View Full Version : Joystick simulation on the gameport.. Which potmeter value ?
01-27-2002, 08:42 PM
Like others, I make my own controllers by simulating joystick axis with Linear potmeters; now, I think officially the gameport was designed for 100 Kohm potmeters, but since you calibrate the controllers I find that any value between 10 and 200 Kohm can be used as well.
Question-1: But, how is the amount of noise ("jitter") on the axis related to the Potmeter value ??
I know that using shielded cable (connected to the connector frame and potmeter housing) greatly reduces noise; but does the noise level increase:
- with higher potmeter values ?
- with lower pometer values ?
- or is the noise level indifferent to the pot value ?
Anybody know this ? And more important, why ?
Question-2: Is the potmeter value important when you use a gameport-to-USB convertor ?
I'm not sure how a gameport-to-USB convertor works, but there must be some analogue-->digital conversion in it. What I discovered: when using a 47 Kohm pot, a full swing of the pot only gave me 7 discrete steps/levels when calibrating the axis in WIndows; using a 100 Kohm pot, I got 15 discrete levels.
Can you explain this ? (or is it indeed the fixed-level
A/D conversion in the convertor ?)
I'd appreciate any answers.
01-28-2002, 10:22 AM
>Like others, I make my own
>controllers by simulating joystick axis
>with Linear potmeters; now, I
>think officially the gameport was
>designed for 100 Kohm potmeters,
>but since you calibrate the
>controllers I find that any
>value between 10 and 200
>Kohm can be used as
>Question-1: But, how is the amount
>of noise ("jitter") on the
>axis related to the Potmeter
>I know that using shielded cable
>(connected to the connector frame
>and potmeter housing) greatly reduces
>noise; but does the noise
>- with higher potmeter values ?
>- with lower pometer values ?
>- or is the noise level
>indifferent to the pot value
The noise is not relevant to the value, it is relevant to the QUALITY of the pot, pay a little more than the cheapo and you will be fine :-)
>Question-2: Is the potmeter value important
>when you use a gameport-to-USB
>I'm not sure how a gameport-to-USB convertor works, but there must be some analogue-->digital conversion in it. What I discovered: when using a 47 Kohm pot, a full swing of the pot only gave me 7 discrete steps/levels when calibrating the axis in WIndows; using a 100 Kohm pot, I got 15 discrete levels.
>Can you explain this ? (or
>is it indeed the fixed-level
>A/D conversion in the convertor ?)
>I'd appreciate any answers.
a good value is 100k linear over the range - ie - if you use only 1 fifth of the range, use a pot that is five times the value. The a to d's in the gameport adapter are probably wired as an r/c circuit on the inputs, in that case, you would need to experiment with the value that gives the best step range. You would also have to use the pot as a variable resistor not as a potential divider.
01-28-2002, 03:02 PM
Excellent questions. Although I know quite a bit about electronics, I'm no expert in joystick devices. The resistor (pot) is part of a multivibrator timing circuit and "significant" noise pickup will cause erratic triggering. The use of gizmos such as light dimmers in your house will produce spikes on the power line which will radiate and may get into your controller wiring.
I tried shielded cable thinking the jitter was due to noise pickup, but I got little improvement. Simply twisting each pair of wires should be sufficient. I believe the jitter is inherent in the circuit. Since there are descrete levels it seems the detector circuit jumps back and forth between the two closest levels as if it can't make up its mind which one to lock on to. I would think that a greater number of levels would reduce the jitter, but I have not tested that theory as the jitter does not appear to interfere with my actual flying.
Potentiometers, either rotary or slide, are available with various resistance tapers. Buy ones with a linear taper which means that the resistance change is constant over the entire range. To determine if you have linear taper pots, set the pot at the center and measure the resistance from the wiper to each end. It should be about the same. Pots with other tapers will work, but you may notice the sensitivity of control is greater on one side than the other when flying.
Check out this website for lots of interesting info:
01-28-2002, 08:11 PM
LAST EDITED ON Jan-28-02 AT 08:18PM (EDT)[p]Hi Guys (assuming you are male :-),
Thanks for your answers.
I will experiment some more with different pot values and ways-of-wiring.
I was just curious to see if someone had a theoretical explanation for the effects I described.
I'am especially interested how the USB convertor works exactly.
Joe, what do you mean with "You would also have to use the pot as a variable resistor not as a potential divider" ?
Yokeman, I found Tomi's site a few months ago as well, and I mailed him several times but didn't get any answers. Doesn't really answer my questions, does it ? And it seems a bit outdated with respect to USB.
01-29-2002, 04:38 AM
hi, if you conect a pot in circuit with the outer pins connec5ted to - say - V+ and V-, then the wiper will ouput a voltage based on the potential divider caused by the circuit
if you connect a source to one pin only of the pots outer 2 pins, the wiper will act as a variable resistor
does that make sense, or do you want me to go a bit deeper ?
01-29-2002, 03:58 PM
I understand what you mean.
And yes, I always use a pot as a variable resistor for this purpose.
02-01-2002, 10:49 AM
In response to your request for info concerning the inner working of the gameport to USB converter, I have researched various chip manufacturers and their application notes to determine if an application specific chip is available for this purpose. I couldn't locate one. A programmable microcontroller is used as an interface. The particular type varies depending upon the maker of the converter. Essentially what it does is sequentially sample each of the 8 data lines of the game plug and sends it serially in binary code to the USB connector the way a keyboard does. It doesn't seem to be the sort of thing most homebrewers will mess with unless they are into writing code. It probably won't be long before Poptronics mag comes up with an article on how to build one.
02-01-2002, 11:27 AM
I should have working microchip PIC (the new 18 series) devices available as usb 8 analogue, devices in a few months. Just waiting for samples of the new 18 series.
I will make the kits available to cockpit builders.
Powered by vBulletin® Version 4.2.0 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.