Stuffed my project up!!

[large_ghostman]

I am in awe of your awsomeness, you awsome youth, you! Well done, have pat on the back and a glass of whatever makes my piddling around with things look like child's play!

Thanks, but honestly it isnt like i wrote the code in C from scratch , it did take a long time and i am pleased with it but I think most people can do it these days with the tools around.

 

[JimB]

honestly it isnt like i wrote the code in C from scratch

I remember one guy that I worked with may years ago had a theory that you only ever wrote one program from scratch, and that is the first one that you write.
All the subsequent ones are variations of what you have done before.

In my case this is fairly true.
Most of my programming activity is in ASM with PIC 16F series, and over the years I have built up a library of routines to do various things which I am likely to need again.
Many of my routines are based on the work of others which I have taken and modified to suit my requirements.

If everything was done from scratch every time, there would be no progress in the world.

We build on what we have done before and upon the previous work of others.

JimB

 

[throbscottle]

It isn't just the code, though, it's the whole thing in it's entirity! That you made all the bits talk to each other, the whole thing talk to some kind of base station, you got the motors working together in an unusual way, you got it to map it's environment (by sonar, no less!) and identify it it's seeing a person! To me, just any part of that is a fairly serious learning experience on its own!

 

[cowboybob]

Well done, LG. Thank you for sharing your success with us.

 

[alec_t]

Incredible accomplishment, LG. Well done indeed. A project that complex deserves at least A***** !

 

[large_ghostman]

Thanks alot guys, you kind of took me by surprise. Maybe its because I worked on it so long and got used to it that I dont see it as anything but a heap of parts I put together.
I did enjoy making it, and its a long way perfect but it was nice to see it fly and finally after attempt 1,0000 it not only missed the wall but avoided it!

The sonar thing is just those distance sensors, on the net is a project that uses them to fake a kind of radar using MATLAB, I used that for the mapping, its not perfect and when testing in my house it got stuck in a room by confusing itself. But worked on the day . Originally the 10 or so were plastic, but saying hello to obstacles soon made me change it, I then switched to fairly heavy duty Aluminum and kept the weight way down, that way I was able to work on the guidance without stopping to stick bits on!
It was also tethered for a long time so I could use PSU power, that was the worse bit of it all, I had to have it frly more than the 10 mins fly time I was getting at the start. The only practical way was to power tether it, but then I wasnt able to work on the guidance just the avoidance, the moment it wasnt tethered the gyros etc were well off, so I then worked on that bit.

So the whole thing was bit by bit, once it was flying ok I was able to add bits, the face thing was added at Christmas again using the Matlab libs. This is where the R PI really scores (and arduino) you can get the student version [*cough* *cough*] cheap and the magic for connecting Matlab and PI etc is free. I dont know Matlab that well, I watched no end of youtube tutorials at night on it and tried things out. I still have no idea what limits there are too it, its a very poweful combination having that and Arduino and PI all talk easily via the same package.

The frame didnt change much until the avoidance was working, I had some hairy moments flying it. I found it was best to stay in one place when working on it so thought I was an obstacle, otherwise if you move too quick you end up making it think its in a different shaped room and it got confused. Most of that is sorted out now, and the moving arms and motors came in around October last year. I did want a 3D representation on the computer screen of it in flight, data from the gyro ect showing what orientation it was in. This was easy to do at the start but as I added bits it started to get hard to handle the huge streams of data coming back. So I opted to do away with stuff like that until I had the other things done.

In the end I didnt do it and I wont be adding it, I will take some pics etc of the dev kits and different bits when I get use of the camera. The bit I had the most trouble from was the frame, I kept making it thinner and lighter as I went, the heavier it got the more I shaved off the frame. I am not set up to do much machining here (not as much as I would like), so that part was a slog at times. The composite stuff they use at School is ok but nowhere near as strong as carbon fiber and weighs only a little less than glass fiber, the batteries came out of a strange vacuum cleaner from China everything on them is in Chinese so I dont know much about them. they are around 38V without load and with load it seems to depend what mood they are in, in an ideal world I would get a cheap petrol RC engine and run a motor from say a heavy duty cordless drill or two as a dynamo, not sure it would be able to run dynamo's for the power but would be interesting to try.

I have no idea how Amazon have a long range quad, I have seen the really thin high powered batteries that are Li Poly, but the price makes them most suitable for use by NASA. There is no way I would pay that for a battery unless the copter could take manned flight. Looking back I wished I had taken more pics while building and if I had worked in say a group then video at different steps would of been good, the out takes would be worth a fortune on you tube!! But working on my own I kind of forget things like that and just plough on.

Did I come top? NOPE, I was beaten to top spot by someone who built a radio. To be fair his write up clinched top place and I knew it might, but we did our projects for different reasons. He wants to goto Art School and I dont, he isnt taking Technology next year at higher level and I am. In some ways I was tactical, I left enough to do to warrant it as a new entry for the next exam. I knew the write up would let me down and I could have done better marks wise had I done a much simpler project, but mine was designed to get noticed by Admission tutors at university, it was designed to come top of an exam that dosnt matter much. So in many ways I made a tactical decision, he just wanted to come top and I wanted something that would give my University application some edge.

Oh I nearly failed completely!! I forgot to tell you that the copter has a small key switch and a tiny dongle thats little more than a resistor in a mini case, they are safety features so it cant be flown without going through some start up procedures, I left them at home day of the Demo!! I was lucky my form teacher came to the rescue and drove to my house to pick them up .

 

[cowboybob]

Oh I nearly failed completely!! I forgot to tell you that the copter has a small key switch and a tiny dongle thats little more than a resistor in a mini case, they are safety features so it cant be flown without going through some start up procedures, I left them at home day of the Demo!! I was lucky my form teacher came to the rescue and drove to my house to pick them up .

When LittleGhostman Industries LTD* goes public, give your form teacher a big chunk of it .

(*LG Industries, LTD is, sadly, already taken... )

 

[large_ghostman]

When LittleGhostman Industries LTD* goes public, give your form teacher a big chunk of it .

(*LG Industries, LTD is, sadly, already taken... )

Meah guess i am going to have to save up and buy them then

 

[cowboybob]

Meah guess i am going to have to save up and buy them then

Chortle....

Actually, wouldn't surprise me .

 

[large_ghostman]

Went to school today (well most of the day) and they are editing the video and will put it on the blog when done. I cant have a copy but I can link to it when its put up . I watched it all the way through today and I have to say I am impressed , not so much with the project but not once in the video can you hear me swear!! Well impressed lol.

Still have no idea what info I can and cannot give, even though I cant see a problem myself, if they are going to post a video of it then I see no harm posting the drawings etc. Only thing I would ask is the mods lock the thread after I post them, just in case someone accidentally goes........hey why didnt you add XYZ to it! I am very paranoid about ruining my chances of submitting MKII, the new air frame will start this weekend with luck. I need to order some more small CNC bits as mine are past there best. I would really like to replace 4 motors (2 I have to replace) but the cost is crippling and would stop 75% of the planned upgrades.

Because I have 12 months less build time now and therefore also less cash I am revising what mods are going on, I have a couple in mind that will be mind blowers anyway. Really annoyed I dropped my bench meter today (actually was a rack mount) and smashed it , so now I am without my faithful 6.5 digit solaratron . It served me well for the short 4 years I had it and although Noah owned it before me I loved it. Shame they dont make that one anymore. It was another super duper ebay score I got cheap because the seller misspelled it

 

[cowboybob]

Thanks for keeping us in the loop, LG.

 

[large_ghostman]

I forgot to mention it earlier, it was while sitting here drawing the new H bridge board I thought about it. I want to try and shave the size of the boards as far down as I can without getting stupid close and difficult to solder etc. Plus my H bridge wasnt great, not sure if it was a mosfet or maybe drive transistor that was out of spec but one of the bridges didnt switch cleanly.

The other thing I need to tackle is noise, it wasnt too bad but I want to up the board to board comms speed. Besides this is the MKII so as many little tweaks as I can make will help add to the write up and points .

Just occurred to me the rack meter might have some decent opamps I could salvage

 

[throbscottle]

You could go for double sided smd layout - or are you doing that already? Surface mount parts aren't hard to solder, especially with young eyes! Printed ribbon cables, difficult but not impossible to work with hand tools, would be lighter than wires. Thinner PCB's would save a few grams too. Lots of decoupling caps, good ground layout and shielding should kill a lot of the noise.
These (or their family, there are various ones) any good to you?: DMHC3025LSD
Re: the meter. Seems such a shame to break it for parts. Is it totally beyond repair? What if you made a new case for it?

 

[large_ghostman]

You could go for double sided smd layout - or are you doing that already? Surface mount parts aren't hard to solder, especially with young eyes! Printed ribbon cables, difficult but not impossible to work with hand tools, would be lighter than wires. Thinner PCB's would save a few grams too. Lots of decoupling caps, good ground layout and shielding should kill a lot of the noise.
These (or their family, there are various ones) any good to you?: DMHC3025LSD
Re: the meter. Seems such a shame to break it for parts. Is it totally beyond repair? What if you made a new case for it?

I am not allowed help with project ideas, but thanks anyway. As for the meter the circuit board snapped on a mounting corner, I think its toast! I am using smd but single sided home made boards that have been shaved even thinner, sound daft but I went with single sided to control weight distribution. I found if you equal the weight as much as you can you save a reasonable amount of power.

I still not 100% why I think it helps with the power but I did notice the motors spend a little less time correcting for a off weight center. I probably have a few grams here and there on the frame I can shave on the CNC, its turning into a rebuild at the moment which I was hoping to avoid. Its all good fun though but will be pleased when its done and handed in, some good spin offs from the project but my next flying machine will be a plane or boat. After Easter I will save up for a half decent ebay RC plane, some really good ones around £275-£300 that are petrol and not nitro. I would like to see what I can do with a plane and something that can fly for 40 mins ish.

The avoidance system would need changing but I have a 5-6km range wireless chip that I can use (range out in the open). I wouldnt fly it out of sight but is good to have plenty of range although I would have GPS and safety on board. I have a old galaxy note II mobile phone that has a 10mp camera on it, that opens up some possibilities for fun and unique pictures. I am just looking at planes at the moment to get an idea of what crops up ebay, that will give me an idea how much I got to save up and how much space you have to play with .

Way out of my league but you never know....**broken link removed** Its got the size etc but the condition is too good, hence the price I guess. But having said that its alot of plane for the money. I sold my Tek analogue scope to fund some of the last project, I am tempted to sell a weller soldering station. The only thing putting me off is once you have used a soldering station the thought of hand soldering again is nasty lol. Its ebay and pay pal that rob you for 25% in fees!! So I might post an add on here if I do decide to sell it.

Anyone fly a petrol plane here (small one and not nitro)? I am wondering what kind of weight they can carry? And how long they can fly for (could add extra fuel I guess, loads of farms and windfarms around here so I naturally have something in mind

 

[4pyros]

If you look back (not even long ago), you will see the grammar police and spacing police giving me a really hard time.

Wow this thread has gone on a long time.
I just wanted to let you know I never hold undeliberate grammar mistakes against anyone.
I for one have always been less then perfect because my brain goes faster then my fingers can type.
That's why I love your tag line.
Can you imagine the things we could share if we were not held back by this keyboard????

 

[large_ghostman]

yeah its just easier to post progress on the project as a whole here. This is a school one though, actually I am on the next one but have alot less time to do it this time. Having said that I dont have to start from scratch this time!

Yeah I keep meaning to post some dev kits etc I have and want to sell. I will try and get it done at the weekend.

 

[throbscottle]

but my next flying machine will be a plane or boat.

I'm really looking forward to seeing you do a flying boat!

I know there are planes that land on water, but doing it the other way around would be awesome!

 

[JimB]

I know there are planes that land on water, but doing it the other way around would be awesome!

In times gone by the Short Sunderland and the Consolidated Catalina (amongst several others) were considered as "flying boats".
The underside of their fuselage was made in the shape of a speed boat "planning hull", rather than relying on floats where the wheels would be.

JimB

 

[throbscottle]

Gorgeous machines

 

[large_ghostman]

If I start in the summer I might be tempted to build one from plans out of balsa wood etc, I am going off the boat idea and instead leaning towards a normal plane, There are some decent 60CC twin cylinder petrol engines for planes, they should be able to lift a decent size plane. ANYONE have the faintest idea where I can get the RC joystick pots from? Please dont say playstation as I hate those