Printable View
Hi all,
Some may be interested in the methods and parts used in my Cessna project so over the next few week I will post lots of pictures, videos and info about it here as it happens. The interesting thing about this project is that I am attempting to have most of the parts manufacturable so if anyone see something they like then gimme a shout.
Chris @ Essex UK
Starting with the largest most expensive part, the glareshield mould as shown below took a couple of weeks to form by the pro's at my local ABS vacuum forming company. The reason for producing a custom glareshield is due to the current availability being too expensive and low quality. A few modifications were made to the traditional design to add strength and style which has worked in my opinion. It is very rigid and is very nice to look at :)
This is the tool made from some kind of resin (hand made!)
http://www.leons-world.com/img/glare_tool.jpg
And here is the first ABS form to arrive from the vacuum forming machine today. Not a very good pic sorry and has a foam panel inserted for perspective.
http://www.leons-world.com/img/glareshield_small.jpg
Chris
Nice... is it cost effective?
Edit:
This glareshield is now available from www.FlightIllusion.com . There does not seem to be much talk about their products in this forum so I will use this thread to evaluate and report on their gauges etc as they arrive. Half of my cockpit is made up of www.homesim.de products too solots to report on here.
Chris
Some more pics now it has been trimmed http://www.leons-world.com/glareshield/
Chris
Now onto what is probably the most difficult part of the project whic is the control knobs.
First they are designed in Pro Engineer and the assembly is checked for correct mating etc.
http://www.leons-world.com/img/knobs_and_blocks.jpg
http://www.leons-world.com/img/assembly.jpg
The first port of call is to have each part machined in aluminium then anodised which will be very satisfying but predictably expensive! This will be done over the next couple of weeks so will post at each stage.
There was only pictures to go on when modelling these parts and my modelling skills are not much so fingers crossed.
Chris
Alternatively, you can just buy one of these panels when they come out, My guess is they will sell around the £100 mark:
http://www.saitek.com/uk/imgs/product/tpm_product.png
You also have nine toggles which you can take out of the casing and use somewhere else, Ie. Switch panel!
I wondered how long it would be before Saitek brought out something like that for the Cessna, but I hope they re-arrange the mix and prop knob because isn't the red knob supposed to be in the centre?
Yes you are correct, it goes, black, red then blue, mainly larger GA aircraft such as the Cessna 182 would need prop pitch. I must admit, regarding the sequence 'yes' they have got it wrong'...but what's to stop you unscrewing the knobs and swapping them round, maybe this is a feature. They have totally killed the design by adding 9 toggles in such an odd looking grid, but they have certainly got the vernier controls right, although the actual black throttle knob looks a bit odd, the white raised label on the black knob should be inverted as to what it is, but altogether it looks 'very' Cessna and I'm liking the product overall!
Did Cessna change something? I'm looking at a poster of a 182 interior and it's black, blue, red (left to right). Here's a link to Cessna's web page showing the same sequence
http://www.cessna.com/MungoBlobs/287..._2010_view.jpg
I think your right smedlik:
http://alexpilot.50g.com/screeny.jpg
My first ever sim was this way too:
http://alexpilot.50g.com/1firstproject.jpg
Saitek is correct, although it seems strange.
I am not sure why you are reinventing the wheel, unless you really just want the challenge. I got real ones from ebay for about $60.00 that were easily interfaced with slide pots. The real ones have appropriate friction locks, handles and verniers for fine tuning of mixture and prop.
http://www.cessna.com/MungoBlobs/287..._2010_view.jpg
http://i885.photobucket.com/albums/a...o/731057be.jpg
Not belittling your design skills, as I really admire people who can do the CAD process well! I just found a different solution to the same problem.
Hmm, it seems the knob order is dependant on preference as it changes from plane to plane regardless of being the same model or not. I do not even know why I am modelling the blue prop knob because it is not a feature in my cockpit anyway?
Yes I reinvent the wheel 10 times a day but there is nothing better to do because I have MS and am stuck to a seat for at least 4 hours a day with nothing better to do. It is definately a challenge no doubt about that! The design process is easy but actually getting local engineers to work on these parts is another matter which is not so easy unless you are already in the engineering game or have your own workshop. I clearly understand why people go to China for work to be done now given my experiences the past year in the UK. There is absolutely no suport for any entrepreneurialism (<-- spell that!) or small business in this country so the challenge of producing quality parts on a budget is definately there.
But on a lighter note, Pro Engineer does not charge me by the hour and is giving me hours of pleasure :) Here is an assembly I produced this morning for the knobs and friction blocks to check everything is cool before submitting the files to the mill.
http://www.leons-world.com/img/control_assembly-sm.jpg
While goofing around with the matte/reflection I found that the knobs can look like they are resin moulded. That is a possibility for crazy looking knobs and the candy apple look is definately a good look! But before concerning about details like that I need to mate up these parts to a custom made PCB that holds the slider switches behind the cockpit. Not going to be easy :(
http://www.leons-world.com/img/resin-look-sm.jpg
Chris
You don't need a pcb if you want to skip it. you can just ,mount a slide pot to a bracket and wire it directly to your interface solution of choice.
Your design skills are outstanding though!
for the control sliders I have reinvented the wheel again and made a slider PCB but with added functions and connections to make it economical for space and tidyness. On one side is the slider positions then on the other side there is a multitude of inputs and outputs to connect various other switches and pots to an FSUIPC interface app directly. It is very clever and works with the RS485 interface so it can connect to slave devices and send their data to FSUIPC interface too without problems.
But my favourite part is the mosfet controlled hour meter terminal. Any FSUIPC offset can turn the 12v supply on and get the meter running. I like the simple things :)
Interfacing directly to FSUIPC with dedicated terminals makes things much more tidy and less fiddling around before flight getting things set up etc.
Unfortunately none of us can get away with the need to boot hundreds of drivers for all the various gizmos on the cockpit whichever method is performed. Maybe one day we will have a standard protocol for all gizmos so just one driver/app required to drive them all.
This will be assembled in a few weeks, 10 available in the first batch.
http://www.leons-world.com/img/RS485-sm.jpg
http://www.leons-world.com/img/RS485_2-sm.jpg
Chris
Wow! This is definitely an overreaction.Quote:
Originally Posted by cjellwood
I have had some terrible upsetting experiences working with UK manufacturers past and present. (could just be me though lol)Quote:
Originally Posted by Leo Bodnar
btw, does anyone know what 'Finis santificat media' means?
Chris
"The end justifies the means"Quote:
Originally Posted by cjellwood
It is indeed a different way of looking at the problem...
Thanks! Someone emailed that quote to me when I was looking for some hard to find PCB components lol. There are plenty of 'means' around at the moment so it looks like I just need some 'ends' :)Quote:
Originally Posted by AK Mongo
Chris
A small animation to show how the PCB attaches to the cockpit via a bracket. Blue knob is in the middle now for the purists
http://www.youtube.com/watch?v=dg2qMhKRruY
Chris
Are you going to have the vernier functions and locks on the prop and mixture? Very cool animation!
no fancy functions unfortunately, just an O ring stretched around each rod to make friction for a realistic feel.Quote:
Originally Posted by AK Mongo
Chris
The airplane that I instruct in has a vernier throttle, which is the first airplane that I've seen that has one.
In this video, watch from 1:27 where the guy has both a lock and vernier function on the controls
<object width="480" height="385"><param name="movie" value="http://www.youtube.com/v/WBBHHKuk1IU?fs=1&hl=en_US"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/WBBHHKuk1IU?fs=1&hl=en_US" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="480" height="385"></embed></object>
At times like this I debate on whether such functions are required in a simulator. A mixture knob control system that has vernier function costs around $120 from Spruce which is good value but at what point does one draw the line between building a simulator as opposed to a real aircraft? The engineering cost for a basic knob with no special function is £30 so a complete set of knobs including electronics and rack will be around £150 (simulator - no special function). Use real knobs and the cost goes to £250+ before you have even thought about connecting it to the PCB and mounting it.
Looking at the features from an engineering perspective it would probably be easier to implement both lock and turn rather than just having turn. I will play with some ideas today to see what is possible.
Chris
I personally, will not invest in the turn and lock feature, I am saving alot of cost and headache by not having this feature. If I was building a cockpit that was specifically for flight instruction, flight school or practicing for my ppl then I would definitely have the function. As my sim is a fun VFR Flyer with the odd bit of IFR equipment, the actual function of my sim does not require the precision of precise tuning where as flight sim simulates the lock and turn function by just providing a push and pull lever which is more than adaquate for my sim's needs. Its my cockpit, and I am totally happy with not having fine tuning on the mix knob, its something that I rarely touch and not worth the added expense where as the money could be spent on other parts!
Alex
Can you place a coaxial encoder on each shaft for precise trimming?
I don't understand why Cessna decided to use Vernier friction nuts in the first place.
Other planes somehow manage to fly without them.
There is definately no middle ground on this issue, the controls either need to be as complex as the original or completely basic. There is also the issue of FSX even responding correctly to minor screw adjustments as it would in a real plane? For teaching purposes a control system with screw function could be dangerously misleading if a student makes adjustements in a real plane according to how it reacted in FSX when turned. Given the sensitivity of the average potentiometer it would probably be a case of [7 turns in FSX = 2 turns in real cockpit] or like effect (if you know what I mean :).Quote:
Originally Posted by Alex Jay
But no harm with the lock function if that can be implemented easily.
Chris
Sensing the turn electrically did cross my mind but there is still the problem of mechanically following the adjustment that has been made.Quote:
Originally Posted by Leo Bodnar
The vernier turn mechanism seems to be popular in marine engine control too so maybe there is the possibility of buying the screw/lock mechanics off the shelf and modifying them to fit the machined Cessna knobs. All depends on price really but there is hope because the Chinese are on it.
http://www.marscable.com.tw/images/Vernier%20Head.jpg
Chris
Things have taken a rather bizarre turn since the vernier issue was raised. Obtaining vernier controls like the one pictured below is not a problem, they cost as little as £5 each from China and can have the knobs replaced with the Cessna type quite easily. The manufacturers of these verniers are also happy to modify them to have 100mm of protruding rod at the end rather than just having a hook for the cable which helps with linking to a sensor. But while going through all that trouble to adapt the vernier i wondered if it would be possible to get away from the slide potentiometer setup which I believe is a wee bit crude considering the effor that is going into other areas. Maybe there is another option?
Well it seems there is another option but it is not an easy one. If this pulls off then there will be a sensor available that will help everyone in every area of sim building, not just throttle controls. No doubt this will be controversial, allow me to explain....
There seems to be a big whole in the market where linear position displacement sensing is concerned. On one end of the market is the slide pots that are pence each and on the other end is the industrial linear sensors that are £150 each at least (below).
http://www.potentiometers.com/images/lt.gif
How perfect are those for everyday sensing of lever positions in cockpits, especially throttle quadrants! But the price makes me heave :(
The solution seems to be Optical Sensors, like the type your mouse probably uses to detect movement. It sounded like a bad idea at first but then I discovered that a steel rod was detectable over the scanning surface and was picked up perfectly. The surface of the rod is ideal due to its reflectivness and all the abrasions that are common with this material. The resolution is amazing too!
http://www.leons-world.com/img/mouse_rod.jpg
The components require to form an optical sensing circuit are simple and cheap. £1 a pop! The output from the PIC can be a voltage of between 0-5v (PWM) depending on the sensor output so acting like a potentiometer and the PIC can be programmed to output that voltage for any distance required! because it senses both x and y axis motion of a rod then it would be great for sterring yokes too. Perfect?
http://www.leons-world.com/img/ADNS-2610Circuit.jpg
So the idea is to create an Optical Position Displacement Potentiometer Module, a bit like this....
(Picture will be here when I make the drawing later)
Of course there is issues such as the position being relative, not absolute. But that is just a matter of having all controls in their default position when the electronics are booted.
Watch this space :)
Chris
Hi Guys correct me if i got this wrong but have you guys just decribe a way to use cheap optical mouses for linear displacement measurement ?? i ask becuase whilst building my motion platform one of the issue i can across was finding good Cheap position sensors which did not really on Rotary potentiomers -- having read this thread i like the concept and can see it having applications for diy project all over the place -- so in a nutshell i need a pic Micro - the internals of a optical mouse and some reflectable medium ( steel rod ) and a little mechanical savvy :-) and a means of starting my platform at a known defualt position
well it is only a theory at this moment but I cant see why there should be any problems with it. I avoid mechanical engineering wherever possible so welcome the thought of simple rods and sensors like this. Here is a pic of how simple it should be i.e. a small box that contains the optical circuit and the rod being feed through the guide so it moves over the sensor camera.
http://www.leons-world.com/img/optical_pot_eng.gif
http://www.leons-world.com/img/optical_pot_eng2.gif
Any size rod or tube can be used but in this case it is a handy 6.35mm rod (1/4"). Quite a catchy name too, "Optical Pot"
Chris
The idea that you have described chris is exactly what I had thought of long ago, but not for a throttle, but for a flight yoke!
This would be perfect technology for a flight yoke!
Same here, an old thought for the yoke and the idea cropped up again for the throttles levers. Both x and y axis are detected by the optic so pretty much perfect for everything.
Imagine this... there is a button on the side of the opti-pot which you press. While it is pressed you push the rod through the range of movement you are measuring (both axis) and it records those parameters in the PIC's EEPROM. Now the opti-pot is configured to measure your exact distance range and output voltage accordingly. That would be complicated for when bi-direction is required such as with a yoke but definately possible.
Chris
The interface PCB is 99% complete, just missing a couple of optocoouplers.
http://www.leons-world.com/img/RS-485_small.gif
Chris
My setup works exactly like the real thing. Twisting on the verniers moves the prop and mixture handles in small increments, and translate into smaller movements on the linear pots. I am using 100mm slide pots, and there is slightly less movement than that in the handles, so I am "wasting" a few mm of the pots capacity. So far, the engine controls are the thing I am most happy with in my "pit", and the realism of pushing in the lock for gross adjustment, or twisting for finer has been a happy surprise.Quote:
Originally Posted by cjellwood
Thanks for that info, the travel distance in general is something I am concerned about at the moment. Are you using original Cessna knobs? It would be great to know the travel distance for the original Cessna throttle, mix and prop knobs; something I forgot to measure when I had time inside a cockpit. The simkits controls look like they are using 60mm pots but that seems a little bit short.
The plan at this end is to reduce the controls cost by modifying cheap vernier controls that are available from China such as this one...
http://www.leons-world.com/img/vernier_taiwan.gif
The stroke/travel available for these controls is 80mm but the 'control end' that connects to this and has the steel rod for connecting is available in 50, 75 and 100mm range. That means if I connect this to a 60mm pot then the 75mm range end will be required and will only travel 60mm due to the restriction of the pot travel. 100mm seems a bit too long but if the optical pot happens then the full 75mm travel can be used. Here is a pic of the Chinese vernier assembly that includes the control end which has a solid rod rather than cable which is handy. The length of the rod can be made longer but keeping the same travel distance.
http://www.leons-world.com/img/control_assembly2.gif
oh and I took a gamble on ebay by purchasing a cheap £2 mouse in the hope that it contained the traditional optical sensors for hacking. It is not the exact chip I required because this has a single chip solution for data/usb but I think data can still be accessed via a couple of pins listed as
Pin Description
4 DM/DATA I/O USB D- / PS2 DATA
5 DP/CLK I/O USB D+ / PS2 CLK
http://www.leons-world.com/img/SP_A2045.jpg
Is there any such software that can read and display PS2 data?
Chris
I can't remember exactly, but can get underneath the hood and measure if you need. All 3 were more than 80 mm but less than 100 in total travel.
Reid
That would be great if you could do a measurement. The vernier control end is restricted to 75mm but it can be modified if needed.Quote:
Originally Posted by AK Mongo
Chris
Does this help:
http://alexpilot.50g.com/abe.jpg
Superb! Thank You! It looks like the travel of the rod is around 80mm when the knob and nut is deducted, so the 75mm Chinese vernier will be fairly accurateQuote:
Originally Posted by Alex Jay
Chris