Bacteria's portable DreamCast handheld console - IntoDream

[bacteria]

Ok, spent about 5 hours today on this soldering - wires from the VMU to the controller, the rumble pack to the controller. Sounds a lot of time, and was, however had to redo some contacts as the continuity tester on my multimeter indicated some of the traces were touching on the wiring, so had to stop that. Also, the wiring is quite fine.

There are three layers on this - the controller board, the rumble pack board, and the relocated part of the controller. The other side needs the relocated part of the other piece of controller, which will be positioned as the 4th layer on the other side. Used a blob of hot glue to secure each board piece to card, and small blob to secure the card to the next board - means they can be easily removed if any repairs needed.



Issue is that the controller has been rewired, the memory card and rumble pack rewired; there is every reason for it to work, however by the same token it might not work, maybe wires too long from the VMU, etc. Also, that is assuming I haven't got the wiring backwards, or will fry the board(s). I have no spares at the moment, so would cause a delay in the project if wiring is done incorrectly at this stage. Only way to know is when done, turning the system on, fingers crossed, hoping all is fine!

[jfe2]

Oh man, I seriously hope it was wired correctly! I have been following this project for a bit now and it would really suck to see such a setback. Good luck!

[bacteria]

Thanks.

Once the controller is finished, including wiring the buttons and d-pad too to it, there isn't much left on the project to do:

* Wire the other connections in - switches, etc
* Fit the console board in place, secure, wire it up to the system
* strengthen the case sides and secure together with a few screws

Obviously there is a fair bit of work for each of the stages above, however as you see, the level of work is getting shorter now!

Haven't done a low battery indicator circuit on this system on purpose - not sure how accurate it will be under such amperage load. Only way to know is to run the system when done until the batteries drain, see what the voltage was under load, and see if a low voltage indicator is going to work properly. When something is under heavy load, it measures a lower voltage than when at rest after all. If I do put one in, then rather than having a hole in the case front (like on the IntoGrafx project), will put a red LED behind the case, so the light will glow through the case when the battery is low - far neater.

That is something I can only ascertain during testing. I want to do testing and tweaking before deeming the project as "completed".

[bacteria]

One thing I discovered today, which is a pain in the butt: nearly wired up the DreamCast controller, and checked the grounds on the buttons and d-pad were universal - they aren't! On a normal controller, you have a common ground connection linking all buttons together, the other connection on the button makes the button press work. Fine. I have however come across some that don't play by these rules, normally some plug'n'play boards like the (crappy) Intellivision one: and now, the DreamCast one!

According to this site I checked link I was expecting a common ground via pin 3. Nada. It doesn't.

Some buttons share a common ground, others don't!

This isn't a big problem, as I used rubber topped tact switches, as each switch is independent as standard, so each side of the contact can be wired to the tacts fine. The d-pad buttons share a common ground, fine.

Issue though is the four separate buttons installed to swap between PSone screen controls and the shifted d-pad controls for Q*Bert. Normally, the grounding from the screen board connects to the grounding on the console and normally, by default, the controller ground connects to the grounding on the console. Not in the DC's case, no, that would be too easy!

The original plan was to connect ground between the d-pad and the shifted d-pad and use the two 2P2T switches to swap the signals between the d-pad and the screen controls (contrast, volume). Can't do that now.

Two options, will only know result when testing: 1) swap ground between the d-pad / screen (that might work), although would mean I only needed to install one switch not two before; or 2) add a third switch 1P2T that also changes the ground.

More work, more time

Nearly finished wiring the controller, intend to upload pics later.

When it comes initially to testing the system out, will leave the DC where it is, wire the controller wires to the DC and also video/audio too; that way, easier to make changes, although it lengthens the project time a bit in other ways.
There are only two options - 1) wire

[bacteria]

Promised update - here it is...

Firstly, getting the last bit of life from my soldering iron, I seem to go through one iron (and a bit on the next one) per project - here you see the iron's protection has gone and the iron is eating itself to death! Never mind, they only cost about £5 each on ebay including postage!



Completed the wiring of the controller, plastic sheet to lower the wiring fits nicely, and also provides a nice ledge for the DC board to sit on.

You will now see how snug everything is:



Here is my (spare) DC board, the plugs trimmed to remove the bits that stick out, slight trim on the opposite end, two corners trimmed and a tiny strip of board (no traces on it) too. All this is needed so the board fits, and the rumble motor doesn't touch the board. It is an extremely tight fit, my calculations ages ago were literally bang on, there was no room for error on this.



You can see the layers on the controller (4 layers), and the "step" between that and the DC board. The controller is secured in place by just one screw, through a piece of card to stop any chance of a short.

Apart from the 5 wires from the d-pad, all the other wires stretch to the opposite side of the case, that is so the controller can be easily removed for any repairs if needed.





All that is needed is the controller 6 wires to be wired to the DC, wire the audio cables to the PSone screen, final inspection, a few misc tests, then I should be able to test the DC working with the joystick, start and d-pad buttons. All being well, the DC can be installed into the case, after testing the system to RGB instead of composite as at the moment. RGB will not only give crisper graphics and more accurate colours, but also be compatible with more NTSC titles.

Not going to do this testing work today, out of time, will do in a couple of days.

[bacteria]

It's almost like a form of courage, although it isn't; like a mixture of fear/excitement/anticipation!

Tested the setup with a normal DC controller, all ok. Great.

Then swapped cable for the controller wired into my portable - shaking a bit, adrenaline....

YAY - all works fine - the memory card binged into action, didn't expect that as it wasn't connected to power, although I knew the VMU works fine without its internal batteries when connected to the controller; giving a pic on the small screen, as you see in the photo. Unlike when I had the other DC controller with no memory pack installed, it gave a message on the screen about needing a VMU for 5 memory blocks, this time it didn't, went straight in - so the VMU works great! Tested the start button, and the red button, d-pad and joystick - all seem to work great!

Have more testing to do, tomorrow or Thursday, however seems to show I did all the wiring right, and system works! YAY



(game BTW is PenPen)

Need to test the other buttons, the orientation of the joystick; rumble pack, other buttons, shoulder buttons (forgot to wire them, will have to do that!).

[jfe2]

I am SO happy to see that thing actually working. Everything looks perfect!

Also, I can't stress enough how much I love the integration of the VMU screen.

[bacteria]

Thanks! I must say, the VMU addon is a nice fun addition. Even though it works fine without connection to batteries, I read somewhere that although it does game saves like that too, it sometimes corrupts if batteries not present, so will just wire it to the DC console (7.4v line): I know it is fine on that voltage, although the board normally takes 6v, as nearly all the testing so far on the VMU was from a 7.4v cell (about 8.2v on full charge).

Only thing I think I did wrong, I think I wired the joystick back to front as got disorientated, so horizontal = vertical; easy to resolve, and minor. One has to expect some errors! Only 6 wires to re-do. Also need to wire the 6 wires to the shoulder buttons too, then can test Quake better (left shoulder = jump, right shoulder = fire). Will probably show this on a quick video.

Will need to remove the controller ports and trim that board a little so it fits in the base, but can be done.

[AmishSamurai]

Make sure to give us the full show of it on a youtube video

[bacteria]

I will do. My previous project videos have been regarded as "a bit short", so I will make sure this one is longer when done, and show a few games being run and played.

BTW, it was commented on ASSEMbler forum that the game above seems to have a border - indeed - however, other games I have played into the intro (Sonic Adventure, Ecco) and into the game itself (Quake 3) are totally full screen, so I therefore think the border is just down to the way PenPen was programmed (intro and menu screens) rather than other factors.

Obviously I can't test everything out to a greater extent game wise until later in the project.