Breaking News
Blog

Why engineers love/hate; embrace/shun reference designs

NO RATINGS
View Comments: Newest First | Oldest First | Threaded View
Page 1 / 3   >   >>
kendallcp
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
kendallcp   4/26/2011 9:59:22 PM
NO RATINGS
The function of a reference design is right there in the name: it's a design to which you can refer. It's a design which is characterized and, if the manufacturer is worth its salt, is repeatable and representative of what can be achieved with the part. It may not do what you want, or do it well enough; that's for you to figure out, when you feel up to taking things to the next level. It should provide some basic didactic elements for the junior engineer to learn from, but it's not the place for smart-ar53 tricks or tweaks in the "wow, you can do that?" category. Do reference designs stifle creativity and differentiation? Only if our education and training system starts failing to deliver engineers who can take on the challenge of rising _above_ the level of the basic reference design and seeing something their peers have not.

the real biff44
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
the real biff44   3/29/2011 7:05:33 PM
NO RATINGS
I find the reference designs to be essential. It is not that I am lazy and can not whip up my own design. It is that the data sheets are often poorly written, with important facts missing. You read the data sheet, you have technical quesitons, then you look over the schematic and layout in the reference design and say "aha! It must be working like so". It also gives you a baseline to go back to when all else fails: circuit sitting there oscilating...go back to the capacitor/inductor/resistor values of the reference design to see if it works then. Yes--you have to recalculate your design for some reason, N0--you have a layout issue, like a via ground that is not connected, etc.

Salio
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
Salio   3/28/2011 3:30:04 AM
NO RATINGS
I think reference design are usually a good starting point. However, they should be used reference only and not absolute design. They should be taken with a grain of salt modified or changed completely based on the design requirements of the task at hand.

jcdrisc
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
jcdrisc   3/26/2011 9:31:05 AM
NO RATINGS
I would echo what kdboyce has said and add : Re analog design issues I trust ALL the appnotes written by industry veterans Bob Pease (ex NSC) and Jim Williams (Linear Technology) I think most good Companies will not issue an app note unless it has been well tested in their labs. After all, they are a marketing tool, and you EXPECT them to work properly !! Evaluation boards similarly.

Sanjib.A
User Rank
CEO
re: Why engineers love/hate; embrace/shun reference designs
Sanjib.A   3/25/2011 5:15:13 PM
NO RATINGS
A few things to be careful about, before using/ incorporating a reference design into the design: - I'd be careful about the reference design for a newly launched component. A new reference design might have a bug! I trust the reference design, which are there for some period of time (proven in use). - I don't think every company test their reference designs for EMC (Electro Magnetic Compliance); So I'd not copy before I evaluate the design. Otherwise the reference designs are very handy for projects with tight schedule commitments.

agk
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
agk   3/25/2011 3:25:37 PM
NO RATINGS
The reference designs from IC manufactureres or the application notes are quite handy many many times. Just reading the chip pin details and their ratings is like hearing a news where as the application notes or reference designs is equal to seeing real model. These help me and make me happy.

MDLX-LD
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
MDLX-LD   3/24/2011 11:12:17 PM
NO RATINGS
From an RF board design point of view we'd like to ask, what if you had a validated simulation model to go with the reference design? That is for those of you who are RF designers would you like to see more examples like those you an find in the following papers? Accurate Models Simplify Reference Designs for RFIC Amplifiers http://www.modelithics.com/paper/228.pdf Non-Linear Simulation of RFIC Amplifier Reference Design Boards http://www.modelithics.com/paper/783.pdf We'd be interested also to see how many would like to see more simulator-ready (and therefore adjustable) reference designs of this basic nature. Larry Dunleavy, President, Modelithics,Inc. Reply to info@modelithics.com

PEIFFER
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
PEIFFER   3/24/2011 4:55:11 PM
NO RATINGS
My experience with reference designs has been quite varied. In some cases, you have no choice but to follow them blindly, bucause the manfacturer can't or won't provide enough information for you to do anything else. In other cases as an early adoptor, I've have seen the refernce design created to fit my specific requirements. You would think these would be quite good, but it isn't necessarily so. The reference designer's goal might have been to get a single unit up and running at room temperature. If that's the case, you are almost guaranteed to run into trouble. Auxillary cuircuits used to support the chip being marketed should never be trusted without careful analysis (e.g. the power supply circuit on a microprocessor reference design is usually worthless, but the memory bus layout is likely quite good). It's a good idea to understand what the reference curcuit designer's objective was before blindly copying anything.

ttt3
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
ttt3   3/24/2011 2:37:48 PM
NO RATINGS
I love reference designs, especially example schematics and layout for complex ICs. They often contain information not covered in the datasheet - you know, the little schematic notes, undocumented pins, etc. Also, they provide a great reference during debugging. When I encounter a problem on my own board design, I sometimes will immediately try to reproduce it on the reference design, to see if it's something I introduced , or perhaps a flaw with the IC (or reference design itself). Also, chip vendors usually take you a lot more seriously when you reproduce problems on their reference design (and are less likely to point fingers at "your" layout, circuit, etc). While I don't think it's a good idea to blindly copy reference designs, they are without a doubt extremely helpful. However, I have a colleague who is of the complete opposite opinion; he usually does not want anything to do with the reference design.....

DutchUncle
User Rank
Rookie
re: Why engineers love/hate; embrace/shun reference designs
DutchUncle   3/24/2011 12:51:37 PM
NO RATINGS
Reference design *software* libraries tend to be bloated and suboptimal. It's hard to be all things to all people, of course, but it's also part of the HW/SW training difference. As a simple example, HW engineers writing software tend to duplicate things they needn't, simply because hardware would need to be fully duplicated in the same situation and that's the way they're trained to think. So while the reference is great to have - presumably it was used for the test development and is known to work - in a performance project it is only a baseline for improvement.

Page 1 / 3   >   >>
August Cartoon Caption Winner!
August Cartoon Caption Winner!
"All the King's horses and all the KIng's men gave up on Humpty, so they handed the problem off to Engineering."
5 comments
Top Comments of the Week
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Flash Poll
Radio
LATEST ARCHIVED BROADCAST
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.