B737 Throttle Quadrant conversion

[Flight_Master]

Axel,

With some smaller motors and good technical skills, you might be able to install the motors on the inside of the base. But there are few challenges if you try to use the original pulleys. 1.) The pulleys are at slight angle directing the cables away from the center. 2.) If you look the picture the location of the motor sprocket  is pretty much the maximum you can towards left (in a pulley direction). If I would move the motor half inch to the left I would not have a full movement of the throttle lever because the chain would get stuck on the pulley. Shortening the chain is not an option too as it already the shortest it can be.


Good luck,

Piotr

   

[jetpilot]

Hi Fred,
I was wondering if you have been able to test yout TQ? Any feedback on the way it is working? Are you happy with the parts you used? would you change something? Motors? anything?

Thank you

Roberto C

[FredK]

Hi Fred,
I was wondering if you have been able to test yout TQ? Any feedback on the way it is working? Are you happy with the parts you used? would you change something? Motors? anything?

Thank you

Roberto C

The thing is I hit a snag with the OC DC MotorPlus Card. The first card I had was defective - it was not recognized by either the SIOC software or the computer as a USB device.  At that point I simply ordered a second card since returning the first was not worth the hassle and shipment cost back to Spain. Well the second card was dead-on-arrival as well.  So I sent back both cards under warranty. My surmise is that the microchips on the cards were never flash programmed in each case, but I do not know that for certain (but OC had mentioned that might have been the problem).  In any event OC has just today informed me that the replacement cards have been shipped. So they will be here next week and I will get the ball rolling again....and I will also have a backup card available to swap in if anything ever goes wrong in the future.

Will keep the thread posted on new updates as they happen.

Fred

[sagrada737]

Hello Fred,

It is interesting to hear of your experiences with the Open Cockpits USB DCMotors Plus card.  I too purchase and received a USB DCMotors Plus Card, and it was defective (It would not operate above 8 volts DC for the external motor supply).  The nice folks at OpenCockpits replaced the card, and I received it today.  The new card now operated correctly at 12 volts DC for the external power supply. 

However, even with the new card, I am seeing an issue that is puzzling to me...   If I use the OpenCockpits test software to check out the USB DCMotors Plus card, ALL the functions (Servos, Inputs, Outputs, and Motors) operate correctly.  Here is a link to their test software:
http://www.opencockpits.com/index.php/en/download

But, when I try to coding a simple function like checking for an Input with SIOC, the card does not seem to operate correctly.  I have been thinking that it is my beginning effort with SIOC, but with your post indicating problems with your DCMotors Plus card, I am thinking that maybe it is the processor on the card that may be the issue.

Would you test run your USB DCMotors Plus card using the OC test software to check its functions, and let me know if you also get everything to work correctly.  This might help us in sorting out this problem.  Also, would you post some wimple SIOC code of how you are testing the DC Motors Plus card.  Thanks.

By contrast, I hear that folks are not having any problems with the standard DC motors card.  It will be interesting to get your feedback on this issue.

Mike

[FredK]

Hi Mike -

At this point it is sort of the blind leading the blind on this.  I just received the replacement cards and both work to the extent that they operate the motors and servos correctly using defined input values in the IOCP console in SIOC.  However within this very limited testing I have observed that the cards will sometimes disconnect unexpectedly. One of the cards seems more prone to this behavior than the other. This has me a bit concerned at this point, but I need to do a bit more testing with wiring in the pots and switches etc. before I come to a conclusion.

For example right now I have no direct cooling on the card (i.e, a fan).  The manual does state that there is a temperature protection circuit and that cooling is required. One of the heat sinks does get quite hot pretty fast I noticed. So perhaps a simple fan will eliminate that problem, but I have to set it all up and test it in that regard.

There is a lot that is packed into this card. It is a neat solution if it can work reliably, but I will be very disappointed if it acts finicky and cuts out periodically.  The fact that you would need a cooling fan at all is a bit disconcerting in and of itself.  My currently installed TQ runs on Phidgets using DVATX and that setup is absolutely rock solid.....never a problem. Perhaps using a separate motor card as others have done is a more reliable path, but I do not know that for certain at this point.  I may all work fine in the end.

Regarding testing the SIOC code. My only suggestion at this point is to reproduce one of the basic input/output codes in the SIOC software manual and test that.  If that does not work (and assumably that is where you are currently at) I would suggest to contact Miguel at OC. He has been very responsive to this point.  Send him the code and ask for his advice. There is also the OC forum, but it seems rarely does anybody answer anything there.

Unfortunately within the next week or so I will not have very much time to play with all this due to my schedule. I do hope to get the basic testing of all wired functions completed and perhaps some simple test coding this weekend so I will report all that as it progresses.

A question for all readers though....Is there anyone here who does have experience with this card in any way that can chime in regarding reliability?  I am keeping my fingers crossed at this point.

Fred

[sagrada737]

Hello Fred,

Thanks for the comprehensive reply.

Regarding the "noise" on this USB DC MotorsPlus Card...   I too have noticed that this card has a high degree of noise on the various circuit lines (as monitored with an oscilloscope).  This is further aggravated when the DC motor is turn ON, and/or its direction reversed.  The result being that sometimes the USB port will disconnect.

Open Cockpits suggests that a 10 mf capacitor be placed on any active potentiometer Analog inputs (across the GND and the +5v connections on the terminal blocks).  For any "unused" Analog inputs (AN1 - AN8),  I simply shorted the GND an the Central connections, thereby forcing the input to zero (0).  All this seems to have eliminated the problem of shutting down the USB port.  I tested this by writing a simple SIOC test code that runs a 12 volt dc motor, a servo and an output LED, with a potentiometer setup to cause active and rapid changes in direction of the motor and servo, and status of the LED.

As far as the heat generated by this card...   Most of the heat is due to the centrally located Voltage Regulator (7805), which supplies 5 volts to the card's components.  This is due to the regulator having to dissipate all the heat in bringing down the 12 volts to 5 VDC.  I have measured the temperature on the 7805 and it gets up to 200 degrees F.  Clearly this regulator will need some cooling - especially if the card is located in an enclosed space.  This will probably be even more important if there the card is fully populated with Servos and Motors.  I am guessing that a better solution will be to use an External 5 volt regulated power supply to handle the card's power requirements.

I think that the USB DC Motors Plus Card is potentially a nice solution, as it provided a lot of various interface on a single card that can be controlled via SIOC.  However, I too wonder, and frankly am amazed by the lack of user feedback on this USB DC MotorsPlus card.  It's like we are the only two people on the planet that are trying to use this card.  A bit exaggerated, but surely someone on this Forum will help us out with some feedback.

Hopefully we will both see some "blue sky" on this promising interface card.

Mike

[FredK]

Mike -

Thanks for that information.  I too have experienced the cutting out on reversal of the throttle levers motors but only when using the highest value (voltage) inputs. Step the values down a bit away from the maximums and it seems to work.  I also saw the capacitor recommendation but did not think of shorting the unused inputs. I'm glad to know those will resolve the problem (hopefully).  And I agree that the external 5V power supply would also be the more logical way to go.

Fred

[fordgt40]

But, when I try to coding a simple function like checking for an Input with SIOC, the card does not seem to operate correctly.  I have been thinking that it is my beginning effort with SIOC, but with your post indicating problems with your DCMotors Plus card, I am thinking that maybe it is the processor on the card that may be the issue.

Mike

I suggest you post your code that you are having problems with and I can have a look at it for you

Regards

David

[FredK]

Mike -

Are you planning to use caps across each DC motor lead as well?  It seems to make sense to me since the DC motors are assumably the primary source of any electrical "noise" in the first place. Can't hurt I would think.  Such is suggested in what I have been researching on the internet as applied to robotics etc. However it is not mentioned in the OC manual.....only for the pots as you noted above. And then there are the servos if you really want to get crazy about it all, but most servos already have caps built in I have read. Again, I have to find the time to sort all this out with my setup.  My DC motors are somewhat remote from where the board is located in my setup so that should minimize electromagnetic interference I would think, but not line interference of course. I have ordered some capacitors to play with but will check with my local Radio Shack to see if they have any so I can test this weekend.

Fred

[sagrada737]

Hello Fred,

I have been running some tests with the DC motors with the USB DCMotorPlus card.  It seems that this card is especially sensitive to motor noise.  If I don't use any filtering on the DC motors, I will get a shut-down of the USB port in a fairly short time.  This requires rebooting the USB device (essentially re-powering the card).  This sort of electrical behavior can't be normal, but I suspect it is known by OpenCockpits, hence their caution on the noise generated by various DC motors.

What I have done to make this card more stable is the following:

- the 10mF capacitor on each potentiometer Analog input.
- Grounding each unused Analog input.
- adding a L-C filter "H-Bridge Circuit" on the active DC motor  (.01 mF for the Capacitors, with 200 mH for each inductor coil).
- added a small fan to cool the 5 volt voltage regulator (L7805)

Here is one of many good links on DC motor noise suppression techniques that might be helpful to you.

http://www.kerrywong.com/2012/01/26/a-short-guide-on-motor-electrical-noise-reduction/

Give all this a try, which might be helpful for you to achieve more stable performance out of this card.

In all this, the Servo operations seem somewhat benign.

As far as sorting out SIOC programming issues...   This is quite a learning curve, as thinking out the logical "Event Driven" strategies is confusing, but I'm sure I'll get better at it.  I think it is helpful to map out a logic strategy, with flow diagrams to help in creating the SIOC code.  I now have all the basic functions working correctly in a single test program (servo, input, output, analog, DC motor).  Having confirmed the basic setup, I can now start on more complex interactions with this USB DCMotorPlus card.

I think it would be a good idea to swap some SIOC code to help in the learning curve.  Although my beginnings are not that impressive.

Mike

[FredK]

Mike -

Thanks for that link. There is a lot of stuff out there about how to suppress DC electrical noise, but it is all over the map.  For example I read somewhere that a simple 10mF capacitor across the motor leads will work, but your L-C filter reference takes things a step further. Capacitor specifications also vary all over the place.  One really has to have more than a basic understanding of electrical circuits to pinpoint what would work best for any particular setup.  That goes beyond my capability which is at a trial-and-error level.

I have been away from this for a few days but am anxious to find the time to explore it all.  I will have some time tonight and tomorrow. After that I have house guests here through next weekend which shuts me down completely. Also hopefully my local Radio Shack has some capacitors I can use, otherwise I have to wait until I receive what I ordered.

I am also going to shoot off an e-mail to OC to see if they have any advice.  As you say it would appear to be a known issue, but placing caps across the motors is not mentioned in the manual at all....just across the analogue input connections. 

Mike - Look for the PM I sent you.

Fred

[Masterploxis]

Dear All,

thanks for nice informative thread. I bought a 737-300 Throttle 2 months ago. Currently i do the research on the components i need. Initially i wanted to buy the OC DcMotor plus card but is seems that this card will not run stable without modification....

So - i might consider a phidgets solution....

But the most important question i have is - what kind of motor should i take. I saw some motors with gearbox, some without gearbox. I do not have a clue which would be the best. Is there a general rule to follow considering torque and rpm ratio ? I mean i could by a DC motor for 40 € or for 100 €-..??!!??

If anybody could give me a short Bill of material of a solid working system - i would highly appreciate it.
Another point of interest would be the potentiometer to use. I saw that string pots are used. What kind to you use., travel and how much impedance is needed ?

Any help would be highly appreciated for this next challenging task of a Throttle conversion.

Thanks and many greetings from Germany !
Andy

www.masterploxis.de

[FredK]

A status update and reply to Andy's questions.......

I have completed some further definitive testing of the OC DC MotorPlus card and can now report some very good results.

The testing I am doing is with the card mounted and wired to all components on the TQ.

I had the setup operating for a full hour without any disconnects, failures, or non-reproducibility whatsoever.  During the test I had all three DC motors running continuously varying speed and direction throughout using value entries in the IOCP console. I also operated the two servos intermittently.  Analog pot readings were steady and reproducible throughout.  Without using caps across the analog connectors I was getting readings of +/- 1 value unit which is plenty good enough.......with using caps the readings appeared even steadier.  Again, reproducibility was very good.

For the pots I am using ports AN 5-6-7-8.  In previous testing I found ports AN 1-2-3-4 to be problematic and to have extremely unacceptable noise interference (however I have not yet tried to wire caps across those).

Important finding:  For the motor operation I did find that there are input value limits at the high end for each motor which if exceeded will stop all activity on the card.  The card stays USB connected (according to the SIOC software) but nothing operates....all motors stop, analogue value readings stop, etc. All goes dead although the led on the card is still illuminated.  Closing and rebooting the SIOC software restores the connection.  I found that the value limits not to be exceeded are specifically reproducible for each motor so thankfully this seems not to be a random event but rather some current or voltage that is being specifically exceeded somewhere in the circuitry.

The following are the acceptable operating value ranges for each of my motors:

Throttle Motor #1:  40-127(the max value input in this case) and reverse 170-240 (higher than 240 cuts everything off)
Throttle Motor #2:  30-120(higher does not work) and reverse 160-255 (the max in this case)
Trim Wheel Motor #3: 40-120 (higher does not work) and reverse 160-240 (higher does not work)

As long as I input values within these ranges there are no failures whatsoever even on changes in direction of rotation.  This range of speed motion is fully acceptable for this application even with this slightly truncated value range.

All switches worked fine when operated randomly within the test.

I did not have any special cooling on the card........The heat sinks did get somewhat hot but not excessively so......I have experienced a lot worse.

For my servos I am using Hitec HS-Y22 for the trim gauge and Hitec HS-815BB for the spoiler arm.  The trim gauge servo works perfectly. The spoiler arm servo never seems to come to a fully neutral state after a value is inputted.  It does move to where it should be but there is a very slight hum that indicates there still is a slight load on the motor.  If I jiggle the lever it stops. That particular servo is a very high torque spec....I might try a different model to see if it makes a difference.  In any event since the spoiler arm motor operates so infrequently I believe I can live with it if necessary.  The servo did get warm during the hour long test but did not fail.

So I have to conclude that given the extreme operating conditions of the test (constant motor activity) the card should work fine for this application given the particular components I am using.  The DC motors I am using are the typical small 12V robot motors. I am using low rpm models for the throttle levers (15rpm/high torque).  Lever motion is very smooth and the motors are definitely not overworked (no evidence of any binding or heat at all) yet I am extremely pleased with the firm feel of moving the throttle levers manually. So these type motors are all you need and may actually be a plus regarding minimizing noise interference versus using more heavy duty DC motors.

I need to find some time to get into the actual SIOC script programming.  I intend to tackle the throttle operation first.   All else will be a piece of cake!

To answer Andy's questions specifically:

Throttle Lever motors:
ServoCity.com
PG12-200P 15rpm precision gear motor

Trim Wheel motor:
Lynxmotion.com
GHM-12 Gear Head Motor
12VDC 30:1 253rpm

Trim Gauge servo:
Hitec HS Y-22

Spoilers servo:
Hitec HS 815BB
(this is a very high torque spec, I may try an alternative relative to my comments above)

Spoilers and Flaps slide pots:
Alps RSAON1119A02
10K ohm thin B taper (linear)

Throttle lever string pots:
Unimeasure LX-PA-10

Sprocket gear for throttle lever drive:
ServoCity.com
RHS-250-24 24 teeth #25 plastic sprocket gear

The main issue regarding using the small robotics type motors specified above (shaft is typically 6mm)  is getting them to mate with the slip clutch (8mm).  For that I found a coupler:
DumpsterCNC.com
Coupler Size "A"
SAC1: Side 1: 6mm bore, Side 2: 8mm bore

I then supported the whole shaft assembly as shown in the pics above with a pillow block:
SDP-SI.com
A 7Z32MPB008MP 8mm Bore, Pillow-Mounted with PTFE
Impregnated Bronze Press Bearing

That is the report for now....

Fred

[fordgt40]

Andy

There is some useful information in this thread

http://www.cockpitbuilders.com/community/index.php?topic=780.msg6206#msg6206

David

[Masterploxis]

Hello dear TQ-Converters,

after i started the initial planning activities - some question came up:

1. The rotation angle of the throttle levers is from Idle to full power approx. 60 degrees. Pulling the reversers will turn the throttle discs additionaly.
Can anybody tell me the total angle - the disc is turned, full throttle -> 60° -> idle -> ? -> revers ??

It is a little bit hard to measure. At the moment i am doing the calculations and the design of my autothrottle system and i need the total angle to calculate the travel way for the cables and sprocket chain...

Or can anybody tell me the travel way ? From Full power to revers - how much distance is a specific point on the cable / sprocket chain travelling ?

2. Another questions is related to the trim gauge servo. Sure you use a HS-Y22 from Hitec ? I was not able to find this partivular model. Only HS225BB was shown. From my understanding - the Trim gauge servo does not need to have as much torque force as the Speed Brake servo.

3. Throttle Lever String Pots LX PA-10: Are they really so expensive ? I get a price of approx. 100 € per piece here in germany. So i might consider an alternative solution to get the postition of the Levers. Maybe sliding pot via some lever construction or a rotary potentiometer which gets the movement of the chain from a sprocket gear. Most Sliding pots have 90mm travel way - so i need to to get complete travel way (from my first question) scaled down to 90 mm to use the available resolution of the sliding pot.

These will not be the last questions i have to ask during this project. ; ))
Cheers
Andy
www.masterploxis.de

[Masterploxis]

Hello Guys,

i need your help. I wired the draw wire sensor to my USB mtor plus card to the AN 1 connector. If i start the Test programm for this card - i can see that the wire sensor basically is working. The pull out distance is 250 mm in total (10inch). So i assume that the resistor value should be 0 and max value when the wire is pulled out completely.

But the problem is: I can pull out the wire for 2 inch - the values are changing, if i pull out more, no change. So it seems that the wire sensor is only working to a certain extent. I have the same issue when i connect the draw wire sensor to a Bodnar 836X card. In the system repferences (calibrate game controller) i can see that the axis indicator is changing.

Also FSUIPC axis shows me changing values until i reach a certain pull out disctance.

I did not buy the LX-PA-10 as you have but a similiar model from a chinese supplier - it is industry quality and i think the sensors are ok. I habe two tand the probelms shows on both.

Maybe th cause for the problem could be the following:

The working voltage is 12-24V. I think the bodnar card and the USB DC motor plus card provide only 5V out. Could this be reason that the encoder only work to certain degree and after that the resistance gets to high to be reocgnized Can i increase the supply voltage to the draw wire encoder somehow ?? How did you do it and did you have the same problems as me ?  I think that the draw wire approach is very common when converting an original TQ and some you might have an input for me.

I would really highly appreciate any help.....

Andy from Germany

[727737Nut]

It is normal to have some line out before it starts reading.  You have to set-up for this mechanically   It sounds like you have a 2" pull set-up   I used one of those on my pitch axis.   
Rob

[FredK]

Andy -

Simply use a multimeter to measure the max and min resistance values over the travel of the string pull (you don't have to hook it up to the motor card card for that).  The min should be "0" and the max should be what your particular model is rated for (for example 1000 ohms). That will definitively determine the working status of the unit. If you get a reading of '1000' at only 2 inches then it really is not a 10 inch pull.  If you only get 20% of max across the entire pull then the unit is defective. This will also rule out whether or not there is a problem with the OC motor card.

Fred K

[Masterploxis]

Dear All,

i took the sensor to my company and let it have checked by our electronics people.

The think is, the sensors do not give out a resistance value from 0 - 1000 Ohm, but 0 - 5.2 V.

So the output is a voltage change and not a resistance change. This is common for machine control systems (they said. So the question is - is there a possibility to use the voltage and have it read by some hardware (Phidgets ?) anmd processed into a value which can be read by SIOC or FSUIPC. So far i know, the hardware we use for homecockpits is processing resistance values for axis assignments and not voltage values.

Unfortunately i cannot get the sensors replaced by the ones with a clear resistance output - so i really need to find way to use a voltage value as position indicator for FSUIPC or SIOC.

Andy

[FredK]

Just wanted to finish the story on this.......

I finally found the time to complete the SIOC script for my TQ conversion project.  The TQ is fully motorized including spd break lever actuation (either via "armed" on landing or via application of reverse thrust on an RTO or after landing).  I am very pleased with the results.  Throttle lever motion is much improved over my old servo-operated TQ. If anyone is interested in the script let me know....it is written specifically for SimA but can be modified otherwise of course.

The Open Cockpits "DCMotorsPlus" interface has operated flawlessly to this point.  I am using caps across the pot connectors to ensure elimination of any noise interference as suggested in the instruction manual.

FredK

[sagrada737]

Congratulations Fred on "Motorizing" you TQ.  That's is no small set of tasks, and it is good to hear that it is working well.

Also, I appreciate your willingness to provide  the SIOC code (hopefully with comments on the functions).  Perhaps you can post it here.  Thanks.

Best Regards,
Mike

[fordgt40]

I should also be interested in seeing this

David

[Nat Crea]

Congratulations  Fred.
Although I'm using Phidgets,  your contribution
is worth gold to future SIOC TQ builders. Thank you.

Nat

[blueskydriver]

Hi Fred,

Could I get a copy of the SIOC file please?

John

[jetpilot]

Fred,
Is there any chance you make a little video of your TQ working??
Also I believe that if you are willing to share the SIOC code, this is the right place.
Thank you
Roberto C