Ping John Larkin: Question on MAX
in [Design]
|
From: Joerg on 1 Jun 2010 14:05 Joel Koltner wrote: > "Joerg" <invalid(a)invalid.invalid> wrote in message > news:86kn17FktaU1(a)mid.individual.net... >> Yeah, but when an employee quits that's usually like a divorce. He or >> she is gone. > > I've seen it both ways... I guess it comes down to both what the > engineer and the company expect, and this is often the kind of thing > that isn't ever written or formally verbalized: Some companies seem to > take an engineer quitting as a form of betrayal and will start > bad-mouthing his work and not want any contact and effectively blacklist > him from being hired again, whereas other companies figure that if an > engineer wants to check out whether or not the grass is greener > elsewhere, if he does so during a lull between projects so that there > isn't any huge disruption and stresses that he's happy to have people > call him to ask for support on some old project he did, there's no need > for bad feelings and the perhaps the guy will be invited to come back if > the new place doesn't work out. > > The later might happen more in non-technical businesses, perhaps? -- My > wife still does a few little side-projects for several of the places > that she's previously worked for. > That's the way to go, she probably has good communication skills. And I am not saying that consultants are always the cat's meouw just because I am one. Once as a manager I did have to fire a consultant, for lack of communications skills. He began fingerpointing and all that, made my engineers unhappy and that just wasn't going to continue. Strange thing is, years before that he used to be an excellent communicator. -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
From: John Larkin on 1 Jun 2010 18:15 On Tue, 1 Jun 2010 10:27:26 -0700, "Joel Koltner" <zapwireDASHgroups(a)yahoo.com> wrote: >> All our resistors and caps are... believe it or not... IN ORDER BY >> VALUE. > >Is that because you wrote a smart sorting routine that understands metric >suffixes? Or because internally you store the raw value so sorting is trivial >(e.g., 100nF is stored internally as 100e-9). We have a document that tells how to assign stock numbers to parts. Resistors and caps wind up being in value order. In part number 104-1600, the 104 specifies an 0805 ceramic cap; 160 specifies 100 pF, per an algorithm. The last digit allows us to have up to 10 different 0805 100pF parts. We currently have two, 1% and 10%. So you get stuff like... 131-6851 RES 0603 76.8K 1% 131-6891 RES 0603 84.5K 1% 131-6921 RES 0603 90.9K 1% 131-6931 RES 0603 93.1K 1% 131-7000 RES 0603 100K 5% 131-7001 RES 0603 100K 1% 131-7002 RES 0603 100K 0.05% 131-7051 RES 0603 113K 1% 131-7170 RES 0603 150K 5% 131-7181 RES 0603 154K 1% 131-7201 RES 0603 162K 1% which aligns part number with stockroom ordering with value. John
From: Joel Koltner on 1 Jun 2010 18:24 "John Larkin" <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in message news:101b06hlvadd656tnr2o6f073fpf81tjnu(a)4ax.com... > We have a document that tells how to assign stock numbers to parts. > Resistors and caps wind up being in value order. > > In part number 104-1600, the 104 specifies an 0805 ceramic cap; 160 > specifies 100 pF, per an algorithm. The last digit allows us to have > up to 10 different 0805 100pF parts. We currently have two, 1% and > 10%. Ah, interesting. ...although I might be tempted to have an algorithm spit out something more like XYYY where the component value is then 0.YYY * 10^X or something (so 113k->6113, 1.12k->4112, etc. -- this still sorts correctly), but I expect you guys kicked around such an idea and decided your way was preferable. Neat...
From: krw on 1 Jun 2010 18:37 On Tue, 1 Jun 2010 10:28:32 -0700, "Joel Koltner" <zapwireDASHgroups(a)yahoo.com> wrote: >"Nial Stewart" <nial*REMOVE_THIS*@nialstewartdevelopments.co.uk> wrote in >message news:86k3j4FrsoU1(a)mid.individual.net... >>> being tied to the database has the advantage that you can choose >>> components you already have >> >> ...and have double sourced and have confirmed footprints! > >Your purchasing peoples' lives will surely become way too boring if you don't >strive to single-source as many parts as possible. > >:-) > >Just kidding, double-sourcing is of course great when it's an option. A big deal is often made about second-sourcing but in reality it's not possible for most things, outside of passives, discretes, and a few unit-logic glue things. So much of a BOM ends up being single-sourced that it's silly to pretend that a multiple-source policy accomplishes much. Yeah, we give it lip service, too.
From: krw on 1 Jun 2010 18:42
On Tue, 01 Jun 2010 08:24:15 -0700, Joerg <invalid(a)invalid.invalid> wrote: >krw(a)att.bizzzzzzzzzzzz wrote: >> On Sun, 30 May 2010 09:34:02 -0700, Joerg <invalid(a)invalid.invalid> wrote: >> >>> krw(a)att.bizzzzzzzzzzzz wrote: >>>> On Sun, 30 May 2010 08:38:54 -0700, Joerg <invalid(a)invalid.invalid> wrote: >>>> >>>>> krw(a)att.bizzzzzzzzzzzz wrote: >>> [...] >>> >>>>>> ... Even when I was >>>>>> contracting (and paid big-$ overtime), management didn't appreciate my ROI >>>>>> calculations on a second monitor (less than a week ;). >>>>>> >>>>> Sometimes they should look, they'd be surprised how much lower in cost >>>>> projects can come in with external engineers. >>>> Just add up the cost hiring of employees. The down side is that sometimes >>>> contractors with the appropriate experience are hard to find, particularly >>>> ones willing to work on-site. >>> >>> Yes, they are often hard to find. But they can be more loyal than >>> employees. If an employee quits after a number of years he's gone. He >>> can't be there for you even if he wanted to because he's got a new >>> fulltime job, plus a family, and so on.. A consultant or contractor can >>> usually be called in again. I had clients from the early 90's who called >>> me in on a problem 15 years later. The fact that I have moved across an >>> ocean in the meantime didn't matter, if you want to be found you can and >>> will be found by them (I gave all of them my new address, just in case). >> >> Consultants are more loyal because they can be bought. ;-) Employees really >> want to be part of the family. They're going to live there, they might as >> well be. >> > >Yeah, but when an employee quits that's usually like a divorce. He or >she is gone. I just had that happen with a company where one of their >key engineers on the project quit and it did cause noticeable project >delays. Certainly, companies tend to rely on a few people too much and documentation too little. Then under-pay them. OTOH, it's also possible that they'll get run over by a bus... >>> On-site, yep, that can be a problem. But when one thinks hard about it, >>> how many times is that really really necessary? Most of the time clients >>> just send me the whole chebang to diagnose in my own lab. Other times I >>> fly out there for a few days. Sometimes it's a long trip, like one to >>> Korea but that was necessary because I needed to train their engineers >>> so that the EMI problems they had wouldn't happen again on their next >>> design. >> >> Tell that to managers. They like to think they're in control. They like to >> think having someone on site puts them in control. I suppose if you don't >> trust the person, at least you can make sure they put in the hours. >> Conversely, being off-site can have its problems, too. The big-picture can be >> missed, or expectations can be completely unrealistic. >> > >That's why consultants must be great communicators and generalists. >"What happens if the signal at port B doesn't .." ... "Oh, dang!" A lot of that is just another set of eyes. That's why good design reviews are so important. I did that several times, even during interviews. |