Saitek Pro Flight Yoke System and Bodner board

[MattiasLind 0]

Hi, 

I am new to this and searched over internet for some information.

I got Saitek Pro Flight Yoke System with a broken circuit board and want to fix this. I found out that a Bodner bu0386x would be good for this, but is it possible to also connect the throttle quadrant to the same card without loosing functionallity as reverse throttle and so? Anyone know where to find some information of how to connect yoke and tq to the card? 

I know that the display on the yoke wont work, but i wont use that anyway. 

Regards Mattias 

[Jim Young 2,356]

1 hour ago, MattiasLind said:

I got Saitek Pro Flight Yoke System with a broken circuit board and want to fix this.

I always wonder how one breaks their circuit boards but the Bodner 386x would definitely work.  I've never tried to connect the throttle quadrant to the same card though and may give it a try and let you know!

[MattiasLind 0]

I have no idea. I am quite tired of logitech atm. I got first one that connected and disconnected to the computer all the time. After a couple of hours it went dead. The other thay i got a new one with same problem, but this went dead after 10min. All i dis was having it plugged in to my computer. Never had any problems with other usb device. 

[flyforever 267]

It should be possible. The board doesn't care whether there's one or more axis coming from different units. For practical reasons, it might be useful to build a small box with the controller and its inputs  in the form of terminal strips, named as axis one, etc. , and same with buttons.

Then you'd run wires from both Saitek units, all labeled.

From here, it's just a matter of connecting the respective axis and buttons to the control box. The control box is the one that's attached to the pc, while the saitek units are attached to the controller.

In Windows, you'd see one "joystick" with all the axis and buttons.

 

Forgot to mention that labeling the wires on the saitek pro yoke is very difficult, especially when it comes to the two hat switches. Running axis wires to a new board is easy. Running button wires to the new board is very, very difficult, but doeable.

tony

Edited December 12, 2018 by flyforever

[Bob Scott 7,720]

As far as the sim is concerned, a joystick axis is a joystick axis...what each one does is controlled by software assignments.  I use several BU0836X boards in a custom setup where I have replaced the pots in a PFC yoke/rudder/quadrant system with Hall Effect (magnetic field) sensors.  The BU0836X (two of them) presents all those inputs over a USB connection as joystick axis values, and what the sim does with each is a matter of configuration.  And, of course, the Swiss Army Knife of hardware configuration is FSUIPC--with that all sorts of advanced configuration is possible.

That said, it's possible your woes are not the Logitech devices, but instead a weak USB controller on the motherboard that can't sink enough current without the voltage dropping off.  It's a common problem with USB ports, and if that's what's going on, a new controller card may not help.

Regards

 

[MarkDH 234]

I have two observations. First, I dismantled a Saitek quad and connected the potentiometers to a BU0836X board but I found the inputs interfered with each other. Maybe I did it wrong, but I don't think so. The pots have three wires each, which I assume should be +/- and centre-tap. I would suggest you open the quad and either trace the wires to the main controller end inside the yoke, or just cut the cable and run new wires out yourself. Either way you will need to do a bit of soldering.

Second, you will likely not get the detent 'switch' functions, as I think these are actually simulated in firmware from detecting the tail end of the analogue range. (In other words there is no separate switch mechanism to be found when you dismantle it.)

Edited December 12, 2018 by MarkDH

[flyforever 267]

I do not have direct experience with a bohener controller, but have considerable experience with other controllers, mostly custom built.

The first thing to do is to plug the controller in the pc and see what's showing in the game controller's windows( the windows that shows the red strips for axis and the buttons.

The controller should  to be configured for a certain number of axis and buttons, and each axis will have a name, ie, x, y, z rot, etc..

Check the controller's software for this.

Once the controller shows up in the game controller window with all of its axis and buttons, then one begins to wire the pots on the various joysticks.

Before connecting the controller to the joystick, shorting the input pin( each axis on the controller has 3 pins) with either ground or positive should show the axis move on the game controller window. The buttons, too, will light up when you short the two pins.

It's important to write down the name and position of each axis .

As to the saitek detente buttons, if I recall, these are two buttons, and they should have some wires somewhere in the unit.

As I've already stated, identifying the wires in one of these units is very difficull if not impossible.

I did retrofit a ch flight pro easily, since this unit has only a few buttons.

tony

[MattiasLind 0]

Good point to put the board in an own box and connect everything there. As i understood it's possible to connect like 32 buttons. Then one can make a box simulate overhead panel. I was more thinking of the throttle if the levelers are directed with pots and if i need to connect them in a special way on the board. I can see like normal x y z but then there are other inputs and not sure if it doesnt matter where to connect those. 

[MarkDH 234]

9 minutes ago, MattiasLind said:

Good point to put the board in an own box and connect everything there. As i understood it's possible to connect like 32 buttons. Then one can make a box simulate overhead panel. I was more thinking of the throttle if the levelers are directed with pots and if i need to connect them in a special way on the board. I can see like normal x y z but then there are other inputs and not sure if it doesnt matter where to connect those. 

There are eight analogue inputs, each of which have three wires (+5V, GND and signal). These are connected to the ends of the potentiometers (+5/GND) and the centre-tap (Signal). The individual analogue inputs are labelled X, Y, Z, etc. but the labels are unimportant because you can assign any axis to any function. BTW, I don't think you will be able to use the buttons on the front of the quadrant either, as these are mounted on a circuit board inside the quad which adds some sort of logic. I tried first removing the individual switches from the circuit board, which proved impossible because they all need to be un-soldered  simultaneously, and then and then I tried leaving them where they are and soldering onto the backs of the terminals. This proved too hard too :(

Look here for what you can do with enough Bodnar boards :)

[flyforever 267]

When you calibrate your joystick in windows, windows assumes the following approach to calibration:

first x, second y, and so on. It's easier, therefore, to think in terms of which lever(axis) should be the first and which should be the last, so that when you're calibrating it, it's a natural process.

In your case, x would be ailerons, y, pitch, etc.

As with all homegrown projects, planning and good documentation is everything.

Get yourself a a 10k pot and see how it works with the controller. The most important thing to remember is that the two extreme pins on the pot represent positive and negative, and the center is the input.

Similarly, the controller should indicate which pin is what.  Also, in some controllers, if the remaining axis are not shorted, the axis that you're trying to play with will fluctuate wildly. So, you need to get yourself some shorting jumpers like the ones found on old harddrives. You can buy  a handful  on ebay.

From this point on, you will will need a continuity tester or a ohmeter to identify and label all the wires.

Before you spend anymoney, keep one thing in mind:

you will most likely succeed connecting the pots to the new controller.

You will most likely fail connecting the buttons to the new controller.

If you can live with the risk of partial failure, you can start spending money.

I say this because after time and money, you may conclude that the price of a new Saitek was not that far off, especially when you consider that if one of the pots wears on the Saitek, you're done.

tony

[oemlegoem 29]

1 hour ago, MarkDH said:

I have two observations. First, I dismantled a Saitek quad and connected the potentiometers to a BU0836X board but I found the inputs interfered with each other. Maybe I did it wrong, but I don't think so. The pots have three wires each, which I assume should be +/- and centre-tap. I would suggest you open the quad and either trace the wires to the main controller end inside the yoke, or just cut the cable and run new wires out yourself. Either way you will need to do a bit of soldering.

Second, you will likely not get the detent 'switch' functions, as I think these are actually simulated in firmware from detecting the tail end of the analogue range. (In other words there is no separate switch mechanism to be found when you dismantle it.)

Mattias: carefully consider the remarks from Mark. I also looked at the Saitek quad. It was a while ago, but the pots are not normal pots. I remember they incorporate a switch function. Together with the built in processor controller, they simulate the reversal function.
The pots in the yoke can be connected straight to the Bodnar board. However, for the switches in the yoke handle, you have to find out which wire at the controller end has a signal function or common ground function. I played around with this but I gave up. My advice is: play with a multimeter first or do not start at all. Even with a multimeter, it is a dog's job.

Edited December 12, 2018 by oemlegoem

[MattiasLind 0]

I found out all the wires on the yoke handle tonight. If i get time i will look at the quad tomorrow. I do work with similar things so it's not too hard. Took me around 30 min. Will look up the pots in the quad carefully. 

[MattiasLind 0]

13 hours ago, oemlegoem said:

Mattias: carefully consider the remarks from Mark. I also looked at the Saitek quad. It was a while ago, but the pots are not normal pots. I remember they incorporate a switch function. Together with the built in processor controller, they simulate the reversal function.
The pots in the yoke can be connected straight to the Bodnar board. However, for the switches in the yoke handle, you have to find out which wire at the controller end has a signal function or common ground function. I played around with this but I gave up. My advice is: play with a multimeter first or do not start at all. Even with a multimeter, it is a dog's job.

This morning i dismantled the quad and the switches does work as switches. One signal in and 2 out (just need one) . The pots seems to be normal pots, reverse lays in a lower range. Must be possible to put the ranges in the configuration with full throttle, none and reverse. I was not thinking of using the circuit boards at all in the yoke and quad. Just run new wires to the Bodner controller board from each switch and pot. That must work i guess. 

If you think i am worng please let me know. 

[MarkDH 234]

1 hour ago, MattiasLind said:

This morning i dismantled the quad and the switches does work as switches. One signal in and 2 out (just need one) . The pots seems to be normal pots, reverse lays in a lower range. Must be possible to put the ranges in the configuration with full throttle, none and reverse. I was not thinking of using the circuit boards at all in the yoke and quad. Just run new wires to the Bodner controller board from each switch and pot. That must work i guess. 

If you think i am worng please let me know. 

The front switches work as switches, assuming you can get to them (see earlier comments). The lever detents don't work as switches, which means you need to do something else if you want to continue to use them. The only way I can see to use those will be to use FSUIPC's 'control sent when range entered' option and it's not clear how reliable this will be because you have no fine control over the range. Perhaps you can edit the range numerically in the FSUIPC config file.