Voltage Divider

Sounds like your divider chain has too low a resistance, & is loading the output from the joystick. What are the highest resistor values you tried? What is the input impedance of your measuring device? What's the input impedance of the arm module?

I suspect that you do not understand what John is quite accurately asking for. So I'll give you a few simple tests that you should do. I suspect also that you do not have a schematic of the potentiometers used in your joystick nor the schematic of what you wish your joystick to connect to. So this is how to test their source and input impedances. The simplest analog joystick signal will be just a potentiometer. It maybe easier to power your joystick with a separate supply for these tests. Now power your joystick without connecting it to your controller electronics. Measure the output voltages with a high input impedance device like a digital volt ohm meter at mid-point setting, fully forward and fully backwards. (This might be what you already did but one needs to know the test circuit , not just the numbers from the test.) Now return to your mid-point position. If you add a fixed resistor now in parallel with your volt meter and the voltage drops precisely in half, the resistor you added will be the output impedance of the joystick at mid-point. But don't go nuts trying to find this by putting in different resistors. As long as a significant voltage change is noticed, recorded and the fixed resistor value is recorded; the output impedance of the joystick can be calculated. With the load resistor still in place, record the new fully forward and fully backwards settings. (Sometimes the output impedance significantly changes with an analog signal.)

Now to measure the input impedance of your electronics will be similar but with less steps. Take a AA battery and record the voltage across the battery. Add a series resistor to the battery and connect this resistor and battery combination to the input of your electronics. Once again when the voltage at the input electronics is exactly at half of the battery voltage, the series resistor has the same value as the input impedance. If a 510K series resistor will not drop the voltage in half, we'll just accept that you have a high input impedance. This may happen.

Once you return this collection of data, I'm certain a simple accurate circuit to meet your needs can be fabricated.

(I do wish that people who start a question would have to sign in with a name.)

A voltage divider is simple but perhaps too simple in this case. If you put the joystick output onto one end of a two-resistor network and take the arm input between the two resistors (the textbook way of using s voltage divider), you don't have isolation between the output of the voltage divider and the arm module. The arm module is parallel to one of the resistors in your voltage divider network, so if the arm has low input impedance it doesn't really matter what resistor is in the voltage divider.

Why not use an opamp instead of a resistor network? That way you have isolation between the joystick and the arm module, and you can choose whatever voltage range you want for the output of the opamp.

So I made a profile now, so i'm not a guest.

Hey that sounds really good. I may just try to do that then. Thanks for the input.

Thanks for registering, and welcome to CR4!

Yes a simple divider maybe too simple. That's why I asked for specific testing to be done to determine what will work before trying a dozen circuit configurations. Just a simple engineering practice I've found useful over the years. When you don't know what you have, find out before you waste your time.

Regards.

It seems that the Out=put Impedance Zout is too high causing a low ''Fan-out'' ie cannot handle more current.

As the supply to Joystick control is built in [rather drawn from internal supply & usually through voltage dividerss so no further dividers can be derived.

Add a non-inverting amplifier with sufficient ''Fan-out'' between Joy-stick control points the control-voltage

or

build your network direcly from a separate power supply for your control network.

Can you run the joy stick on 3.3V?

GUEST,

IF THIS THE TYPE OF JOYSTICK THAT YOU HAVE THE OUPUTS DEPEND ON THE LOAD OF THE EQUIPMENT INPUTS FROM THE RHEOSTAT WIPERS AT PINS 3, 6, 11 & 13 TO PROVIDE A VARIABLE CURRENT.

YOU WOULD HAVE TO GROUND THE OPEN ENDS OF THE RHEOSTATS TO MAKE THEM INTO POTENTIOMETERS AND REDUCE THE DC VOLTAGE APPLIED TO PIN 1 AND 9 TO GET 0 TO 3.3 VOLTS OUTPUT FOR A VARIABLE VOLTAGE.

THAT'S THE BEST I CAN DO WITH THE LIMITED INFO PROVIDED.

JON

The joystick needs the +5Vdc input. and the pinout does not work that way.

Its a Sidewinder 3D Pro running as a simple analog joystick without the drivers for it.

heres the pinout.. http://www.allpinouts.org/index.php/PC_Gameport

Ok cool, if the analog circuit diagram is precisely what your link shows (particularly that the potentiometers are 100k) then building one +3.3V reference supply (the maximum input voltage) to replace the +5V into the joysticks will solve your problem for both axes. The simplest reference would be a resistor zener combination. A 1N4728A zener diode will do. The anode of the diode goes to ground. The cathode (color band) of the diode goes to the +3.3V joystick input (formerly the +5V pin 1). Lastly the resistor goes from the +5V supply to the common node of the diode cathode and the joystick supply. This resistor should be 5.6Ω<R_s≤R_l*1.7/6.6. Now R_s is the series resistor that you are adding. R_l is the input impedance of your electronics. To measure the input impedance you can just use your volt meter's resistance measurement capability instead of the complicated method I previously described. To minimize power consumption I would choose R_s≈R_l/4.

Like Redfred said:

Rs 5.6Ω

Jon

Wait a minute, you gave us the data needed to calculate your input impedance if the potentiometer values are 100K. 4.4V/R_l=5.0V/(R_l+100K) → Rl= 440K/0.6≈730K. So with that high of an input impedance you must get into the joystick and ground the unconnected potentiometer end as kudukdweller9 drew here. (Nice work by the way ) But to save some power and to not overheat anything I would make Rs≈18K. This will keep the zener always in the avalanche region providing sound regulation.

Red,

That's better.

Jon

ChicoEE,

What I said before still applies.

If this is the Joystick that you have it depends on the load of the equipment inputs from the Rheostat wipers at Pins 3 & 6 to provide a variable current.

You would have to ground the open ends of the Rheostats to make them into Potentiometers and reduce the DC Voltage at Pin 1 to about 3.3 Volts using "Vreg" for the desired variable Voltage in the 0 to 3.3V range.

Jon

kudukdweller9,

You will get a more accurate response by attaching the unused potentiometer end to ground. But this will require minor rewiring of the joystick. Also without knowing the input impedance of the sensing electronics one cannot design or select a suitable +3.3V regulator.

Red,

"attaching the unused potentiometer end to ground"

I said grounding the unused Rheostat end makes it a potentiometer and the wiper will then go from 5 volts to zero volts.

The input impedance of the electronics is usually quite high for the sake of linearity.

Jon

Quite correct. Notice the GA I gave you and my slight change of Rs to keep things cool.

Thanks Fred,

Well I took apart the 15pin connector and there were only 10 wires connected. The pins at 6,8,9,13,15 have no connection with this pin or the circuit. And I tried to get some sort of a datasheet for the Sidewinder 3d pro, but to no avail. Even without those grounded I have obtained outputs.

So you're pretty much saying not to use an opamp to get the voltage drop? and change the input voltage just to 3.3V while grounding those pins that are not connected?

David

Yep. See the schematic that kud.9 drew up. You will have to ground the unused potentiometer end though.

yes I misread his response and it through me off. Cuz I did not follow right away.

So I'm grounding out which side?or which pins?11,13? or must I open up the joystick and ground that?

I wouldn't have thought this joystick would be beyond me...lol

Sorry to say but you have to open the joystick. You will find for each axis of motion a three lead part with only two wires on it. You must add a ground wire to the unused lead.

ChicoEE,

Is this your sidewinder?

http://home.utah.edu/~dlg7/sidewinder/repair.html

or http://zapp.koolbear.com/index.html

Jon

yes that is, where did you find that? I've been trying to find something similar. so i took it apart and I do not see where the pin I need to set to ground is.

ChicoEE,

Google as usual.

Jon

Well thanks a bunch for the ideas i do appreciate it. and sorry to have misled, I did not understand how the sidewinder actually operated.

Back to the drawing board.

Thanks again,

David

ChicoEE,

Forget it.

The Joystick needs the 5 volts to operate optical devices and ICs in it.

After digging through the net I found that you had no idea what was in the Microsoft Sidewinder 3d pro controller and having been mislead by you we gave you wrong information.

The Joystick uses optical position sensors and the outputs are from some kind of processing ICs.

Your task is not simple.

Jon

Well I misled myself, I was assuming it was a basic analog joystick. Yea I had no idea what was entailed.

So Ideas? or would using the opamp idea be ok for that voltage drop?

CicoEE,

The whole mess requires drivers to operate on a computer.

Jon

and that's not what I am going for. Trying to use this to control a turret based on the voltage given and then writing code to appropriate the PWM output on the motors.

zero to min to max input changes speed, direction and then duration of input to output.

Thanks for the input though.

ChicoEE,

Using an O'scope it would be interesting to see what kind of outputs follow movement of the stick.

Jon

Well from what I have measured on the x&y axes before I took it apart it was a steady 4.67V at zero point then a max and min when moving left to right at 4.36V to 4.99V.

and the firing trigger is a digital high until pulled.

ChicoEE,

X and Y sounds like an Analog to digital function at work.

Jon

Kud,

So are you thinking that I could not translate that to perform the actions I desire?

But the voltages do change depending on the angle as well, so a smaller angle say to the left would represent 4.7V up until the max angle giving the max voltage.

I spoke with some others in another forum about the joystick and I got responses saying if I did not use the drivers then it would function as an analog joystick with digital high buttons.

David

David,

I think you could "jerry rig" a translator for the analog to get it in the range you need.

Sounds like an interesting project.

Jon

_{(I think we've talked about this before ...)}

In the light of new information: without the drivers it will emulate a simple analog joystick - but only when connected to a gameport.

A circuit with a 5V supply and an op-amp or two - & maybe a couple of comparators for the button signals - should be able to give you, say, 0.3V to 3.0V, centered on 1.65V, for the joystick output, and maybe button signals switching between 0.3 and 3V, all suitable for input to your arm module (whatever that may be).

Maybe someone here has an hour or two to spare?

Well I'm going to try to use a difference amplifier and see how that works. Theoretically by calculation it gives me centered at 1.5V a min of .45V and a max of 3.1V

Just need to go to a lab and test it since I do not have all the equipment here at my house.

Thanks,

David