Ok, I think I can finally (more or less) call this finished. The DB9 (Joy Stick) sockets arrived today. I thought it would be a quick bit of soldering a bit of drilling and mounting the sockets at the left hand side of the case, reassembling the case, putting the screws back in and enjoy! As it turns out, not quite!
Here's how I connected up the sockets for the joystick/controllers. You may notice that there appears to be some sockets at the bottom right of the N-Go board and you are correct, those are some joystick sockets
Unfortunately they are going to be deep inside of the case so I can't really use them easily.
Fortunately, the Spectrum Next or in this case, N-Go has two options for joysticks. The sockets are duplicated on the bottom left. Since I want to install the sockets where they were originally situated in this Spectrum case, all I need to do is solder each of the connection on the vacant socket footprints with a bit of wire to bring the sockets out to the side of the case which basically looks like this...
I drilled some holes so I could use some small nuts and bolts to secure them in place. So that's it right? Everything is connected so I just need to button it all up and start to have some fun using it right? Well that's what I thought. I carefully reassembled everything, plugged in a joystick and booted the machine up. This is when I discovered the machine had other plans!
The menu appeared on the screen and I began to use the joystick to scroll to the "Browser" option to load a game, but I over shot the option and that's when I had to press UP on the joystick to move back one option. Buuuuuuuuut, it just refused to move up. All other controls worked except UP
So I thought "Drat! I must have a loose connection!" So all those screws had to come back out and the case opened. I tested the connections with my multimeter and they were all fine. I decided to resolder them just in case.
I assembled the machine enough to test the joystick, but still UP just wasn't working. So I rechecked everything again, and again. I tried jumping the connections together with wire. Basically when you press UP on the Joystick what you are doing is closing a switch between pin 1 on the joystick connector and a GROUND connection. The computer sees a zero volt signal asserted on the connection to the Joystick bus and responds appropriately. But doing this manually didn't work, yet all other signals responded normally!
I tested for continuity everywhere but could not figure out where the "UP" signal was connected. Eventually I started applying pressure to different parts of the N-Go board while holding the Joystick in the UP position and watching for a response from the computer. Eventually I saw the cursor on the menu move up and start looping round, so that's got to be a dry solder joint!
These were the chips that I applied pressure to (U6 and U7)...
I couldn't tell for sure which was causing the problem specifically, but logically, it's likely it's the one on the right, closest to the affected joystick socket. So I applied some flux and resoldered both chips...
I cleaned everything up and tried the joystick again, and thankfully, this time everything was now working normally!
One thing I did notice was that the realtime clock module didn't seem to be working. It's not super important, but it's something that's nice to have as you can see easily when you created a file etc. I tried resoldering the realtime clock chip, but this time it made no difference
I mentioned in an earlier post that I thought maybe the previous owner had been a bit rough with the N-Go board as one of the keyboard connectors was a bit flakey. I think the evidence is mounting as I find it very unlikely they wouldn't have noticed a problem with the joystick. I suspect they sold it hoping the new owner would assume they'd broken it. It's possible I might have damaged it, but I don't think so, I have been handling it very carefully.
My theory is that the seller damaged the keyboard connection and the joystick problem was just a manufacturing defect. They did mention in the eBay listing something about a (none existent as it turns out) firmware setting to enable use of an original Spectrum keyboard. I looked high and low for instructions on this but there was no mention of this "Setting" anywhere.
You can use an old style PS/2 PC keyboard with this machine, so I think maybe this was an attempt to push the new owner towards that option, or at least muddy the waters as to why an OG keyboard wasn't working. Just a theory as I say.
So anyways! It's working and all screwed back together and I'm pretty happy with it!
Here's a picture of the back of the machine before I made any modifications...
And here's how it looks now...
Not super tidy, but I can deal with that in the future by creating a backing plate, for now it's fictional. The internal cassette deck is connected to the cassette socket using a right angled 3.5mm jack plug and this makes it easy to disconnect it if I ever want to use another audio source to load software, quite unlikely, but useful
Here's how the left hand side of the case looked before modification...
And here it is now...
The reset button functions as the label denotes and the other two are "Drive Select" and "None Maskable Interrupt"--which just really means that it allows you to pause the computer and bring up a menu that allows you to change the settings.
So that's it, done for now! At least until I want to have a look at what's causing the problems with the Real Time Clock, but I think I will leave that for another day!
Hopefully someone might find this all vaguely interesting