Mouse joystick swapping kills power

[dschallock]

Anyone have an idea what would cause the power to the system to die like it’s being shorted when swapping mouse or joystick when Amiga is on?

[duga]

Short answer: never plug/unplug anything while your Amiga is powered on.

[dschallock]

Originally Posted by duga

Short answer: never plug/unplug anything while your Amiga is powered on.

That sounds like good advice and I appreciate you taking the time to respond, But then the only way to play 2 player games is if you have a multi joystick controller device? It’s hard for me the believe that is how it is for everyone for all these years playing 2 player game on Amiga.

[jotd]

yes it's not recommended to do so, but I never had any issues in more than 26 years. Maybe it depends on the joystick/mouse?

I'm not sure but would a cable doubler do the trick, so you can connect mouse & joystick at the same time? I'd be interested in a solution as well, as it's tedious to plug/unplug.

[zipper]

I think there were mice or/and joysticks where the connector had a metal sleeve - when misplaced it could short the pins.

[solidcore]

I built a mouse/joystick switcher because of this risk. Not worth hot-swapping on an amiga! you can short the 5v line to gnd and or other pins and kill CIA?!

[dschallock]

Originally Posted by solidcore

I built a mouse/joystick switcher because of this risk. Not worth hot-swapping on an amiga! you can short the 5v line to gnd and or other pins and kill CIA?!

I don’t suppose you have the schematic for that switcher build?

[dschallock]

Originally Posted by solidcore

I built a mouse/joystick switcher because of this risk. Not worth hot-swapping on an amiga! you can short the 5v line to gnd and or other pins and kill CIA?!

I didn't realize you made and sold units for this. Sorry for asking for the plans, I didn't realize. I would love to buy one of your units if you have any for sale still. Sounds like a great product!

[dschallock]

NEW DEVELOPMENT

I checked the power on pin 5 of the port which is supposed to have 5VDC apparently and it only shows 3.75VDC. Any ideas what could cause this?

Thanks!
Dan

[Daedalus]

First off, hot-plugging as a general rule is not a good idea. But in reality it's a little more nuanced than people seem to realise. The real danger is from connecting peripherals that have their own power supply, e.g. printers, monitors, modems, other computers, in which case both items should be powered off before connecting. But stand-alone devices like mice and joysticks don't present any risk other than shorting out 5V by connecting it clumsily when the plug has a metal shield. If it's not shorting out likethat, there must be something else wrong for a machine to reset that way - loose pins or conductive debris around the port internally for example, or damaged insulation under the mainboard and loose shielding. Does it do the same thing simply removing connectors? What about wriggling the plugs? Or any of the other connections you have to the machine that might be getting moved by the action of connecting or disconnecting?

A simple cable doubler won't work - you need to actively disconnect the four direction inputs from the mouse for a joystick on the same port to work. This is what solidcore's board does electronically.

Now, Pin 5 isn't a power supply pin on the Amiga. It is on the Sega Megadrive/Genesis (is that where the idea came from perhaps?), but that's non-standard. Pin 7 should have 5V present, pin 8 should be ground. Pin 5 isn't used for mice except where they have 3 buttons or a scroll wheel, and isn't used by standard joysticks.

[dschallock]

Originally Posted by Daedalus

First off, hot-plugging as a general rule is not a good idea. But in reality it's a little more nuanced than people seem to realise. The real danger is from connecting peripherals that have their own power supply, e.g. printers, monitors, modems, other computers, in which case both items should be powered off before connecting. But stand-alone devices like mice and joysticks don't present any risk other than shorting out 5V by connecting it clumsily when the plug has a metal shield. If it's not shorting out likethat, there must be something else wrong for a machine to reset that way - loose pins or conductive debris around the port internally for example, or damaged insulation under the mainboard and loose shielding. Does it do the same thing simply removing connectors? What about wriggling the plugs? Or any of the other connections you have to the machine that might be getting moved by the action of connecting or disconnecting?

A simple cable doubler won't work - you need to actively disconnect the four direction inputs from the mouse for a joystick on the same port to work. This is what solidcore's board does electronically.

Now, Pin 5 isn't a power supply pin on the Amiga. It is on the Sega Megadrive/Genesis (is that where the idea came from perhaps?), but that's non-standard. Pin 7 should have 5V present, pin 8 should be ground. Pin 5 isn't used for mice except where they have 3 buttons or a scroll wheel, and isn't used by standard joysticks.

Thanks for such a great reply!
So I need to go check pin 7 now and see if I get 5v there. I definitely am reading 3.75v on pin 5 though. Should it be zero there? I wonder what in the heck is going on to be getting voltage at pin 5 if I'm not supposed to.

So the new Amigakit optical scroll mouse (I think it uses cocolino) anyway, I tested plugging that mouse in last night when the system is powered off, and then tried turning the power on. I see the red optical light flash for a second under the mouse but it goes immediately out and the Amiga will never actually power up or boot. I have to unplug that mouse and plug an old commodore tank mouse in and then the system will power on and boot. Does this information help determine what might be going on?

I will also check for debris in the ports and loose connections as you mentioned.

Thank you again for taking the time to respond!

[Daedalus]

Originally Posted by dschallock

Thanks for such a great reply!
So I need to go check pin 7 now and see if I get 5v there. I definitely am reading 3.75v on pin 5 though. Should it be zero there? I wonder what in the heck is going on to be getting voltage at pin 5 if I'm not supposed to.

It really depends on what the machine is doing. Pin 5 is configurable as an analogue input or a digital input or output. If it's left floating, it might be high if in digital mode or some weird in-between value if left unconfigured. The Cocolino software could also be pulsing that pin in a similar way to the CD32 controller, resulting in a multimeter getting an in-between value. I wouldn't be too concerned with that for the moment anyway.

So the new Amigakit optical scroll mouse (I think it uses cocolino) anyway, I tested plugging that mouse in last night when the system is powered off, and then tried turning the power on. I see the red optical light flash for a second under the mouse but it goes immediately out and the Amiga will never actually power up or boot. I have to unplug that mouse and plug an old commodore tank mouse in and then the system will power on and boot. Does this information help determine what might be going on?

That is a little weird, and sounds like something is partially shorting somewhere. None of the pins in the port should stop the machine from booting other than an actual 5V short... Maybe try measure the resistance between pins 7 and 8 on the mouse while unplugged, see what it is.

[idrougge]

This is why joysticks have all-plastic connectors. Many mice do not, and therefore it's easy to short a pin when hot-plugging the connector. The PSU detects the short and tries to save the computer by cutting the power.

[dschallock]

Originally Posted by Daedalus

It really depends on what the machine is doing. Pin 5 is configurable as an analogue input or a digital input or output. If it's left floating, it might be high if in digital mode or some weird in-between value if left unconfigured. The Cocolino software could also be pulsing that pin in a similar way to the CD32 controller, resulting in a multimeter getting an in-between value. I wouldn't be too concerned with that for the moment anyway.


That is a little weird, and sounds like something is partially shorting somewhere. None of the pins in the port should stop the machine from booting other than an actual 5V short... Maybe try measure the resistance between pins 7 and 8 on the mouse while unplugged, see what it is.

Ok, tested some things and this is what I got.

I tested resistance and I have a complete circuit between pins 7 and 8 on both ports.

also I have:

Voltage between 8 and 1 = 4.85VDC
Voltage between 8 and 2 = 4.85VDC
Voltage between 8 and 3 = 4.85VDC
Voltage between 8 and 4 = 4.85VDC
Voltage between 8 and 5 = 3.75VDC
Voltage between 8 and 6 = 4.85VDC
Voltage between 8 and 7 = 5.00VDC
Voltage between 8 and 9 = 4.81VDC

[dalek]

So this is in the 3000 that you replaced the PSU with an ATX one?

If the PSU is cutting out and restarting it's likely detecting a short. I'd recheck your wiring - something sounds wrong.

Checking the schematics (page 9): http://www.amigawiki.de/dnl/schematics/A3000.pdf probably something to do with your '+5V User' connection on PSU pin 12.

[dschallock]

What is Amiga PSU pin 1 = +VID? What is +VID? Should it have voltage?
The schematic shows that it comes from some source just called +VID and then it connects to the positive side of a fixed capacitor. Maybe not important but I don't see any of the amiga PSU to ATX conversions doing anything with that pin.
Except for pin 1, the whole top row of pins (2 through 5) are +5VDC
Then the middle row (pins 6 through 10) are all GND
On the Third row on PIN 12 it asks for +5V_User (like you mentioned @dalek ) but what is the difference between +5V_user and the top row of +5V? Is the _user pin 12 just +5VDC like pins 2-5 ?

[dalek]

If you check the schematics +VID is also used on the video hybrid and video clock regenerator circuit. Ian's guide suggests it should be connected to a regular 5V rail.

Ian's guide also suggests 5V_User be connected to a regular 5V source. I would check that it is coming off a separate 5V rail on your PSU if possible. Are you sure there are no shorts on that adapter you have on pin 12?

[dalek]

Oh, I saw in your other thread "...Including I made the little circuit to produce the -5v and everything seems to be working great.
EXCEPT..."

How are you generating the -5V? As a test, try disconnecting your mod and see if the problem still happens.

[dschallock]

Originally Posted by dalek

Oh, I saw in your other thread "...Including I made the little circuit to produce the -5v and everything seems to be working great.
EXCEPT..."

How are you generating the -5V? As a test, try disconnecting your mod and see if the problem still happens.

I used a little circuit made by TI called a lm7905 http://www.ti.com/lit/ds/symlink/lm79.pdf
and it take negative voltage between -7 and -30 and outputs -5v.
Are you saying to try running the system with that pin not connected to any power?

[dschallock]

Originally Posted by dalek

If you check the schematics +VID is also used on the video hybrid and video clock regenerator circuit. Ian's guide suggests it should be connected to a regular 5V rail.

Ian's guide also suggests 5V_User be connected to a regular 5V source. I would check that it is coming off a separate 5V rail on your PSU if possible. Are you sure there are no shorts on that adapter you have on pin 12?

So when I connected pin 1 of the PSU (+VID) to +5V it seems to have made pin 7 of the mouse port produce a clean +5V. That change makes the Amigakit optical mouse now work. All would seem cool except pins 1-5 are also producing 5V. It doesn't seem to affect the functionality of the mouse, but I'm guessing that's not good. hahaha.
I haven't tried disconnecting the -5V yet. That would be next to try.