Author Topic: AR488 Arduino-based GPIB adapter (Read 286341 times)

WaveyDipole · « **on:** February 07, 2019, 01:09:29 pm »

sundance · « **Reply #1 on:** February 07, 2019, 02:34:10 pm »

Thank you very much for your excellent work!
Last year I also decided to write my own GPIB controller software for the Arduino (when I bought an HP-3478A). It worked - although I didn't finish it yet. And it's always extremely interesting to see how others do it...

-sundance

metrologist · « **Reply #2 on:** February 07, 2019, 05:37:12 pm »

Spectacular. I was concerned because I recall Girlando did not want derivatives. You mention he reviewed and authorized this.

Inspired by the original work of Emanuele Girlando, licensed under a Creative
Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Any code in common with the original work is reproduced here with the explicit
permission of Emanuele Girlando, who has kindly reviewed and tested this code.

Thanks also to Luke Mester for comparison testing against the Prologix interface.
AR488 is Licenced under the GNU Public licence.

I'm not that versed on the licensing though and it is not clear to me if your code is protected by Creative
Commons Attribution-NonCommercial-NoDerivatives 4.0 International License or the GNU Public licence, or if those are the same references as you mention both. Sorry to bring up licensing topic, but I am not sure now what is restricted to modify and if possible what, and thanks for the work to advance the code.

sundance · « **Reply #3 on:** February 07, 2019, 06:05:26 pm »

Quote

(...) but the sketch is an almost complete re-write of Emanuele's original code

That sounds more than just some adjustments/patches to Emanuele's code. And if you compare both WaveyDipole's to Emanuele's code there are not too many sections that are similar, except the obvious parts that cover hardware compatibility.
It's like with my own code: I took ideas from Emanuele and others (like here: http://www.rudiswiki.de/wiki/GPIBtoUSB_Nano3) who wrote GBIB software for the Arduino. But since I also wrote it from scratch, the final result is completely different. The only thing (concerning licence and patents) that worried me was the use of the (quasi standard) Prologix command syntax.

kutte · « **Reply #4 on:** February 07, 2019, 06:16:14 pm »

thank you,
but does not compile. Errors are
AR488-0-45-10:758: error: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
plenty of them..
AR488-0-45-10:758: error: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
AR488-0-45-10.ino: In function 'bool gpibSendData(char*, uint8_t)':
AR488-0-45-10:1964: error: return-statement with no value, in function returning 'bool' [-fpermissive]
invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
Kutte

maxwell3e10 · « **Reply #5 on:** February 10, 2019, 12:36:17 am »

I tried it with an Atmega328p NANO board. It works OK if I don't use ++auto command. If I use ++auto 1 command, everything gets slowed down, so even ++ver response is delayed by the timeout. Also the ++rst command makes it hang.

To make EZGPIB recognize the adapter, it needs to have "GPIB-USB" string in the ++ver response.

texaspyro · « **Reply #6 on:** February 10, 2019, 05:09:20 am »

Quote from: maxwell3e10 on February 10, 2019, 12:36:17 am

To make EZGPIB recognize the adapter, it needs to have "GPIB-USB" string in the ++ver response.

Years ago I had to make a GPIB interface that emulated a Prologix adapter (at the time the Prologix code wrote to EEPROM every time certain commands were sent and this killed the EEPROM rather quickly if you talked to multiple devices). To get my code to work with the program that was controlling it I had to output the ++ver (or was it a help command?) verbatim to what Prologix was sending.

kutte · « **Reply #7 on:** February 10, 2019, 04:23:06 pm »

Quote from: kutte on February 07, 2019, 06:16:14 pm

thank you,
but does not compile. Errors are
AR488-0-45-10:758: error: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
plenty of them..
AR488-0-45-10:758: error: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
AR488-0-45-10.ino: In function 'bool gpibSendData(char*, uint8_t)':
AR488-0-45-10:1964: error: return-statement with no value, in function returning 'bool' [-fpermissive]
invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
Kutte

error handling seems to have been downgraded to warnings after updating Arduino IDE fro 1.6.5 to 1.8.19
Kutte

WaveyDipole · « **Reply #8 on:** February 10, 2019, 04:33:37 pm »

Quote from: maxwell3e10 on February 10, 2019, 12:36:17 am

I tried it with an Atmega328p NANO board. It works OK if I don't use ++auto command. If I use ++auto 1 command, everything gets slowed down, so even ++ver response is delayed by the timeout. Also the ++rst command makes it hang.

To make EZGPIB recognize the adapter, it needs to have "GPIB-USB" string in the ++ver response.

The string can be changed with the ++setvstr command and saved with the ++savecfg command.

Thanks for the feedback regarding the NANO, including slowdown and ++rst hanging.
I will need to investigate these further.

WaveyDipole · « **Reply #9 on:** February 10, 2019, 10:17:42 pm »

Quote from: kutte on February 07, 2019, 06:16:14 pm

thank you,
but does not compile. Errors are
AR488-0-45-10:758: error: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
plenty of them..
AR488-0-45-10:758: error: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]

Quote from: kutte on February 10, 2019, 04:23:06 pm

error handling seems to have been downgraded to warnings after updating Arduino IDE fro 1.6.5 to 1.8.19
Kutte

I am using Arduino IDE version 1.8.7 (Linux)/1.8.8 (Win) standalone (.zip package) downloaded directly from the Arduino.cc site for development and neither seems to generate these errors or warnings. I haven't tested it with v1.8.8/1.8.19 downloaded from the Win 10 store though. I will investigate further.
Issue logged: https://github.com/Twilight-Logic/AR488/issues/2

EDIT: Tested with Arduino IDE 1.8.8/1.8.19.0 downloaded from the Win10 store and it seems these warnings do indeed appear in that version. I will investigate and see whether the code can be adjusted to avoid them. For further updates, please follow the linked issue on GitHub.

Quote from: kutte on February 07, 2019, 06:16:14 pm

AR488-0-45-10.ino: In function 'bool gpibSendData(char*, uint8_t)':
AR488-0-45-10:1964: error: return-statement with no value, in function returning 'bool' [-fpermissive]
invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]
Kutte

I believe I know what the problem is and it will be fixed in the next update.
Issue logged: https://github.com/Twilight-Logic/AR488/issues/1

WaveyDipole · « **Reply #10 on:** February 12, 2019, 12:01:45 pm »

Version 0.45.11 uploaded. This should address the compiler warnings in Arduino IDE v1.8.8/1.8.19.0.

kutte · « **Reply #11 on:** February 13, 2019, 10:34:13 am »

Quote from: WaveyDipole on February 12, 2019, 12:01:45 pm

Version 0.45.11 uploaded. This should address the compiler warnings in Arduino IDE v1.8.8/1.8.19.0.

thank you again, but a lot of warnings still exist:

Y:\USB_16GB\Voltage reference\GPIB\ar488-arduino\ar488\AR488.ino:153:0: warning: "SPE" redefined

#define SPE 0x18

In file included from c:\program files (x86)\arduino\hardware\tools\avr\avr\include\avr\io.h:272:0,
from c:\program files (x86)\arduino\hardware\tools\avr\avr\include\avr\pgmspace.h:90,
from C:\Program Files (x86)\Arduino\hardware\arduino\avr\cores\arduino/Arduino.h:28,
from sketch\AR488.ino.cpp:1:
c:\program files (x86)\arduino\hardware\tools\avr\avr\include\avr\iom328p.h:294:0: note: this is the location of the previous definition
#define SPE 6
Y:\USB_16GB\Voltage reference\GPIB\ar488-arduino\ar488\AR488.ino:758:1: warning: ISO C++ forbids converting a string constant to 'char*' [-Wwrite-strings]

Y:\USB_16GB\Voltage reference\GPIB\ar488-arduino\ar488\AR488.ino:758:1: warning: ISO C++ forbids converting a string constant to 'char*' [-Wwrite-strings]

Y:\USB_16GB\Voltage reference\GPIB\ar488-arduino\ar488\AR488.ino:758:1: warning: invalid conversion from 'void (*)()' to 'void (*)(char*)' [-fpermissive]

Y:\USB_16GB\Voltage reference\GPIB\ar488-arduino\ar488\AR488.ino:758:1: warning: ISO C++ forbids converting a string constant to 'char*' [-Wwrite-strings]

and many !!! more, Kutte

coromonadalix · « **Reply #12 on:** February 13, 2019, 10:53:46 pm »

Witch arduino version do you have when you try to compile it

, please be more specific with your errors to help WaveyDipole

On my 1.8.8 IDE it compiled and sent just fine for an duemilanove and a uno, for tests purposes ??

kutte · « **Reply #13 on:** February 14, 2019, 04:08:37 pm »

compiled with version 1.8.8 this time without any erros. Don't know what happend the last time, sorry for any confusion.
Kutte

WaveyDipole · « **Reply #14 on:** February 14, 2019, 05:53:51 pm »

kutte, thanks for the update. The warning is interesting since it indicates a definition as it indicates that 'SPE' is defined elsewhere in the Arduino IDE support files. I will need to have a think about how to deal with this - probably will have to change all of my #define names.

Glad you got it compiled without error in the meantime. Curious that the IDE should behave in such an inconsistent manner.

Kean · « **Reply #15 on:** February 15, 2019, 10:58:28 am »

Quote from: WaveyDipole on February 14, 2019, 05:53:51 pm

kutte, thanks for the update. The warning is interesting since it indicates a definition as it indicates that 'SPE' is defined elsewhere in the Arduino IDE support files. I will need to have a think about how to deal with this - probably will have to change all of my #define names.

Yes, you may want to add a prefix in front of your shorthand definitions just to be on the safe side.

In particular SPE is also defined in various Arduino hardware includes as the "SPI Enable" bit in one of the AVR Special Function Registers.
See for example hardware/tools/avr/avr/include/avr/iom328p.h (included via hardware/tools/avr/avr/include/avr/io.h)

Thanks for this updated code by the way. It is something I think I can use in the future for some HPIB CS/80 disk interfacing I want to do.

WaveyDipole · « **Reply #16 on:** February 23, 2019, 06:24:01 pm »

Quote from: maxwell3e10 on February 10, 2019, 12:36:17 am

I tried it with an Atmega328p NANO board. It works OK if I don't use ++auto command. If I use ++auto 1 command, everything gets slowed down, so even ++ver response is delayed by the timeout. Also the ++rst command makes it hang.

To make EZGPIB recognize the adapter, it needs to have "GPIB-USB" string in the ++ver response.

Could you please provide the exact sequence of commands you used to reproduce the issue?
Also, what equipment are you driving please?

I am using quite a slow meter, so perhaps not noticing the issue.

Thanks.

maxwell3e10 · « **Reply #17 on:** February 23, 2019, 10:26:49 pm »

I was testing it with HP3457 meter. I use a serial terminal to issue ++addr 22 command and then start sending commands through it. If I set ++auto 1 and the meter is send command TRIG SGL, then there is a significant delay before it gives a reading. Also there is the same delay for ID? command or ++ver command.

On the other hand, if I set the meter to TRIG AUTO, the program gives a continuous stream of voltage readings in the ++auto 1 mode. This maybe is useful, but its not the behavior one would expect. I think the problem maybe that the program issues a continuous stream of ++read commands in the ++auto 1 mode. My guess is that the buffer is getting delayed by ++read commands when no answer is supposed to come from the meter.

WaveyDipole · « **Reply #18 on:** February 24, 2019, 08:15:04 pm »

Quote from: maxwell3e10 on February 23, 2019, 10:26:49 pm

On the other hand, if I set the meter to TRIG AUTO, the program gives a continuous stream of voltage readings in the ++auto 1 mode. This maybe is useful, but its not the behavior one would expect. I think the problem maybe that the program issues a continuous stream of ++read commands in the ++auto 1 mode. My guess is that the buffer is getting delayed by ++read commands when no answer is supposed to come from the meter.

Thank you for taking the time to feed back the details of the problem. The section on the auto command in the Prologix manual says that it "saves the user from having to issue read commands repeatedly", which sounds like a process that performs repeated read commands automatically. Having read that section again in conjunction with your observations, I am now wondering whether what was meant is that it should perform ONE read automatically after a command is issued, saving the user from having to issue a ++read after each command sent to the instrument? In that context the note regarding the "Query unterminated or -420 error" starts to make sense. However, the example still doesn't seem to fit in somehow so I'm still not sure. Moreover, how does one set parameters for this automatic read (e.g. ++read eoi)? It would be appreciated if you or someone else reading this could explain what behavior one would expect from the auto 1 command?

I will be happy to adjust the program to comply with the expected behavior.

maxwell3e10 · « **Reply #19 on:** February 24, 2019, 11:19:19 pm »

I don't have Prologix's original software, but the behavior I observed with Girlando's code is that in ++auto 1 mode just one ++read command is issued after each command sent to the device. This generally make sense, since most GPIB instruments (except some meters in auto mode) don't have new data available in the buffer for reading at all times, only in response to a specific command.

On the other hand, for the case when data are available, it maybe useful to keep the auto mode as you have implemented it, simulating "Talk only" GPIB mode. So perhaps it would make sense to keep the continuous reading option that you have already implemented with something like ++auto 2 setting.

An even fancier implementation would be to look for "?" as the last character in the command sent to the meter. If there is a ?, then execute one ++read command, if not then don't execute the read command. Otherwise, the ++auto 1 mode can still cause errors if a command doesn't generate a response from the device.

GregDunn · « **Reply #20 on:** February 27, 2019, 02:27:14 am »

I appreciate the effort in producing a GPIB compatible interface for Arduino. My boards are on the way and should be here soon; meanwhile, I'm trying to assemble the resources so that I can put them to use.

I have downloaded the AR488 code and it compiles properly, so shipping it to the Uno board (standard FTDI interface, not the CH340G) should be straightforward.

Now, here's where I need some guidance: I have one PC, but it's in a different room from the workspace and would be very inconvenient. I have several Macs, of which two are portable and can be set up on the bench. I understand that EZGPIB and the GPIB Toolkit only run on Windows, but that having the FTDI drivers I can talk directly to the AR488 with GPIB commands via a terminal through USB, correct?

Is it possible to script communications to the interface without the special software, using the terminal and a shell script? I realize I'd be giving up some of the capabilities, but all I really need to do is read and write a few things from the target device (DMM data and in the case of my HP 3478A, cal data). It doesn't have to be anything fancy, and I'm comfortable with *nix shell scripting.

maxwell3e10 · « **Reply #21 on:** February 27, 2019, 05:03:18 am »

Yes, all of these serial-interface GPIB don't require a special driver. A simple serial terminal program can be used interactively or any program (pythod, matlab) that can talk to the serial port.

GregDunn · « **Reply #22 on:** February 28, 2019, 08:02:40 pm »

Well, that was anticlimactic.

Plugged the Uno into a USB port, uploaded the sketch, connected via 'screen' and it just worked. Unfortunately, there's a bug in 'screen' that prevents setting local echo, so I've been using CoolTerm to verify that the interface is communicating to the Mac. (I really want to use *nix scripting to do this, rather than CoolTerm's Applescript.

I may end up using another screen emulator)

Now, to wire up the GP-IB plug and try it out on my HP 3478A...

Krampmeier · « **Reply #23 on:** February 28, 2019, 08:11:12 pm »

Many thanks for creating and sharing this!
I built the adapter with a chinese Arduino Nano clone, using the latest "nightly" Arduino compiler. Got a lot of warnings (including the one about the duplicate definition of 'SPE' ), but no errors.

Communication with my HP 4192A impedance analyzer sees to work fine, but when I sent *IDN? to a Keithley 2001, the first 1 to 3 characters of the response were always wrong. Same with a Keithley 2010.

I just tried it again:

Quote

*idn?
++read
OEITHLEY INSTRUMENTS INC.,MODEL 2010,0632912,A15 /A02
fetch?
++read
m9.59916086E-01
:SENS:FUNC?
++read
bVOLT:DC"

Edit: I just noticed that the Keithley's display twitches when I send the ++read command. The instrument also clears the "REM" indicator, so it goes out of remote mode. Hmmm. Did I make a mistake in the wiring?

The behaviour with ++auto 1 does not make sense with the DMMs, as the DMM starts showing Errors when the adapter tries to read something when nothing is there. It does work with the 4192A. I agree with maxwell3e10 that there are two options for "auto read" which can make sense depending on the instrument.

coromonadalix · « **Reply #24 on:** February 28, 2019, 09:14:38 pm »

would be nice to have a final schematic / wiring to be sure we dont miss something ??


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: AR488 Arduino-based GPIB adapter (Read 286341 times)

Share me