Is a USB to GPIB dongle/convertor a difficult project ?

[Fred Bartoli]

Joerg a écrit :

Fred Bartoli wrote:
Joerg a écrit :
JosephKK wrote:
Joerg wrote:
legg wrote:
On Wed, 20 Feb 2008 01:27:15 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:


Whatever works to get the HPGL image out would be fine.

For the software side, look at the program from KE5FX called
hp7470.
It captures plots over the gpib and lets you manipulate, compare
and
store them. I use this with my 8753, 8757, 8569 and 8562
analyzers.
It supports the NI and Prologix gpib boards. I have been
meaning to
try it with one of the NI gpib-serial boxes I have sitting around.
Those boxes are cheap and plentiful. Since the Prologix uses a
serial port emulator, there may be a chance.

By the way, on your other post about ebay, search for either E2050*
or E2050A. If you leave the A off, you will not find any matches.

Thanks. Yep, that showed it. Along with fashion socks and other
stuff ;-)

Seems they now run silent auctions there as well. It appears the
Prologix solution would be a bit simpler though. I'll never
understand why HP picked that dreaded GPIB bus in the first
place, with its expensive garden hose cables and all that. The
site of KE5FX is very interesting. In case others want to look:
http://www.thegleam.com/ke5fx/gpib/readme.htm

They not only picked it, they developed it before there was GPIB,
which always explained a lot of things, to me.


Still doesn't make sense. Around the time it became popular others
had a much simpler serial bus figured out, for example the
institute where I did my master's project. Instead of prohibitively
expensive garden hose cables we could use cheap telephone wire and
did not have the length restrictions.


Well there were two issues, HPIB (later renamed GPIB) was available
on HP instruments and serial was not. And with the advent of
logging DMMs, VNAs, digitizing 'scopes, digital SAs and such, the
1000 to 1 data transfer speed advantage added up. GPIB also has
group execute trigger (GET) which can cause a group of instruments
to go through a sequence of measurement and switching accumulating a
chunk of measurements is a short time. Followed by reading the
instrument data at a more reasonable pace.


Sure the bus has a few (very few) advantages but let's face it, 99%
of users wanted to do one thing: Document the results.


I guess not, at least at the time it was invented (no GUI OS,...)
Almost all what people wanted was automation. For documentation
purpose they had either... cameras, printers/plotters,...


Nope. My Dolch Logic Analyzer always provided nice files via its RS232
port which I could then import into DOS-Word. You do not need a GUI at
all to create nice-looking documents with graphics and instrument screen
shots in there. The drill was always the same: Find problem such as bus
contention or glitch, insert picture to show the problem, type up
solution, draw solution on OrCAD SDT or in some cases Futurenet Dash,
insert schematic into document, done.

Of course then there were those instruments such as HP that didn't have
RS232 and we had to use the old Polaroid cameras. The goo from the back
of those instant pictures could cause really nasty stains on clothing.
Afterwards I used a Logitech hand scanner (ScanMan or something like
that) in order to create an image file that could be imported into
DOS-Word.

All that back in 1978?


--
Thanks,
Fred.

 

[Joerg]

Fred Bartoli wrote:

Joerg a écrit :
Fred Bartoli wrote:
Joerg a écrit :
JosephKK wrote:
Joerg wrote:
legg wrote:
On Wed, 20 Feb 2008 01:27:15 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:


Whatever works to get the HPGL image out would be fine.

For the software side, look at the program from KE5FX called
hp7470.
It captures plots over the gpib and lets you manipulate,
compare and
store them. I use this with my 8753, 8757, 8569 and 8562
analyzers.
It supports the NI and Prologix gpib boards. I have been
meaning to
try it with one of the NI gpib-serial boxes I have sitting around.
Those boxes are cheap and plentiful. Since the Prologix uses a
serial port emulator, there may be a chance.

By the way, on your other post about ebay, search for either
E2050*
or E2050A. If you leave the A off, you will not find any matches.

Thanks. Yep, that showed it. Along with fashion socks and other
stuff ;-)

Seems they now run silent auctions there as well. It appears the
Prologix solution would be a bit simpler though. I'll never
understand why HP picked that dreaded GPIB bus in the first
place, with its expensive garden hose cables and all that. The
site of KE5FX is very interesting. In case others want to look:
http://www.thegleam.com/ke5fx/gpib/readme.htm

They not only picked it, they developed it before there was GPIB,
which always explained a lot of things, to me.


Still doesn't make sense. Around the time it became popular others
had a much simpler serial bus figured out, for example the
institute where I did my master's project. Instead of
prohibitively expensive garden hose cables we could use cheap
telephone wire and did not have the length restrictions.


Well there were two issues, HPIB (later renamed GPIB) was available
on HP instruments and serial was not. And with the advent of
logging DMMs, VNAs, digitizing 'scopes, digital SAs and such, the
1000 to 1 data transfer speed advantage added up. GPIB also has
group execute trigger (GET) which can cause a group of instruments
to go through a sequence of measurement and switching accumulating
a chunk of measurements is a short time. Followed by reading the
instrument data at a more reasonable pace.


Sure the bus has a few (very few) advantages but let's face it, 99%
of users wanted to do one thing: Document the results.


I guess not, at least at the time it was invented (no GUI OS,...)
Almost all what people wanted was automation. For documentation
purpose they had either... cameras, printers/plotters,...


Nope. My Dolch Logic Analyzer always provided nice files via its RS232
port which I could then import into DOS-Word. You do not need a GUI at
all to create nice-looking documents with graphics and instrument
screen shots in there. The drill was always the same: Find problem
such as bus contention or glitch, insert picture to show the problem,
type up solution, draw solution on OrCAD SDT or in some cases
Futurenet Dash, insert schematic into document, done.

Of course then there were those instruments such as HP that didn't
have RS232 and we had to use the old Polaroid cameras. The goo from
the back of those instant pictures could cause really nasty stains on
clothing. Afterwards I used a Logitech hand scanner (ScanMan or
something like that) in order to create an image file that could be
imported into DOS-Word.


All that back in 1978?

No, mid 80's. But the Dolch is older, got it used. Back in the late 70's
you could not really do bitmap stuff on the small IBM pre-PC computers
or on the Commodore but you could on the Apple (and on IBM mainframe
systems). Possibly that's the reason why Apple became so popular with
journalists.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[John Larkin]

On Fri, 22 Feb 2008 08:43:37 -0800, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

legg wrote:
On Wed, 20 Feb 2008 01:27:15 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:


Whatever works to get the HPGL image out would be fine.

For the software side, look at the program from KE5FX called hp7470.
It captures plots over the gpib and lets you manipulate, compare and
store them. I use this with my 8753, 8757, 8569 and 8562 analyzers.
It supports the NI and Prologix gpib boards. I have been meaning to
try it with one of the NI gpib-serial boxes I have sitting around.
Those boxes are cheap and plentiful. Since the Prologix uses a
serial port emulator, there may be a chance.

By the way, on your other post about ebay, search for either E2050*
or E2050A. If you leave the A off, you will not find any matches.

Thanks. Yep, that showed it. Along with fashion socks and other stuff ;-)

Seems they now run silent auctions there as well. It appears the
Prologix solution would be a bit simpler though. I'll never understand
why HP picked that dreaded GPIB bus in the first place, with its
expensive garden hose cables and all that. The site of KE5FX is very
interesting. In case others want to look:
http://www.thegleam.com/ke5fx/gpib/readme.htm

They not only picked it, they developed it before there was GPIB,
which always explained a lot of things, to me.


Still doesn't make sense. Around the time it became popular others had a
much simpler serial bus figured out, for example the institute where I
did my master's project. Instead of prohibitively expensive garden hose
cables we could use cheap telephone wire and did not have the length
restrictions.

HPIB/GPIB was invented before computers or microprocessors were common
in test labs. With just a heap of TTL and some switches, a beastly
counter or DVM could be piped to a more beastly mechanical printer,
and measurements could be logged without a computer. That's why it has
the goofy talker/listener orientation.

John

 

[Joerg]

John Larkin wrote:

On Fri, 22 Feb 2008 08:43:37 -0800, Joerg
notthisjoergsch@removethispacbell.net> wrote:

legg wrote:
On Wed, 20 Feb 2008 01:27:15 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:


Whatever works to get the HPGL image out would be fine.

For the software side, look at the program from KE5FX called hp7470.
It captures plots over the gpib and lets you manipulate, compare and
store them. I use this with my 8753, 8757, 8569 and 8562 analyzers.
It supports the NI and Prologix gpib boards. I have been meaning to
try it with one of the NI gpib-serial boxes I have sitting around.
Those boxes are cheap and plentiful. Since the Prologix uses a
serial port emulator, there may be a chance.

By the way, on your other post about ebay, search for either E2050*
or E2050A. If you leave the A off, you will not find any matches.

Thanks. Yep, that showed it. Along with fashion socks and other stuff ;-)

Seems they now run silent auctions there as well. It appears the
Prologix solution would be a bit simpler though. I'll never understand
why HP picked that dreaded GPIB bus in the first place, with its
expensive garden hose cables and all that. The site of KE5FX is very
interesting. In case others want to look:
http://www.thegleam.com/ke5fx/gpib/readme.htm
They not only picked it, they developed it before there was GPIB,
which always explained a lot of things, to me.

Still doesn't make sense. Around the time it became popular others had a
much simpler serial bus figured out, for example the institute where I
did my master's project. Instead of prohibitively expensive garden hose
cables we could use cheap telephone wire and did not have the length
restrictions.

HPIB/GPIB was invented before computers or microprocessors were common
in test labs. With just a heap of TTL and some switches, a beastly
counter or DVM could be piped to a more beastly mechanical printer,
and measurements could be logged without a computer. That's why it has
the goofy talker/listener orientation.

Hmm, my father computerized the first lab and production line (cold
rolled steel factory) in the early 60's, more than 10 years before GPIB
made it into the labs. Don't know what they used for connections,
probably some kind of TTY bus. And all he had was 2K of RAM which was
considered huge in those days.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Joel Koltner]

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
newsJ_vj.7535$Ru4.1259@newssvr19.news.prodigy.net...

Sure the bus has a few (very few) advantages but let's face it, 99% of users
wanted to do one thing: Document the results.

For engineers, sure... but I suspect you'd be surprised at just how many
GPIB-capable instruments were sold to be used as part of production line
setups. Tektronix had a LOT of test equipment on their production lines, all
being synchronized over GPIB to perform automated testing.

Some of our automated test setups use GPIB as well, although it's primarily
because we still have some older, viable instruments around that we don't want
to "put out to pasture yet" (not because we like it all that much!) -- you can
bet that when we do the replacements will have Ethernet interfaces on them.

 

[Joerg]

Joel Koltner wrote:

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
newsJ_vj.7535$Ru4.1259@newssvr19.news.prodigy.net...
Sure the bus has a few (very few) advantages but let's face it, 99% of users
wanted to do one thing: Document the results.

For engineers, sure... but I suspect you'd be surprised at just how many
GPIB-capable instruments were sold to be used as part of production line
setups. Tektronix had a LOT of test equipment on their production lines, all
being synchronized over GPIB to perform automated testing.

Some of our automated test setups use GPIB as well, although it's primarily
because we still have some older, viable instruments around that we don't want
to "put out to pasture yet" (not because we like it all that much!) -- you can
bet that when we do the replacements will have Ethernet interfaces on them.

I have only seen very occasional use of GPIB in production. Lots of gear
had it but there was nothing connected. Somehow I grew up with TTY and
the serial bus just like my father, then completely skipped the GPIB era
and went to Ethernet in the 90's. The few occasions where I did
encounter GPIB it worked but in terms of enterprise management it was a
drag. The topper was a guy some time around 2005 who was busy restoring
a really old PC and installing Win3.2 or something on it. That was the
only box that was able to talk to some of the GPIB. Trying to get any
data from production to the central server was a major pain in the neck.

I really don't understand why HP didn't provide RS232 in addition. It's
so easy and cheap. I mean, Dolch did that even on their lower-end
analyzers so it wasn't rocket science. I used to take that on the road
along with a Wang 8086 laptop. That plus a cheap D-Sub cable and I came
back with perfect test data on file. The Wang had a built-in printer but
I never really needed it.

GPIB drove me nuts. Here you have a $10k analyzer that has all the comms
features one could wish and right next to it sits a $45k HP-machine that
can't talk even RS232. Pathetic.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Joel Koltner]

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:903wj.10303$0o7.9766@newssvr13.news.prodigy.net...

I really don't understand why HP didn't provide RS232 in addition.

I wouldn't be surprised if it was something as simple as being so proud of
their "new and improved" bus that they saw little need to add the "old
fashioned" RS-232.

It's so easy and cheap.

Plus, back in those days, they'd be using Jim's 1488/1489 and would have
increased his retirement funds just a bit!

---Joel

 

[Joerg]

Joel Koltner wrote:

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:903wj.10303$0o7.9766@newssvr13.news.prodigy.net...
I really don't understand why HP didn't provide RS232 in addition.

I wouldn't be surprised if it was something as simple as being so proud of
their "new and improved" bus that they saw little need to add the "old
fashioned" RS-232.

Yep, to me it always looked like NIH syndrome. Or
Not-Proprietary-Enough-Phobia.

It's so easy and cheap.

Plus, back in those days, they'd be using Jim's 1488/1489 and would have
increased his retirement funds just a bit!

I did use those a lot. Some of that stuff is still in production.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Jim Thompson]

On Sat, 23 Feb 2008 17:18:08 -0800, "Joel Koltner"
<zapwireDASHgroups@yahoo.com> wrote:

"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:903wj.10303$0o7.9766@newssvr13.news.prodigy.net...
I really don't understand why HP didn't provide RS232 in addition.

I wouldn't be surprised if it was something as simple as being so proud of
their "new and improved" bus that they saw little need to add the "old
fashioned" RS-232.

It's so easy and cheap.

Plus, back in those days, they'd be using Jim's 1488/1489 and would have
increased his retirement funds just a bit!

---Joel

Hurrumph! I was a direct employee of Motorola when I designed the
1488/1489. No royalties, though I did get $1K each for the patent
applications.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

America: Land of the Free, Because of the Brave

 

[Jim Thompson]

On Sun, 24 Feb 2008 01:28:07 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Joel Koltner wrote:
"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:903wj.10303$0o7.9766@newssvr13.news.prodigy.net...
I really don't understand why HP didn't provide RS232 in addition.

I wouldn't be surprised if it was something as simple as being so proud of
their "new and improved" bus that they saw little need to add the "old
fashioned" RS-232.


Yep, to me it always looked like NIH syndrome. Or
Not-Proprietary-Enough-Phobia.


It's so easy and cheap.

Plus, back in those days, they'd be using Jim's 1488/1489 and would have
increased his retirement funds just a bit!


I did use those a lot. Some of that stuff is still in production.

Quite a few of my "legacy" parts are still manufactured, either by
Motorola/OnSemi, or by Lansdale or Rochester who bought the rights
from Motorola.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

America: Land of the Free, Because of the Brave

 

[Joerg]

Jim Thompson wrote:

On Sun, 24 Feb 2008 01:28:07 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:

Joel Koltner wrote:
"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
news:903wj.10303$0o7.9766@newssvr13.news.prodigy.net...
I really don't understand why HP didn't provide RS232 in addition.
I wouldn't be surprised if it was something as simple as being so proud of
their "new and improved" bus that they saw little need to add the "old
fashioned" RS-232.

Yep, to me it always looked like NIH syndrome. Or
Not-Proprietary-Enough-Phobia.


It's so easy and cheap.
Plus, back in those days, they'd be using Jim's 1488/1489 and would have
increased his retirement funds just a bit!

I did use those a lot. Some of that stuff is still in production.

Quite a few of my "legacy" parts are still manufactured, either by
Motorola/OnSemi, or by Lansdale or Rochester who bought the rights
from Motorola.

Last time I was at that client they used TI and National for those
chips. It all depends where purchasing gets the best deal. ONSemi is
usually quite competitive though, and now about the only semi mfg with a
web site that works well.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[krw]

In article <47c097a4$0$25894$426a74cc@news.free.fr>, Fred Bartoli
<" "> says...

Joerg a écrit :
Fred Bartoli wrote:
Joerg a écrit :
JosephKK wrote:
Joerg wrote:
legg wrote:
On Wed, 20 Feb 2008 01:27:15 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:


Whatever works to get the HPGL image out would be fine.

For the software side, look at the program from KE5FX called
hp7470.
It captures plots over the gpib and lets you manipulate, compare
and
store them. I use this with my 8753, 8757, 8569 and 8562
analyzers.
It supports the NI and Prologix gpib boards. I have been
meaning to
try it with one of the NI gpib-serial boxes I have sitting around.
Those boxes are cheap and plentiful. Since the Prologix uses a
serial port emulator, there may be a chance.

By the way, on your other post about ebay, search for either E2050*
or E2050A. If you leave the A off, you will not find any matches..

Thanks. Yep, that showed it. Along with fashion socks and other
stuff ;-)

Seems they now run silent auctions there as well. It appears the
Prologix solution would be a bit simpler though. I'll never
understand why HP picked that dreaded GPIB bus in the first
place, with its expensive garden hose cables and all that. The
site of KE5FX is very interesting. In case others want to look:
http://www.thegleam.com/ke5fx/gpib/readme.htm

They not only picked it, they developed it before there was GPIB,
which always explained a lot of things, to me.


Still doesn't make sense. Around the time it became popular others
had a much simpler serial bus figured out, for example the
institute where I did my master's project. Instead of prohibitively
expensive garden hose cables we could use cheap telephone wire and
did not have the length restrictions.


Well there were two issues, HPIB (later renamed GPIB) was available
on HP instruments and serial was not. And with the advent of
logging DMMs, VNAs, digitizing 'scopes, digital SAs and such, the
1000 to 1 data transfer speed advantage added up. GPIB also has
group execute trigger (GET) which can cause a group of instruments
to go through a sequence of measurement and switching accumulating a
chunk of measurements is a short time. Followed by reading the
instrument data at a more reasonable pace.


Sure the bus has a few (very few) advantages but let's face it, 99%
of users wanted to do one thing: Document the results.


I guess not, at least at the time it was invented (no GUI OS,...)
Almost all what people wanted was automation. For documentation
purpose they had either... cameras, printers/plotters,...


Nope. My Dolch Logic Analyzer always provided nice files via its RS232
port which I could then import into DOS-Word. You do not need a GUI at
all to create nice-looking documents with graphics and instrument screen
shots in there. The drill was always the same: Find problem such as bus
contention or glitch, insert picture to show the problem, type up
solution, draw solution on OrCAD SDT or in some cases Futurenet Dash,
insert schematic into document, done.

Of course then there were those instruments such as HP that didn't have
RS232 and we had to use the old Polaroid cameras. The goo from the back
of those instant pictures could cause really nasty stains on clothing.
Afterwards I used a Logitech hand scanner (ScanMan or something like
that) in order to create an image file that could be imported into
DOS-Word.


All that back in 1978?

By '78 I had a Tektronix Signal Processing System that ran their SPS
-BASIC on a PDP-11, that did all that and more; no GUI - BASIC. I'd
added a pile of HP stuff to it. Never had any trouble with the Tek
gear or their interface but HPIB was the pits.

--
Keith

 

[JosephKK]

Joerg wrote:

JosephKK wrote:
Joerg wrote:
Klaus Kragelund wrote:
On 20 Feb., 03:57, Joerg <notthisjoerg...@removethispacbell.net
wrote:
none wrote:
Joerg wrote:
Winfield wrote:
On Feb 19, 12:48 pm, "Joel Koltner" <zapwireDASHgro...@yahoo.com
wrote:
"John Larkin" <jjlar...@highNOTlandTHIStechnologyPART.com
wrote in
message
news:8f2mr3t13pel5aobev2ijao7rvs4b3gmhd@4ax.com...
Seems to me that GPIB is, and should be, dying. Everything
should just
be Ethernet.
USB to GPIB and Ethernet to GPIB converters are the probably the
last products
that will ever be commercially developed for GPIB.
Ever read the actual IEEE GPIB spec?
No, although from having written test programs to talk to GPIB
devices I
suspect it's quite massive and somewhat convoluted.
---Joel
Compared to USB or Firewire, it's simple.
But the hardware sure isn't.
You can actually do a simple gpib interface with just a parallel port
but then you have to deal with drivers.
Is there any link where somebody did that? Drivers would otherwise be
the big roadblock. Me and the world of software are miles apart.

--
Regards, Joerg

http://www.analogconsultants.com/- Skjul tekst i anfřrselstegn -

- Vis tekst i anfřrselstegn -

I did something simelar once. I did not have the money to buy the NI
cards and cables and got stubborn.

So I bought USB to serial converters for the instruments and just
hooked up a USB hub. The USB to serial converter costs 20USD.
(prolific are good).

It worked well with the HP instruments, but my tek scope needed
precise timing of TX and handshaking. The simple USB serial convertes
do not handle handshaking well, so I needed to slow down the transfer
rate and thus transferring scope pictures took to long

So, I bought a 4 serial card and ran 4 serial cables instead. This way
some of the NI drivers work right off the bat


But the problem with much of the lab equipment, especially HP, is
that they do not have any serial interface. There is only the GPIB.


Of course HP keeps using it, they invented it.


And the users around them have moved on. I bought a Taiwan-engineered
scope. Has three (!) USB ports. Don't want to schlepp the laptop? No
problem. Stick a little USB flash memory into the front connector, hit
"Save". It lets you select whether to store just an image file or the
whole works plus CSV data. Sweet.

If you store in PNG each screen shot is around 5kB. Comes out crisp and
clean when imported into a PDF report. You still can store thousands of
them even if there is only a really old 16MB stick in your tool box.
AFAIR there is also a way to pipe them to a cell phone. "Hey, Joe, can
you take a look at the weird spike about 40% up the slope?"

I suppose that it could be done though. PCCA Std. 101 defines all the
necessary AT commands. Getting the scope to dial the phone may be an
issue. Mixing voice and data streams is definitely an issue.

 

[JosephKK]

Joerg wrote:

JosephKK wrote:
Joerg wrote:
legg wrote:
On Wed, 20 Feb 2008 01:27:15 GMT, Joerg
notthisjoergsch@removethispacbell.net> wrote:


Whatever works to get the HPGL image out would be fine.

For the software side, look at the program from KE5FX called hp7470.
It captures plots over the gpib and lets you manipulate, compare and
store them. I use this with my 8753, 8757, 8569 and 8562 analyzers.
It supports the NI and Prologix gpib boards. I have been meaning to
try it with one of the NI gpib-serial boxes I have sitting around.
Those boxes are cheap and plentiful. Since the Prologix uses a
serial port emulator, there may be a chance.

By the way, on your other post about ebay, search for either E2050*
or E2050A. If you leave the A off, you will not find any matches.

Thanks. Yep, that showed it. Along with fashion socks and other
stuff ;-)

Seems they now run silent auctions there as well. It appears the
Prologix solution would be a bit simpler though. I'll never
understand why HP picked that dreaded GPIB bus in the first place,
with its expensive garden hose cables and all that. The site of
KE5FX is very interesting. In case others want to look:
http://www.thegleam.com/ke5fx/gpib/readme.htm

They not only picked it, they developed it before there was GPIB,
which always explained a lot of things, to me.


Still doesn't make sense. Around the time it became popular others
had a much simpler serial bus figured out, for example the institute
where I did my master's project. Instead of prohibitively expensive
garden hose cables we could use cheap telephone wire and did not have
the length restrictions.


Well there were two issues, HPIB (later renamed GPIB) was available on
HP instruments and serial was not. And with the advent of logging
DMMs, VNAs, digitizing 'scopes, digital SAs and such, the 1000 to 1
data transfer speed advantage added up. GPIB also has group execute
trigger (GET) which can cause a group of instruments to go through a
sequence of measurement and switching accumulating a chunk of
measurements is a short time. Followed by reading the instrument data
at a more reasonable pace.


Sure the bus has a few (very few) advantages but let's face it, 99% of
users wanted to do one thing: Document the results.

Correct. That is why we are seeing USB on instruments all over the place.

 

[Michael A. Terrell]

Joerg wrote:

Joel Koltner wrote:
"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
newsJ_vj.7535$Ru4.1259@newssvr19.news.prodigy.net...
Sure the bus has a few (very few) advantages but let's face it, 99% of users
wanted to do one thing: Document the results.

For engineers, sure... but I suspect you'd be surprised at just how many
GPIB-capable instruments were sold to be used as part of production line
setups. Tektronix had a LOT of test equipment on their production lines, all
being synchronized over GPIB to perform automated testing.

Some of our automated test setups use GPIB as well, although it's primarily
because we still have some older, viable instruments around that we don't want
to "put out to pasture yet" (not because we like it all that much!) -- you can
bet that when we do the replacements will have Ethernet interfaces on them.


I have only seen very occasional use of GPIB in production. Lots of gear
had it but there was nothing connected. Somehow I grew up with TTY and
the serial bus just like my father, then completely skipped the GPIB era
and went to Ethernet in the 90's. The few occasions where I did
encounter GPIB it worked but in terms of enterprise management it was a
drag. The topper was a guy some time around 2005 who was busy restoring
a really old PC and installing Win3.2 or something on it. That was the
only box that was able to talk to some of the GPIB. Trying to get any
data from production to the central server was a major pain in the neck.

Or was it that the only drivers they had for their GPIB card was
Windows 3.11? The early EISA NAtional Instruments cards were a pain in
the ass. OTOH, Scientific Atlanta wrote their SATE software to run
under windows 2.0, but failed to transfer the source code to Microdyne
after they lost the patent infringement lawsuit and they contracted with
Microdyne to maintain equipment that had been shipped to their existing
customers.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Michael A. Terrell]

Joerg wrote:

Hmm, my father computerized the first lab and production line (cold
rolled steel factory) in the early 60's, more than 10 years before GPIB
made it into the labs. Don't know what they used for connections,
probably some kind of TTY bus. And all he had was 2K of RAM which was
considered huge in those days.

Real time for that system, was probably on the order of several
seconds. Armco Steel, in Middletown Ohio, built a computerized hot
strip mill at that same time, using a huge 5 MB hard drive (40 inch
platter?) with a 5 HP three phase motor. (I ended up with the three
phase breaker box, with the ENGRAVED warning plate "DO NOT ENGAGE DISK
DRIVE FOR FIVE FULL MINUTES").

Serial communications HAD to be used, because the mill was over a
mile long. The computer system was replaced in the mid '80s when they
had purchased the last remaining Westinghouse hard drive in the world,
and no one was rebuilding them.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Joerg]

JosephKK wrote:

Joerg wrote:
JosephKK wrote:
Joerg wrote:
Klaus Kragelund wrote:
On 20 Feb., 03:57, Joerg <notthisjoerg...@removethispacbell.net
wrote:
none wrote:
Joerg wrote:
Winfield wrote:
On Feb 19, 12:48 pm, "Joel Koltner" <zapwireDASHgro...@yahoo.com
wrote:
"John Larkin" <jjlar...@highNOTlandTHIStechnologyPART.com
wrote in
message
news:8f2mr3t13pel5aobev2ijao7rvs4b3gmhd@4ax.com...
Seems to me that GPIB is, and should be, dying. Everything
should just
be Ethernet.
USB to GPIB and Ethernet to GPIB converters are the probably the
last products
that will ever be commercially developed for GPIB.
Ever read the actual IEEE GPIB spec?
No, although from having written test programs to talk to GPIB
devices I
suspect it's quite massive and somewhat convoluted.
---Joel
Compared to USB or Firewire, it's simple.
But the hardware sure isn't.
You can actually do a simple gpib interface with just a parallel
port
but then you have to deal with drivers.
Is there any link where somebody did that? Drivers would otherwise be
the big roadblock. Me and the world of software are miles apart.

--
Regards, Joerg

http://www.analogconsultants.com/- Skjul tekst i anfřrselstegn -

- Vis tekst i anfřrselstegn -

I did something simelar once. I did not have the money to buy the NI
cards and cables and got stubborn.

So I bought USB to serial converters for the instruments and just
hooked up a USB hub. The USB to serial converter costs 20USD.
(prolific are good).

It worked well with the HP instruments, but my tek scope needed
precise timing of TX and handshaking. The simple USB serial convertes
do not handle handshaking well, so I needed to slow down the transfer
rate and thus transferring scope pictures took to long

So, I bought a 4 serial card and ran 4 serial cables instead. This way
some of the NI drivers work right off the bat


But the problem with much of the lab equipment, especially HP, is
that they do not have any serial interface. There is only the GPIB.


Of course HP keeps using it, they invented it.


And the users around them have moved on. I bought a Taiwan-engineered
scope. Has three (!) USB ports. Don't want to schlepp the laptop? No
problem. Stick a little USB flash memory into the front connector, hit
"Save". It lets you select whether to store just an image file or the
whole works plus CSV data. Sweet.

If you store in PNG each screen shot is around 5kB. Comes out crisp
and clean when imported into a PDF report. You still can store
thousands of them even if there is only a really old 16MB stick in
your tool box. AFAIR there is also a way to pipe them to a cell phone.
"Hey, Joe, can you take a look at the weird spike about 40% up the
slope?"


I suppose that it could be done though. PCCA Std. 101 defines all the
necessary AT commands. Getting the scope to dial the phone may be an
issue. Mixing voice and data streams is definitely an issue.

They usually don't do that simultaneously but send the image to the
email address of someone back at headquarters, then call a few minutes
later to discuss it.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Joerg]

Michael A. Terrell wrote:

Joerg wrote:
Joel Koltner wrote:
"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
newsJ_vj.7535$Ru4.1259@newssvr19.news.prodigy.net...
Sure the bus has a few (very few) advantages but let's face it, 99% of users
wanted to do one thing: Document the results.
For engineers, sure... but I suspect you'd be surprised at just how many
GPIB-capable instruments were sold to be used as part of production line
setups. Tektronix had a LOT of test equipment on their production lines, all
being synchronized over GPIB to perform automated testing.

Some of our automated test setups use GPIB as well, although it's primarily
because we still have some older, viable instruments around that we don't want
to "put out to pasture yet" (not because we like it all that much!) -- you can
bet that when we do the replacements will have Ethernet interfaces on them.

I have only seen very occasional use of GPIB in production. Lots of gear
had it but there was nothing connected. Somehow I grew up with TTY and
the serial bus just like my father, then completely skipped the GPIB era
and went to Ethernet in the 90's. The few occasions where I did
encounter GPIB it worked but in terms of enterprise management it was a
drag. The topper was a guy some time around 2005 who was busy restoring
a really old PC and installing Win3.2 or something on it. That was the
only box that was able to talk to some of the GPIB. Trying to get any
data from production to the central server was a major pain in the neck.


Or was it that the only drivers they had for their GPIB card was
Windows 3.11? The early EISA NAtional Instruments cards were a pain in
the ass. OTOH, Scientific Atlanta wrote their SATE software to run
under windows 2.0, but failed to transfer the source code to Microdyne
after they lost the patent infringement lawsuit and they contracted with
Microdyne to maintain equipment that had been shipped to their existing
customers.

Well, teh best thing would be not to need GPIB at all. The next best
thing is a hardware converter to RS232 or USB.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Joerg]

Michael A. Terrell wrote:

Joerg wrote:
Hmm, my father computerized the first lab and production line (cold
rolled steel factory) in the early 60's, more than 10 years before GPIB
made it into the labs. Don't know what they used for connections,
probably some kind of TTY bus. And all he had was 2K of RAM which was
considered huge in those days.


Real time for that system, was probably on the order of several
seconds. Armco Steel, in Middletown Ohio, built a computerized hot
strip mill at that same time, using a huge 5 MB hard drive (40 inch
platter?) with a 5 HP three phase motor. (I ended up with the three
phase breaker box, with the ENGRAVED warning plate "DO NOT ENGAGE DISK
DRIVE FOR FIVE FULL MINUTES").

Serial communications HAD to be used, because the mill was over a
mile long. The computer system was replaced in the mid '80s when they
had purchased the last remaining Westinghouse hard drive in the world,
and no one was rebuilding them.

This one was actually realtime. Stopping the cold-rolled steel line was
not an option, ever. The contract specified that it be run as usual.
Then they had to detect flaws in the material while it was passing
through. My father said the real challenge was the data acquisition
because there would be the occasional splat of grease falling off the
gears of overhead cranes and right onto the steel.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Michael A. Terrell]

Joerg wrote:

Michael A. Terrell wrote:
Joerg wrote:
Joel Koltner wrote:
"Joerg" <notthisjoergsch@removethispacbell.net> wrote in message
newsJ_vj.7535$Ru4.1259@newssvr19.news.prodigy.net...
Sure the bus has a few (very few) advantages but let's face it, 99% of users
wanted to do one thing: Document the results.
For engineers, sure... but I suspect you'd be surprised at just how many
GPIB-capable instruments were sold to be used as part of production line
setups. Tektronix had a LOT of test equipment on their production lines, all
being synchronized over GPIB to perform automated testing.

Some of our automated test setups use GPIB as well, although it's primarily
because we still have some older, viable instruments around that we don't want
to "put out to pasture yet" (not because we like it all that much!) -- you can
bet that when we do the replacements will have Ethernet interfaces on them.

I have only seen very occasional use of GPIB in production. Lots of gear
had it but there was nothing connected. Somehow I grew up with TTY and
the serial bus just like my father, then completely skipped the GPIB era
and went to Ethernet in the 90's. The few occasions where I did
encounter GPIB it worked but in terms of enterprise management it was a
drag. The topper was a guy some time around 2005 who was busy restoring
a really old PC and installing Win3.2 or something on it. That was the
only box that was able to talk to some of the GPIB. Trying to get any
data from production to the central server was a major pain in the neck.


Or was it that the only drivers they had for their GPIB card was
Windows 3.11? The early EISA NAtional Instruments cards were a pain in
the ass. OTOH, Scientific Atlanta wrote their SATE software to run
under windows 2.0, but failed to transfer the source code to Microdyne
after they lost the patent infringement lawsuit and they contracted with
Microdyne to maintain equipment that had been shipped to their existing
customers.


Well, teh best thing would be not to need GPIB at all. The next best
thing is a hardware converter to RS232 or USB.

Microdyne built IEEE-488 ports into all their telemetry products,
because their customers demanded it. After that, it became a legacy
interface, even though the receivers had RS-232, RS-422 and 10/100
Ethernet interface a lot of customers insisted on being able to use
IEEE-488 in existing systems. It appears that they still use it in
current production RCB-2000.

<http://www.l3com.com/products-services/productservice.aspx?type=p&id=167>


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida