Amstrad CPC WiFi

[yaworski]

Hi everyone,

This is my first post on the forum and I'm bringing gifts .

AFAIK there were no publicly available models for 3d printed cases for M4 board rev 2.5c, so I decided to create one myself. After many tries and iterations I've written a parametric PCB enclosure generator for OpenSCAD and used it to generate cases for rev2.5c and rev2.5b (both available for boards with edge and IDC connectors).

I've only tested a case for rev2.5c with edge connector, as it's the only M4 board a have access to. According to Duke, the difference between b and c is the addition of the hack button, so 2.5b version should also work well. I don't know if the IDC is going to be OK. I couldn't take any measures myself, but I've used IDC (male with shroud) connector size from the datasheet.

The case is not perfect yet. Tweaks are welcome (preferably as pull requests). Also measurements of other revisions are very welcome. It's very easy to create definitions for them.

The code is licensed under GNU GPL v3 and can be found here: https://github.com/yawor/m4board-enclosure

[Gryzor]

Very very nice design!! The buttons look great too, but will they last? Does the material have enough elasticity?

Also please someone make some to sell


PS Oh, and welcome here, thanks for sharing!

[yaworski]

Very very nice design!! The buttons lookl great too, but will they last? Does the material have enough elasticity?

Also please someone make some to sell


PS Oh, and welcome here, thanks for sharing!

Regarding the button long run durability, it's also something I've been worried about, but after printing the top part and trying it with board underneath I've realised there's probably no need to worry.
The buttons need really small travel distance to activate microswitches on the board, because the actuator (a cylinder protruding down from the button, which presses the microswitch) on each button is on the opposite side of the hinge, at the very end. If it was closer then the travel distance would be greater.
I've also thinned the hinges on the point where it connects to the rest of the case, making it similar to living hinge designs. The hinge is also along the printed layers, not across, so there's no worry of button breaking off because of delamination. PLA I've printed it with allows much much more travel distance than is needed (I've tested this by printing only a small test button first and playing with it).


All above is just my opinion, as I'm a programmer, not a material expert, but I've tried to think about the failure modes and couldn't find any obvious ones.


Much more worrying is the embossed text on the surface. The top should be printed top down so there's quite a lot of bridging required to cover the holes for the text. My Prusa mini was able to do this, but with small issues on the first layer of bridge, which was covered with inserts anyway. But all depends on how good a printer is at bridging longer distances. It can also be printed without texts.

[Gryzor]

I was actually intrigued and impressed by the embossed text - I know very little about printing but it did look very pretty to my eyes.

A very small travel distance makes sense I suppose, probably worth the risk!

[yaworski]

3d printing is also quite new to me. I only have my printer for less that two weeks now . I've been doing some 3d design in my spare time for some time though.

If you go to my github link, you can open stl directory and open one of the stl files. Github has a stl 3d viewer implemented so you can see how the models look. If you check to *_top.stl files, you can see how it's placed on the printing bed. If you manipulate the view with a mouse and look under the model, you can see that the text layers start higher than the bed level. So the printer leave a hole for the letters during the first few layers as it's building upwards and then it must bridge the gap when it finally reach a layer where there's no hole anymore. As the gap is quite large, the bridges may be deformed and may need so processing after the print is finished (for example some cutting and cleaning may be required) so the printed letter/symbol inserts fit properly.
There're also *_texts.stl files, which you can print separately with different filament color (like I did) and then glue them into the embossed parts.

Regarding the button travel distance, I wanted to record a video how it works but the movement is so small it's barely visible . The buttons themselves are really comfortable to press, as they are 12 mm in diameter (not counting the hinge). They also work great as a fidget toy - you just want to press them all the time  .


I'll also include ready to use STL files without embossing.

[Gryzor]

Ah, you have protrusion below to actually push... ok.

I just had more time than normal playing with the github viewer

[zhulien]

This is my first post on the forum and I'm bringing gifts .

I love the designs.  I wonder though, these are perfect for someone using M4 directly plugged into a CPC and look perfect too.  Could you design variations of these for sitting vertically within the MX4 board - and perhaps similar cases for other common MX4 board hardware so they are all encased instead of wibbly wobbly circuit boards?

[yaworski]

I love the designs.  I wonder though, these are perfect for someone using M4 directly plugged into a CPC and look perfect too.  Could you design variations of these for sitting vertically within the MX4 board - and perhaps similar cases for other common MX4 board hardware so they are all encased instead of wibbly wobbly circuit boards?

I see that MX4 uses IDC connectors. There are versions for IDC prepared (scad source and rendered STL files) on the github, but not tested yet if they fit correctly. The potential problem I see is that the buttons may not be easily accessible when multiple extensions are next to each other. My enclosure generator script doesn't support buttons on other faces than the top one - I would like to add this feature for other faces, but other than the top and bottom faces, it's not that easy because of few reasons: there's a split between top and bottom part around the box; printing button of this design on a vertical wall (regarding the printing direction) may not even be possible (it could melt together with the rest of the case and not work correctly). Also moving the case buttons on the side walls when the microswitches are pressed from the top may pose some challenge . But there's a whole subject of compliant mechanisms, so it's not impossible. A side button could still press the microswitch through some kind of lever on a living hinge.

I'm already thinking about rewriting the generator script for next version using a really good library I've found which would replace a lot of my own functions and making the code a lot more manageable. I could then maybe think of some solutions, for example a few designs of buttons selectable for each button (for example a simple round or square piston button).

As for the enclosures for other extensions, I'm not against it . All the code is publicly available and uses GPL license. Please take a look at the m4board_*.scad files. There's no documentation in them, but you can also take a look at the pcb_enclosure.scad, where all the arguments are documented to some extent. You need calipers to take measures of the board and its features and you need to have OpenSCAD installed. I can help with explaining the arguments and with measures (without physical access I can't do it myself), but I think this is beyond this thread .

[Muddyfunster]

I've been having some trouble getting CRTC3 and a few other demos to work on the M4. (CRTC3, Eerie Forest to name 2).


I'm using a 6128+ with an M4 with 2.05 firmware.


When I try to launch CRTC3 for example, I get a black screen with flowlib1.3 shown for a split second then black with white dots and dashes in the lower part.


Pretty much everything else runs fine.


I have the M4 rom in 7 and M4FE in 4.


I assume I'm doing something wrong my side or I have a setting that's incorrect.


Would appreciate any advice.

[Duke]

I assume I'm doing something wrong my side or I have a setting that's incorrect.

Would appreciate any advice.

Have you enabled 512KB cartridges in the webinterface?

[Muddyfunster]

Thanks Duke,


I don't see any option or button to enable 512k carts in the web interface. Cartridge just reads 256kb/512kb. I don't see anything like that on Control or Settings either.

[Duke]

You have to set the option "Use only 16 slots" then it will allocate the remaining 16 slots to the cartridge aswell so size increase from 256 to 512 KB.
Now reboot the M4 board and reup the 512KB cartridge demo.

Thanks Duke,


I don't see any option or button to enable 512k carts in the web interface. Cartridge just reads 256kb/512kb. I don't see anything like that on Control or Settings either.

[Muddyfunster]

You have to set the option "Use only 16 slots" then it will allocate the remaining 16 slots to the cartridge aswell so size increase from 256 to 512 KB.
Now reboot the M4 board and reup the 512KB cartridge demo.

Thanks Duke, that did the trick. I knew it was going to be something I'd not done or was missing.


Every day is school day!

[Animalgril987]

Hi Duke.
I've just acquired a 464.
How is your internal version of the M4 progressing?  Will you be making them for the 464?  If not are you still making the external one?
Thanks, Alan.

[Duke]

How is your internal version of the M4 progressing?  Will you be making them for the 464?  If not are you still making the external one?

Tbh. I have not touched it for a very long time, so no progress I am afraid.

The external one is still being made, with a lead time of 4-6 weeks after committing to an order (see details in first post of this thread).

[genesis8]

Hi,
Is there any interest in replacing the 8266 by the ESP32-C3 WiSoC ?
It's pin to pin compatible with ESP8266, works with ESP32 development framework (e.g. ESP-IDF), supports Wi-Fi & Bluetooth LE 5.0 connectivity, and integrates 400KB SRAM & 384KB ROM. It features a single32-bit RISC-V (RV32IMC) core @ 160 MHz, and consume as little as 5uA in deep sleep mode. Pricing will also be similar to ESP8266.
See https://www.cnx-software.com/2020/11/22/esp32-c3-wifi-ble-risc-v-processor-is-pin-to-pin-compatible-with-esp8266/ for more informations.

[Duke]

Is there any interest in replacing the 8266 by the ESP32-C3 WiSoC ?

Sorry for late reply.
I don't really see a reason for it. What would be the advantage?
Also it is only pin compatible with simple ESP8266 module communicating via serial lines. The module  (ESP12F) used on M4 board exposes almost all the 8266 pins, and both SPI paths are used and some gpios.
My ESP firmware is "bare metal" to give a BSD like sockets interface to the CPC aswell as a webserver at the same time (with custom functions to control the CPC), unlike the simple "AT commands" firmware supplied by Espressif.
So my firmware would need to be completely re-written for ESP32 from the ground up. And then the only advantage would be addition of BLE (not sure what interesting devices we would find for that, ie. most gamepads are BT classic).
If I decided to make a brand new product, I would probably use the ESP32 though (the full version with all pins exposed), it is a powerful and cheap chip.

Aside from that, I am working very hard to fulfill all current M4 board orders this weekend, so everybody can get their M4 board in time for XMAS
- Imperium Solo orders will have to wait for next weekend.
No new orders will be ready this year as I just used the last M4 board PCB.

[GUNHED]

Let us know when you ship unit #1000

[XeNoMoRPH]

Hey @Duke !!!

As a curiosity, how many units of M4 have you already sold? , it has to be close to 1000 !!! 

[Duke]

As a curiosity, how many units of M4 have you already sold? , it has to be close to 1000 !!! 

952 units shipped... So almost there!

[XeNoMoRPH]

Awesome !!!! ... and ... thanks to colleague @hERMOL , we have another new frontend for the M4 board !!! , it's called RulezCharge !!!!

download from CPC : |httpget,"cpcrulez.fr/m4/m4beta.bin"

now we can download games directly from CPCrulez !!!


https://youtu.be/EOqbmtsVHps

[Gryzor]

This video requires an epilepsy warning, but wow!

I also love the interface -the font, the header, mode 2 rules!

[Skunkfish]

Really impressive, the speed of the search and download is so quick!
But is it possible to change the download folder?
And execute a program directly from the .DSK without dropping back to BASIC?

[XeNoMoRPH]

Really impressive, the speed of the search and download is so quick!
But is it possible to change the download folder?
And execute a program directly from the .DSK without dropping back to BASIC?

not possible with this beta version, but these options may be included in the next version, the author is improving it little by little.

[VincentGR]

@XeNoMoRPH Please update the progress of the launcher here or in a new thread