B737 Throttle Quadrant conversion

[Jan737]

Hi Fred,

Could I get a copy of the SIOC file please?

Jan

[Sudden81]

Hello !

I bought OC interface card , I have been dissatisfied.

I have been dissatisfied with :
1) SIOC . Have to get help from a frien to program SIOC
2) If you unplug the USB and switch connections with other USB devices , the card get a new address , resulting in the need to change in the script.
3) 7 segment displays are lit even when the computer is off
4) The cards are unstable for me

This is just what I heard from others , have no personal experience :
5 ) Their servo cards do not have much control options , Axle Ration update frequency, filters etc..
6) Their DC motor card also has shortcomings
7) Steppermotor card is not very precise? Or many possibilities.
Servo card is not particularly exact if compared with Pololu.


Anyway why I wonder because I have a real TQ and if it is really worth buying this DC Plus card.

Is not 10 bit and 8 bit incredibly low resolution ?

Leobodnar has 12-bit . =4096 steps

If compared with other DC motor card (Phidget , Pololu) potentiometer cards, etc. How does this benefit besides that everything is in one card?

Anyone can show a video with this card? Or Multiple? TO / GA. N1 indication when you slowly pull the handles forward and how N1 slowly and gently moves up. Yes such a precise demonstration that possible? Please!

Sincerely, Jacob!

[fsaviator]

Jacob,

Ian Sissons talks about using a polulu I think for motorizing a TQ on his site.  It is doable.  There are several threads here that deal with modifying real TQs and motorizing others and not all are using OC.

My RSP 1.5 Proline (the works) has three cards.  Two OC DC motor cards, and a joystick card.  If I ever have to, I'll probably replace the two DC cards with the DC plus card.  The TQ works great.  it needs a tune-up right now which is frustrating for me as I don't SIOC very well, but I'm plugging away at it and learning slowly.

I keep my USB plugs in the same place so not too many issues.  It is easy enough to change the assignment in the SIOC.INI though if for whatever reason you have to.

If you are dissatisfied or have issues using OC cards, I would suggest you not go that route in the first place.  As you know, SIOC requires some learning.  It can be very frustrating troubleshooting a hardware issue, while trying to learn the software for that particular issue at the same time.

I can tell you that if you take the time to learn it (I haven't, so I speak from experience) SIOC, and OC interfaces are the most versatile, and the best bang for the buck.  In fact, OC even provides the ability to make your own cards.  I had a faulty key card for my second CDU.  The money it would have cost me to send it back, I was able to buy a chip programmer and a chip, then clone the good chip from the other CDU keycard...  back up and running.

It may appear the most "clunky" and large in comparison to newer interfaces, but it set the standard early and I find more and more that you can do ANYTHING with the right mixture of cards and scripts and it isn't tied to one type of airplane.

I started with OC, went away and have a mixture of other cards in my sim, but with my overhead, and TQ, I find myself going back to OC little by little.  My RSP TQ, my two CDU's, my Forward and aft overhead are all run with SIOC now.  My only complaint?  Lack of a clear A-Z manual in English.

[fordgt40]

Jacob

Yes, SIOC is challenging but worth the effort. It is extremely powerful and quick. I run my whole pit on OC cards 2 usb Expansion, 6 mastercards, 5 display cards, 3 servo cards, 2 Output cards , 1 dcmotors card and I USBkeys. All run steadily and with no problems. It is important to use adequate power supplies, the correct gauge wire and do not use cheap chinese USB Hubs

I have the earlier DCmotors card running my throttle levers and the 8 bit resolution is no problem - movement is as smooth as silk. The key is to use a good quality pot with say a minimum of three turns from min to max and then connected via gearing to the throttle levers. This way you avoid jitters and/or hunting. The beauty of SIOC is that you have easy control of the motors and all ancillary functions on the quadrant - just needs a bit of brain effort

With regard to your "friends" comments, if you are using servos for gauges then you do not need exact movements or indeed other control options

David

[FredK]

Attached is my TQ SIOC script file for all who may be interested.  If you have any questions let me know.

To address Jacob's questions.....

Obviously it is hard to troubleshoot from afar.  All I can say is that so far the OC card is running very stable for me.  Yes, if you do unplug the USB connection you will need to adjust the ID in the sioc.ini when you plug it back in.  However this is a very simple procedure (takes less than a minute) and once you are all connected it is not often that you will need to reconnect anyhow.  Precision is more than enough for a TQ application and of course use only quality pots.  Regarding lever movement I am using high torque spec DC motors (the small typical robotics type).  Since they are high torque their motion is relatively slow which helps with smoothness and also provides enough force leverage to move the lever mechanisms.

I am in the process of removing my old TQ and installing the new one which requires some reconfiguration of my physical setup.  I am working on that as we speak.  Once it is back together I will post a video.

Fred

[Sudden81]

Words is one thing.

But can't any one show it on youtube?

How the tq move?

Hard to se OC on the tube!

[fordgt40]

If I knew how to post a video I would.
I am totally confident in my statement about the smoothness of the lever movement. Like Fred, I also use high torque very low speed motors with gearing so that it takes approx 5 secs to go from Idle to fire wall

[Sudden81]

Do you have a Iphone and a register on youtube that is simpel! Best Jacob!

[sagrada737]

Hello Fred,

Thanks for posting the SIOC code.  It's probably Greek to most on the forum, but invaluable for those wanting to see a working set of code.  Much appreciated!

When you have time... Post a YouTube video of your motorized TQ working.  Thanks again - nice work!

Mike

[fordgt40]

Fred

Thanks for posting the code. It is always interesting to see how others tackle it 

David

[blueskydriver]

Thanks very much, Fred.

John

[Flying_Fox]

Thank you very much Fred for your input.
I also work now on programming my motorized TQ, however I do the interfacing with Pokeys and Phidgets cards and programming in C#.

My testbed for autothrottle logic functions looks roughly like this:



I model it programmatically.
In the middle - the TQ state (Not connected to the actual TQ at the moment - Pokeys box is red).
On the right in yellow box - autothrottle imitator - it is easier to play with it than work with actual FSX input. Can be switched to the actual FSX input at any moment. Current FSX values are also monitored in the screen.

Nick

[Masterploxis]

Hello Guys,

i finally got my original TQ working in my pit. Big thanks goes out to Fred K. He helped me a lot and it is his SIOC script that i am using for the autothrottle and trim system (in a little modified version).

The question came up about the smoothness of the lever movement. I made a video showing the autothrottle working with the DC motors driven by SIOC.

Original 737 Throttle finally working in my Homecockpit....

If you are interested, here is also a video i made about my DC motor holders for the TQ.

How to make nice DC motor holders for a B737 TQ with a CNC machine

Andy
www.masterploxis.de

[hexpope]

Which type of aluminium did you use and how did you cost ? I really love the new CNC also. Is that a Kress Spindle on it ? Great job with the DC Motor brackets also.

[Masterploxis]

Hi Pat,

the aluminium i used is Al-Mg 4,5. 200x200x8mm approx. 15 € for 2 pcs.

And yes - it is a Kress spindle. Totally sufficient for all jobs i have for my homecockpit as long as i go with aluminum.

Thanks

Andy

[hexpope]

Thanks for that, and that's pretty cheap too !

[Jan737]

Hi All,

Just a question, why are the levers not moving together, but more or less after each other?
Looks not very real to me.

Can that been changed?

Best regards

Jan Geurtsen

[Masterploxis]

Jan,

the autothrottle system is a closed loop circuit which means that the SIOC permanently checks the actual position of the lever with a target value and activates the dc motor incase of a deviation.

So from Software side everything should be perfect.

The reason why the levers do not move synchronized at the same time lies within the mechanical setup of the system. The steel cable tension cannot be made the same by 100 percent and the slip clutch setting on the dc motors is also not 100% the same. Maybe also the cheap chinese dc motors show a slightly different acceleration graph when powered on.

All these factors contribute to very small timing misalignments for the DC motor activity resulting in slightly different lever movement.

With better aligned steel cable tension, better slip clutch adjustment and better motors, the problem could be minimized (at least that would be my starting point to improve the system).

Andy

[FredK]

Jan,

the autothrottle system is a closed loop circuit which means that the SIOC permanently checks the actual position of the lever with a target value and activates the dc motor incase of a deviation.

So from Software side everything should be perfect.

The reason why the levers do not move synchronized at the same time lies within the mechanical setup of the system. The steel cable tension cannot be made the same by 100 percent and the slip clutch setting on the dc motors is also not 100% the same. Maybe also the cheap chinese dc motors show a slightly different acceleration graph when powered on.

All these factors contribute to very small timing misalignments for the DC motor activity resulting in slightly different lever movement.

With better aligned steel cable tension, better slip clutch adjustment and better motors, the problem could be minimized (at least that would be my starting point to improve the system).

Andy

Hopefully by the weekend I will be finished with the major part of the construction modifcations I am making to my cockpit.  I will post a video of my TQ operation then when things are back together.

Fred K

[FredK]

Hi All -

These are some video links demonstrating the operation of my TQ.

Regarding the motorized motion of the thrust levers I did make a change in the SIOC script eliminating separate slow and fast speed zones. I am now just using a single speed (slow) across the full range of motion and a null (dead) zone. What I have found is that if the motorized motion of the levers is too fast it will overrun the null zone causing a back-and-forth "hunting" phenomenon. This can be eliminated by widening the null zone but that will be at the expense of synchronized movement for the two levers. So by keeping the lever speed slow you can optimized both paramters.  I am using a 15rpm rated high torque motor operating at the lowest possible speed.  A 10rpm or even lower speed would probably be even better, but the way it is currently operating meets my satisafaction so I'm not going to bother changing the motors.  As you can see from the video the constant slow speed is plenty fast enough.  The OC DCMotorsPlus card is continuing to work fine.

http://youtu.be/KPF-UHSOye0


http://youtu.be/d6DtCHm1MMI


Fred K

[fsaviator]

Very nice Fred.  Thanks for sharing

[hexpope]

Great work Fred, you've done a fantastic job at converting the TQ.

[matta757]

Fred,

Wow, that looks awesome. You have done a great job!

Matt

[Sudden81]

Hi All -

These are some video links demonstrating the operation of my TQ.

Regarding the motorized motion of the thrust levers I did make a change in the SIOC script eliminating separate slow and fast speed zones. I am now just using a single speed (slow) across the full range of motion and a null (dead) zone. What I have found is that if the motorized motion of the levers is too fast it will overrun the null zone causing a back-and-forth "hunting" phenomenon. This can be eliminated by widening the null zone but that will be at the expense of synchronized movement for the two levers. So by keeping the lever speed slow you can optimized both paramters.  I am using a 15rpm rated high torque motor operating at the lowest possible speed.  A 10rpm or even lower speed would probably be even better, but the way it is currently operating meets my satisafaction so I'm not going to bother changing the motors.  As you can see from the video the constant slow speed is plenty fast enough.  The OC DCMotorsPlus card is continuing to work fine.

http://youtu.be/KPF-UHSOye0


http://youtu.be/d6DtCHm1MMI


Fred K

I have to ask.

Why do the handles not move smoother?

Why will they not follow each other?

Why do it seems that they can only move in one speed?

Hope I did not offend you with my little too "straight on" questions?

The reason I ask is that if you have put $ 4000 on a tq must move absolutely perfect if you build on it.

Wondering how it works in a real simulator?

Hmmm... just ide' WAM Arm Cable Drives

Animatronic Character Basics - Cable-Control Tentacle Mechanisms - PREVIEW - Special FX Tutorial

More ide's:

Super Duty Belt Drive Pan System

Sincerely, Jacob!

[fordgt40]

Jacob

If it is important to you to have precisely synchronised throttle levers then a simple amendment to the sioc script will achieve this. I coded a routine that defined the maximum "difference" between the current and required throttle positions for each lever. If the difference was within an "acceptable" range then I coded both levers to move to the same position at the same time. This solved the problem of latency in the code.

If I could find the video proving it then I would post it 
In any event if anyone wants my SIOC code with this routine built in, then let me know

David