't select the text in another
scanned image PDF it must be the former.

Are you saying a reader will convert images to text?

If the images look like readable ASCII characters, yes. I don't think
size makes much difference, but I haven't done much experimentation
into how badly I can butcher the text and the OCR will still work. I
also haven't tried to edit the text after reading to correct OCR
errors.

Maybe a demo will help. Note that the initial scan and file saves
were done in Irfanview, while the OCR and subsequent saves were done
in PDF-Xchange:

Original document scanned to JPG using Irfanview 4.44:
http://802.11junk.com/jeffl/OCR%20Demo/JPG.jpg
This is not searchable.

Same document saved to PDF using Irfanview 4.44:
http://802.11junk.com/jeffl/OCR%20Demo/PDF-no-OCR.pdf
This is also NOT searchable.

Same document in PDF-Xchange 6.0 build 322.4 after OCR:
http://802.11junk.com/jeffl/OCR%20Demo/PDF-after-OCR.pdf
This one can be searched.

PDF-Xchange screen grab showing a typical search result:
http://802.11junk.com/jeffl/OCR%20Demo/PDF-Xchange-screen.jpg

--

Rick C