I have been sitting here for the past several hours laying out a PC Board and I realise that I absolutely DESPISE PC Board routing. I ran the Autorouter just for kicks and it routes itself into a corner... Usually does. And trying to force fit something never works so I end up routing the board manually anyway.
So, my Electronics Pet Peeve is routing PC Boards.
Any Freaks care to share their Electronics Peeve?
Jim
Jim,
I LOVE laying out pcbs; must be the puzzler in me.
My gripe if any is tolerating the customers who initially agree on the design approach including enclosure format and size and then after the prototypes have been done change their minds (yes plural) and want the design made smaller. I am thinking of one particular UK based client... grumble, grumble.
"Can you make it in the form of a kitten?"
|
"Experience is what enables you to recognise a mistake the second time you make it." "Good judgement comes from experience. Experience comes from bad judgement." "Wisdom is always wont to arrive late, and to be a little approximate on first possession." "When you hear hoofbeats, think horses, not unicorns." "Fast. Cheap. Good. Pick two." "We see a lot of arses on handlebars around here." - [J Ekdahl] |
 |
I LOVE laying out pcbs;
+1 ...in fact I would love any work at all....what am I saying? I'm a pensioner....
But my Electronics Peeve starts with a C, well not really electronics I guess.
I hope all my PCB laying prevents me from getting some form of dementia, it's as good as playing solitaire or doing jig saw puzzles or playing cards, not as good as dancing though.
People who don't properly document pinouts; giving you almost - but not quite - enough information to work it out.
A common one is 2-row connectors: pin 1 may be identified - but then no indication of whether pin 2 is the next in the "row" or the next in the "column".
My peeve, is when I tell a client that something they want can't reasonably be built and they reply, "All you gotta do is ...".
They don't seem to have an understanding of the laws of physics, regulatory compliance, or just common sense.
My peeve, is when I tell a client that something they want can't reasonably be built and they reply, "All you gotta do is ...".
They don't seem to have an understanding of the laws of physics, regulatory compliance, or just common sense.
Like I said:
"Can you make it in the form of a kitten?"
;-)
It's a common issue across the spectrum of human endeavour. Those who decide are, generally, ignorant. Their skill set includes decision making, not thing making.
William S. Burroughs has an excellent piece on this subject... "Sure, start fakin' it. Sure, and leave the details to Joe..." ;-)
|
"Experience is what enables you to recognise a mistake the second time you make it." "Good judgement comes from experience. Experience comes from bad judgement." "Wisdom is always wont to arrive late, and to be a little approximate on first possession." "When you hear hoofbeats, think horses, not unicorns." "Fast. Cheap. Good. Pick two." "We see a lot of arses on handlebars around here." - [J Ekdahl] |
|
+1 - Love laying out boards. Have never used autorouter. Do not trust them to get bypass or crystal traces short or long parallel traces that should not be long and parallel.
Pet Peeve? Complex devices (thinking wireless transceiver ICs) with gobs of registers and gobs of settings and explanations about how this works and that works, but nothing about setting it up for any of the standard operating modes.
Jim
My peeve, is when I tell a client that something they want can't reasonably be built and they reply, "All you gotta do is ...". They don't seem to have an understanding of the laws of physics, regulatory compliance, or just common sense.
It's a common issue across the spectrum of human endeavour. Those who decide are, generally, ignorant. Their skill set includes decision making, not thing making.
Certainly feasibility and practicality has to come in somewhere, and I rankle at edicts with y'all.
That said, nothing wrong with brainstorming and "blue sky".
Steps
- Defer judgement. You never know where a good idea is going to come from. The key is make everyone feel like they can say the idea on their mind and allow others to build on it.
- Encourage wild ideas. Wild ideas can often give rise to creative leaps. In thinking about ideas that are wacky or out there we tend to think about what we really want without the constraints of technology or materials.
- Build on the ideas of others. Being positive and building on the ideas of others take some skill. In conversation, we try to use “and” instead of “but.”
- Stay focused on the topic. Try to keep the discussion on target, otherwise you can diverge beyond the scope of what you're trying to design for.
- One conversation at a time. Your team is far more likely to build on an idea and make a creative leap if everyone is paying full attention to whoever is sharing a new idea.
- Be visual. In live brainstorms we write down on Post-its and then put them on a wall. Nothing gets an idea across faster than drawing it. Doesn't matter if you're not Rembrandt!
- Go for quantity. Aim for as many new ideas as possible. In a good session, up to 100 ideas are generated in 60 minutes. Crank the ideas out quickly and build on the best ones.
Obviously this can go back-and-forth forever. Finding the best balance point is tricky.
My peeve, is when I tell a client that something they want can't reasonably be built and they reply, "All you gotta do is ...". They don't seem to have an understanding of the laws of physics, regulatory compliance, or just common sense.
If a scientist tells you something is possible he is probably correct
If a scientist tells you something is impossible, there is probably another one somewhere already working on it.
Mine is:
Companies discontinuing chips without providing a reasonable alternative (Xilinx!!).
I'm with the guys who like laying out circuit boards. I have the Eagle autorouter (expensive as it was) and I occasionally let it take a swipe at a board to see what it can do, but I've NEVER shipped an autorouted board. I always rip them up and do them again by hand.
S.
+1 on the layout
love the puzzling and see that something that seems to not fit when you start ends up with room to spare.
An autorouter is just an automatic way to route yourself to the asylum. Routing by hand let's you make the call when important considerations need to be implemented--it's possible to set up a lot of routing rules, but they often don't "get" what you are trying to do or what you consider vital, in terms of tradeoff (if you put this close to that, then these other things must be separated further).
Pet peeve is someone placing a scope probe on a sig generator output, to provide the signal into a circuit.
Hi,
Since many of you mentioned laying out PCBs, I have a question - hopefully not disturbing the topic too much...
While I too enjoy the puzzle of it, I feel that I lack good method. My experience is basic (about 10 boards so far, two layers at most, a few of them with just a couple of components), and when I need to create a board with several ICs (especially with narrow pitch, of course) things get tangled up fast. Eventually I'm finding myself starting from scratch several times, hitting walls until it somehow looks reasonable (usually not as clean or beautiful as I'd like). I even resort occasionally to placing components on both sides of the board - it works fine but again, I feel it could be done better.
So, are there any tips, resources or ideas on how to improve my routing skills? Beyond practice-practice-practice of course :-)
Thanks,
Hi,
Since many of you mentioned laying out PCBs, I have a question - hopefully not disturbing the topic too much...
While I too enjoy the puzzle of it, I feel that I lack good method. My experience is basic (about 10 boards so far, two layers at most, a few of them with just a couple of components), and when I need to create a board with several ICs (especially with narrow pitch, of course) things get tangled up fast. Eventually I'm finding myself starting from scratch several times, hitting walls until it somehow looks reasonable (usually not as clean or beautiful as I'd like). I even resort occasionally to placing components on both sides of the board - it works fine but again, I feel it could be done better.
So, are there any tips, resources or ideas on how to improve my routing skills? Beyond practice-practice-practice of course :-)
Thanks,
Best thing I can offer is some moral support - I understand, I'm about the same place you are.
I love laying out boards, but sometimes I feel like a deer in headlights. There's a hundred variables that depend on a hundred other variables and the second set alters the first...
But like all things, decisions have to be made - make the important ones first, then work around those to make the next and so on and so forth. You WILL have to make compromises at some point, it's the nature of the medium. Save them for last, then make them knowing that the most important aspects of your design have not been compromised.
I think a large portion is experience.
to me how I start depends on if there is space enough or not.
We always have a number of components that have to be at a specified location. those you put down first and lock them in place.
Then I start with grouping components that need to be together outside the PCB. in this way with the ratsnest on you can quickly see how to best place components to avoid signal crossings.
Then if there is space enough you just put the blocks in place and start routing.
If space is an issue, I start from one side of the PCB placing the components and routing the tracks and so fill the board up from one side.
If in the end there is space left I spend time in moving components a bit to create better grounding(keep that in mind all the time when routing).
when all is routed, normally you can do a cleanup as lines might be better off taking another route in the end(also ground improvement)
I just etched 2 of the prettiest little home brew boards I've ever done. The toner transfer was PERFECT, a rare achievement.
Fried the chips onto them. One of them works perfectly, the other has the programming data pin shorted to plus.
Examine - examine - examine
Clean - clean - clean
Magnify - magnify - magnify
Finally, back to the frypan and lift the chips off. There it is! A tiny copper bridge under the chip.
The largest known prime number: 282589933-1
It's easy to stop breaking the 10th commandment! Break the 8th instead. 
So, I guess that qualifies as a re-fried board!
JC
Should have said, "Forget it! Make another one." Would have saved a lot of time, but it's the principle of the thing.
The largest known prime number: 282589933-1
It's easy to stop breaking the 10th commandment! Break the 8th instead.
Use only a good quality oil when frying boards
Getting back to the thread title's topic, I just found another peeve:
Forum threads that have more posts about something in no way concerning the thread topic than about the original topic!
Forum threads that have more posts about something in no way concerning the thread topic than about the original topic!
But that's one of my favourite things! ;-)
|
"Experience is what enables you to recognise a mistake the second time you make it." "Good judgement comes from experience. Experience comes from bad judgement." "Wisdom is always wont to arrive late, and to be a little approximate on first possession." "When you hear hoofbeats, think horses, not unicorns." "Fast. Cheap. Good. Pick two." "We see a lot of arses on handlebars around here." - [J Ekdahl] |
|
Stupid error messages with don't give any useful information to enable the problem to be found or solved.
Like this:
How hard would it have been for the author to actually say what component it was looking for at the time?
https://www.avrfreaks.net/comment...
Surely, the people writing this stuff must be familiar with what it takes to debug such issues - so must know how useless this is?!
EDIT
Now I can understand why one might want to just have a "headline" message for the non-technical user, or for when the user realises immediately what's gone wrong - but there always needs to be a 'full details' option somewhere
Forum threads that have more posts about something in no way concerning the thread topic than about the original topic!
Nah, those can sometimes be more informative than the answer to the original question.
Mine would be forum posts asking for urgent help, which receive a load of useful replies, only for the OP to disappear off the face of the planet forum only to reappear months later asking another urgent question.
Mine would be forum posts asking for urgent help, which receive a load of useful replies, only for the OP to disappear off the face of the
planetforum only to reappear months later asking another urgent question.
Oh. You mean like my threads
The largest known prime number: 282589933-1
It's easy to stop breaking the 10th commandment! Break the 8th instead.
