Some time since I have had time to explore KiCAD, and in the meantime something failed in my version 5.10 use. I were having an eye on trying to use ver 5.99 when I came to a stop.
As I am a Linux user mainly I found that my distro do not have an easy way to handle Nightlies. Spent time working out the quickest was to drag a small laptop with win 10 into use.
Have ver 5.99 installed, the machine is slow but useful. Small screen with possibility for using external screen.
Of course I now find so much forgotten, old age plus memory problems significant slow down for me.
I would however like to speed up my re-learn of KiCAD related to making 555 timer projects. To save a lot of time I am wondering if there is any members that has finished projects using 555 timers as base who are prepared to share so I can learn from exploring those.
My thinking is that a lot of decision would already be made and I could carry on from that stage experimenting?
I have looked on the internet in general but not had much luck, though I do remember coming across a couple of projects but can’t find my links either. One did have finished board and project available for download.
Any hints where I am likely to find such layout projects will be much appreciated as well
I think that 555 projects are generally not layout critical. You are rather in kHz not GHz.
So learning how somebody have connected all nets he needed to connect will not give you much.
I would suggest to try to do any simple (even completely not useful) KiCad project from schematic to PCB (you can even start with 2 resistors connected parallel). When you will be able to do it then it should be no problem to do any 555 project you will want.
I am using 5.1.10 but I think that it is good idea to learn 5.99. If I have a Windows 10 PC I would be probably trying to use 5.99, but I don’t have.
If you want to learn about how a NE555 works, then just read the datasheet, do a general search for some schematics and build a few different circuits on a breadboard.
If you know what schematic you want to use and know how KiCad works, then entering the schematic and making the PCB layout is less than half an hour of work. Silly simple things like choosing the right connector you want to use can easily become a relatively big time consuming part of such projects.
I think that if you use an original bipolar type NE555 you may need to be careful to place the power supply bypass capacitor close to the chip. I think that the output totem pole may exhibit some cross-conduction, causing a <500 nSec spike in the supply current. Similar to old TTL logic.
EDIT: Actually this is a standard operating practice which has served me well over ~46 years of engineering. It is more critical for some ICs than for others. It is good practice to place power supply bypass capacitor(s) as close as possible to the chip - any chip - as the next thing you do after placing the chip itself. It is possible that there are some exceptions but I cannot think of any.
When engineers were learning to use TTL, https://en.wikipedia.org/wiki/Transistor–transistor_logic
inadequate power supply bypassing would permit power supply glitches to propagate. These glitches could easily flip a flip flop (or flop a flop flip?) and screw up an otherwise good design.
Traditional NE555 is infamous for this, it’s much much worse than normal TTL logic actually. I suggest a few µF cap right next to it. Or use a CMOS variant, they don’t do this
You are probably correct. One thing…is that stray inductance can be lower with a new SMT design using an NE555 in SO8 and chip capacitors than it was in a 1980 through hole design using an NE555 in PDIP and radial leaded capacitors. An 0603 or 0805 16V X5R (maybe 2.2 uF) would be better than what I remember from the 1970s: a 1 uF 50V Z5U radial leaded ceramic.
Thank you all for posting. I should just clarify a bit. I am not looking for design of the circuit, that is no problem for me (as long as I can remember what I am doing) it is merely layout handling I am looking for.
For example building your own foot prints, selecting suitable layers apart from top and bottom copper.
How to go about maximising number of designs on the 100mmx100mm footprint etc.
I have just downloaded project that ’ [franzee]’ linked to.
I will come back on that later when I have actually loaded what I got down.
There are already footprints in the library for all the usual 8-pin packages as well as the usual discretes. Which reminds me that decades ago we had a joke that IC stands for Indiscrete Circuit.
** retiredfeline**
Thanks for posting. Yes I am familiar with the basics of KiCAD offerings. I just wrote the first thing that came to mind. I am pondering a sneaky little way to make some to me interesting ‘modifications’ to a project which will require modified footprints but that should not be a problem. May be shown all in good time when I get comfortable again in KiCAD.
If you want practice making footprints why practice on a 50-year old IC? (Unless you have a box full of them.) There are lots of designs using components that don’t have footprints. As a bonus you can also create a (or modify an existing) 3D model for it and get practice in FreeCAD and KiCad StepUp.
I just forgot that placing capacitor close to the chip is also covered by layout term. I was thinking about all other connections / placements except that one which seems so obvious to me that out of any discussion or consideration.
The disturbances should be filtered as close to their source as possible. So I powered SN65HVD72 (RS485 interface IC) by 1k ferrite bead (0.25 ohm DC resistance) and 0402 100n placed of course close to IC. At RS485 A,B lines I use 22 ohm series resistors so even when bus is shorted output current is limited to less than 50mA (it is 3V3 IC so at differential output about 2V) and normally when bus has 2x100 ohm (we use not special RS485 cables but ethernet cables) the output current is max about 2V/(44+50) = 22mA. We use 115200 bps and specify bus max length of 300m.
That works well in many our products since many years. How do you think - is there some bug in that solution?
retiredfeline
Well I do actually have many available, and they can be handled easily and I do not need a 3D magnifier to see what I am doing.
I am not looking to spend more time learning FreeCAD etc either, I have been there already.
What I don’t have worked out yet how to make use of many pcb options. The ones I did managed some many months ago I have forgotten. A couple of my own more complex KiCAD projects are now stuck in mangled form in v 5.1.6.
General
I will change subject slightly. On the internet I found a project that suited me fine, the problem is it ain’t working for me. I would acknowledge the originator but I don’t remember where it came from at the moment.
I will attach the project, I think it is complete. It is simple but the schematic has a mapping problem to library and the pcb is probably complete.
I would be happy if somebody would try and fix it, so it is working or explain what has gone wrong. I think I succeeded zipping and uploading, fingers crossed.
PCB options are driven by needs. E.g. if the circuit is simple and undemanding you don’t need a 4 layer board. (I’ve yet to do one, but now that the price difference from 2 to 4 is lower, I might use it to get ground planes.) You could look over the more demanding designs and see if you can work out why they chose those PCB options. Squeezing lots of copies inside 100x100 mm is useful in some cases such as making lots of little breakout boards, but panelising can incur extra charges which might negate savings on separate jobs.
