Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On Fri, Dec 20, 2024 at 09:49 AM, Dave_G0WBX wrote:
There are similar software (controller) projects that are Rasp' Pi based too, just needing the bus driver and interconnect components.
There are open source hardware...and software... solutions available for implementing a GPIB controller... directly on a Raspberry Pi.
Mileage might vary.
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On Thu, Dec 19, 2024 at 08:57 PM, James Morris (W7TXT) wrote:
Over $1000 for a USB/IEE488 adapter is interesting
You don't say what you are doing... or what you want to control.... and with what.
But depending on how much fooling around you want to do... and time you want to spend... you might get something that works... for a lot less.
And then... depending on what you are doing... you might spend a lot of money...for a hobbyist... and not get something? that works either.... at least without spending time too.
Off course... what works for Tom, might not work for Dick... and might not work for you either.
There's lots of options.
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On Fri, Dec 20, 2024 at 07:11 AM, Jared Cabot wrote:
IMO... and AFAIK... the original physical layer specification of HP GPIB (at least including the connectors, cables, control signals, signal timing and synchronization) was largely standardized, and adhered to, in both the IEEE 488.1 and IEEE 488.2 standards... making GPIB, and IEEE 488, interfaces relatively consistent... when it came to... at least those things.
HP GPIB is old system. The physical layer hardware is not complex. It is not a hard system to understand.
However... as it has worked out... over the 50 plus years...the software...the controllers... and, the commands... they were a different story.?
And so.. in that context... began the decades long obfuscation... by NI... but not only NI... and just IMO... to make the controller hardware... and the controller software ... as convoluted... and as proprietary... as possible...just, for 'reasons' ? ??
On top of that, there has been a 'desire' to make a half century old... relatively simple... 'parallel bus system' ...'work' with ever more complex... and faster... serial buses... as well as...and in that regard... faster controllers, computers, OS, and software.
What we have here, as hobbyists, cheapskates, no budget businesses, and the 'brave' ... is that huge pile upon us... when we dare to want a GPIB controller... that does everything everybody wants... for next to nothing... and on modern hardware and software too.
And that's why... I would agree... you do need a 'selection'
?
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On Fri, Dec 20, 2024 at 10:16 AM, Matt Huszagh wrote:
I agree with you on price: the e5810a is usually a little more expensive than your device. But if you¡¯re patient, getting one for around $200-300 is often possible.
Of course, mileage may vary... say... if you run a surplus sales site... and buy and sell that stuff... and vacuum up a lot of other junk... with a winning bid on a lot.
But, IMO, you're going to wait a long time (never -e) to see a e5810a for 200USD, and (never -2e) to see one for 300USD.
If someone has one for 200USD, I'd buy it.
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On 2024-12-20 1:16 PM, Matt Huszagh via groups.io wrote: That looks nice Steve. For the record the e5810a is also OS-independent as it can be controlled from a browser (like your device) and supports VXI-11 so mostly I run it from Python (on Linux) using python-ivi. Can yours be used outside a browser? Yes. It supports Telnet as well as a USB/serial port, so is completely OS-agnostic. Also I noticed yours can emulate a 7470, which is cool. Does this support instrument-initiated plots? One use case for me is the calibration routines and verification of my HP 3048 require ?the plots be initiated from the 3048 and there¡¯s actually no other way to get this info as you can¡¯t save the logs to a file. That would be a little tricky, but it might be possible to command the plot (even with a null command) from KISS-488 and then within the IEEE-488 bus timeout period, initiate the plot from the instrument side. I think I see a feature I need to add to my "soon as I get a round 'tuit" list. I agree with you on price: the e5810a is usually a little more expensive than your device. But if you¡¯re patient, getting one for around $200-300 is often possible. I'm asking half that, basically covering my direct costs with not much left over to pay back all the hours on firmware. Steve Hendrix
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On 2024-12-20 12:49 PM, Dave_G0WBX via groups.io wrote: There are similar software (controller) projects that are Rasp' Pi based too, just needing the bus driver and interconnect components.
I have yet to try one of those. But the idea of such a thing does appeal, especially as LAN connectivity comes as standard, plus the possibility of diagnostic logging to a USB stick being easy to do. There are lots of "roll your own" possibilities out there. So much so that about 5 years ago I decided to take that "easy" route to be able to get screen dumps from my oscilloscope, SA, VNA, etc. A couple thousand hours of firmware work later, I've learned that there are a whole lot of unanticipated pitfalls. But my KISS-488 provides access to IEEE-488 / HPIB / GPIB via any of a USB/serial port, Telnet over Ethernet, and HTTP over Ethernet (with its own web server), and doesn't care if you're running Windows 2.0, Windows 11, Unix, or even MSDOS. I'm at this moment engaged in one more firmware battle to hunt down a bug, but it's available at my www.hxengineering.com, with firmware rev 2.31 that passed all of my beta testers. They did, however, miss some obscure things relating to the Prologix-compatible commands, and that's what I'm hunting now. Steve Hendrix
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
That looks nice Steve. For the record the e5810a is also OS-independent as it can be controlled from a browser (like your device) and supports VXI-11 so mostly I run it from Python (on Linux) using python-ivi. Can yours be used outside a browser?
Also I noticed yours can emulate a 7470, which is cool. Does this support instrument-initiated plots? One use case for me is the calibration routines and verification of my HP 3048 require ?the plots be initiated from the 3048 and there¡¯s actually no other way to get this info as you can¡¯t save the logs to a file.
I agree with you on price: the e5810a is usually a little more expensive than your device. But if you¡¯re patient, getting one for around $200-300 is often possible.
Matt
On Thu, Dec 19, 2024 at 11:12 AM Steve Hendrix via <SteveHx= [email protected]> wrote:
On 2024-12-19 2:04 PM, Matt Huszagh via
wrote:
Just
to throw another option in the mix: the Agilent e5810a is
awesome and they can be found for reasonable prices if you
wait around a bit. Adapting to Ethernet is, in my opinion,
much nicer than adapting to USB. But I haven¡¯t managed to get
this working nicely with the GPIB toolkit, so if you want to
use that I¡¯d find an NI adapter.
Just to throw one more hat in the ring, my newly-introduced Rev 2
of my KISS-488 adapter is a LOT cheaper, and handles USB/serial as
well as Ethernet (both Telnet and HTTP). Its real claim to fame is
that it's OS-agnostic, with no special drivers or firmware to load
(and to go obsolete with the next rev of Windows). If you're
interested, check out or email me privately.
Steve Hendrix
Steve Hx each HxEngineering period com
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
Fir those who gave never used the AR488 Arduino based devices.? They work as GPIB controlers (or devices.)
Like the usual controllers, they can handle more than one GPIB "device" as is, up to perhaps 4 instruments, subject to bus driving and loading limits.
With additional bus driver IC's and terminating resistors, up to the full complement of what you can have on a Bus electrically at the same time.
Or, if you have one AR488 per instrument, and connected via a USB Hub, you can have as many as the USB subsystem can support.
Handling the different addressing requirements in your software is your responsibility, but an advantage could be, that other than the actual commands etc, each instruments own unique bus command termination/controll byte sequences could be easier, perhaps.
(EOI, CR, LF, CRLF or other methods etc.)
Horses for courses.
The AR488 devices themselves, can also work as un-addressed Listen Only things, spitting out over the USB Serial emulation to the PC any/everything they hear on the bus.
I have found that to be useful as a crude bus monitor/analyser, but it does NOT include the bus control line state info' but is quite useful as it is. (The Genuine Prologix devices can AFIK do that too.)
Though, as all the AR488 code is open source and well documented, I suspect a full analyser mode could be added (or a new indipendant project entirely created based on it) subject to enough codespace being available, and dedicated presentation/controll software for the PC used.? Idealy, thay would be created with a cross platform compatible coding tool, such as QT Creator, FLTK or similar.
The Arduino hardware is capable and very low cost. Especially the ESP32 based versions.
There are similar software (controller) projects that are Rasp' Pi based too, just needing the bus driver and interconnect components.
I have yet to try one of those. But the idea of such a thing does appeal, especially as LAN connectivity comes as standard, plus the possibility of diagnostic logging to a USB stick being easy to do.
Seasons greeting to all.
Dave 'KBV.
|
Re: HP859x GPIB read cal constants
Thanks Christian,
?
that's a relief, so nothing wrong! :-)
?
Best regards,
Tom
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
On 12/20/24 01:56, Wilko Bulte via groups.io wrote: I also built a couple of AR488, Arduino based/open source adapters. Dirt cheap work like a charm and emulate Prologix adapters. Cheap enough to have multiple, one per T&M box, instead of running fat GPIB cables just run a thin flexible USB one. Does the AR488 have a USB hub so you can daisy chain them? Do you have extra boards to sell? I want to make a few AR488 too.
|
Re: HP859x GPIB read cal constants
Hi Tom.
that "ERROR" message irritated me as well when I first did the
flatness calibration. It actually means just that the value
presented is the "amount of error" that you want to compensate. It
is all about the value below the ERROR title. A bit misleading, I
think.
Best wishes,
Christian
Am 20.12.24 um 15:22 schrieb tom_iphi
via groups.io:
toggle quoted message
Show quoted text
Hi folks,
?
I have meanwhile played with reading flatness data of my
8593E.
Using the TRA? command I do indeed get the data out that I
had written down on paper 8 years ago.
?
One thing worries me:
When I hit the softkey "EDIT FLATNESS", then ERROR shows up,
see attached screenshot.
What does ERROR mean to tell me? The flatness data appears to
be correct. I do measure correct levels over the full frequency
range.
?
Sidenote:
There seems to be a little flaw in the KE5FX plotter tool
that I used to capture the screenshot.
The label on the analyzer screen actually reads:
?
ERROR
12.000 MHz
-.33 dB
?
The plot only shows two lines, the second crippled.
?
Thanks, Tom
?
?
|
The 34401A will also transfer more digits via GPIB than specified. Contrary to the 3457A, you don't have to read the 7th digit via a separate command and add it to the result, the 34401A will output more digits on its own if resolution is not specified via command. Similar to what others have observed on the 3457A, it is mostly noise but resolution could probably be increased by averaging the results.
?
I did a record of my 3456A and my 34401A after a 2h warmup while starting to heat up the room to see how much drift I have to expect when making measurements at cold room temperature vs. a higher one e.g. with the heater on or in the summer. The surfaces surrounding both instruments heated up by about 5¡ãC. Both instruments have a different instrument below and above that form a kind of insulation for their self-warmup. Both instruments recorded a voltage source I built in March this year consisting of an MAX6250 (5V) with reference design components that has been running since then 24/7. Measurements were taken at maximum resolution and with 100NPLC@50Hz each. To me, the variation on the 34401A seems low enough to give it additional resolution over the step size of the 3456A. ?
|
Addin front banana terminals is indeed easy. Do keep in mind that if you want sense lines you have to grab those from the back of the unit! I did that on my units.
Check out the NFM Youtube channel if you want to perform this modification.
Wilko
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
I have found a lot of older (and some current) applications only support NI hardware. Especially apps for Tek scopes from around the turn of the century.
However, I find the Keysight software MUCH nicer and polished making it significantly easier to use.
The NI suite is an insanely bloated mess (moreso than the Ketsight package) that is not nearly as intuitive to use.
All IMO anyway.
?
My solution? I have both. :)
?
?
Jared
|
Re: HP6624A Power Supply - screws on the terminal strips - what size?
Yes the older stuff definitely uses 5-40 screws for the terminal bars (rated for 15A max), had this problem with a 6186 recently, eventually found enough to replace the missing ones.
?
https://www.farnell.com/datasheets/2344901.pdf
?
David
|
Re: HP859x GPIB read cal constants
Hi folks,
?
I have meanwhile played with reading flatness data of my 8593E.
Using the TRA? command I do indeed get the data out that I had written down on paper 8 years ago.
?
One thing worries me:
When I hit the softkey "EDIT FLATNESS", then ERROR shows up, see attached screenshot.
What does ERROR mean to tell me? The flatness data appears to be correct. I do measure correct levels over the full frequency range.
?
Sidenote:
There seems to be a little flaw in the KE5FX plotter tool that I used to capture the screenshot.
The label on the analyzer screen actually reads:
?
ERROR
12.000 MHz
-.33 dB
?
The plot only shows two lines, the second crippled.
?
Thanks, Tom
?
?
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
The original NI USB GPIB adapters are usually more expensive then the 82357B. I suppose it's all upto personal choice.
?I always used NI version, found the NI Max software is easier to use, and the .Net DLL for programming is extremely easy to figure out.
The Chinese replica of NI USB GPIB, is surprisingly good for the price point as well.?
On Fri, 20 Dec 2024, 02:50 James Morris (W7TXT) via , <morrisjl= [email protected]> wrote: Use Agilent 82357B adapters are similar in price to brand new NI devices. I'd prefer to use Agilent/HP generally, but also seem to recall reading that the Agilent devices are troublesome. Is that correct? What do folks recommend?
James W7TXT
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
I bought a Keysight adapter from eBay, which keeps on trucking. Genuine or not, not sure, but it works. YMMV as always.
I also built a couple of AR488, Arduino based/open source adapters. Dirt cheap work like a charm and emulate Prologix adapters. Cheap enough to have multiple, one per T&M box, instead of running fat GPIB cables just run a thin flexible USB one.
Wilko
|
Hi,
I've got some different models of this generation. All make quiet 'terrible' noise. Indeed most of the times the reason is to be found in the run hours of the fan itself and the rather extreme specs these machines have to comply with.
The solution I practice is as follows:
?
- The fan is a standard size fan, replace it with a low noise type but keep in mind the CFM rating. You can go somewhat lower if ambient temperature is similar to office room but don't push it too far down. Personally I scavenged fans from top of the line computer power supplies.
?
- Additionally a temperature regulation was added.
My notes tell me that I used these settings for the modules:
? ? ? ? - Set minimum low at first level that the fans starts spinning
? ? ? ? - Acceleration temp 50¡ãC
? ? ? ? - Acceleration Width 20¡ãC
? ? ? ? - With this setting the fan will be running at 100% at 70¡ãC
?
Careful, the big heatsink in the power supply is at fairly high voltage, make sure to electrically isolate the temperature probe.
?
The newer models indeed have a temperature regulated fan. Better but still not all so good as the fan spins up quiet quickly.
?
Important: These power supplies have some RIFA caps. These don't age well at all and by now will likely be cracked. If you see one of them is cracked don't switch on the power supply anymore as those caps will eventually explode without warning and leave you with a terrible mess to clean. Replace at all cost. I've never had to replace any other caps.
?
Something else: Most models have holes on the front right in the housing to accommodate front terminals. It was a factory option. You only have to poke through the front plastic and fit your terminals. There are also connections foreseen on the PCB to connect the terminals.
?
?
?
|
Re: Opinions on the 82357B vs NI USB to GPIB adapters?
Thanks for the info -- indeed, these are ebay finds. Over $1000 for a USB/IEE488 adapter is interesting.
On Thu, Dec 19, 2024 at 11:04?AM Matt Huszagh via <huszaghmatt= [email protected]> wrote: Hm, where are you getting that data? Keysight is selling new?82357B devices for $750. The NI adapters new are $1000+, depending on where you get them.
If you¡¯re referring to the new devices listed on eBay, be careful. It¡¯s not entirely clear what these are, whether they¡¯re knockoff clones, or factory rejects, or something else. But I¡¯ve been bitten in the past. My last one lasted a couple months and then died.
Apparently there¡¯s a similar authenticity issue with used 82357B devices.
More generally, John Miles¡¯s useful GPIB toolkit apparently works best with the NI adapter. If you plan to use that, that¡¯s a vote in favor of the NI adapter. Aside from that, I can¡¯t say much about the quality difference of the genuine products.
Just to throw another option in the mix: the Agilent e5810a is awesome and they can be found for reasonable prices if you wait around a bit. Adapting to Ethernet is, in my opinion, much nicer than adapting to USB. But I haven¡¯t managed to get this working nicely with the GPIB toolkit, so if you want to use that I¡¯d find an NI adapter.
Matt On Thu, Dec 19, 2024 at 10:50 AM James Morris (W7TXT) via <morrisjl= [email protected]> wrote: Use Agilent 82357B adapters are similar in price to brand new NI devices. I'd prefer to use Agilent/HP generally, but also seem to recall reading that the Agilent devices are troublesome. Is that correct? What do folks recommend?
James W7TXT
|
Jared Cabot
factory
