English Amiga Board


Go Back   English Amiga Board > Support > support.Hardware

 
 
Thread Tools
Old 23 October 2019, 23:22   #1
jimbob
Registered User

 
Join Date: May 2006
Location: Kilmacolm
Age: 41
Posts: 625
A600 Right mouse button fault

Hello, got an odd problem with an A600 here. In short the right mouse button either doesn't work at all or occasionally sort of works but seems to have a delay between pressing it and detection.


1st odd thing is the button detection works perfectly if you disconnect the 5 V from pin 7 of the mouse port, (This stops the motion detecting from working but the buttons still work.)


When the mouseport power pin is disconnected, Paula pin 36 sits at ~3.7V and button presses are detected properly , (this pin being shorted to ground by the button press).


When the mouseport power pin is connected, Paula pin 36 sits at 5V and the button press does not work reliably but shorting pin 36 manually to ground does read reliably as a button press.


2nd odd thing is that on some occasions, with the mouseport power pin connected, a button press will be detected after a short delay that seems to coincide with the floppy drive click.


Anyone any idea whats going on here? Does the Amiga do some detection of what is plugged into the mouse port?


I've eliminated usual suspects, replaced mouse port and passive components to pin 9 and also resoldered relevant Paula pins. Tried various different mice which brings up odd thing number 3. When using a Micromys USB mouse adaptor motion is usually not properly detected, though occasionaly it works as normal, with the same button issue remaining. Button issue also remains when I use a mouse port switcher.



jimbob is offline  
Old 24 October 2019, 00:17   #2
solarmon
Registered User

solarmon's Avatar
 
Join Date: Dec 2018
Location: UK
Posts: 686
Have you used Amiga Test kit to check that CIA timings are OK?
solarmon is offline  
Old 24 October 2019, 00:22   #3
jimbob
Registered User

 
Join Date: May 2006
Location: Kilmacolm
Age: 41
Posts: 625
Not familiar with the Amiga test kit, I have a DiagROM chip which iirc has a CIA timing test, will try it out. Thanks.
jimbob is offline  
Old 24 October 2019, 00:33   #4
solarmon
Registered User

solarmon's Avatar
 
Join Date: Dec 2018
Location: UK
Posts: 686
Quote:
Originally Posted by jimbob View Post
Not familiar with the Amiga test kit, I have a DiagROM chip which iirc has a CIA timing test, will try it out. Thanks.
I highly recommend it. Obviously, it requires that your Amiga can boot and load Amiga Test Kit.
solarmon is offline  
Old 24 October 2019, 01:05   #5
jimbob
Registered User

 
Join Date: May 2006
Location: Kilmacolm
Age: 41
Posts: 625
Amiga seems to be working perfectly otherwise, can you give me a link? EDIT(found it!)


Just tried DiagROM. CIA tests all passed, with or without mouse plugged in or button pressed. Following results from the mouseport tester.


POTINP reads $5500 with no button pressed.


Button press has no effect but when I short pin 9 to ground. (Have verified the switch does work!)



POTINP reads $5100 and RMB is detected, I assume this is normal.


Also, and not sure if this is normal, when POTINP is registering a RMB press POT0DAT upper byte looks like it may be cycling through a counter?

EDIT:Tried Amiga test kit too. CIA tests all passed. Only new information now is that joystick port when configured as mouse also fails to detect RMB but either mouse or joy port, when configured as a gamepad, pressing the RMB will cause detection of all gamepad buttons except the D-Pad.

Last edited by jimbob; 24 October 2019 at 02:49.
jimbob is offline  
Old 24 October 2019, 04:18   #6
Shatterhand
Warhasneverbeensomuchfun

Shatterhand's Avatar
 
Join Date: Jun 2001
Location: Rio de Janeiro / Brazil
Age: 36
Posts: 3,437
Loose pin?

My mouse has a loose pin for the left button. I need to keep something below the connector so it's forced a little up, this way the pin works properly.

It has been like this for like a couple of years now. I have a new DB-9 connector to change but haven't yet brought it to someone who can do it (because I won't get anywhere near my A600 with a solder iron )
Shatterhand is offline  
Old 24 October 2019, 10:33   #7
Daedalus
Registered User

Daedalus's Avatar
 
Join Date: Jun 2009
Location: Dublin, then Glasgow
Posts: 4,388
The right mouse button is nothing to do with the CIAs - it's handled by Paula as the pin is one of the A/D inputs. It's possible that there's damage to a resistor or ceramic capacitor on that line, or some PCB damage from capacitor leakage. If you've already checked over the ceramic capacitors, I would guess it's a damaged trace that's offering a high resistance to the pin. There are several electrolytic capacitors near the pin 9 traces that can be damaged by leakage.

The voltage on pin 9 of the ports should be relatively high (well over 4V) when no button, and 0V when the button is pressed.
Daedalus is offline  
Old 24 October 2019, 14:08   #8
Vypr
Registered User

 
Join Date: Dec 2016
Location: East Kilbride, Scotland
Posts: 182
I had a very similar issue on an A2000 which had me pulling my hair out. What I owuld notice is that the right I replaced all the components between the mouse and Paula to no avail until the day I decided to clean some leftover corrosion from when the battery was removed. After scrubbing the area where the corrosion was with a fibre pen and recoating the exposed traces the problem vanished so I can only guess that the issue is what Daedalus was describing, the corroded areas were causing signal leakage on the 5V line
Vypr is offline  
Old 26 October 2019, 19:07   #9
jimbob
Registered User

 
Join Date: May 2006
Location: Kilmacolm
Age: 41
Posts: 625
Well, I'm a bit stumped for the moment, must be a marginally failing component or trace somewhere as suggested. Here is the current state -

Both standard and micromys mouse tested fully working on another A600. Both 9 pin connectors replaced with new, all passives in the mouse port to Paula circuit replaced. Measures 68R to Paula pin 36 as expected. Resistance to ground and 5V comparable with working board. Hmm, there are a few other 470pF caps which shunt other pins to ground at the same area, must suspect them next.

Micromys mouse movement usually does not work properly in either mouse port, standard mouse motion works. This may be a good clue, what is different about micromys compared to standard mouse?

Neither right mouse button is detected, except when setting gamepad mode in Amiga Test Kit, this activates all non D-pad switches, not sure how the port is configured to detect all those buttons in an actual gamepad?

With mouse unplugged, shorting pin nine to pin 8 does register a right click. With the mouse plugged in, making this circuit elsewhere on the board also detects a right click.

Board is Rev1 A300. Cleaned thoroughly during a recap some months ago when it did not have this problem.
jimbob is offline  
Old 27 October 2019, 04:54   #10
jimbob
Registered User

 
Join Date: May 2006
Location: Kilmacolm
Age: 41
Posts: 625
Well, managed to get the mouse buttons working but I'm not too happy with the solution.

I replaced the 68R resistors from both pin 9s with 47R.

I measured the Paula pin voltages with RMB pressed at just over 1V and guessed this was too high. But at the mouse port side of the 68R resistors was at 0V. So lowering the resistance, lowers the voltage drop here enough to detect the button press. Now around 0.8V which is comparable with my other working board. Question is should Paula be drawing ~15mA in this case? Can anyone explain how the Paula pins are supposed to work and do you think they are a bit faulty now and likely to get worse after my "fix"

Also, Micromys mouse motion still usually doesn't work properly so something else is also on the edge I think. Thanks
jimbob is offline  
Old 27 October 2019, 07:06   #11
Bruce Abbott
Registered User

Bruce Abbott's Avatar
 
Join Date: Mar 2018
Location: Hastings, New Zealand
Posts: 292
Quote:
Originally Posted by jimbob View Post
Question is should Paula be drawing ~15mA in this case?
The A500 has no resistance between PAULA and pin 9 on the mouse port. I measured the short circuit current to ground on my A500 with 3 different Paula chips installed. The results were 19.5mA, 17.8mA, and 16.1mA.

I also tested my A600 (14.9mA) and A1200 (12.0mA). These values may be slightly reduced due to the 68 Ohm resistor in series.

So 15mA appears to be quite normal. Different chips will have different values due to process variation (doping levels etc.). This can also cause different logic threshold voltages, so perhaps your Paula has a lower than normal threshold voltage on the RMB input.

Anyway, since the A500 doesn't have any series resistor at all it should be quite safe to use 47 Ohms.
Bruce Abbott is offline  
Old 27 October 2019, 23:20   #12
jimbob
Registered User

 
Join Date: May 2006
Location: Kilmacolm
Age: 41
Posts: 625
Thanks Bruce, that's good enough for me.

Confirmed that the current with button pressed is now ~14mA and if make the circuit after the 47R resistor it is ~20mA.

Micromys actually worked properly last few tries also. No clue why this should be but this one is going home now anyway.

I guess the resistors were added as some protection against hotplug accidents or pin configuration/port device mismatches.
jimbob is offline  
Old 28 October 2019, 12:31   #13
Daedalus
Registered User

Daedalus's Avatar
 
Join Date: Jun 2009
Location: Dublin, then Glasgow
Posts: 4,388
There are EMI filters on the A500 between Paula and the port pins (5 and 9), which will present a small resistance. Later models replaced these with discrete resistor and capacitor elements, presumably for cost reasons.

The difference in movement with the Micromys and the standard mouse is that, typically, the standard mouse uses photodiodes to produce the pulses, whereas the Micromys will output logic levels. These voltages will differ depending on the additional circuitry. As for the right button situation, this is complicated by the fact that the inputs on Paula aren't logic inputs at all, but analogue inputs, so the threshold is determined by the software reading the port, not the normal TTL thresholds as for the directional inputs.
Daedalus is offline  
 


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Similar Threads
Thread Thread Starter Forum Replies Last Post
Amiga mouse cable fault alphagemini OT - Technical 2 01 July 2018 19:22
A2000 left mouse button use in workbench causes mouse to stop responding. gary_nz support.Hardware 5 17 September 2017 11:03
A600 right mouse button not working protek support.Hardware 48 18 March 2013 15:57
A600 Level 7 button and/or adapting Nordic Power to A600 internally? Photon support.Hardware 16 01 October 2009 23:06
a600 video fault elmasta support.Hardware 14 03 January 2009 15:34

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +2. The time now is 03:46.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, vBulletin Solutions Inc.
Page generated in 0.07930 seconds with 15 queries