B737 Throttle Quadrant conversion

[FredK]

Jacob -

To try to answer your questions and to provide some perspective of how things work......

To start with, the native operation of the lever motion in my TQ is perfectly smooth and can be synchronized perfectly.  What I mean by that is if the levers are operated independent of FSX via the SIOC test utility (by input of target program values) they move perfectly smooth across the complete range of motion.  You can also set speed to match precisely.  So that part is not the issue.

What the issues are.....(1) FSX itself and (2) the interfacing communication with FSX.

Regarding FSX.....Take a flight and pull up the TQ window in FSX.  What you will see is that the throttle movement is not smooth to begin with.  The throttle levers move in a kind of incremental stepwise fashion, albeit the levers always move in perfect synchronization (but I ask is that realistic?).  The SIOC code of course ultimately draws from what FSX is dictating so can it be any better?  For the most part my lever movement is generally even smoother than FSX particularly across longer ranges of motion, perhaps having to do with some smoothing by the avionics program (SimA in this case) but I cannot say for sure. 

There are also inherent communication lags in software communication with FSX that contribute to the syncing issue. I agree that synchronization is not perfect like in FSX, and could in fact be better.  Synchronization is more of an issue across long ranges of motion...for example on a flight level descent retard.  But you can judge for yourself in the videos.

I cannot speak as an authority regarding how the real 737 operates.  However based on cockpit videos I have viewed the throttle levers appear to move perfectly smooth but not necessarily in perfect synchronization with one another.  I would guess this has to do with the attitude and yaw configuration of the aircraft at the particular moment, but I'd be interested in knowing what the experienced pilots here have to say about that.  So perhaps one can argue that perfect synchronization is not necessarily a priority goal.  FSX does not take that into account, but is it different in X-Plane which is modeled differently from what I understand? But all that gets us off on a different tangent here....off topic.

From videos I have seen of commercially offered motorized TQs on the market the smoothness issue appears to be the nature of the beast.  You can easily Google those examples.  Perhaps others with commercially purchased TQs can comment though.

So Jacob, to be sure I am not here to sell you a $4000 TQ, and I realize that is a significant investment that one has to consider carefully.  My only purpose is to do this project for my own enjoyment with the satisfaction in having accomplished something. This hobby has many nuances...what is important to some may not be to others etc.  It is what rings your bell.  But what is the alternative here?  Is a non-motorized TQ really a better option I ask?

David -

I did play with some code to try to better sync the levers just as you have suggested but encountered problems with "hunting" given the more complicated script.  Would appreciate if you could send me that part of your code....thanks very much.

Fred K

[archen]

On my last jumpseat ride in a fairly new 738 the levers did sync 100%. I noticed PF always put his hand on the levers just before a decent or climb was initiated (still on AT). He just had his hand there and did not do anything to the levers. I asked why and they told me there had been an accident when AT changed thrust mode and retarded the levers but not synchronized. Ended up in a roll and crash. So now they must always, by putting there hand on the levers when AT changes power to confirm they not get seperated.

So they CAN differ but it's not usual and it's definitely not good when it happens in the real plane. In the sim I guess it's more important in the visual way that it looks good when they synchronize.

Btw: I like the smoothness you have achieved when they are moving. If you can get them to sync and don't make stops on the way to the target position also it would be fantastic.

I have a 733 TQ myself and looking into how to motorize it the best way. SIOC is interesting but I'm waiting to see what prosim comes up with.

Sent from my iPhone using Tapatalk

[sobsidian]

First, I have to thank all of you for all this wonderful information.

I will most likely be getting a real 737-500 TQ in the next week. My fabircation skills are horrible, so as much as I can find pre-fab'd the better. I am decent with electronics and excellent programmer. So I may even offer to help trade services if neccesary to get some of those pre-fabircated parts.

Anyway, I have a few questions that I have not found yet:

* Fred, your part list was great, however the one piece I did not see you give a part number for was the slip clutch. Could you please provide the model for that?

* I have read 2 or 3 other detailed TQ builds and none of them have mentioned the reverse thrust potentiometer or how the motor is tied for the trim wheel. Are these extra things that need to be added as well?

Thanks in advance!

[FredK]

First, I have to thank all of you for all this wonderful information.

I will most likely be getting a real 737-500 TQ in the next week. My fabircation skills are horrible, so as much as I can find pre-fab'd the better. I am decent with electronics and excellent programmer. So I may even offer to help trade services if neccesary to get some of those pre-fabircated parts.

Anyway, I have a few questions that I have not found yet:

* Fred, your part list was great, however the one piece I did not see you give a part number for was the slip clutch. Could you please provide the model for that?

* I have read 2 or 3 other detailed TQ builds and none of them have mentioned the reverse thrust potentiometer or how the motor is tied for the trim wheel. Are these extra things that need to be added as well?

Thanks in advance!

I purchased the slip clutches from Sterling Instrument (www.sdp-si.com)......Part # S98CA6MMOC320838.  They are made by Polyclutch I believe (they look identical).

A single potentiometer is used for both forward and reverse thrust for each lever. That is, a portion of the pot travel is assigned for reverse deployment, a portion for forward travel.  The reverse thrusters on the TQ conveniently operate that way.  FSUIPC handles reverse thrust as negative input values so the SIOC script is written to account for all that.

See my pictures earlier in this thread that show the motor mounting to drive the trim wheels.

[sobsidian]

Thanks Fred!

I was watching Scott's in-depth videos and got a much better idea of what I was looking for, and your comments are invaluable.

A few more quick questions, I was looking to order those string pots online and found one or two places that I could conveniently order. Most others looked like I had to call and place an order. And are they really around US $100-$150/ea?

Any advice on metal working tools I might need. My tools are all wood tools at the moment. I noticed a lot of metal work. Sorry if this sounds silly, but could I purchase some metal cutting saw blades and use those same tools? And no, I won't hold you liable 

I'm thinking about having to cut out metal spacers for pulleys, or fabricating my own metal brackets. What type of medal did you use and how did you cut it for the most part? On the brackets, did you have to do any welding?

Again, I appreciate all your help! This last bit of info should be plenty to get me started.

[FredK]

Thanks Fred!

I was watching Scott's in-depth videos and got a much better idea of what I was looking for, and your comments are invaluable.

A few more quick questions, I was looking to order those string pots online and found one or two places that I could conveniently order. Most others looked like I had to call and place an order. And are they really around US $100-$150/ea?

Any advice on metal working tools I might need. My tools are all wood tools at the moment. I noticed a lot of metal work. Sorry if this sounds silly, but could I purchase some metal cutting saw blades and use those same tools? And no, I won't hold you liable 

I'm thinking about having to cut out metal spacers for pulleys, or fabricating my own metal brackets. What type of medal did you use and how did you cut it for the most part? On the brackets, did you have to do any welding?

Again, I appreciate all your help! This last bit of info should be plenty to get me started.

Yes, the string pots are expensive....I paid about $150 each.  They are quality products though and work very well.

Regarding the metal work.......You certainly do not need to get involved with welding anything unless you are capable and experienced doing so of course. All of my fabrication is either working with wood or aluminum which is easy to cut or drill.  Aluminum can be cut with an inexpensive reciprocating jig saw....I also use a metal cutting blade on my power mitre saw for straight cuts. For the pulley assembly I fashioned I used aluminum sheet metal that I cut and then assembled together using pop rivets with a cheap hand tool...very easy to do if you have not tried that yet.  Most everything else was drilled and screwed together.  The actual metal body of the TQ is made from a soft metal so it is easy to drill and cutout openings where required.  In my case the TQ came with the bulky understructure that sits under the NAV pedestal.  I trim all that off for my needs using a reciprocating Sawz-All kind of saw.  Those can even be rented at Home Depot if necessary.  Bottom-line....all you need is a little ingenuity....no special tools required!

I hope that helps.

Fred K

[sobsidian]

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

Hey Fred,

I FINALLY got my 737 TQ after 4 months of waiting. I am about 85% complete, just bench testing everything now. I would love to get a copy of your script that you configured. My build is almost entirely based on your parts and design...so THANK YOU! Saved me a lot of time to get this going. I've spent about 2 weeks off and on and can't believe how fast its going.

Also, I'm running into a weird scenerio where, when I activate the Spoiler Servo, the DC card shuts off. I have to reload the SIOC program to get the card back. It appears to be a power draw issue. If I test with the trim servo, I don't have any issues. I've also done the recommended caps across the POT terminals, and shorting the GND from the unused analog inputs to the GND terminal on the power block. Not sure how to solve the issue for the spoiler servo. Any advice would be helpful.

Also, for everyone else, I will be posting pictures and videos of my build once I get it complete. Along with a MOSTLY complete parts list (I know Fred did a great job of getting the big parts listed, but there were a lot of small things that I ended up having to re-order).

[FredK]

Hi Shawn

I'm glad most everything is working out.  I have not experienced the servo problem so I really do not know what to say about that.  Does the problem happen with the servo connected to the board without any load on it?  That is, with the  spoiler arm disconnected.  Also, is there any binding in the motion travel of the spoiler arm?  It took a fair amount of trial and error to get the pot linkage and servo linkage for the spoiler arm not to fight each other.  I did finally work all that out to assure there was no binding and to have the least amount of resistance anywhere in the travel.  In fact there is very little force required to move the lever.  I am away from home right now....will post my latest script when I get back.  An earlier script is posted earlier in this thread.

Fred

[sobsidian]

The problem does not happen while the servo is not commanded to a position. So just plugged into the board, no issue. Its once I send the command to travel that it will activate for less than a second, then the board drops signal. There was a bit of binding I think on the metal slide plate that contacts the spoiler lever. I added a dab of grease and is much smoother. However it appears that has not fixed the problem. I added a spacer to prevent the spoiler arm from contacting the metal slide plateand it doesn't make a difference either. I may also not be sending the proper commands to the servo motor. I'm testing this sending values between 1-800 through the IOCP console to simulate movement. Sometimes the servo will work more than a second, sometimes not. It always will always pull the spoiler back to the down position, but trying to motion it forward is usually when it disconnects right away. But its sporadatic. Sometimes it will move for longer periods of time, sometimes instant off (anywhere from .5 sec to 1.5 sec before it quits). So strange.

I should also point out that I'm using an external 5V supply from a computer power supply. Thats the only load on it besides the 12V supply for the motors. I've tried play with the jumpers to switch the power supply and doesn't make a difference either.

And I did find the copy of your older script. That should be a good starting point for me. Thanks! I'm going to be converting it to prosim.

[FredK]

Shawn

Could it be a bad servo?  Does the same thing happen if you connect the servo lead to a different port, that is, the known good one for the trim servo?  Does the same thing happen if you disconnect the spoiler lever?  I'm just trying to think of all the obvious things.  I would also suggest to contact open cockpits to see what they say....usually they are pretty good responding about technical issues like this.

Fred

[sobsidian]

It could be a bad servo. I have another OC servo card I can test it with. I will try that and also contact OC.

I did try completely removing the arm from the servo and tried changing the port, no go. I'm thinking maybe adding a capacitor might help but not sure how to wire that up since I believe the servo only gets power when commanded to a position, so it would constantly discharge between servo commands. It seems to me like the power draw is more than the card can handle. I think the servo itself works because sometimes it does work. I wish I had another way to test the servo besides the OC cards. It's just so where because it's the same servo you are using. Anyway, I'll reach out to them and see what they suggest.

[sobsidian]

Okay...so I'm now on my THIRD! HS-815BB servo (I fried the last two)  . This time I bought a separate Servo tester which verifies the servo is in perfectly working condition before I hooked up anything. I still have not been able to connect to servo to the DC Motor board, and I've tried two different boards. So...I found a decent work around. The workaround was to wire the servo directly to a standard OC USB servo card. Now all is right in the world!

However, I'm still having issues with my outputs. OC has yet to respond to my repeated emails for support (10 days so far). But I'm unable to figure out how to properly wire the outputs for 12V for the parking brake light (and eventually a solenoid/relay for the parking brake). I fried the IC chip for the outputs on the first board and have been waiting for OC to respond on how to properly wire for 12V on the outputs before I try out the new board. Any help would be appreciated. Their manual makes it sound like GND and Postive are reversed (which would make sense because I cannot get more than 0.61V when I connect it the same way as their other output cards).

I've also started a new thread so as not to hijack this one any longer, which includes some videos of what I've accomplished so far. Its 90%. Here is the new thread: http://www.cockpitbuilders.com/community/index.php?topic=4350.0

[FredK]

Shawn -

That is understandably frustrating about your OC card issues particularly if you have not received a response from OC.

I assume your basic jumper and power connections are what they should be, but just to be sure:

Jumper positions:

5V power: Set jumper to Pos2/JP7 (Internal 5V power)
Power output: Set jumper to Pos3/+V Mot

12V power connection:

12V power supply leads to "+V Motors" (for  + lead) and to "GND" (for - lead)

Also:

I have my parking brake light wired directly to a 12V source bypassing the OC board.  There are dual micro switches for the PB lever - one is connected to the switch input on the OC board, the other operates the PB lamp (simple light switch on/off).  Therefore there is no output command for PB lamp per se.  My strategy was to minimize as much current flow through the board as possible although I really do not think that should matter.

There continues to be no issues for me with the OC board and I have been doing a fair amount of flying of late.

FredK

[carlostxnl]

hi Fred
are you using prosim? I ordered a TQ from" cockpit for you " I havent recieved yet but i think shouold be here pretty soon
I,m not good in SIOC, and  a lot of people that ordered the TQ is having problems with the programing
do you think your script will work with this TQ. this TQ also has the motorized parking break
do you have to program part in fsuipc?

GREAT  JOB
Thanks for sharing

[FredK]

Hi Carlos -

I am using SimA.

The basic program logic in my TQ script could work but you would need to reassign all the inputs and outputs of course.

I do not have my TQ set up for a motorized parking brake release so my script does not include that aspect.

FredK

[carlostxnl]

Thanks Fred
I will talk to the guys and see
I keep you posted


Sent from my iPhone using Tapatalk

[carlostxnl]

Hi Fred
I assuming that the dc motors are 12v. How many amps ?
I ordered a TQ from cockpit-for-you but I haven't received an answer to my question regarding the amperage uds by the dc motors. I want to use the correct power supply
Thanks
Regards
Carlos

[Flying_Fox]

Hi Fred
I assuming that the dc motors are 12v. How many amps ?
I ordered a TQ from cockpit-for-you but I haven't received an answer to my question regarding the amperage uds by the dc motors. I want to use the correct power supply
Thanks
Regards
Carlos

Hi Carlos,

You really don't need to consider more than 1 amp for DC motor.

My whole TQ eats no more than 25 watts at peak when trim and levers are moving. Backlighting included. I have standard 200 watt computer PSU for it, feeding 12v and 5v, which by itself eats ~6 watt in idle mode. So any computer PSU will do - and there will be plenty of power left for other needs.

For Trim motor I just use the 25v power brick from the old electric drill. It's 1A only but that's more than enough. When trim motor works - it eats only 6 watts.

[carlostxnl]

Thanks for the info
I hope i an make it work

[tkellogg]

Fred,

First, let me tell you how impressed I am with the motorization of your 737 throttle quadrant. Second I want to thank you for sharing your expertise and photos of your hard work. If not for your hard work and sharing, along with numerous others, I would never have been able to build my sim. I have a few questions that I have not been able to answer in reviewing this post. The sprocket you used in the drive assembly is attached to the assembly but I cannot tell how. It almost appears as if it's sandwiched between the components of the slip clutch. Would you mind telling me how it's attached and what if any additional components are used to attach the sprocket? It also appears as if there is a shaft collar between the pillow block and the slip clutch. Is that the case? Also do you know how long the shaft is that is used to connect all of the components together? And lastly, based one the information in your post, the sprocket is made out of acetal. Did you use a metal or plastic chain? Again thanks for all of your hard work. Look forward to hearing from you.

Tom Kellogg

[FredK]

Hi Tom

The sprocket gear is made out of a plastic.  It is fixed to the plates of the slip clutch with pins that come with the slip clutch assembly.  I choose the plastic sprocket since it was simply easier to drill accurate holes in the plastic as required to accept the pins. There is then a metal hub spacer that also sandwiches the sprocket gear flush with the gear sprocket.  It is fixed to the shaft with a set screw.

I do not remember exactly, but the shaft size diameter varied slightly between the assembly components. So I used a shaft collar to make it all work by building up the narrower diameter with some metal tape.  Sorry, but it is difficult for me to get an accurate measurement of the shaft components since it is all buried deeply within the center pedestal assembly, difficult to get to. I am not motivated to take it all apart.

My TQ has been operating flawlessly for 7 years now.  The only modification I have made was to replace the the lever motors with slower 5 rpm units about 3 yr ago.  I found slower movement minimized "hunting" and synced better with the standard motion sweep time for the real TQ.

There is no wear issue for the plastic sprocket gear since there really is not a lot of strain involved.  The chain drive is metal. So plastic has withstood the test of time....but metal would be better of course IF you are confident that you can get the pin holes drilled correctly.

You may want to check with Mike Sherick.  Mike took the basics of my design a few steps beyond converting his TQ.  He has a home machine shop to make custom parts and brackets etc.  He is using my SIOC code straight out.

Fred K

[tkellogg]

Thank you very much

[tkellogg]

Fred I see in your post where you attached the SIOC program for your throttle quadrant. I am unable to figure out how to download the file. I click on the paper clip but don't get a response. Am I doing something wrong? Perhaps you could just email me the SIOC program to my email at kkellogg@bellsouth.net Thanks.

[FredK]

Fred I see in your post where you attached the SIOC program for your throttle quadrant. I am unable to figure out how to download the file. I click on the paper clip but don't get a response. Am I doing something wrong? Perhaps you could just email me the SIOC program to my email at kkellogg@bellsouth.net Thanks.

Sent!

[Masterploxis]

Fred I see in your post where you attached the SIOC program for your throttle quadrant. I am unable to figure out how to download the file. I click on the paper clip but don't get a response. Am I doing something wrong? Perhaps you could just email me the SIOC program to my email at kkellogg@bellsouth.net Thanks.

Sent!

Hi Fred,

It has been a while since we had been in contact about the sioc for the tq... Can you sent your current sioc file again to me?

We have moved into a bigger house and I did reassemble the simulator, the motorized Tq anyhow is still not working properly and I would like to start again with a clean script from you.. Adress is
Masterploxis@web.de

Your support is highly appreciated

Cheers
Andy