If you are measuring clock and data timing you need to be focusing on Jitter measurements. You need to focus on time interval error measurements aka TIE. All Keysight scopes can do this measurement You should choose a scope with a bandwidth that represents what the receiver will “see”. some people make the mistake of choosing a very high bandwidth scope. that will most likely have higher bandwidth, reveal more noise and jitter but does no the mimic what the receiving side will see. in addition some standards like PCIE can have the clock extracted from the data and there is no need to transmit the ref clock at all. in such case you have to recover the clock in the receiver with a CDR and there are requirements for pll filter settings which you must emulate with the scope. so I hope this gives you some direction to get started. Possibly I can do a detailed presentation and video demo on that topic. I wouldn’t say it falls into the strictly automated category. It is a measurement that is quite common. Automation is more about automating repeated measurements that you would start out doing manually.
Ah. I see, I remember in college, we designed a robot for my senior project. It was such a pain in the ass to have to rerun the robot every time after we made small changes to either the circuit or the code to verify that it worked correctly. Currently, I do automation tests for websites, how hard would it be for me to jump to hardware automation given most of my experience is working with running tests on other software? I haven't applied to these jobs yet, but it's been a while since I was in college, where I studied computer engineering.
Hardware automation is not difficult. if you have a good handle on coding it’s just a matter of knowing how to set up the resources and access them. There are some universal scpi commands as well. would be a good idea if you are serious tomaybe get your hands on some old or used hardware that you can connect to and experiment like a power supply and dmm ok’d scope or something like in the project shown. I may soon do a generic scpi example to show how you can leverage anything you can find and use documentation to do the bulk of the coding. I am a big fan of not throwing things out. being able to make use of older equipment can bring a bit more life when funds just do not exist. I also may do a c# example. Just remember that a scpi controlled instrument is just another thing to read from and write to that has behaviours you want to access like stimulus, power or some form of measurement. meaning if you write something your screen with “print” words come on the screen. if you write to a power supply that is powering a device you turn something on. Sounds to me like you would do fine. a lot of the challenges come in with waiting for an instrument to complete a step in a multi instrument setup. It’s best to have a lot of write and read back checks to get past those. once you figure out those issues then you are well on your way.
check out this link in the keysight library edadocs.software.keysight.com/kkbopen/programming-example-how-to-perform-sequencer-function-using-python-programming-683775606.html I apologize ahead of time for the robovoice in that one but it gets you going.
ive used labview for years. Now im learning Python to do the same thing. apparently Some company's don't use labview but do data acquisition ..with python. maybe its cost effective ..not exactly sure why.
a lot of the Big name companies have gone the Python route. Labview works well but has costs and a learning curve for new users. I come from the C world and was surprised the adoption rate on Python which is a pretty loose way to code. but. it gets things done quicker. I think anyone working with instruments should be versed in Python.
Tim....Does Keysight provide seminars, or WebX sessions...anything on this subject of Python/ instrument automation? I would love to be able to learn as much as possible on this.
Obviously not by the no response to your question. HPVEE is a great language and they let it die. Its no wonder LabView took off, even though VEE was why more practical and efficient for automation.
I though I had answered this in another thread. I have done during a regional Tour in central Canada and numerous on-site walkthroughs. Generally Local AEs will choose topics for seminars which are based on request. Do you have something you would like to see here with specific instrumentation and objectives you are targeting?
