What's missing from manufacturer websites?

Why don't manufacturers don't offer device libraries? Surely they've got the most to gain by making it easy to drop-in their products. Maybe they don't think it's a factor in product selection; maybe they're right. (I'm not just talking about the package footprint, but the pin assignments, ERC attributes, etc.)

It sounds common to set aside a library of "trusted" devices that survived a board spin without a glaring defect. I don't know that I'd change this approach, but I'd trust a library from the manufacturer more than one created from a datasheet by me or the EDA vendor.

What about devices for Spice models? I'm not a Spice user (yet), but this is something else that's traded on the underground instead of just provided by the manufacturers...

I talked recently with the (marketing) guy that owns the website for a large mfr. They're re-designing the site to be more "designer-centric", and shoving the marketing / corporate glitz to the background, which seems like a smart move. They claim to be doing the whole thing based on focus groups with real designers, supposedly even ignoring internal opinions. I'll be interested to see the result.

I mentioned the idea of providing device libraries for download, and he swears not one person mentioned it as a nice-to-have during their research. Is this just so status-quo that nobody thinks to ask for them, or am I missing something?

Richard

Reply to
Richard H.
Loading thread data ...

[snip]
[snip]

I suspect that the use of simulators is NOT as universal as you might think... particularly with designers using off-the-shelf IC's.

The heavy-duty users of simulators are mostly microchip designers.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.
Reply to
Jim Thompson

This is partly a positive-feedback effect. Op amps and so forth do exactly what you expect except when they don't, and the crummy macromodels that manufacturers shovel out are no help, because they crap out even sooner. They don't predict the details of slew limiting correctly, nor CMR problems, just for a start. Last time I used them (admittedly awhile back) only National's included input capacitance, and none had good noise models. I did once extract a typical value for input capacitance for an LF357, which wasn't in the datasheet, but that was by *reading* the macromodel, not by executing it. Still, that shows that they're not *entirely* useless.

When pressed, the chip house will always say that a transistor-level simulation would run too slowly. This is of course a fib, because they've been saying the same thing for 15 years at least, during which time op amp complexity hasn't increased that dramatically, whereas computing power has got quite a bit cheaper. The reality is that device-level models would reveal more about the chip internals than they'd like. Device level simulations are quite different, but those are mainly used by chip designers, as you say.

The result is that the usual models are useless for simulations requiring accuracy, so their only remaining use is as a substitute for thought.

Cheers,

Phil "do the algebra" Hobbs

Reply to
Phil Hobbs

I'd have been happy yesterday if Epcos's site recognised the part number I'd got from a distributor !

After some guesswork, I found a similar series part with a part number 1 digit out but 'not recommended for new designs'.

It's rubbish if the distributors have the new part but you can't find the data for it on the manufacturer's site .

Graham

Reply to
Pooh Bear

Manufacturers are driven by that ONE custeomer who is going to buy 10 million units. That customer will get right royal treatment. Application engineers will design and debug the circuit for them, create device libraries etc. etc., and wipe their arse too. The rest of us are lucky to be allowed to buy a few.

Paul Burke

Reply to
Paul Burke

Hello Paul,

Companies that subscribe to that mantra usually get choked to death soon by a few of those "select customers", for example from the cut-throat automotive sector. Ignoring young start-ups can lead to a quick puff of smoke. On the stock market.

There is a reason which TI and others are doing well while Infineon often complains about their biz.

Regards, Joerg

formatting link

Reply to
Joerg

Hello Graham,

With European companies I found that to be pretty common. A few times I had to Google for one of their parts and then could only find the data sheet on some Russian site. Pathetic. IMHO a lot of the top S&M staff at those companies needs to be let go and replaced. If I would be in charge there wouldn't even be severance payments beyond what is required by law.

Regards, Joerg

formatting link

Reply to
Joerg

Slightly OT, but I find connectors are a PITA,recently I was trying to layout a PCB for an EDAC516(?) connector, why don't they post a gerber file that I can import into my PCB prog? preferably in the simplest minimalist format. DXF files seem to screw my ancient Accel /altium software.

Then I was trying to pick a D type from the Cannon range, jeez, 10 MBytes PDF file, 301 pages of options from the 1998 Dtype download.

I JUST WANT A D TYPE FOOTPRINT...........(that i can trust) Or that ferrite pot core pinout, with all the pins tilted at 45degrees to reality

martin

Reply to
martin griffith

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.