Breaking News
Comments
pierreohm
User Rank
Author
EE
pierreohm   8/1/2014 7:54:40 AM
NO RATINGS
fascinating report Glen, blame the io esd :) for the sram error rate 

It seems most epic work gets done when various skilled people team-in as long as the analog ressource has more character and skills than ego and the digital guy is fast and verstalile in coding stuff out of his comfort zone.

After all it makes you wonder if sales&marketing driving the engineers crazy while keeping the communication channel open is one recipe for success.

 

 

MeasurementBlues
User Rank
Author
Re: EE
MeasurementBlues   8/1/2014 9:32:07 AM
NO RATINGS
When I was at GCA, the standard way of doing business was to sell a product with anunreasonable delivery date, then tell engineering.

Company is long out of business.

zeeglen
User Rank
Author
Re: EE
zeeglen   8/1/2014 8:30:51 PM
NO RATINGS
Hey Martin, I once worked for a corp like that.  The marketeers would wait until someone else proved a concept and started making a few bucks off it, then would jump on the bandwagon and tell engineering they had to deliver in 6 months or miss the market window.

To which the standard eng response was "Then why didn't you morons ask for it 6 months ago?"

And yes, that place too went down the crapper.

David Ashton
User Rank
Author
Nice work
David Ashton   8/1/2014 9:20:32 PM
NO RATINGS
Love your work Glen. Really good faultfinding.  Goes to show - Analog skills will never really go out of fashion!   We deal with similar boards in PABXs and suchlike and I wouldn't even think of deliving as deep as this, we just junk them or return them to the manufacturer for repair if they go wrong. 

AZskibum
User Rank
Author
Re: Nice work
AZskibum   8/3/2014 3:08:29 AM
NO RATINGS
The solution to some problems requires both analog & digital expertise -- either an analog engineer & a digital engineer, or a "digilog" enginer :)

zeeglen
User Rank
Author
Re: Nice work
zeeglen   8/3/2014 6:40:48 PM
NO RATINGS
@AZskibum either an analog engineer & a digital engineer, or a "digilog" enginer :)

Somehow that terminology sounds better than "analdig", which is more like something one might experience in a doctor's examination room...


kfield
User Rank
Author
Re: Nice work
kfield   8/4/2014 2:36:34 PM
NO RATINGS
@glenchenier "Somehow that terminology sounds better than "analdig", which is more like something one might experience in a doctor's examination room..."

OMG - that was my first instinct too!

elctrnx_lyf
User Rank
Author
Re: Nice work
elctrnx_lyf   8/6/2014 6:41:03 AM
NO RATINGS
It owuld be great if microcontroller them selves shows up a message that wrong voltage level is applied to their pins instead of deep digging into the datasheet. I hope the Microcontollrer companies are looking at this.

green_is_now
User Rank
Author
Re: Nice work
green_is_now   8/17/2014 12:50:16 PM
NO RATINGS
Or even better

Analogic

But that's all ready taken...by management

docdivakar
User Rank
Author
Re: Nice work
docdivakar   8/8/2014 2:09:57 PM
NO RATINGS
@AZskibum I too like "digilog" engineer title, how ever the word is already in vogue. There are many businesses with "digilog" in their title.

There is also another word sysnthesis like "Anadigit" engineer, sounds close to Anadigics, a struggling RF Devices company (stock @67 cents!).

MP Divakar

AZskibum
User Rank
Author
Re: Nice work
AZskibum   8/11/2014 10:02:43 AM
NO RATINGS
The more traditional title would probably be "analog/mixed-signal engineer," but that one has the connotation of an analog engineer who also knows digital. There are many engineers who might be better described as "digital/mixed-signal."

j_b_
User Rank
Author
Sorry
j_b_   9/1/2014 2:18:32 PM
NO RATINGS
Sorry, but I really think it was a bad decision to use that PIC board in the first place. Since you had to order other material anyway you could have ordered some really working and proofen hardware with enough I/Os and simple programming model. That way you would have eliminated all the risks which inevitably come with such unprofessional homebrew boards as the one described. I mean - even the soldering of the missing caps took more time than the ordering of a nice SoM. OK, OK, it's always easy in hindsight. I agree. But really the quality of the PIC board should have let to doubts concerning its developer. My experience - do not attempt to go cheap when doing something which is complicated on its own. Use the right tools.

zeeglen
User Rank
Author
Re: Sorry
zeeglen   9/2/2014 10:19:20 AM
NO RATINGS
@ j_b_ OK, OK, it's always easy in hindsight.

Yes it is, and with no reason ahead of time to anticipate any problems the beauty of the PIC board was that several were already in house and ready to use, with no additional cost. The physical interface to the SIMM had to be constructed regardless, no matter what I/O board was used. One cannot solve any resulting problems until they crop up and make themselves known.

j_b_
User Rank
Author
Re: Sorry
j_b_   9/2/2014 10:33:44 AM
NO RATINGS
Surely you're right. It's just that I have seen things like that more often than should be expected ... instead of planing first and then using what's best there was some stuff at hand and that was used because one could start right away and continuing down that road one compromise led to another and when, after some considerable time, people looked back and wondered how this little problem could have cost so much work it turned out that a cleaner approach with "real" tools and a bit more planing could have led to a direct solution. Of course any real-life situation carries its own logic and limitations, so - sorry if I sounded judgemental, no offense intended.

zeeglen
User Rank
Author
Re: Sorry
zeeglen   9/2/2014 12:23:28 PM
NO RATINGS
@j_b_ It's just that I have seen things like that more often than should be expected

I as well, no offense taken. That is one of the risks that need to be juggled when making this sort of decision.  Many times things do go right on the cheap, so the increased cost of better tools would have then been the wrong decision.  "Do more with less..."  If the successes outweigh the failures then one is still ahead in the long run.

Bad decisions do impart the accumulated experience needed to better judge the difficulty level and try to make a more accurate prediction the next time.  Or at least not get bitten by the same bug twice.

Of course,there are always those situations where one knows the tools are wrong.  But then one must convince management to spring for the cost of better tools.  ("Well, you came through last time with that same scope, why do you need a new one for this project?")



Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Radio
LATEST ARCHIVED BROADCAST

What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.

Brought to you by:

Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...
Like Us on Facebook
Special Video Section
Once the base layer of a design has been taped out, making ...
In this short video we show an LED light demo to ...
The LTC2380-24 is a versatile 24-bit SAR ADC that combines ...
In this short video we show an LED light demo to ...
02:46
Wireless Power enables applications where it is difficult ...
07:41
LEDs are being used in current luxury model automotive ...
With design sizes expected to increase by 5X through 2020, ...
01:48
Linear Technology’s LT8330 and LT8331, two Low Quiescent ...
The quality and reliability of Mill-Max's two-piece ...
LED lighting is an important feature in today’s and future ...
05:27
The LT8602 has two high voltage buck regulators with an ...
05:18
Silego Technology’s highly versatile Mixed-signal GreenPAK ...
The quality and reliability of Mill-Max's two-piece ...
01:34
Why the multicopter? It has every thing in it. 58 of ...
Security is important in all parts of the IoT chain, ...
Infineon explains their philosophy and why the multicopter ...
The LTC4282 Hot SwapTM controller allows a board to be ...
This video highlights the Zynq® UltraScale+™ MPSoC, and sho...
Homeowners may soon be able to store the energy generated ...
The LTC®6363 is a low power, low noise, fully differential ...