Skip navigation

Community

Blog Posts

Blog Posts

Items per page
1 2 Previous Next
0
This feed URL is no longer valid. Visit this page to find the new URL, if you have access: <a href="http://www.facebook.com/profile.php?id=357696587589030">http://www.facebook.com/profile.php?id=357696587589030</a>

Originally posted by Facebook.com at http://www.facebook.com/feeds/syndication_error.php#5bb1329e84a0d7cae566d1161054 2b7c_
0

First Background

 

I use modbus in production almost daily. The average bus I work on will contain 5+ devices, and most systems will have 2-3 buses per computer resulting in normally 10-15 devices in total per a single computer normally. My company does nearly all of its digital data reporting via modbus, which is normally very possible to push 200+ data points per second (on a single channel, over wifi mind you, do the math to figure out your use case).

 

I've done this for nearly 3 years now. I enjoy my job most days, and I 'generally' enjoy MODbus. The Community MODbus API I saw as a god sent, since at the time I had to use an older 'arguably' more open. Library in question I used at the time has not been updated since labview 8.2 which I was okay with using. Its slightly buggy shoots you in the foot once in a while but works generally nonetheless. The main attraction is its completely open source and transparent, you know what it does when why and how.

 

In Jan 2013 a 'newer' Community API was created, which I took as a godsent since the older library had some glaring flaws. I've attempted to stay current on Community Modbus API and I've used in product for nearly 8 months. Here is my report on it.

 

DO NOT EVER,  USE THE COMMUNITY API FOR SERIAL COMMUNICATION IF YOU WANT YOU APPLICATION TO WORK.

 

Now I'm talking about its serial half NOT its TCP half. i have had absolutely no problems what so ever with the TCP half, it functions perfectly.

 

The serial communication side is literally useless in every single way shape and form. Here are the issues I've had with the serial communication master API.

 

1) Don't ever use in a multi-thread environment. If you have multiple master instances (on seperate channels) the serial interface flat out fails. It falls on its face and die. Its VI calls will block for AGES. Even if the data they recieve is literally 100% correct. I can not explain this problem for the life of me. I get massive time outs, and no actual serial I/O I dont understand.

 

2) Unit ID's shouldn't be changed witha single function. If you poll multiple devices at the same time (or the same register on several devices). You shouldn't need to call 2 functions in a loop. Why do I have obj.setUnitID(int val);obj.readHoldingReg(); why isn't UnitID a field of reading/writing? MULTIPLE DEVICES PER BUS is why serial RS-485 exists. If you only need a single UnitID per bus you should be using RS-232 and a BETTER protocol. Unit ID is set per packet in MODbus/RTU, so why is do I set it seperately of per call?

 

3) Setting Unit ID on start up is idiotic. See above for why.

 

4) Random errors. Serial Master Starrt sometimes gives : Error Code 4 <Function 4>:Failure. For no reason, simply restarting the application solves the problem. I don't know what this error is or why it happens. If I use the library I have to build a special case to detect and re-initialize the communication object, it just get thrown at random.

 

4.5) Random Errors 2: Error 1556, "???" this error is great. I still have no clue why I get it, I can't reproduce it from time to time. I have no clue why I'm getting this error, I'm passing valid data to the VI.It just happens on certain computers.

 

5) 5second time out is way to long, even for testing. I mean its great I can take a sip of my coffee while waiting for a single function call to time out but if I have multiple functions the time out is dumb. Read 4 registers before reporting an error? It'll take 20 SECONDS to time out.

 

These even happens when the serial library COMMUNICATES WITH ITSELF VIA DIFFERENT APPLICATIONS ON DIFFERENT COMPUTERS.

 

Changes I would suggest:

 

1) Make the community api, actually involve the the community open source it so I know whats going on, so I can make changes and submit them. This is clearly broken, and has no Q&A testing, and I think its maintainer doesn't even use it on a daily basis.

 

2) Don't use it in the wild. Prosoft PLC's, MMI flow meters, Fuji, Omega, Newport, anything by Mitsubshi it fails MISERABLY on.

 

3) Impliement some form of QA testing. Their clearly is NONE. This library is buggy, and largely useless.

 

Thank you for your time.

 

William Cody Laeder, Software Specialist

cody.laeder@re-sol.com


0

Er sieht nicht wie der Standardroboter aus, den sich jeder vorstellen würde, sondern eher wie eine Puppe. Seine Fähigkeiten jedoch sind außergewöhnlich. Ausgestattet mit einem NI Single-Board RIO und programmiert mit NI LabVIEW, wird Zeno, der von Hanson RoboKind erstellte Roboter, von einem Forschungsteam eingesetzt, um Kinder mit Autismus besser betreuen zu können.

 

https://decibel.ni.com/content/servlet/JiveServlet/downloadImage/38-10439-56982/436-202/zeno2.jpg

 

Unter der Leitung von Dr. Dan Popa setzt das Team die Körpersprache des Roboters ein, um mit autistischen Kindern nonverbal zu kommunizieren und somit ihre Lebensqualität zu verbessern.

 

» Erfahren Sie mehr

0
This feed URL is no longer valid. Visit this page to find the new URL, if you have access: <a href="http://www.facebook.com/profile.php?id=357696587589030">http://www.facebook.com/profile.php?id=357696587589030</a>

Originally posted by Facebook.com at http://www.facebook.com/feeds/syndication_error.php#5bb1329e84a0d7cae566d1161054 2b7c_
0


In Massachusetts, the governor is proposing to ban noncompete agreements.  Go Deval!

My first experience with such agreements stretches back to 1996.  This tech company wanted to hire me, but one of their conditions was I sign a piece of paper stating that if I left the company I couldn't work for any other place doing similar work for a couple of years.  When I first read that I was thrown for a loop and almost turned down the offer.  But then I talked to a friend in law that said those things were hard to enforce - such docs were used more for scare tactics in the tech industry.

Why would a company want to use intimidation tactics to keep their employees from jumping ship?  I can understand not wanting someone to go to a competitor with trade secrets, but there are already some laws in place to prevent that.  But wouldn't a more positive solution be make working there too good to quit?  When I worked at Hewlett Packard in the late 90s they called it "golden handcuffs."  All the perks of working at HP just made it too depressing to leave for some other company.

Another thing that comes to mind is that the company using a noncompete agreement is being hypocritical about experience.  When they look to hire someone, they prize that person's experience in the industry.  But then they turn around and damn someone who wants to take the experience they've acquired to a different company.  That really irks me.  Any experience I've gained in the course of my job (writing programs, building test systems, going to conferences, etc.) is MINE.  The company pays me for the work I do, and they own the product of that work.  They do NOT own the knowledge that I've gained in the process.  That knowledge and experience I've gained by being a good engineer and deliberately learning new things.

On a related subject, you could always watch that bad Ben Affleck movie Paycheck.  Then again, maybe not.

Originally posted by noreply@blogger.com (R. Greg Dudgeon) at http://savvytestengineer.blogspot.com/2014/04/noncompete-agreements-and-who-owns .html
0

Introduction

A fuse is a temperature sensitive device that plays a critical role in circuit protection. Since the operating temperature has an effect on fuse performance and lifetime, it is extremely valuable to be able to model fuse thermal derating in a SPICE environment while the operating temperature should be taken into consideration when selecting the fuse current rating.

 

This new model implements a very fast acting fuse that includes its thermal derating curve. Simulation will show how the operating current and temperature have an effect on the fuse’s performance and lifetime.

 

Fuse Model with Thermal Derating Parameters

 

The Multisim model is developed based on the approach introduced in this paper and is shown in Fig. 1. The implemented fuse model is for the 0603SFV050F/32-2 very fast-Acting chip fuses by TE Connectivity. The parameters used inside the models are as follows:

  • the rated current: irs=0.5A
  • the nominal cold DCR: RFref=0.860Ω
  • the nominal I2t: I2tref=0.0093A2sec


The model requires the fuse resistance at the end of the pre-arcing time, rfm, which is usually provided in the datasheet. This resistance can be determined through laboratory measurements. Another variant to determine the rfm is to consider that the heating process of the fuse is rapid and adiabatic. Knowing the fuse material, the fuse resistance at room temperature, RF, using equation (1), rfm can be calculated considering that the resistance versus temperature is linear:

 

Picture10.png

 

where α is the temperature coefficient of the fuse material, Tm is the melting temperature of the fusing material, and T0 is the reference temperature. The length and section area of the fuse are considered constant.


The thermal derating curve is implemented using a lookup table, while the values above were taken from the fuse manufacturer datasheet. The variation of the fuse resistance with temperature and also the variation of the I2t in percent with temperature are all taken into account inside the model.

 

Picture1.jpg

Figure 1. Fuse SPICE model internal structure

 

Note that in the model above the W2 current controlled switch is used for removing the path for Cfuse discharge after the fuse is blown. To view the model of the fuse double click on it and then click on the Edit Model button under the Value tab.

 

Running the Fuse Model

 

The fuse is placed in a simple test circuit (attached to the article) to verify its operation in an interactive simulation setup.

 

Picture3.jpg

Figure 2. Test circuit setup

 

Picture2.jpg

Figure 3. Load voltage during when over current passes through the fuse

 

 

The Parameter Sweep Analyses can be used, to show the fuse’s response at different currents. In the Analysis Sweep field, the Transient Analysis option must be chosen. Both the Transient and Parameter Sweep Analyses configurations are shown below in Figure 4.

 

Picture4.jpg

Picture5.jpg

Picture6.jpg

 

Figure 4. Analysis setup in Multisim

 

 

Running this sweep analysis, the transient response of the fast acting sweep can be evaluated under different current conditions. This is valuable in determining how fast the fuse will melt and how much energy is needed to do so. The results are shown in Figure 5.

 

 

Picture7.jpg

Figure 5. Analysis results in Multisim

 

 

 

Moreover, the effect of the ambient temperature could be analyzed in Multisim. It is important to understand how differently the fuse will behave at 75 degrees from a normal 25 degrees room temperature. For this purpose a temperature sweep from 25 to 125 degrees is setup in Multisim to run the same circuit under a constant load current.

 

Picture8.jpg

Figure 6. Temperature sweep setup in Multisim

 

Picture9.jpg

Figure 7. Simulation results of the temperature sweep

 

 

The ismulation results show that the higher the environment temperature is the faster the switch will respond to the over-current.

 

You can download this advanced fuse model and start using it in the attachment.

 

Clcik here to download a 45-day free evaluation of the latest release of Multisim.

0

NI has announced an integrated software defined radio solution for rapidly prototyping high-performance, multichannel wireless communication systems. The NI USRP RIO platform is built on the NI LabVIEW RIO architecture and combines a high-performance RF transceiver with an open LabVIEW programmable FPGA architecture. USRP RIO is ideal for a wide range of application areas, including 5G wireless communications research and communications intelligence.

 

USRPRIO.jpg

 

Key Features:

  • Shorten time to results significantly
  • Prototype wireless and algorithm systems faster
  • Transmit and receive signals from DC to 6 GHz


>> Learn more about the NI USRP RIO platform.

0

NI has announced an integrated software defined radio solution for rapidly prototyping high-performance, multichannel wireless communication systems. The NI USRP RIO platform is built on the NI LabVIEW RIO architecture and combines a high-performance RF transceiver with an open LabVIEW programmable FPGA architecture. USRP RIO is ideal for a wide range of application areas, including 5G wireless communications research and communications intelligence.

 

USRPRIO.jpg

 

Key Features:

  • Shorten time to results significantly
  • Prototype wireless and algorithm systems faster
  • Transmit and receive signals from DC to 6 GHz


>> Learn more about the NI USRP RIO platform.

0
This feed URL is no longer valid. Visit this page to find the new URL, if you have access: <a href="http://www.facebook.com/profile.php?id=357696587589030">http://www.facebook.com/profile.php?id=357696587589030</a>

Originally posted by Facebook.com at http://www.facebook.com/feeds/syndication_error.php#5bb1329e84a0d7cae566d1161054 2b7c_
0

Think back 40 years ago, when cell phones didn’t exist and PCs were just surfacing. In order to reactivate the first microprocessors and first electronic memory of the 1970s, you would have to recreate a complete software studio from original specifications of Russian and American CPUs. Not an easy feat.

 

Développement Ingénierie (DI), a company that designs and builds data acquisition systems and test benches, has created an exclusive collection of 12 unique artworks based on vintage components for contemporary art of the Eiffel Tower.

 

eiffeltower.jpg

 

The pieces of art are made from beams and decorative original arches of the Eiffel Tower structure and include a clock showing the stylized Eiffel Tower and Paris. The replicas start to time automatically and show the exact same time as the Eiffel Tower through an onboard GPS receiver.

 

DI used NI PXI hardware and the NI Vision Development Module to develop and validate the display electronics for the clocks. Using LabVIEW system design software, DI recreated the development of the first CPU in history in a reasonable time while creating innovative artwork. 

 

>> Read the full case study.

0

Hallo,

 

how can I create fixed-point number using LabVIEW VI scripting? And set the word length and integer word length?

0
0
This feed URL is no longer valid. Visit this page to find the new URL, if you have access: <a href="http://www.facebook.com/profile.php?id=357696587589030">http://www.facebook.com/profile.php?id=357696587589030</a>

Originally posted by Facebook.com at http://www.facebook.com/feeds/syndication_error.php#5bb1329e84a0d7cae566d1161054 2b7c_
0
This feed URL is no longer valid. Visit this page to find the new URL, if you have access: <a href="http://www.facebook.com/profile.php?id=357696587589030">http://www.facebook.com/profile.php?id=357696587589030</a>

Originally posted by Facebook.com at http://www.facebook.com/feeds/syndication_error.php#5bb1329e84a0d7cae566d1161054 2b7c_
0
This feed URL is no longer valid. Visit this page to find the new URL, if you have access: <a href="http://www.facebook.com/profile.php?id=357696587589030">http://www.facebook.com/profile.php?id=357696587589030</a>

Originally posted by Facebook.com at http://www.facebook.com/feeds/syndication_error.php#5bb1329e84a0d7cae566d1161054 2b7c_