First Amiga - Mouse Control Issues

[Simon Carter]

Hi folks,

 

I've recently picked up my first Amiga 500, seems to be in decent working condition all round with the exception of the mouse/joystick ports. Currently using a mouSTer adapter to use a USB mouse with the computer (don't have any working DB9 pin 'real' mice to test with), and the only results I get are a jerky, random style cursor movement, totally unusable. Joysticks use is also the same, unreliable, with random reads from input.

 

I've picked up the ATK v1.21, and whilst I'm sure it is a great piece of software, I cannot for the life of me find any documentation for it, so any results I'm getting might as well be written in Mandarin. Most useful, would be a range of 'what good looks like' images so I could begin to troubleshoot.

 

Through basic searching I've read that the twin CIA chips could be the issue, I've swapped these over and had the same results and there don't seem to be any physical issues with the actually DB9 sockets themselves.

 

Checking the controller ports through ATK, on the analog setting, the system does seem to be getting some form of input even when there is nothing attached to either port. As soon as I reconnect the mouSTer adapter the input stops.

 

Appreciated this is very little to go on at the moment, and if required I can provide photos/videos.

 

Any help would be greatly appreciated

 

Si

 

[+OLD CS1]

I cannot find my 500 schematics right away, but in the A2000, there is an LS157 (U202 in the 2000) which sits between the ports and the CIAs.  ISTR this is the same arrangement in the 500.

[Daedalus2097]

Yeah, it's the same arrangement in the A500, but let's step back for just a moment. The CIAs have very little to do with the controller ports. They only handle the fire button / left mouse button, and nothing else. So it's expected that any movement-based issues are the same when they're swapped.

 

The movement pins are routed through a multiplexer circuit as mentioned, and then on to the Denise chip for reading. The mouse and joysticks both use the same 4 pins for movement so if they're not working correctly it'll affect both.

 

AmigaTestKit is indeed an excellent bit of software, and while it can give you a lot of deep access and information, the part you need is the controller ports part, and you seem to have found that already. Switching to analogue will give some random input when an analogue controller isn't connected; analogue controllers don't use the same pins for directions as standard (digital) joysticks or mice. The main thing to look at is how the joystick mode responds to a joystick in each port. The directions will light up white when they're active, and green when they have been active. This should give you a good idea if the movements are any way related to the inputs. If the inputs are triggering without pressing any direction, are they still triggering when the joystick or mouse is removed? Is the behaviour the same on both ports? What if you put the mouse in the other port (it won't work in Workbench but ATK can show you it)?

 

The small logic chip mentioned above (74LS157 or similar) is used to gather all the directional pins (1-4 on each port) into a single data stream, so it can affect some or all of those pins. It's not unheard of for that chip to fail - unfortunately it's soldered in so it's slightly more work to swap it out. There are also pull-up resistors assigned to each pin, and failure of these (or the 5V supply to these) could cause similar random inputs. These resistors are in a resistor pack designated RP401, and should each be 4.7K to 5V. It's worth checking that this is working as it should - when running properly and with nothing connected to the ports, pins 1-4 on each port will have a steady 5V present.

 

The schematics are usually found in nice vector PDF format on amigawiki.org, but it seems to be down at the moment.

[+OLD CS1]

When you find the A500 schematics, also check for two resistor packs between the '157 and Denise.  The Amiga 2000 schematics show RP204 and RP205, both seemingly positioned near their relative components on the data lines M0H, M0V, M1H, and M1V.  RP204 looks to be an in-line resistor pack, so it might be a bank of resistors on some mainboard revisions rather than an actual pack (array) with a common connection.  I am frustrated by not having all of my schematics handy 😠

[Daedalus2097]

Those resistor packs don't (or at least, shouldn't) exist on the rev. 6 or rev. 8 A500s anyway, so don't worry about them. On an A2000 they could indeed cause some funky issues.

[eightbit]

Look for damage to R401 and R408. I had a stupidly made mouse adapter knock out the mouse and burn one of these out some time ago. Be careful with some of those homebrew adapters online. 

 

You can see R401 in my instance was replaced to get the mouse functioning again:

 

 

 

And the damaged original:

 

 

 

Not saying it is your issue but it's a good idea to check to see if these look burnt/damaged in any way. They are underneath the floppy drive BTW.

[Simon Carter]

Looks like I have a rev.6 A500 so the resistors packs shouldnt be an issue, thanks guys for helping out with that. I'll need more spare time than I have at the moment to fully investigate the avenues that @Daedalus2097 has mentioned. I'll also check out the resistors that @eightbit identified.

 

Many thanks folks, assuming, I get a chance I'll update the thread on what I find

 

Si

[Daedalus2097]

R401 and R408 protect the 5V supply to the controller ports, and will burn out shorted for any significant length of time, and should really be a heavier power rating that most resistors on the board - maybe 1 or 2W. When they fail you won't get the 5V supply on pin 7 of the relevant port. That's not needed for basic joysticks, so you should get normal behaviour for basic joysticks, but no or spurious activity from active devices, e.g. mice, mouse adaptors, CD32 pads.

[Simon Carter]

I'm no electronics expert but I'm going to go out on a limb here and thank @Daedalus2097 and @eightbit for definitely giving me a place to look. My multimeter is buried in my loft somewhere, but I don't think I need it...

Edited September 4 by Simon Carter

[+OLD CS1]

1 hour ago, Simon Carter said:

Yeah, looks cooked, both literally and as the kids would say!  Nice to see it did its job.  That is a 1/4W 4.7 Ω resistor, which means something pulled more than (ideally) 50mA through the port for long enough to burn it out.  Usually, you can get away with a little over and it will just get hot.  As @Daedalus2097 pointed out, likely a short at the port so the resistor acted as a fuse.

 

Calculating back from pure 5V over 4.7 Ω gives a wee bit over 1A.  At a dead short, that is 5W, way way way over the limit of the resistor.  I wonder if it glowed before it burned out.

 

EDIT: Taking a second look, that is more likely a 1/2 W resistor, limiting the circuit to 100mA.  @Daedalus2097 out of curiosity, what does your schematics show?

[eightbit]

Be VERY careful with that "mouSTer adapter". It may have been the culprit.

 

Like I said, been here and done that. If you decide to test it again you might as well test with the board out of the machine just in case you zap the replacement resistor

 

Here's the 10 pack I has purchased when this happened to me back in 2019: https://www.amazon.com/gp/product/B01D0GEWC2, but you being in the UK you may have a different local source. 

 

Good luck!

 

 

Ohh, and to answer @OLD CS1's question, I didn't see it glow. But I was right there with the machine open when it happened and I can tell you it let out some magic smoke and an electrical burning stench. Since they were underneath the floppy drive I seriously thought it was coming from the drive itself. All caused by some $11.99 mouse adapter on eBay. I wasn't happy!

Edited September 4 by eightbit

[eightbit]

Here was the original resistor PN:

 

 

4.7 Ohm 1/2w 

 

Edited September 5 by eightbit

[Daedalus2097]

The schematics I have (redrawn vector versions of the originals) don't give a power rating for the parts; that's in the BOM, and appears to be 0.5W as listed above, though they look to be heavier (maybe 1W) on the machines I have to hand here. Ultimately, you'll need a huge resistor and a heatsink to indefinitely survive a dead short on the port, so what the power rating of the resistor does is determine how long it will last in short conditions before burning out. A 0.5W resistor will handle more abuse than a 0.25W one.

 

If an adaptor from eBay was shipped with a dead short, that's pretty damning on whoever was selling it. I'd kick up a fuss about that. There's a particular retailer that seems to sell poorly soldered devices (and not always cheap ones either - think TF accelerators) that have on occasion failed to work and potentially risked damage to machines, and has the audacity to put a "QC Passed" sticker on such boards despite having clearly never tested them. I wonder if it was the same seller as that dodgy adaptor...

Edited September 5 by Daedalus2097

[+OLD CS1]

4 hours ago, Daedalus2097 said:

that's in the BOM, and appears to be 0.5W as listed above, though they look to be heavier (maybe 1W) on the machines I have to hand here.

I would stick with 1/2W resistors rather than 1W.  I get they can take more abuse, and maybe modern power supplies will not suffer, but I would prefer a lower limit on the power which can be shorted through a port.  If only to help save a faulty, but not failed, device.

4 hours ago, Daedalus2097 said:

(and not always cheap ones either - think TF accelerators)

This makes me shudder and rage at the same time.  Like you say, accelerators are not cheap, and the risk of taking out my system with a dodgy device is not comforting.  Especially since so much of this stuff comes from outside of our country, where we are unlikely to be able to easily seek remedy for damages if the seller says "too bad, so sad," and terms like "not liable for damage to your device" exist.

[eightbit]

I purchased one of these because I know quality is behind what this seller makes:

 

https://www.ebay.com/itm/303872542435

 

But, he's had shipping problems. His shipping used to be free (via postal service) but things kept going missing. Now he is using FedEx which I am sure is more reliable, but the cost for shipping is now almost the cost of the adapter

 

You just have to be careful who you purchase from and ask around in Amiga forums (or here) and get good recommendations. There are too many "hobbyists" out there just looking to cash in on the craze and some of these people really don't care how well the product is built. 

 

By the way, that adapter and this made the perfect replacement for my Amiga 500 which was missing the OG tank mouse when I got it:

 

https://www.amazon.com/gp/product/B09F18BZSZ

[Daedalus2097]

I make hardware myself and sell it, and I'd be utterly ashamed to ship something to anyone without having 100% tested it on my own hardware first. To not do that is lazy and negligent at best, and possibly a cynical cash grab as you say.

[Simon Carter]

I replaced the resistor, like for like, but sadly no joy. Haven't put the mouSTer anywhere near the computer this time, although I'm pretty certain the issue was present before I even used one. Plugging a known working joystick in port 1 when playing Arkanoid 2 as a simple test game had similar results to before the resistor replacement, and I don't think I did too bad a job of soldering it in. I'll keep looking when I have more time, work is about to get crazy busy, so wont have spare time to deep delve solutions. 

 

Loads of thanks to you guys for helping out

 

Si

[eightbit]

On 9/8/2024 at 4:50 PM, Simon Carter said:

I replaced the resistor, like for like, but sadly no joy. Haven't put the mouSTer anywhere near the computer this time, although I'm pretty certain the issue was present before I even used one. Plugging a known working joystick in port 1 when playing Arkanoid 2 as a simple test game had similar results to before the resistor replacement, and I don't think I did too bad a job of soldering it in. I'll keep looking when I have more time, work is about to get crazy busy, so wont have spare time to deep delve solutions. 

 

Loads of thanks to you guys for helping out

 

Si

 

Can you take pictures of the entire board in high res? Maybe we can scour over it and see if we can see something visually. When you have the time of course...