BTW, the problem is that you are using (or you received a schematic) with parts not in the PSpice libraries. the netlister is looking at that wire, and there is only one part there that has a PSpiceTemplate associated with it to connect to.
Charlie
Thanks, Charlie. And yes, I should have looked at your web site but didn't know you had PSpice hints on it. It's like seeing the correct freeway exit from the left lane at 65mph. "Dang, there it is, should have ..." :-)
This may be useful for other, too: If you do a min-max simulation does it really show the absolute limits (tolerances and offsets assumed at their respective worst cases)? I know it varies only one parameter at a time and then sets it to the respective worst case but does it always catch all?
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Yesterday night I figured that out, by trial and error as usual. Puts green cloverleafs or something next to them. Which don't go away when you replace with a SPICE library part but that's another story ...
But why on earth can't it assume that when there is a resistor and it says 10k in the value field, that it is a 10k resistor?
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Doesn't Crapture have the feature where you can "archive" the "workplace" and pass to someone without your libraries? PSpice Schematics certainly does. ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
We always find these resources AFTER we beat our heads against the wall for a few hours...(or days!)
Yes, the worst case analysis goes to the limits. Basic description: System does a set of sims for sensitivity, i.e. which was does varying the value of this component change the desired output. It then does a final sim with each component set to the limit of the tolerance in that direction.
Now, this won't show absolute worst case, especially where resonances and other tuned effects come in, but it can sure be handy. It can also be fun to first do the worst case, and then do a bunch of Monte Carlo analyses and see if any of the MC results go outside the bounds set by the WC. (You append the sim results from the WC analysis to the results of the MC!)
Charlie
In my case there won't be any sharp resonances, no tuned LC or any of that. If it really checks each and every component for impact then it would go to the limits. Probably one of those things that you'd want to let run overnight ;-)
Right now I am still learning the basics which seems to take a lot longer than with LTSpice. Things like zooming in the sim results. With LTSpice you just draw a square, let go, and whoopdidou. Not with PSpice. Somehow the old Microsim DOS version was easier. Like a stick-shift car without all the fancy on-board computers, you hopped in and drove :-)
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
In the '70s, a friend was designing an I/O Interface driver chip for an IBM box (don't remember if it ever made it to a product...). They demanded that he do statistical-transient[*] analysis (the norm was statistical DC on the appropriate levels and nominal and worst-case transient for timing) on it before they'd have the chip fabbed. He got special permission to take down one of the interactive (VM) mainframes on the weekends and reboot it to MVS to run his simulations as a single user. It took several months and *boxes* of paper to complete what they demanded; several hundreds of thousands (of funny-money) in computer time.
[*]All values and process parameters were described in terms of absolute, lot, chip, and geometrical tolerance distributions.I tried LT spice but couldn't find the starter. I had work to do, so used the crap I've been using.
I could walk you through an LT Spice circuit entry and sim run in about 5 minutes on the phone. I even use it for simple stuff, like resistive voltage divider calcs and RC charging curves, things I could do with a calculator.
John
[...]
Ouch! But my case is different, and nowadays computer time is almost free.
The starter is the little marathon runner with the sweat beads, on the top left :-)
I found it very intuitive, got the hang of it with very little manual reading. Even though I am not very good with GUIs and pictograms. The only things I had to study was how to do some fancy stuff with the Y-axis and things like that. IMHO it's the best SPICE ever.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...
Well, I don't use that function. But there is a hold time (t10
It was then well into six digits of funny-money. It was stupid then. My friend was the first to admit it, but he just sent the bill to the customer.
Like I said, I couldn't even get a resistor and a source hooked together. I've been using Tina, which has rather simple controls, unless you want to add a model. It takes me at least an hour to add a simple op-amp, through their convoluted process and then it rarely ends up in the library I want it. I have some simulations to do tomorrow but it's sort of an emergency (boards just came back and there is a strangeness that we didn't anticipate - need to model a fix) so I doubt I'll have time to try again.
to
lot,
the
In LT Spice I can draw a simple circuit and run a transient analysis in a couple of minutes. Adding one of the library opamps (all LTC parts, of course) takes a minute maybe, pulling in the part and a couple of power supplies. LT Spice is so easy that I often use it instead of a calculator and a bunch of algebra.
A few things are annoying:
The ideal opamp model doesn't work right out of the box. I generally use a VCVS instead.
There's no ideal diode; you have to make your own.
The "help" could be a lot better. Specifically, there could be more examples and less meta-language. Some things, like building parts with temperature coefficients, take a while to figure out their syntax.
John
the
Probably best to kick it around on a cold and rainy day. Of which we have a heck of a lot lately :-(
Since you have hardware hopefully you can just reach in and find the problem with bench tools. Sometimes I just build the fix and roach in on there.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
to
lot,
the
What about other parts? It does take SPICE models, right? How easy are they to integrate?
No biggie, as long as you can do a piece-wise linear I-V model.
I didn't find the help useful for getting started, even.
the
Not here. It's been beautiful. It's 79F and sunny, now, with highs in the
70s and lows in the 50s as far as the forecasts go. People say the Winter was cold (if highs 30s can be considered "cold") but there wasn't six weeks of it. By the middle of February it was in the 60s. ... I'veThe problem isn't fixing it, that's a simple BOM change. Rather, making sure the ramifications of the fix the other engineer wants to use isn't worse than the problem. It makes me nervous fixing an offset problem by loading the node down. I did all the simulations a couple of years ago (this circuit was lifted from another product) and I remember that the impedances are high for more than one reason. I can't remember all the reasons, though. ;-)
the
We have one of those winters that simply won't end. 40F was the top today and our firewood of over four (!) cords is now nearly gone :-(
With every year winter tends to become longer and colder. In retirement, maybe we should move to Arizona or some other place more warm. We both do not like winter.
... I've
I'd be nervous, too. Sounds like a duct tape fix.
That's why I am constantly preaching that people shall start writing the module spec before even firing up the CAD. Otherwise it's never done right, or not done at all. "Why was this transistor added?" ... "I dunno, Joe did that" ... "Well, can we ask Joe?" ... "No, he quit a year ago to sail around the world, he's somewhere on the Pacific right now".
For every project I ever did my clients received a module spec, whether they requested it or not. I just don't work any other way. This has saved the bacon many times. And when I am gone from this earth the next guys can read up on how it works. The VP Engineering at my first job used to say "If you didn't document it, it didn't happen".
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
that
MVS to
of
lot,
the
I've
Might this be helpful?
used the
...
to
A few years ago I was tasked with creating a full analysis and CAD package and specifications for a "design" that existed only as a schematic, no notes whatsoever ;-) And it was medical... the imaging end of an X-ray machine of some kind :-(
Took me about 6 months. I had to roll my own Spice models for photo-multipliers, etc, but I got it all working, matching their lab measurements!
Actually sort of fun... lots of WTF moments :-) ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Remember: Once you go over the hill, you pick up speed
used the
...
Our largest electric bill was $175 this Winter. It'll probably be $100 with no heat, so $2/day isn't bad. We did use some gas for the fireplace but we didn't use that much more without it. We just weren't as toasty when watching TV. ;-)
I really didn't mind Winter so much, until about March 1st. Then I was in agony until June. Now Winter is no big deal. There were only a couple of weeks where it was too cold to work in the attic (still building a shop up there).
You'll like the South. I don't know if I'd like PHX, but I like the weather here, even with the Summer humidity.
to
It is, sorta. The problem crept in because the other guy wants to get rid of the output transformers. The transformers masked the offset (I always knew it was there but it didn't matter). The test board worked because the shorted output cap did pretty much the same thing. The proposed fix is to lower the receiver impedance to load the driver. Ugh.
I'd agree but I can't even get a useful product spec.
I'm used to working with pretty detailed specs, too. The real world is different. Very different.
that
MVS to
of
lot,
the
What I do is plop a diode on the schematic and change its type to "DIDL" or some such dummy name. Then add a Spice directive to the schematic (there's an icon for that) somewhere nearby
.model DIDL D(Vfwd=0 Ron=1u Roff=1G)
As I said, I can walk you through a simple circuit, on the phone, in a few minutes.
John
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.