6502.org Forum  Projects  Code  Documents  Tools  Forum
It is currently Tue Nov 12, 2024 3:46 am

All times are UTC




Post new topic Reply to topic  [ 178 posts ]  Go to page 1, 2, 3, 4, 5 ... 12  Next
Author Message
PostPosted: Sun Jul 30, 2023 4:54 pm 
Offline

Joined: Fri Mar 18, 2022 6:33 pm
Posts: 490
O well. I hadn't planned for my next project to be a "blue month," but the project box I found is blue, the protoboard I decided to use is blue, and I'm off work in August, so why not. I guess this can be my thing now.

Blue August will be a more full-featured system than what I've built up to now. I plan for it to be more like a "real" 8-bit micro home computer. It will live in an enclosure, be configurable by front panel switches, and have I/O ports on the back. As mentioned elsewhere, I learned a lot from Blue April, but I'm ready to move on from the backplane of many tiny cards. Blue August will be more of an MBC, "multi-board computer." It will have a main board that is technically an SBC, but without any I/O. The main board will have CPU, RAM, ROM, glue and decoding. I really like the "deck of cards" style method used in Amiga prototyping and in Daryl Rictor's SBC-4, so instead of expansion slots the main board will connect to the I/O boards with IDC cables, as Garth describes in the Primer.

Anyway, let's talk construction. I am a fan of "wire wrap on top" by soldering male pin-headers next to IC sockets in order to get wire-wrap connection points without having to use expensive wire-wrap sockets. Project boards with two-hole strips are convenient for this. However, .6 wide IC sockets span 5 holes, while .3 skinny IC sockets only span 2 holes, which means that three-hole strips, or strips separated by power rails are necessary. I built Peanutbutter-1 on this stuff:
Attachment:
File comment: BusBoard POW3U PowerBoard
20230725_144133.jpg
20230725_144133.jpg [ 3.81 MiB | Viewed 20650 times ]
It's very nice, but one problem I ran into with PB1 was chip density. Using this board as designed (with the ICs between the rails like a breadboard) limits the size of the project. In order to get everything on board I ended up having some ICs straddle the rails and some ICs go between the rails. This meant I had to do quite a bit of drilling out traces. In the long run I might as well have just used regular old strip-board. Another drawback of this stuff is that it's pretty expensive, about $11 for a single Eurocard size board. In the long run, I decided that limitation in flexibility didn't make up for the convenience of of having power and ground rails already laid out.

So I bought a variety of inexpensive project boards from Amazon and Ali Express to see what I could come up with:
Attachment:
20230725_144057.jpg
20230725_144057.jpg [ 4.73 MiB | Viewed 20650 times ]
Not pictured is another blue stripboard, with 5 holes per strip, separated by single power rails. I ended up picking the blue board in the lower left. It would be nice to have larger boards with hole patterns like these, but I couldn't find any. This board is scored so you can break it into two smaller boards along the center seam. I will just be wiring the two halves together across the gap, but having the smaller boards available means that I can make stackable modules that aren't full length if I have something small I want to add.
Attachment:
20230725_151441.jpg
20230725_151441.jpg [ 3.69 MiB | Viewed 20650 times ]
Attachment:
20230725_152556.jpg
20230725_152556.jpg [ 4.22 MiB | Viewed 20650 times ]
I tried a variety of "dry run" layouts for a few days, and then started soldering the ground return network. I will use the bus strips on the board for VCC.
Attachment:
20230726_163835.jpg
20230726_163835.jpg [ 4.23 MiB | Viewed 20650 times ]

_________________
"The key is not to let the hardware sense any fear." - Radical Brad


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 12:52 pm 
Offline

Joined: Sat Oct 09, 2021 11:21 am
Posts: 718
Location: Texas
Paganini wrote:
O well. I hadn't planned for my next project to be a "blue month," but the project box I found is blue, the protoboard I decided to use is blue, and I'm off work in August, so why not. I guess this can be my thing now.


A good start to a good project! I like that you are expanding on what you have already learned.

You do like wire-wrap it seems! What advantages have you found using that, vs breadboards or soldering? I have never wire-wrapped, but I figure it is a good in-between, having the durability of soldering but the ease of breadboards. Am I right?

What software have you been thinking of running on it? I personally have found that an important question early on, as it decides: how much RAM vs ROM you will have available, what your I/O will look like, etc. I know you are capable of 80-column text, so will this be used for programming or text games? Will all of the software be on ROM, or will you have some other storage device (like an external hard drive or SD card)?

Thanks for the pics, always fun to see what you are doing :)

Chad


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 2:18 pm 
Offline

Joined: Mon Jan 19, 2004 12:49 pm
Posts: 959
Location: Potsdam, DE
I've had wire-wrapped boards in production video systems which have lasted decades - and in one case over thirty years (and was still running when they dismantled the studio control room it lived it!).

The joint on a wirewrap - done correctly - is multiple point contact welds, so very robust electrically. Plus it's easy to modify existing wire-wrapped systems, for some value of 'the wire you need to replace is at the bottom of a stack of three on the same pin'.

Neil


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 2:28 pm 
Offline

Joined: Fri Mar 18, 2022 6:33 pm
Posts: 490
sburrow wrote:
You do like wire-wrap it seems! What advantages have you found using that, vs breadboards or soldering? I have never wire-wrapped, but I figure it is a good in-between, having the durability of soldering but the ease of breadboards. Am I right?
Hi Chad! That is a multilayered question! I think, first of all, there's a kind of aesthetic component: I enjoy doing the wire wraps. The best analogy I can think of is tool sharpening. I really like sharpening my hand tools with old-fashioned oilstones. I like the stones, the oil, the sound and feel, the smell. Some people really hate sharpening and look at it as a chore they want to get over with as quickly as possible. Other people see it as a kind of meditative activity to be done at the end of the day in the shop and looked forward to as a nice way to finish up. Wire-wrap for me is more like that second group; I'm perfectly happy to sit in silence in my workshop for hours making wraps. It's very satisfying when you get good ones!

That being said, there are some frustrations with wire-wrap, primarily related to the wire. Kynar wire is really hard to strip compared to other wire, and quality control seems to be hit or miss. My first spool (50 feet of blue wire) required a bit of effort to strip, and I would occasionally mess up a wire. But my second spool (1000 feet of white wire I got on sale) is like a totally different thing. Often it WILL NOT strip, and just snaps off in the stripper. Or sometimes the bare wrapping part of the wire refuses to turn and breaks off in the wrap tool. It seems like the insulation thickness, wire thickness, and "adhesion" between them is really inconsistent along the length of the spool. It's not uncommon for me to waste 3 or 4 wires before I get a good one. This is annoying, because this wire is expensive! Faults can also be hard to spot. Yesterday I was doing some initial testing with my scope, and the R/W\ line appeared to be grounded. I couldn't see anything wrong, but I replaced that wire and R/W\ suddenly started working, so I must have had a bad wrap.

When it comes to soldering, I have gotten a lot better at it since becoming an electronics hobbyist. I enjoy soldering if it means putting a pin through a hole, soldering it to that hole, and having it be instantly connected to the things it needs to be connected to by strips on the project board. I don't mind bending component leads around to make more distant connections if I have to, but I generally despise point-to-point soldering on pad-per-hole perfboard where you have to try and get two or three different things through the same hole and hold them steady while soldering. I especially loath plated through hole. I do make mistakes, and it's just nearly impossible to get the solder in the hole heated up enough to use the solder sucker without lifting the pads off the protoboard. I had this problem frequently with Blue April (my first foray into soldering hobby electronics) and it was really frustrating.

As far as durability goes, I think both methods are fine for hobby projects if you do them right. Wire-wrap took us to the moon and gave our high end fighter jets RADAR, so it's probably fine for my hobby computers! ;)

As far as breadboards go, I love my breadboards, and have learned how to get good results with them. However, I don't look at a project on breadboards as finished. For me, breadboards are a kind of sandbox. I wouldn't build a thing on a breadboard if I already knew exactly how it worked. (You know, those projects that are like following a recipe: you buy the components, solder them together in order, and boom, you have a computer.) Breadboards are so I can do hands on experimentation... how does my VGA circuit behave if I reset the line counter on HBLANK rather than HSYNC? What does my clock signal look like with AHC logic vs. HC logic? My breadboard VGA "card" is about twice the size of the project box I have to hold Blue August, so if I want Blue August to have VGA I will have to make a wire-wrap version of that too. :lol:

Quote:
What software have you been thinking of running on it? I personally have found that an important question early on, as it decides: how much RAM vs ROM you will have available, what your I/O will look like, etc. I know you are capable of 80-column text, so will this be used for programming or text games? Will all of the software be on ROM, or will you have some other storage device (like an external hard drive or SD card)?
This will be like an expanded version of Peanutbutter-1. It has the same 128-byte movable I/O window, so that it can be configured for I/O in Zero Page if you want that. It also has 64K of SRAM and 128K of Flash ROM (in four 32K banks) using the scheme discussed in the "Period Appropriate" thread. The ROM bank will be selected by front-panel switches, but RAM can be banked in and out by software writing to a register, so it can boot from ROM, copy ROM to RAM, then run from RAM. I should be able to run PAGIMON on it with minimal effort. I haven't done a serial port yet, so I plan for the first I/O board to have an ACIA on it. I might see if I can get it to run Ben Eater's port of WozMon over that. In the long run there will, of course, be VGA text. A long-term plan I have had for a while is to eventually study Doug Comer's XINU book and port it to the 6502. I also have Loeliger's "Threaded Interperative Languages" that I would like to study and create a FORTH-like environment. It can live in a different ROM bank from the XINU boot loader so you can choose your execution environment with the flip of a front panel switch! :D As far as mass storage goes, I'd like to give it a filesystem on Compact Flash. I already have a leftover IDE / CF adapter that has the right kind of pin header for interfacing to breadboards and IDC connectors. That's a bit in the future though; I don't know much about filesystems, so I'm not sure how challenging that project will be.

_________________
"The key is not to let the hardware sense any fear." - Radical Brad


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 3:04 pm 
Offline

Joined: Sat Oct 09, 2021 11:21 am
Posts: 718
Location: Texas
Paganini wrote:
I enjoy soldering if it means putting a pin through a hole, soldering it to that hole, and having it be instantly connected to the things it needs to be connected to by strips on the project board. I don't mind bending component leads around to make more distant connections if I have to, but I generally despise point-to-point soldering on pad-per-hole perfboard where you have to try and get two or three different things through the same hole and hold them steady while soldering. I especially loath plated through hole. I do make mistakes, and it's just nearly impossible to get the solder in the hole heated up enough to use the solder sucker without lifting the pads off the protoboard. I had this problem frequently with Blue April (my first foray into soldering hobby electronics) and it was really frustrating.


I agree. That is why I print all of my boards at this point. Also, I have stepped past the "fearful" stage with printed boards, and I am now perfectly fine cutting traces with a knife, soldering bodge wires in odd places, bending IC pins outside of their sockets and soldering to them directly, and I even drilled through a board once! So, yes, I agree, putting a pin through a hole and soldering it is great. That's what I strive for! When it doesn't work as expected, or I want to try different configurations, things get interesting but not terrible. Though it seems you like wire-wrap, it seems to not be perfect either (what is perfect though?). Maybe a combo/mix of different methods would work best going forward?

Quote:
This will be like an expanded version of Peanutbutter-1. It has the same 128-byte movable I/O window, so that it can be configured for I/O in Zero Page if you want that. It also has 64K of SRAM and 128K of Flash ROM (in four 32K banks) using the scheme discussed in the "Period Appropriate" thread. The ROM bank will be selected by front-panel switches, but RAM can be banked in and out by software writing to a register, so it can boot from ROM, copy ROM to RAM, then run from RAM. I should be able to run PAGIMON on it with minimal effort. I haven't done a serial port yet, so I plan for the first I/O board to have an ACIA on it. I might see if I can get it to run Ben Eater's port of WozMon over that. In the long run there will, of course, be VGA text. A long-term plan I have had for a while is to eventually study Doug Comer's XINU book and port it to the 6502. I also have Loeliger's "Threaded Interperative Languages" that I would like to study and create a FORTH-like environment. It can live in a different ROM bank from the XINU boot loader so you can choose your execution environment with the flip of a front panel switch! :D As far as mass storage goes, I'd like to give it a filesystem on Compact Flash. I already have a leftover IDE / CF adapter that has the right kind of pin header for interfacing to breadboards and IDC connectors. That's a bit in the future though; I don't know much about filesystems, so I'm not sure how challenging that project will be.


I don't know how much you like the software side of things, but I *massively* enjoy it. I am a programmer at heart, and the hardware side of things is just a means to an end. You once said you like systems programming, so this XINU and Forth stuff sounds definitely down your alley. Once you have that stuff, what then? I know that seems way way into the future, but just wondering. Some of us like to build things for the sake of building things. I personally don't like to admit that, but I find myself doing it more often than not. I guess when I make something, I'm either making a tool to help me keep going (like a system monitor), or I'm making a tool for my daughter (like BASIC), or I'm making a game for the next Math Day.

I'm sure you like playing music as a career, but have you considered using the skills that you are developing here to maybe one day switch careers? Just thinking out loud :)

Good project!

Chad


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 3:58 pm 
Offline

Joined: Mon Jan 19, 2004 12:49 pm
Posts: 959
Location: Potsdam, DE
Paganini wrote:
Kynar wire is really hard to strip compared to other wire, and quality control seems to be hit or miss. My first spool (50 feet of blue wire) required a bit of effort to strip, and I would occasionally mess up a wire. But my second spool (1000 feet of white wire I got on sale) is like a totally different thing. Often it WILL NOT strip, and just snaps off in the stripper. Or sometimes the bare wrapping part of the wire refuses to turn and breaks off in the wrap tool.


It has to be said that the difference between hand wrapper and a powered cut-strip-n-wrap tool is chalk and cheese; the latter makes it all a doddle. But they ain't cheap these days!

Neil


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 7:13 pm 
Offline

Joined: Fri Mar 18, 2022 6:33 pm
Posts: 490
sburrow wrote:
That's what I strive for! When it doesn't work as expected, or I want to try different configurations, things get interesting but not terrible. Though it seems you like wire-wrap, it seems to not be perfect either (what is perfect though?). Maybe a combo/mix of different methods would work best going forward?
I do plan to learn how to make PCBs before too long. It's pretty clearly a good way to go, and even the Primer suggests that now PCBs are cheap just skipping wire wrap and going straight to custom PCBs is a viable option. I haven't done that *yet* for a couple of reasons. The first one is sort of philosophical: from a hobbyist point of view, the purpose of a PCB is to take a complicated thing and abstract it into something that appears simple. A routed PCB can be a rats-nest of traces, but in appearance it's all nice and flat and tidy so all I have to worry about is soldering the components on. *For now,* that's just what I want to avoid. For every wire, I want to know that wire is there because I put it there. If I have to connect A0 to 4 or 5 different things, I want to have to think about what order I connect them in, and I want to be able to take them back off and try a different order and see what changes. I can get that kind of low-level control over a PCB only theoretically (while I'm drawing on my computer screen) but not actually - at least not without making another batch of PCBs and waiting for them to ship! The second reason is more practical. I imagine that learning to layout PCBs with KiCAD is a lot like learning to use EXCEL or EMACS. Just learning to use the design tool itself is a heavy cognitive load. On top of that, I gather that there's a lot of mechanical knowledge involved in laying out a PCB - things like hole size, trace widths, component footprints, etc., that comes with its own learning curve. All of that is stuff that it's worth taking the time to learn, but it's not stuff that I want to be thinking about right now.

Quote:
I don't know how much you like the software side of things, but I *massively* enjoy it. I am a programmer at heart, and the hardware side of things is just a means to an end. You once said you like systems programming, so this XINU and Forth stuff sounds definitely down your alley. Once you have that stuff, what then? I know that seems way way into the future, but just wondering. Some of us like to build things for the sake of building things. I personally don't like to admit that, but I find myself doing it more often than not. I guess when I make something, I'm either making a tool to help me keep going (like a system monitor), or I'm making a tool for my daughter (like BASIC), or I'm making a game for the next Math Day.

I'm sure you like playing music as a career, but have you considered using the skills that you are developing here to maybe one day switch careers? Just thinking out loud :)
It's sort of the reverse of that. Long ago I went to school for computers to be my career, didn't like it much, and quit to play the violin for a living. This hobby is sort of a way to come to grips with those old interests: I'm learning on my own terms now what I wanted to learn back then at NJIT but didn't.

I'm a bit the opposite of you when it comes to software, too. Software is fine; you have to have it to get anything done! But writing it is sort of like doing philosophy. You solve problems by *describing them.* The hardware is where the magic is for me - actually making real things perform real actions. If I just want software I don't need a 6502 for that. A few years ago I went through SICP and wrote a Scheme interpreter in Scheme. More recently I went through "From NAND to Tetris." Both those texts are great in their own ways; really good for your mind. But at the end of them I really didn't know much more about computers than I did when I started; I just knew more about math.

I do really like building things, as you allude to. My other hobby is luthiery. I've done some restoration work, but I'd really like to make a whole violin from scratch!

There is an element of utility involved, too. I like making things for the sake of making things, but I'm less likely to do it if the thing truly has no function after it's built. I used to put together models with my Dad from time to time, and I still think of models as a "me" kind of thing. When we went to the U. S. S. Midway museum I bought the U. S. S. Midway Revell model. I still like it, but it's also still in the shrink wrap. :)

I don't know about the future. I think it will take a long time before I get finished with the 6502. Maybe I'll be satisfied after that. Or maybe at that point I'll go back to the Intel bus. I really like the 286. It's such a weird little chip! I've always thought it would be cool to build an alternate universe 16-bit UNIX workstation that's non-PC-compatible.

_________________
"The key is not to let the hardware sense any fear." - Radical Brad


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 7:40 pm 
Offline
User avatar

Joined: Fri Aug 30, 2002 1:09 am
Posts: 8539
Location: Southern California
Paganini wrote:
Kynar wire is really hard to strip compared to other wire, and quality control seems to be hit or miss. My first spool (50 feet of blue wire) required a bit of effort to strip, and I would occasionally mess up a wire. But my second spool (1000 feet of white wire I got on sale) is like a totally different thing. Often it WILL NOT strip, and just snaps off in the stripper. Or sometimes the bare wrapping part of the wire refuses to turn and breaks off in the wrap tool. It seems like the insulation thickness, wire thickness, and "adhesion" between them is really inconsistent along the length of the spool. It's not uncommon for me to waste 3 or 4 wires before I get a good one.

I address these and lots more in my page about answering WW questions and doubts, at http://wilsonminesco.com/6502primer/WireWrap.html .  The little stripper that comes inside the handle of the silver-colored wrapper shown and linked there is the only one I use for stripping, and it works great, every time, on every roll of WW wire, every brand, and lets me get in tight spaces, unlike the stripper built into the middle of the colored WSU wrappers shown on the same page.

As for breaking the wire off, I just have to be sensitive to possible binding, and if it feels like there's binding, back up a bit or let up on the downward pressure, feeling for what's going on inside the tool.  The friend/neighbor who kind of got me into wire-wrapping told me that the tool may have burrs or sharp edges inside when it's new, and that they will smooth out with use and it will get better.  I don't have any trouble anymore, except that I have to be extra careful if I remove a wrap (like if I got it on the wrong pin, one next to where it should have gone) and re-wrap.

The silver-plated Kynar wire and the copper wire have very different feels though, and I'll get into trouble if I don't adjust my thinking when I go from one to another.  I have the copper ones in colors that were not available in Kynar when I bought them.  If I wrap the Kynar wire as if it were the copper wire, I'll break it or make it wrap on previous layers instead of being in contact with the pin.  If I wrap the copper wire as if it were the Kynar type, I'll get gaps between the turns, exposing the pin and wasting pin space.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 8:27 pm 
Offline

Joined: Fri Dec 21, 2018 1:05 am
Posts: 1114
Location: Albuquerque NM USA
I did lots of wire wrapping early in my career. Around the time I retired, I bought a 2000’ roll of 30-ga kynar wire thinking I’ll be doing more wire wrapping. But I actually did very little wire wrapping in 10+ years of retirement; beside sockets being expensive, I found wire wrap prototypes rather bulky, so my prototyping solution is now point-to-point wiring. Sometime I’ll design a board with address/data hooked up but the control signals left out to be completed with point-to-point wires-90% pcb, 10% manual wiring.

This is what I did this morning-wiring in a fast SMT RAM to replace DIP RAM; I’ll talk about it in a separate post.
Bill
Attachment:
7065710C-7719-4B60-A65C-1A1D4AC51C1D.jpeg
7065710C-7719-4B60-A65C-1A1D4AC51C1D.jpeg [ 2.61 MiB | Viewed 20523 times ]


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 8:37 pm 
Offline
User avatar

Joined: Fri Feb 17, 2023 11:59 pm
Posts: 163
Location: Lviv, Ukraine
So you went for that name, nice! :)

paganini wrote:
It will live in an enclosure, be configurable by front panel switches, and have I/O ports on the back

Ah, another fellow fully-contained SBC with a case, sweet!

I've always wanted to try wire-wrapping since I really love soldering (I too find it to be a kind of meditation). However the tools are really expensive these days, so one day I'm going to grab something cheap on a local flea market. I saw some videos from Bill Herd on this topic, so I have a very basic idea what to look for.

paganini wrote:
[...] but I generally despise point-to-point soldering on pad-per-hole perfboard where you have to try and get two or three different things through the same hole and hold them steady while soldering.

That's what I always end up having when using stripboards for something more complex than a board with 2 chips, haha. No matter what and how I planned, I was always either too short on space or had too much spare space. For that reason, PCBs were a saver for me: soldering bus lines was just too much for me to be considered a relaxing meditation, plus I really enjoy the process of laying out the parts and tracing the board, while letting someone else do the actual manufacturing. PCBs also give me a pleasant sense of anticipation while the board is made and shipped, kinda like waiting for Santa. And they also force me to stop adding more and more stuff after I've laid my parts out, so that I can have at least one simple functional project instead of an endless unfinished mess of jumper wires. Time is a scarce resource nowadays anyway!

sburrow wrote:
Some of us like to build things for the sake of building things

True. I have countless number of unfinished projects, each of them having started from a flood of excitement caused by discovering some new tool, language, library, or concept. Sometimes it comes to performance software, which eventually doesn't save me much time in a long run considering the time I spent on setting it up just for the sake of it. :D

paganini wrote:
I do plan to learn how to make PCBs before too long. It's pretty clearly a good way to go, and even the Primer suggests that now PCBs are cheap just skipping wire wrap and going straight to custom PCBs is a viable option.

If you're going to have slots for various extension cards, you'll have a nice flexibility to have different boards made with different processes! (However I myself prefer a single board with all my stuff.)
paganini wrote:
Just learning to use the design tool itself is a heavy cognitive load. On top of that, I gather that there's a lot of mechanical knowledge involved in laying out a PCB - things like hole size, trace widths, component footprints, etc., that comes with its own learning curve. All of that is stuff that it's worth taking the time to learn, but it's not stuff that I want to be thinking about right now.

Fair enough. However I must admit that I found KiCad really easy to use in terms of PCB design, and I do main routing with DipTrace (it does a phenomenal job with auto-routing - that's the only thing I use it for at this time, however lots of people say DipTrace is superior to KiCad in all terms). If you have some spare time, I'd really recommend giving it a try: most of stuff is available out of the box, and some forum members have generously provided some missing parts (namely 65xx parts - I think it was Alarm Siren). I often use KiCad (both schematic & PCB designer) even when I'm doing a perfboard build: it really helps to avoid rat nests in advance, decide on most optimal component placement, and have a 3D preview rendered (the latter really helps when you're planning to mount components or boards on top of each other, this way it's easy to see if stuff is going to fit).
(Not trying to push you in any sense to go for a PCB - just sharing my positive experience.)

_________________
/Andrew

deck65 - 6502 slab with screen and keyboard | ПК-88 - SBC based on KM1810VM88 (Ukrainian i8088 clone) | leo80 - simple Z80 SBC
nice65 - 6502 assembly linter | My parts, footprints & 3D models for KiCad/FreeCAD


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 31, 2023 9:36 pm 
Offline
User avatar

Joined: Fri Aug 30, 2002 1:09 am
Posts: 8539
Location: Southern California
paganini wrote:
g. It's pretty clearly a good way to go, and even the Primer suggests that now PCBs are cheap just skipping wire wrap and going straight to custom PCBs is a viable option.

I just added to that indented note at the top of the page, "WW still has the advantage in being able to modify or augment; and compared to a thru-hole PCB, WW still has the advantage in density, since you can put sockets right up against each other with no space in between, and you can still get everything routed."

Quote:
Just learning to use the design tool itself is a heavy cognitive load. On top of that, I gather that there's a lot of mechanical knowledge involved in laying out a PCB - things like hole size, trace widths, component footprints, etc., that comes with its own learning curve. All

Learning is one thing, but the things you mention here are more like just getting set up, particularly the matter of component land patterns, ie, the size and placement of SMT pads to use for various packages.  I probably have hundreds, which is a major reason I still use my old CAD rather than migrating to something more modern.  I made these land patterns to get greater density than what I could get with the standard libraries supplied; and then there also are lots of custom parts that are particular to one manufacturer's switch, connector, whatever.

As for trace width and spacing, almost all traces, and definitely the signal traces, can be as narrow as the PCB house can reliably do.  I think they can all do .006"/.006" (0.15mm/.015mm) now without charging extra.  I saw a little board made for a tiny hard disc from one of our suppliers 30 years ago that had .002"/.002"!  Keep in mind that 1oz copper is .0014" thick; so a .002" trace is almost as tall as it is wide, and yet under the microscope, you could see that its cross-section was pretty much a perfect rectangle.  How they can do that is beyond me.

For holes, we used to keep them no smaller than one-quarter the thickness of the board, so for .062"-thick (1.6mm) board, we didn't go below .015" for vias, to make sure they got plated thru reliably.  Now apparently they can go to an 8:1 aspect ratio, meaning an .008" via in a .062"-thick board, although I have not dared to try it yet myself.  You don't gain much by going below a certain point in via size though, because you should still have a pad that presents a .010"-wide ring all the way around the hole, for several reasons.  The drilling indexing may not be perfect, and the drill bit can also wander a little as the hole is started, and they have to start with a bigger drill bit than specified and then the internal plating in the hole reduces its diameter to what you specified.  If the hole isn't well centered in a too-small pad, there's the possibility of a breakout, and if a trace comes to the pad right where the breakout is, the reliability of the connection suffers.  The trace really needs to connect to a ring around the hole before diving down into the hole.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?


Top
 Profile  
Reply with quote  
PostPosted: Tue Aug 01, 2023 2:58 pm 
Offline

Joined: Fri Mar 18, 2022 6:33 pm
Posts: 490
GARTHWILSON wrote:
I address these and lots more in my page about answering WW questions and doubts, at http://wilsonminesco.com/6502primer/WireWrap.html .  The little stripper that comes inside the handle of the silver-colored wrapper shown and linked there is the only one I use for stripping, and it works great, every time, on every roll of WW wire, every brand, and lets me get in tight spaces, unlike the stripper built into the middle of the colored WSU wrappers shown on the same page.
I use that exact same brass stripper, which I leaned about from the Primer. I think I've just got a bad spool.

Quote:
As for breaking the wire off, I just have to be sensitive to possible binding, and if it feels like there's binding, back up a bit or let up on the downward pressure, feeling for what's going on inside the tool.  The friend/neighbor who kind of got me into wire-wrapping told me that the tool may have burrs or sharp edges inside when it's new, and that they will smooth out with use and it will get better.  I don't have any trouble anymore, except that I have to be extra careful if I remove a wrap (like if I got it on the wrong pin, one next to where it should have gone) and re-wrap.
My wire wrap tool is a few years old and I've done 3 - 5 projects with it, so I would think that its break-in period has expired. However, your comment caused me to investigate a little more closely. It seems that the OK INDUSTRIES wire wrap tools that are commonly available in the $20 - $30 range are actually knockoffs. OK INDUSTRIES was bought by Jonard, and they changed the design of the tool (maybe to try and combat counterfeiting). The "official" Jonard WSU 30-M now has a plastic handle.

With my particular (knockoff) tool, I've never been able to get the insulation up inside for making "modified" wraps. Yesterday, to see if I could smooth off any burrs, I tried flossing the tool with some spare wire, and actually blew out a small chunk of metal! I think what happened is that "modification" wasn't properly completed - was only partially drilled, or punched, or whatever, leaving debris clogging the aperture. Now I can feed insulation up to the channel, and it seems to be working more smoothly. Fingers crossed! I went ahead and ordered a real Jonard one, too, although I'm not thrilled about the idea of a plastic handle.

and3rson wrote:
So you went for that name, nice! :)
I did! I had your comment to encourage me. :D

Quote:
Ah, another fellow fully-contained SBC with a case, sweet!
This thread has kind of taken off, so I'm behind on build log posts, but in the next one I'll include a picture of the enclosure and reveal my plans for it.

Quote:
I've always wanted to try wire-wrapping since I really love soldering (I too find it to be a kind of meditation). However the tools are really expensive these days, so one day I'm going to grab something cheap on a local flea market. I saw some videos from Bill Herd on this topic, so I have a very basic idea what to look for.
The "Wire Wrap 101" article and accompanying Bill Herd videos are great. Blue August's ground return network is inspired by one of the pictures from that article. The best part is the coffee cup, though. :lol:

Quote:
I often use KiCad (both schematic & PCB designer) even when I'm doing a perfboard build:
Now that's an interesting notion. I never would have thought of that, but it makes sense. I did find the schematic side of KiCAD easier to learn than I expected. It only took about a week before I was able to produce some passable diagrams.

_________________
"The key is not to let the hardware sense any fear." - Radical Brad


Top
 Profile  
Reply with quote  
PostPosted: Tue Aug 01, 2023 3:53 pm 
Offline

Joined: Fri Jul 09, 2021 10:12 pm
Posts: 741
Paganini wrote:
Quote:
I often use KiCad (both schematic & PCB designer) even when I'm doing a perfboard build:
Now that's an interesting notion. I never would have thought of that, but it makes sense. I did find the schematic side of KiCAD easier to learn than I expected. It only took about a week before I was able to produce some passable diagrams.

I've even used it to help lay out breadboards sometimes! If you have a large-ish circuit to build you can benefit from the ERC checks in the schematic editor, and then drag the parts around to try to minimise the amount of wiring between them - really the same as what you do on a PCB but of course with much more restricted options for part placement etc.

Edit: here's an example I did the other day, a breadboard layout for a cut-down version of my recent project, to test the basics will work before ordering PCBs. I didn't connect all the wires, the ratsnest is good enough to show roughly where they'd go, and you can spot long-distance connections and check they're not going to cause any issues, consider reassigning pins/units/etc to simplify things, etc.
Attachment:
kicad-breadboard-layout.png
kicad-breadboard-layout.png [ 240.29 KiB | Viewed 20448 times ]


I wouldn't be too afraid of using it for PCB layouts, if you have at least a little time and mental bandwidth to learn. You can make life easier for yourself if you start with large boards, spaced-out components, and four layers with ground and power planes, and don't get too hung up on best practices until you're comfortable with the tool. Maybe you won't actually send them off to manufacture, but it's a good way to learn the ropes and maybe have more confidence to come back to it when you have something you do want to get manufactured.


Top
 Profile  
Reply with quote  
PostPosted: Tue Aug 01, 2023 7:23 pm 
Offline

Joined: Fri Mar 18, 2022 6:33 pm
Posts: 490
Here's the box Blue August will live in. I think it has a nice retro look.
Attachment:
20230725_144230.jpg
20230725_144230.jpg [ 3.38 MiB | Viewed 20430 times ]
The front/rear panels are removable, but they're made of aluminum. I don't have good metal-working capabilities (I don't even have a nibbler), so I got a sheet of 1/8 inch DUCO ABS plastic. One side has a kind of rough surface that looks exactly like something from the 1970s.

Here's the main board with some IC sockets and bypass caps soldered.
Attachment:
20230727_175729.jpg
20230727_175729.jpg [ 3.93 MiB | Viewed 20430 times ]
The open space in the upper left is for the ROM ZIF socket. It's pretty thick, so I saved it for last. Also because it's so wide, I couldn't use the same plan of not having to solder socket pins to anything (the ZIF socket covers up the rows I would normally have put the header pins in). So, I had to make one million tiny staples to connect it to the neighboring rows.
Attachment:
20230728_161750.jpg
20230728_161750.jpg [ 4.8 MiB | Viewed 20430 times ]
Soldering finished, ready for wrapping:
Attachment:
20230728_172855.jpg
20230728_172855.jpg [ 3.53 MiB | Viewed 20430 times ]
I'll populate the sockets so that I can test as I go. I'll have the scope nearby so I can check each connection as I make it. I should be able to make sure that, e. g., the clock stretcher is working, as soon as the control signals are wired up.

_________________
"The key is not to let the hardware sense any fear." - Radical Brad


Top
 Profile  
Reply with quote  
PostPosted: Tue Aug 01, 2023 7:42 pm 
Offline
User avatar

Joined: Wed May 31, 2023 10:46 pm
Posts: 26
Location: Sebastopol, California
gfoot wrote:
Paganini wrote:
Quote:
I often use KiCad (both schematic & PCB designer) even when I'm doing a perfboard build:
Now that's an interesting notion. I never would have thought of that, but it makes sense. I did find the schematic side of KiCAD easier to learn than I expected. It only took about a week before I was able to produce some passable diagrams.

I've even used it to help lay out breadboards sometimes! If you have a large-ish circuit to build you can benefit from the ERC checks in the schematic editor, and then drag the parts around to try to minimise the amount of wiring between them - really the same as what you do on a PCB but of course with much more restricted options for part placement etc.

Edit: here's an example I did the other day, a breadboard layout for a cut-down version of my recent project, to test the basics will work before ordering PCBs. I didn't connect all the wires, the ratsnest is good enough to show roughly where they'd go, and you can spot long-distance connections and check they're not going to cause any issues, consider reassigning pins/units/etc to simplify things, etc.
Attachment:
The attachment kicad-breadboard-layout.png is no longer available


I wouldn't be too afraid of using it for PCB layouts, if you have at least a little time and mental bandwidth to learn. You can make life easier for yourself if you start with large boards, spaced-out components, and four layers with ground and power planes, and don't get too hung up on best practices until you're comfortable with the tool. Maybe you won't actually send them off to manufacture, but it's a good way to learn the ropes and maybe have more confidence to come back to it when you have something you do want to get manufactured.


I too am working on designing my own breadboards/protoboards in Kicad and getting them fabricated. I once got into wire wrapping but felt it got a bit messy to deal with, and have moved back to soldering. I like the breadboards that mimic the solderless breadboards with the 5-wide strips perpendicular to the long pairs of bus lines. These are not as dense as the 2-wide strips allow, but make it super convenient to connect multiple wires to a pin.
Attachment:
File comment: first Kicad breadboard build
pcb_breadboard_001_PXL_20230730_225618232.MP.jpg
pcb_breadboard_001_PXL_20230730_225618232.MP.jpg [ 10.23 MiB | Viewed 20428 times ]

I'm happy to share the files -- design still needs some cosmetic cleanup on the silkscreen layers. The four edges are designed to support stacking edge connectors to make a a vertical bus of connected boards (the silkscreen is misleading there currently).

_________________
--Mark


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 178 posts ]  Go to page 1, 2, 3, 4, 5 ... 12  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: