Whether intended or not, I am receiving bad service

[_user32]

Dear Euler's Identity,

I will isolate and try out the address/decode logic going to the mux and compare this with the circuit simulation that I am running on your circuit.   Also, just to be sure we are comparing apples to apples here, are you using Multisim v10.0.1?

With respect to the additional commentary expressed about NI, our products and our level of technical support, I can understand some level of frustration regarding finding a bug - this is understandable.  However we are working to put our best possible effort forward and myself and the rest of the National Instrument team is working hard to overcome your issue and come to a solution and/or possible work around.  Many customers have been successful using NI products and there is no reason why you cannot be successful as well.

I will continue to give you support on the forums.ni.com site, as I have seen that you have provided some level of value to the NI community, however I would like you to review the acceptable use policy before considering additional postings.

http://www.ni.com/legal/termsofuse/unitedstates/us/ 

If you cannot adhere to this, please contact me directly and I would be happy to discuss ongoing support of this matter via phone.   My cell is 207 415-7754.

Regards,

Patrick Noonan
Business Development Manager
National Instruments - Electronics Workbench Group

[Euler's_Identity]

I'm using Multisim Version 10.0.267

As I understand it, I am using the forum in an acceptable manner.  If I am not, then please explain how I am not.

I just want to get this problem solved, and thus far the forum has done me the most good.


P.S. "With respect to the additional commentary expressed about NI, our products and our level of technical support, I can understand some level of frustration regarding finding a bug - this is understandable."  You make it sound like I'm wigging out because I've found A bug, but the fact is this is simply the LATEST bug I've found.  I've been finding bugs in Multisim for many many years now.  And no, this isn't the first time I've been returned an unusable circuit (not my desired circuit) and had it called "fixed."     

Message Edited by Euler's Identity on 09-10-2007 06:47 PM

[Euler's_Identity]

As it may help in solving this problem, in this post I'm uploading a version of this circuit that does work.  However, it's using the oscilloscope and it's using a lower input impedance op amp, the LM324, as I could never get the LF347 to work in this circuit.

What you see on the scope is correct at all times (that I have seen), but it's supposed to be a multimeter so that I can see the precise voltage from the diff amp, a cumbersome task using the oscuilloscope.

Also, remember that you must also download the hierarchical blocks from earlier in order for this circuit to work.

[_user32]

Dear Euler's Identity,

You are correct, I have experienced the same issue that you have seen when trying to use the DMM with this type of circuit - again I am assuming it is similar or related to the same root cause as problems similar to before.   I have have also sent our R&D team this circuit to confirm this fact and will post the findings. 

I believe an alternative (and acceptable) work around - that will give you the resolution you need, is to use the interactive probe.  It should be a part of both the Multisim PRO and EDU versions.  Please confirm or refute that using this interactive probe is a viable work around.

To use this and set the resolution, place the probe on the wire coming from the output of the diffamp.  If you double click on the 'Parameters' tab, their will be a field called resolution that you can modify to increase the resolution of the voltage reading you are interested in.

Regards,

Patrick Noonan

Business Development Manager

National Instruments - Electronics Workbench Group

[Euler's_Identity]

I loaded the original circuit and deleted the multimeter. Instead, I added the static interactive probe and set up to only look at voltage. I got the same error upon simulation. I tried simulating again and got the same thing.

What would you like me to try next?

[GarretF]

Hi Euler:

I read one of you earlier posts, and I thought it would be useful to clear up some version numbering because you are not running the latest version and the version numbers you will see are not always intuitive.

The original Pro and Edu are 10.0.0. In the Help > About Multisim, you will see a version number 10.0.144.
There was a small patch to these, known as 10.0.0.1. In the Help > About Multisim, you will see a version number 10.0.267. This is the version you are running
There was a new release, 10.0.1, which includes ~170 defect fixes. In the Help > About Multisim, you will see a version number 10.0.343. You can see a list of the defects fixed in the Readme file, and this version is available free of charge to anyone who has 10.0.0.

I understand from Tien that the 10.0.1 release does not resolve the particular issue you are seeing, however, I still strongly urge you to upgrade to this latest version because this version fixes a number of issues in the original 10.0.0 release.

You can get this upgrade through the Support and Upgrade Utility, or through our website (ftp://ftp.ni.com/support/circuitdesignsuite/10.0.1/). If you download from the FTP site, be sure to select the appropriate version: Professional, Educational, or Student.

Message Edited by koekje on 09-11-2007 08:18 AM

[Euler's_Identity]

I'm guessing that you're speaking of the 290.218 MB upgrade.  I'm curious, who was responsible for deciding a 290.218 MB upgrade is reasonable?  Does that person have any idea how long it takes to download 290.218 MB using a dialup connection?  Isn't there any other option besides making a person suffer through such a long download?  For instance, could the upgrade be done in parts, without the recipient having to use another piece of software from another company to have any chance of success?  That is one huge upgrade!  And it still won't fix this particular bug??

(I tried to download the upgrade some time back, but when I checked many hours later (overnight) I'd been bumped off.  I reasoned that I'd bite that bullet when I got wind that Multisim was working, as opposed to bending over backwards to download an upgrade that large and still have problems.  Yes, I realize that sounds harsh, but if you only knew.  But, that said, I do still thank you for explaining the upgrade nomenclature.)

I'll see what I can do.

I'd like to understand something.  Why am I the bad guy when all I want is a piece of software that's supposed to work to actually work?  Why is that?  Why am I wrong to think that I should not have to work to fix problems with software which I am in no way responsible?  Why should I be expected to feel guilty for giving feedback when I get little and am left hanging when I need a solution to a problem?  If I'm responsible for anything, I'm responsible for getting Multisim to the point it's at, thanks to all the feedback I've given to Multisim over the years!  It is this mindset that users and customers should have to work for Multisim that causes the need for others to have to download a 290.218 MB upgrade.

       

 

[_user32]

Euler's Identity,

We would be happy to send you the CD for the upgrade rather than have you download this.  Please let me know and I will locate your address information and send the CD directly to you.

Also I am posting this circuit so that others can try out to see if the probe is working. 

I was able to corrupt the circuit using the DMM instrument and then inserting the probe would cause same error, however starting from your original circuit from scratch and then inserting the probe is working for me.

Regards,

Patrick Noonan

Business Development Manager

National Instruments - Electronics Workbench Group 

[Euler's_Identity]

Yes, I would like the CD sent to me immediately.  Please send it quickly, such that I get it next week.  In fact, bill me for the shipping expense if the expedience cost cannot be justified on your end. 

About the circuit, that's been my experience.  I'll have a working circuit and then something happens which takes out prior work, forcing recreation that is just shy of completion.   The best I can say is that it seems to involve the placement (the use) of the LF347 and the multimeter, but it's a corruption (as you've said) such that simple removal of these doesn't always yield a successful simulation.

I wish I knew what I did that caused this.  I just built and built and then whammo!  Then there was no dependable path backward.

Concerning hierarchical block usage, you say the subcircuit would have been more appropriate for the mux's.  How will I know when I should or should not use a hierarchical block in the future?  As I see them, integrated circuits are hierarchical blocks, circuits previously designed and used as blocks with inputs and outputs.  Whereas a subcircuit is something smaller, like an optoisolator or a few-stage amp.  In no way did I get the impression that the 1:16 mux should've been a subcircuit.  It's a gray area.

Later I intend to try out the circuit you've uploaded.  However, are you saying that the multimeter won't work where the probe does, or could I go with your circuit and a multimeter?

[_user32]

Dear Euler's Identity,

I will send the CD so that it will get to you in the next few days.

Also, the circuit I sent with you has a probe in it with the precision set to 9 digits.   If it doesn't work, please don't spend any additional time on this.  I would also like others in the forum to try my previously posted circuit to verify this simulates.

As to the actually issue, I am having R&D look at the before and after - so that we can determine what is happening when the DMM or OpAmp is placed, and the details on what is causing the corruption.  I will report the results but it may take a few days.

Regards,

Patrick Noonan
Business Development Manager
National Instruments - Electronics Workbench Group