Printable View
Hi,
I am building a 737 throttle assembly with A/T servos and I have come up with a way to disconnect the servos when the A/T is not active so that you can move the throttles manually. I made a drawing that describes how I plan to do it and I'm wondering if my idea will work or if I have missed something crucial.
I would appreciate any comments from anyone who thinks this might or will not work for whatever reason.
Thanks,
Maurice
http://i32.photobucket.com/albums/d3...rvoswiring.jpg
Maurice,
Your design has been used by a number of builders.
i.e. using A/T arm status to control power to servo.
Using 8/8/8 to control Phidget 0/0/4 (relays) is redundant.
Phidget 0/04 is already digital output.
i.e. will respond directly to A/T Arm status.
Regards,
Alan.
Hi Alan,Quote:
Originally Posted by alandyer
I realize that, but A/T arm status is not the only variable here. I need A/T arm AND either IAS or Mach inputs to actuate the relays since I don't want power to the servos just when I arm the A/T. Arming is not the only thing that is needed befofe the A/T functions. You still have to turn on the IAS switch or the Mach switch, otherwise the A/T does nothing.
Am I out to lunch here ? :)
Thanks,
Maurice
Maurice,
One thing missing from your diagram is input for servos.
i.e. THROTTLE POSITION
Your servos would be tracking throttle position in MS-FS, right ?
i.e. your real throttles will track virtual throttles.
When your virtual autopilot (controlled by real IAS/Mach and N1 switches) changes virtual throttle your real throttles will track.
Design consideration: When should real throttle track virtual throttle ?
All the time ? Nope. Hence, requirement for Relay to disengage virtual world from real world. You understand this, Neo :)
The answer that most designers arrive at:
When virtual autothrottle is driving virtual throttle levers !
Add a block for virtual AUTOPILOT in your diagram.
If you are super fastidious you might put a dotted line through that block and label one half MS-FS and the other half PM MCP/PM CDU.
Digital input for MACH/IAS and N1 will run to that block.
Coming out of that block will be THROTTLE POSITION controlling servo.
Coming out of block will by A/T ARMED controlling relay to servo power and possibly electromagnetically controlled clutch.
Regards,
Alan.
[QUOTE=alandyer]Maurice,
One thing missing from your diagram is input for servos.
i.e. THROTTLE POSITION
QUOTE]
Hi Alan,
Just some clarification to my earlier question. I already have the servos & the throttle handles moving in response to the A/T and that is working great thanks to Rob Howard's excellent article AND your excellent software. So, what is 'missing' from my diagram was intentional since all I am really concerned about at this time is how to disconnect the power to the servos when A/T is turned off. The diagram wasn't meant to show all the wiring necessary to achieve the full A/T or manual operation.
I'm not going to try & use a clutch or some other mechanical way since this would complicate things un-necessarily in my opinion. I have the powered throttles working OK with A/T control; all I need is to disable power to the servos when flying manually. I still think my diagram allows me to do that (I will add the N1 input).
So, I really hate to question your answer again :o , but I'm still not sure that I need to do anything else if all I want to do is cut-off the power to the servos when I don't want the A/T to control the throttle.
Sorry if I am too dense :confused: , but can you please look again at my diagram and let me know exactly what you think may be wrong with it based on this new information.
Thanks again
Maurice
I would use A/T ARM offset to control power.
SPEED, N1 and VNAV are inputs to virtual A/T (MS-FS and/or PM).
These set the mode for A/T unit (virtual in your case).
A/T in turn manipulates power levers which you track with servos in real world.
When A/T is disengaged it does not matter what the mode is.
i.e. A/T is not driving power levers.
Alan,Quote:
Originally Posted by alandyer
I'm sorry but I have no idea how your reply relates to my question in any way. Obviously, I haven't made myself very clear and I don't know how else I can put it.
Anyway, I don't want to waste any more of your time. Thank you for trying anyway.
Best regards,
Maurice
Hi M - did you find the answer to this one yet? I'm guessing the power would have to be switched off somehow when any none of the at modes are active - ie, the at thrust annunciation in the fma is blank - what happens if you have the porwe to the servo running, but are not in any AT mode on your set up? Will the servo hold the last position it was in?
cheers,
FC
This is how I resolved my problems. This works perfectly for me and although it may not be the best solution, it does everything I need it to do.Quote:
Originally Posted by farrokh747
Basically, power is supplied to the servos if the autothrottle switch is on and if either TOGA, Airspeed or Mach hold are on. The A/T disconnect on my throttle assembly turns off the autothrottle and that automatically removes power to the servos and allows me to move the throttle handles & control the power manually.
The spare inputs to the OR gate can be used if there are other conditions that require power to the servos, but so far I haven't found any (I'm sure there must be other times when that would be required, and if I find out what it is, I will use the spare inputs).
http://entertainment.webshots.com/ph...51510320bvfAHE
Regards,
Maurice
Hi maurice - thanks for the diagram - will see how I can make it work for me -
Happy New Year -
c
FC
Hello all,
My autothrottle works fine. The only problem I have now is that it always works - can not disengage the servos so it will move freely again by the pilot.
I saw all the posts here about using the 0/0/4 relay card and understand the idea of this implementation.
I was just wondering - When I double click on the Ph icon (the Phidgets driver) and then the AdvancedServo Phidgets card, I noticed it was possible to click or unclick the engage option (using FS2Phidgets version 4.3.8 - last version?).
When the servo is disengaged, it is possible to move the throttle freely.
I am asking myself why does one have to use the Phidgets 0/0/4 interface card, if its possible to give a software command to engage/disengage the servos???
I guess your answer will probably be - there's no such option in FS2Phidgets. But then again - is it really impossible to do that automatically via FS2Phidgets? That FS2Phidgets looks at a specific logical variable (coming from FSUIPC for example) and that the value TRUE or FALSE will determine if the specific servo will be powered or unpowered?
If this option is not available in FS2Phidgets - how much work will it be to add this feature to FS2Phidgets?
I was wondering since I am coming from the software side of computers - and if the simple Ph driver can engage/disengage the power to the servo, any software can - am I wrong?
Thanks in advance for your ideas about that thought,
Eyal
What I actually mean to ask is:
Would it be possible for FS2Phidgets to simply call the function
CPhidgetAdvancedServo_setEngaged
Which is found in the Phidgets21 package (see phidgets21.h in C:\Program Files\Phidgets\)
Whether it calls this function with a TRUE or FALSE in the engagedState parameter is up to the decision logical variable, that has to be possible to define through the user interface in FS2Phidgets (my impression until now is that it is still impossible to do this via the user interface of FS2Phidgets).
Is it lots of work to add this feature to the user interface and the implementation of FS2Pidgets.
Thanks again,
Eyal
Eyal,
Would take a about weekend of my time.
I have lost track of the number of times I have given up my weekends and never heard back from a builder or had to send a number of "Did update work?" emails to evoke a response.
Give me a good reason why I sould give up another of my weekends?
Regards,
Alan.
Hi Alan,
Thank you very much for your reply.
Well, there might be some good reasons for giving up another weekend:
1) I promise to let you know if the update worked out.
2) I think this small change can make many builders quite
happy (if this will work, the state-of-the-art
FS2Phidgets will be even more versatile).
In my opinion this servo engage/disengage problem
really cries out for an elegant software solution.
3) I'll be more than thankful ?
Curiously waiting for your reply,
Eyal Halm
Id like to take this opportunity to thank Alan for his work on FS2Phidegets.
We(people with TQ'S) would be history if it wasnt for this marvelous software.
I'm sure Ive thanked Alan in the past, but if I haven't for a while....then...THANK YOU!
Any chance of a Xplane2Phidgets?;):D
Nat
I agree,
This is indeed a very fine piece of software Alan!
You did a very good job!
Many thanks in the meantime,
Eyal
Eyal,
The folks at Phidgets have added some nice properties to servos.
The acceleration and speed properties, when fully utilized should make servo operation a lot smoother.
I already use the Engaged property, you highlighted, to disengage servos when the fs2phidget program terminates.
I may currently also disengage when the servo is "idle" to prevent servo overheating. Been a while since I looked at that code.
But disengage-engage functionality driven by user definable rules is on the cards. (see below).
You are now officially an alpha tester. We'll do servos and autopilots first.
Currently I am reworking the FS2Phidget program to move away from the current clumsy circuit and logic variable implementation to a more intuitive interface where a user will be able to construct simple circuit "schematics" using a combination of real devices (phidgets) and virtual devices (electrical, hydraulic or pneumatic) that function like their real-world counterparts in a graphical real-time interactive representation.
Also, exploring X-Plane integration.
Next week is my daughter's 21st birthday and my Mom is visiting for a few week's from Africa, and I am currently working away from home.
How soon do you need upgrade? I would like to suggest that I have something in, say, 3-4 weeks? Will that work for you?
Ek kan Hollands goed verstaan.
Ek gee nie om nie as jy wil jou huis taal gebruik in die toekoms.
Groete,
Alan
Hi Alan,
Again thanks for your reply and the extra info.
Do you mean the Phidgets 0/0/4 should be used for that?Quote:
But disengage-engage functionality driven by user definable rules is on the cards. (see below).
I am more than glad to be defined as an alpha tester :-)Quote:
You are now officially an alpha tester. We'll do servos and autopilots first.
Sounds really innovative! I wonder how this interface will look like...Quote:
Currently I am reworking the FS2Phidget program to move away from the current clumsy circuit and logic variable implementation to a more intuitive interface where a user will be able to construct simple circuit "schematics" using a combination of real devices (phidgets) and virtual devices (electrical, hydraulic or pneumatic) that function like their real-world counterparts in a graphical real-time interactive representation.
Also, exploring X-Plane integration.
Greetings for you daughter's birthday!Quote:
Next week is my daughter's 21st birthday and my Mom is visiting for a few week's from Africa, and I am currently working away from home.
Well, the sooner the better but of course your family matters will go first :-)Quote:
How soon do you need upgrade? I would like to suggest that I have something in, say, 3-4 weeks? Will that work for you?
I also read somewhere that it is possible to download the source code somewhere on the site but couldn't find it.
The best solution will be if you can add the "user defined engage-disengage feature" to the software, so that all other 737TQ builders can enjoy that feature - but if you can tell me where I can download the code / you send it to me - maybe I can look into it as well (as a temporary solution).
Grappig, is dat afrikaans trouwens? Woont je moeder in Zuid-Afrika? Wat een toeval :-)Quote:
Ek kan Hollands goed verstaan.
Ek gee nie om nie as jy wil jou huis taal gebruik in die toekoms.
Goed - we kunnen gewoon in het Engels verder - helemaal geen probleem - maar wel grappig dat je nederlands kan!
So thanks Alan for your reply,
Looking forward for your next one,
Groetjes,
Eyal
Your choice.Quote:
Originally Posted by eehalm
You're also an end-user providing input to the design and functionality.Quote:
Originally Posted by eehalm
Will send you a screen shot. Shoot me an email to alan@coconutair.comQuote:
Originally Posted by eehalm
I don't use that code base any longer, and no, you may not have my current version. I hear what you're saying - my family may be important but you still want it now. You're pushing it, bud. Here is the deal - I have not seen my 82 year old mother in 3 years - you will get software when I am good and ready. Take it or leave it.Quote:
Originally Posted by eehalm
JaQuote:
Originally Posted by eehalm
Quote:
Originally Posted by eehalm
Regards,
Alan.
Ooh, jy's ook 'n Jaapie van oorsprong! Dis baie snaaks!Quote:
Originally Posted by Alan Dyer
:)
Okay back to english :)