| Pee Cee joystick [message #382847] |
Tue, 09 April 2019 22:21  |
|
|
Originally posted by: Anthony Adverse
I can get my hands on a reasonably spiffy super old joystick, similar to the plastic box with 2 buttons on it, and a button on top of the stick itself.. Doesn't seem to expensive, but the question is......
If you were going to modify this, would you strip it down and replace the lead with adding in whatever resistance is required to balance it out, or would you make a 15 -> 9 adapter and fit your balance resistors in there? I guess with an adapter you could probably use it with most other PC joysticks.
A
|
|
|
|
| Re: Pee Cee joystick [message #382849 is a reply to message #382847] |
Wed, 10 April 2019 02:48  |
Michael J. Mahon
Messages: 1767
Registered: October 2012
Karma: 0
|
Senior Member |
|
|
Anthony Adverse <the.ertceps@gmail.com> wrote:
> I can get my hands on a reasonably spiffy super old joystick, similar to
> the plastic box with 2 buttons on it, and a button on top of the stick
> itself. Doesn't seem to expensive, but the question is......
>
> If you were going to modify this, would you strip it down and replace the
> lead with adding in whatever resistance is required to balance it out, or
> would you make a 15 -> 9 adapter and fit your balance resistors in there?
> I guess with an adapter you could probably use it with most other PC joysticks.
>
> A
>
What will be required if you want to get 0..255 readings is not resistors,
but capacitors.
The maximum paddle reading is actually a reading of the RC time constant
formed by the maximum resistance of a joystick potentiometer and a fixed
0.022uF capacitor built into the Apple.
Since the maximum resistance of your joystick is less than the Apple spec
of 150K, you will need to add some parallel capacitance to produce a 255
reading with a lower resistance.
The total capacitance required with a joystick maximum resistance R is:
C = 0.022 x 150K/R, and since 0.022uF is already internal to the Apple, the
capacitance that you will need to add in the joystick is:
Cjoy = C - 0.022
So, for example, if your R is 100K, then the total C required is 0.033, and
the external capacitor required is 0.011uF.
Since capacitors are only available in discrete values, and have rather
large tolerances, you may need to parallel two standard values to ensure
that the RC time constant is large enough to return a value of 255, yet
small enough to make most of the travel range useful.
In the example given, I’d try a 0.01uF capacitor and check the range the
Apple reads. If it reaches 255 you’re done. If it falls short, try
paralleling an additional 0.001uF or 0.002uF. Use the smallest capacitance
that causes the Apple to read 255 so that the range of 0..255 covers almost
all of the joystick’s physical range.
The capacitor(s) can be connected inside the joystick or an adapter,
between the wiper and the connected end of the pot.
Use the same technique on the other pot and your joystick should serve
perfectly!
--
-michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
|
|
|
|
| Re: Pee Cee joystick [message #382851 is a reply to message #382849] |
Wed, 10 April 2019 07:31 |
|
|
Originally posted by: Anthony Adverse
etween the wiper and the connected end of the pot.
>
> Use the same technique on the other pot and your joystick should serve
> perfectly!
>
> --
> -michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
That is a lot more complicated than I expected. The ones I have seen in pieces in the past, have had the pot output linked straight into another resistor, and I would have expected to either parallel or serial another resistor into that to pull it into line with the A2 joystick. ergo the either increase or halfing on the resistance values. The capacitance is straight out of left field there... for me anyway.. given there's only ~50 ohms difference between the two. Although I guess capacitance is essentially the opposite of resistance :)
A
|
|
|
|
| Re: Pee Cee joystick [message #382852 is a reply to message #382851] |
Wed, 10 April 2019 14:27 |
Michael J. Mahon
Messages: 1767
Registered: October 2012
Karma: 0
|
Senior Member |
|
|
Anthony Adverse <the.ertceps@gmail.com> wrote:
> etween the wiper and the connected end of the pot.
>>
>> Use the same technique on the other pot and your joystick should serve
>> perfectly!
>>
>> --
>> -michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
>
> That is a lot more complicated than I expected. The ones I have seen in
> pieces in the past, have had the pot output linked straight into another
> resistor, and I would have expected to either parallel or serial another
> resistor into that to pull it into line with the A2 joystick. ergo the
> either increase or halfing on the resistance values. The capacitance is
> straight out of left field there... for me anyway.. given there's only
> ~50 ohms difference between the two. Although I guess capacitance is
> essentially the opposite of resistance :)
>
> A
>
Not the opposite, but quite different!
I gave a fairly complete explanation of why you can get the result with a
pretty simple modification. The reason for the theory is to allow you to
compute the added capacitance for *any* reasonable pot resistance less than
150K. (Adapting pots greater than 150K is much more complicated if
linearity is to be maintained. Fortunately this situation seldom occurs.)
Adding a small capacitor in parallel with the joystick pot is actually
easier to do than adding a series resistance, since no existing connections
need to be broken.
Procedure: 1) measure maximum joystick pot resistance, 2) compute
correction capacitor value, and 3) solder correction capacitor across the
two connected joystick pot terminals, 4) repeat for additional axes.
Some may wonder why I suggested simply paralleling the variable resistance
rather than connecting one side of the capacitor to ground. The reason is
that everything works as long as one side of the capacitor is at AC ground,
and +5v is a good AC ground. Connecting the capacitor to the two connected
terminals of the pot eliminates the need to determine which of those
terminals is the +5v terminal. This is straightforward for *any* variable
resistance type joystick.
If, instead, you were to add a series resistance of, say, 50K to a 100K
joystick pot, then the operating range of the joystick would be from about
85 to 255, which is certainly not what you want. Adding a parallel
resistance would lower the maximum resistance (and reading) even more, and
would also make the joystick response nonlinear.
I’ve “guessed” 100K for the joystick pots because that is the common PC
value.
--
-michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
|
|
|
|
| Re: Pee Cee joystick [message #382853 is a reply to message #382852] |
Wed, 10 April 2019 18:35 |
|
|
Originally posted by: Anthony Adverse
On Thursday, April 11, 2019 at 4:27:09 AM UTC+10, Michael J. Mahon wrote:
> Anthony Adverse <the.ertceps@gmail.com> wrote:
>> etween the wiper and the connected end of the pot.
>>>
>>> Use the same technique on the other pot and your joystick should serve
>>> perfectly!
>>>
>>> --
>>> -michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
>>
>> That is a lot more complicated than I expected. The ones I have seen in
>> pieces in the past, have had the pot output linked straight into another
>> resistor, and I would have expected to either parallel or serial another
>> resistor into that to pull it into line with the A2 joystick. ergo the
>> either increase or halfing on the resistance values. The capacitance is
>> straight out of left field there... for me anyway.. given there's only
>> ~50 ohms difference between the two. Although I guess capacitance is
>> essentially the opposite of resistance :)
>>
>> A
>>
>
> Not the opposite, but quite different!
>
> I gave a fairly complete explanation of why you can get the result with a
> pretty simple modification. The reason for the theory is to allow you to
> compute the added capacitance for *any* reasonable pot resistance less than
> 150K. (Adapting pots greater than 150K is much more complicated if
> linearity is to be maintained. Fortunately this situation seldom occurs.)
>
> Adding a small capacitor in parallel with the joystick pot is actually
> easier to do than adding a series resistance, since no existing connections
> need to be broken.
>
> Procedure: 1) measure maximum joystick pot resistance, 2) compute
> correction capacitor value, and 3) solder correction capacitor across the
> two connected joystick pot terminals, 4) repeat for additional axes.
>
> Some may wonder why I suggested simply paralleling the variable resistance
> rather than connecting one side of the capacitor to ground. The reason is
> that everything works as long as one side of the capacitor is at AC ground,
> and +5v is a good AC ground. Connecting the capacitor to the two connected
> terminals of the pot eliminates the need to determine which of those
> terminals is the +5v terminal. This is straightforward for *any* variable
> resistance type joystick.
>
> If, instead, you were to add a series resistance of, say, 50K to a 100K
> joystick pot, then the operating range of the joystick would be from about
> 85 to 255, which is certainly not what you want. Adding a parallel
> resistance would lower the maximum resistance (and reading) even more, and
> would also make the joystick response nonlinear.
>
> I’ve “guessed” 100K for the joystick pots because that is the common PC
> value.
>
> --
> -michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com
Well there you have it, I hadn't particularly considered linearity.. and wouldn't have thought of going near it with a capacaitor. Lives and learns. Thank you.
A
|
|
|
|
Members
Search
Help
Register
Login
Home
