An Improved MENSCH™ Microcomputer

For discussing the 65xx hardware itself or electronics projects.
User avatar
AndrewP
Posts: 368
Joined: 30 Aug 2021
Location: South Africa

Re: An Improved MENSCH™ Microcomputer

Post by AndrewP »

Quote:
... Kicad ... auto-routing ...
So I decided I'd try Freerouting again and after a bit of googling I also realised there's a new version of Kicad. Specifically Kicad 7.

Couple of cool things:
  • The Freerouting plugin just works from inside Kicad now. I had previously also had to do the export to spectra, freeroute, import back into Kicad dance. It still struggled quite a lot with that board I posted above but if I had setup my netlists properly I could probably have persuaded it to do the data traces completely.
  • Shift F - Automatically Complete Trace Route. Basically selective, simple auto-routing.
  • Pack and Move Tool. Places selected components near each other. Very useful for initially placing components.
Somehow upgrading Kicad major versions is still a pain in the proverbial but 7 seems worth it.
barnacle
Posts: 1831
Joined: 19 Jan 2004
Location: Potsdam, DE
Contact:

Re: An Improved MENSCH™ Microcomputer

Post by barnacle »

My company (and therefore me) used the Number One Systems Easy PC Pro for five years, but changed to Altium when I retired (and before I returned part-time) so that's what I use now.

Neil
User avatar
Michael
Posts: 633
Joined: 13 Feb 2013
Location: Michigan, USA

Re: An Improved MENSCH™ Microcomputer

Post by Michael »

I like the free version of DipTrace which is limited to 200 holes but when I bumped into that limit several years ago they were kind enough to provide me with a 500 hole license. I have not tried KICAD, yet.

Recently I've been using the free browser based EasyEDA package in order to take advantage of JLCPCB's once-per-month $8 coupon for PCB orders from designs created in EasyEDA. With the coupon I pay $1.50, including shipping, for five 4x4 inch PCBs. I'm currently hand routing an EasyEDA version of a simple and relatively inexpensive development SBC with integrated ROM Emulator/Programmer (see below).
Attachments
easyeda project.png
easyeda project 2.png
Beater 08.png
Martin_H
Posts: 837
Joined: 08 Jan 2014

Re: An Improved MENSCH™ Microcomputer

Post by Martin_H »

I am bumping this topic because I finally sat down and taught myself enough of KiCad to layout a schematic for my ideas. The project uses the W65C265QBX as the starting point, but I expanded the headers to two rows of forty pins (gaining access to every CPU pin). I also added RAM and ROM to the board to avoid the nuisance of wiring those with each project.

My idea is to create a breadboard friendly breakout board for the w65c265 to test how it works before design a PCB's with more features. This PCB allows access to the full address bus, chip select lines, and control signals. Attached are the two pages of my schematic. Also, the KiCad project on Github: https://github.com/Martin-H1/KiCad/tree ... 20Reloaded

Any feedback on my work so far would be helpful.

I am now trying to create the PCB from this schematic, and I hit a stumbling block. I need to assign footprints to components. A sample error is below:

Code: Select all

Error: Cannot add C2 (no footprint assigned).
Error: Cannot add J4 (no footprint assigned).
Error: Cannot add C3 (no footprint assigned).
Error: Cannot add J9 (no footprint assigned).
C2 would likely be aa small disc capacitor, and J4 would be a 1x40 pin header. So how do I determine their footprints?
Attachments
Mench_reloaded_p2.png
Mench_reloaded_p1.png
barnacle
Posts: 1831
Joined: 19 Jan 2004
Location: Potsdam, DE
Contact:

Re: An Improved MENSCH™ Microcomputer

Post by barnacle »

Ah, that's an easy one!
Screenshot from 2025-08-31 07-49-39.png
Screenshot from 2025-08-31 07-49-39.png (4.32 KiB) Viewed 1626 times
Three of these buttons are essential for every Kicad design.

The leftmost assigns numbers to components; you will have got numbers automatically when you placed a part so it will be ok without this, but it changes the numbers so they're sequential - either left->right or top->bottom - on the diagram.

The next checks that all the rules have been met: that you have for example power connections, no outputs driving other outputs, no unused pins etc.

Ignore the next: it's for circuit analysis. I've never used it.

The next one is the one you want: that will give you a list of every component on the board and let you select a package for it. It's not always obvious, and you might have to play around a bit, but you'll get there. Most of the ICs are in the 'package' library.

Hope this helps,

Neil
gfoot
Posts: 871
Joined: 09 Jul 2021

Re: An Improved MENSCH™ Microcomputer

Post by gfoot »

For C2 you need to physically look at your capacitor, check the pin pitch, and select a footprint that reflects that pitch and the approximate size of the disc. Technically only the pin pitch matters but it can be useful to have it check the actual sizes of components to make sure they don't obstruct each other, and you can get it to do a 3D render of your assembled board as well if you like, and sometimes looking at the 3D model for a footprint (in the footprint viewer) can help you understand what it is and how it differs from similar ones.

The pin header will need you to select based on similar criteria - what pin pitch, male or female, upright or right-angled? If you already have the component then of course you pick the thing that matches your one, otherwise you can decide how you want it and buy the corresponding component.
barnacle
Posts: 1831
Joined: 19 Jan 2004
Location: Potsdam, DE
Contact:

Re: An Improved MENSCH™ Microcomputer

Post by barnacle »

For pin headers, note that the standard libraries contain headers with three different pin possibilities (anticlockwise, a/b, and left-right) none of which will generally suit the pinout you've already selected in the part... you may have to edit whichever part you choose.

Neil
gfoot
Posts: 871
Joined: 09 Jul 2021

Re: An Improved MENSCH™ Microcomputer

Post by gfoot »

That's a nice point - while you can get away with using the generic symbols and footprints, for large more complex connectors it makes a lot of sense to create your own symbols and footprints just the way you want them (especially which pin is which, and maybe custom silkscreen) - at least be aware that it is an option.
Martin_H
Posts: 837
Joined: 08 Jan 2014

Re: An Improved MENSCH™ Microcomputer

Post by Martin_H »

@barnacle, thanks, those tools are handy. I had been renumbering by hand, and the other found five errors I corrected.

@gfoot, ok I will measure the capacitor I am planning to use.
Martin_H
Posts: 837
Joined: 08 Jan 2014

Re: An Improved MENSCH™ Microcomputer

Post by Martin_H »

I have assigned footprints to every device except the ds1813 econo reset. I found a library for that device, which specifies ds1813_revised-TO-92 as the footprint. But apparently KiCad doesn't recognize it. I looked for a generic TO-92 equivalent but couldn't find one. Editing a library is something I am not quite up to.

The other thing that concerns me is that while I have soldered 0805 SMD components. I haven't soldered anything like the 100 lead QFP w65c265. I have watched several videos on SMD soldering and honestly trying it is probably the only way to know if I can do it.

The w65c265 also comes in a PLCC-84 package can use a socket. Although the data sheet encourages the QFP 100 lead package. It might be a safer bet.
barnacle
Posts: 1831
Joined: 19 Jan 2004
Location: Potsdam, DE
Contact:

Re: An Improved MENSCH™ Microcomputer

Post by barnacle »

PLCC sockets are easy to mount (at least, the through hole types. They also come in SM types, which seems to defeat the object of the exercise). You can get QFP 100 sockets, but I don't think they're particularly economic. I suppose it depends on whether you need the extra pins or not...

(Soldering QFP is an exercise in frustration, unless you have a solder stencil and a heat gun... I can do it by hand, with fine solder and a fine tip, but I've been soldering fifty years or more. Use leaded solder, not lead-free.)

Neil
User avatar
Dr Jefyll
Posts: 3525
Joined: 11 Dec 2009
Location: Ontario, Canada
Contact:

Re: An Improved MENSCH™ Microcomputer

Post by Dr Jefyll »

Martin_H wrote:
100 lead QFP w65c265 [...] trying it is probably the only way to know if I can do it.
If you go the QFP route, just be sure to conclude by running a continuity check to detect the case of two or more adjacent pins shorted together by excess solder. In that circumstance, solder wick is a reasonably easy remedy.

What's tougher to fix is the case where you've accidentally bent some of the pins, and need to straighten them. In a worst case scenario, this might mean scrapping the chip and buying a replacement.

I don't wanna coax you too far out of your comfort zone. But if you can avoid this sort of mishap, you should be fine -- just keep trying 'til you get it right.

ETA: Neil mentioned fine solder and a fine tip iron, implying a "one pin at a time" approach. I've been successful with that, and also with the method where you run a big blob of solder over multiple pins at once, then wick away the excess. (And using leaded solder, regardless of the approach.)

-- Jeff
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html
Martin_H
Posts: 837
Joined: 08 Jan 2014

Re: An Improved MENSCH™ Microcomputer

Post by Martin_H »

Thanks for the suggestions.

I may buy an SMD practice kit, a QFP 100 breakout board, and a w65c265. It's a small investment and would let me know if I can do it. That way I can change approach before having a PCB fabricated.

Meanwhile I will keep trying to figure out the footprint issue for the DS1318.
Update: I finally figure out the footprint issue and was able to export everything to the PCB. Now the hard part which is positioning components and routing traces.
barnacle
Posts: 1831
Joined: 19 Jan 2004
Location: Potsdam, DE
Contact:

Re: An Improved MENSCH™ Microcomputer

Post by barnacle »

The gloopy flux that comes in syringes is _very_ useful for separating accidentally shorted SM pins, and in general for cleaning things up. Even though it can leave a residue... (I'm still trying to find a good way to clean it... IPA seems to do all right, and also tap water (er, distilled, honest!) seems to work, but with warm air to clean it.

To sort your DS1318 problem, um, I know a few ways to do it but it's hard to explain in text. There are probably videos showing how to (a) design a new footprint or (b) choose an existing TO-92 footprint and edit it. It's not difficult, but it's not immediately obvious.

Neil

Edit: if there is an existing symbol for DS1318, use that on the schematic. Then allocate a TO-92 footprint to it. Then when you position it, check that the right pins go where you expect them to (check the datasheet) and if necessary edit the footprint _for that circuit only_ (you'll get a choice).

Edit 2: Solder wick _and_ gloopy flux. :)
Martin_H
Posts: 837
Joined: 08 Jan 2014

Re: An Improved MENSCH™ Microcomputer

Post by Martin_H »

I updated the PCB from the schematic and chose the Eurocard size. I am hoping that standard makes it easier to fabricate.

I placed the components working left to right in places that seemed reasonable to me. But I kept a wider spacing to make assembly of the SMD components easier.

One thing that concern me are the oscillators for FCLK and CLK. The data sheet said keep them close to their respective pins. I hope what I did was reasonable as I know how twitchy oscillators can be.

Before I start the hard work of laying down tracks. Does anyone have feedback?
Attachments
Board layout draft.png
Post Reply