That's where rapid communication comes into play. My layouter and I are in constant email contact during layout and I always prepare a brief write-up for him. Like "U3, 6, 23, 15, 23 section swap ok but keep those in U15 within same chip". So he even knows before the layout what he can swap. Then at some point he sends me back what he did swap and I update the schematic. However, I tend to draw schematics "layout-friendly" so he usually finds that the allocation is already close to what he'd have liked. Hindsight section swaps in an Eagle schematic are a breeze, just a few mouse clicks.
Yes, but I never do that myself because I don't do layouts. However, others have said it's pretty easy, AFAIR you just have to reconnect in the schematic so it looks clean.
I find them to be a bit on the expensive side so I haven't used tiny logic yet on anything that went into production. Most of my stuff is quite cost critical.
One more problem with tinylogic: flimsy package. It is easy enough to lift a pin of a generic SC or SOT, but if you try to bend a pin of tinylogic it breaks off immediately. Very annoying when debugging a hardware :-)
Vladimir Vassilevsky DSP and Mixed Signal Design Consultant
If you ever choose to move to southern Oregon we've got plenty of projects that don't have particularly tight cost constraints. :-) (Although as far as tiny logic goes, we generally use the "universal" NC7SZ57/58 parts whenever we need a 2-input/1-output device.)
(We're still looking for an RF engineer and a tech writer, last I heard.)
Try SC-75 diodes. They break off when you sneeze within five feet. Had the "pleasure" to reverse-mount BAT199 because the layout CAD had pins 1 and 2 reversed. AFAICT this whole EDIF standardization attempt was a complete joke. There simply is no standard.
My wife already thinks it's getting too cold here in northern California. Every year seems to be colder than the previous one. We have "migrated" from two cords of wood to five :-(
Those UHS gates sure are nice. I like their wide VCC range. Very zippy, too. Unfortunately they can't be used single cell, AFAIK they require at least 1.7V or so. Also, I believe there isn't a 2nd source (yet).
I only know an FPGA guy who might be available (not sure if he still is). Sometimes I get the impression that most RF engineers are retired by now. Or worse ....
I'm sure Jim wouldn't mind some company either then. :-)
As far as I can tell, most RF engineers are now doing IC design work and it's those who are used to more "distributed" systems that are pretty hard to come by these days.
Oh, I'd sure like it down there. But when we move I want to make that a move to a state that is very biz-friendly. Less red tape, less biz taxes. Problem is that where most of those states are the winters are rather harsh.
Still looking for that tropical island with reliable Fedex service ....
Yes, in RF chip design I even met some young and bright minds right out of university. But the minute you present a real-world tapped inductor that is controlled by easing its core in and out of DC saturation nearly all of them freak out.
Assuming that the environment continues to maintain quality standards.
Its also useful to review the output of the process a few steps down the road to see if what you had intended is actually what's getting built. Or if something about your design could stand some improvement to get a better product.
Why not? I've seen many intricate designs come together perfectly with the team splattered out across half a dozen states, and occasionally countries. I had one client where the boss jetted up here. For no particular reason other than wanting to see this guy that did their designs for so many years.
Sometimes it's better to take a risk than to hold out for the perfect candidate who may never materialize.
The processor development team was spread over four states and three continents. It can work, but communications is more difficult.
I did a couple PCB designs several years back where the contract assembly guy was 150mi. away. I drove down to see him once or twice a week to make sure things were going right (and get appropriate priority). It worked out very well.
I told John Larkin that I was available a while back but wasn't willing to move to SF. He said that design was a full contact sport. Some see it that way, unfortunately.
Time zones can get nasty. But it can be done, especially if both sides are willing to scoot their work times so the overlap is larger. Three continents can become an issue though.
My layouter and I have done lots of layouts. Last time we met was at his office when his cat was still alive, probably about 10 years ago.
John and I seem to agree on most things but not on that one. Often when team members from my projects finally do meet it's by coincidence. At a conference, for example, or an airport. Then you can hear expressions like "Man, your hair's turned gray!" or "What's that accent you've got?". Yes, there have been folks who designed the other half of the board but whom I've never spoken to until we finally met. It was all emails, files and FTP.
What's really cool is VPN. A business visitor used it from here. Also I've seen it at another client this year. At that client it was a bit spooky, there was a huge monitor behind me and the cursor seemed to be operated by a ghost (from hundreds of miles away). I had to sometimes darken the room to see noise on an analog scope. That made it even more spooky ;-)
You are not alone. The PCB is an unique component in the RF system; PCB's I'm familiar with often contain a lot of printed filters and inductors.
I work with a lot of engineers who themselves layout PCB's, do RF design, and DSP development. Sometimes in the same day. I usually do my own PCB layout in the RF signal chain on anything above 3 GHz.
It's a problem but fortunately all three didn't have to get together at the same time. The German folks and the Indians worked on different aspects of the design. Most communication was via email with a status telecon once a week.
You killed his cat? No wonder he hasn't had you back. ;-)
I can understand his point. Close is more efficient but there are a lot more fish in the ocean than the "Bay". It's his nickle, though, and I can respect that.
It certainly is strange how distorted mental pictures can be.
I took a class via VPN about a year ago. Quite cool.
Sure, no problem. Twice every other week (usually Mon/Thurs) and once the odd week (usually Tuesday or Wednesday). It's only a two hour drive each way and a night in a hotel. No big deal.
I believe the idea is that they're looking for full-time employees and there's a perception that "full-time" means largely parked at a desk. I am somewhat speculating here, though; personally I think a large amount of work could be done off-site, although occasionally even a tech writer would need to be on-site so that he or she could sit around and play with The Real Hardware (which tends to be big rack-mounted equipment surrounded by test equipment).
My fiance and I have found that southern Oregon is that it is perhaps a decade "behind" when it comes to things like "employment expectations." She has it far worse than I do -- where she works, even the salaried employees (who mainly prepare reports, crunch numbers, etc. -- nothing that couldn't be done (a) on a flexible schedule and (b) off-site in nearly all cases) are required to be there 8-5, no ifs, ands, or buts. Any deviations from that schedule has to be taken as vacation or sick leave; they're not allowed to make up hours on weekends or in the evenings, and at least in theory they can be terminated if they're more than 15 minutes late more than a half-dozen times in a year... aieee!!! The pay is quite respectable though, which is uncommon down here (she kept her job in Portland for about a year while looking): jobs that pay over, say, $50k/year are nowhere near as common as they are in Portland, yet the cost of living is almost as much. Housing is just as much, gasoline is a little higher, food and medical is about the same, although veternarians are noticeably cheaper.
As for me... the folks I work for are quite generous and reasonable. Time-wise, they go with the notion that there are "core" hours of perhaps 10-3 where it's expected most people will generally be around, but they have no qualms about people taking time off in the middle of the day to, e.g., visit a health club. Coming in late is fine, making up time on weekends is fine, etc. It's the usual routine that I've found at nearly all the engineering companies I've worked for: Actually getting stuff *done* is a far more important than what your daily schedule is.
We have used consultants -- sometimes for many months in a row -- for getting jobs done where we find we don't have the manpower or the expertise necessary. Most of the good ones we've used I think we'd generally be more than happy to have as employees, but of course most of them aren't interested in that option. :-)
Hmmm... I work full time, and I'm largely parked at a desk. However, the desk doesn't happen to be in the same timezone as anyone else in my group (we're spread across a couple of continents).
Bonus! This year I got to meet with my boss in person once. First time in many years, since he doesn't even live in the same country as I do.
"BJ" skrev i en meddelelse news: snipped-for-privacy@e9g2000prf.googlegroups.com...
In former times I would do prototype PCB's of the tricky bits, test them and glue them onto a big piece of wood to make up the board/system ;-). Then show and explain the tricky areas to the "PCB-person". The person then gets back if there is a problem and I would pop in for quick chat every 3 days or so. Worked Ok.
Yep - Because the "PCB-persons" know stuff that I do not care to know!! Like which solder processes to use, how to correctly orient and pad for reflow/wave soldering, components where the "canned" layout does not work, things that will confuse Production or Test .... and lots & lots of other itty-bitty details that I would surely also have missed!
"Joel Kolstad" skrev i en meddelelse news: snipped-for-privacy@corp.supernews.com...
Don't ask - *do* ;-)
Most decent engineers I have been around *scorn* the formal process for prototype work because it totally breaks the flow to wait for weeks and weeks for desicions to be made by the mis-informed - and the boss usually will not care if you can sneak whatever-it-is-you-need onto a small value order (or the credit card - like most evaluation boards and development systems these days; Microchip really broke the mold with the PIC development system at USD 80,--).
As long as you get results you will get away with it.
PolyTech Forum 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.