After reading a recent blog I have decided I REALLY need to get out there and fly this year. I spend far too much time in front of my computer... So thanks to Max Levine's wonderful blog, http://diydrones.com/profiles/blogs/ardupilotcopter-mega-6 I've decided to go ahead and modify my Turnigy 9X to add a 6-position switch. This is going to be more or less a mini-build log as all I've done so far is order the parts....and they're all available from a single supplier, so save money on shipping!!!
Again, props to Max Levine for his original blog on this topic.
Parts ordered (the resistor values came from an experiment with a Turnigy 9X and potentiometer):
(1) X ALPS 6-position switch = $7.06/ea or (1) X ALPS 6-position switch = $9.40/ea
7 resistor solution:
(1) X 1.3K ohm resistor - $0.11/ea
(5) X 1.5K ohm resistors = $0.11/ea = $0.55
(1) X 1.0K ohm resistor = $0.15/ea
5 resistor solution:
(1) X 2.8K ohm resistor = $0.15/ea
(3) X 1.5K ohm resistors = $0.11/ea = $0.33
(1) X 2.5K ohm resistor = $0.15/ea
EDIT Sorry to keep changing the resistor values on you, but I'm trying to get as close as possible to the middle of the band for everyone. AR Projects read a min and max of 1085 and 1921, where I read a min and max of 1047 and 1878.
Current Cost (not including shipping all ordered from Mouser.com) = $8.46
Shipping is "estimated" at $6.40 so I'm in for $14.86 so far.
For those that don't want the hassle, you can buy a 6 position switch pre-made from one of our fellow DIYdroners - AR Projects - http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=120731012907&fromMakeTrack=true&ssPageName=VIP:watchlink:top:en
UPDATE 6/11/2011
I didn't wait for Mouser to get the switch in stock, so I ended up spending $9.40 including shipping to get the switch from eBay. If you can wait, the original price of $8.46 plus shipping still stands. If not, you're looking at $3.40 for the knob plus $6.00 shipping and $1.40 for the resistors and knob plus Mouser charged me $5.20 in shipping. Total cost with shipping was $16. If I could have waited I would have saved about $2.00 on the total cost.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=140481713934&ssPageName=STRK:MEWNX:IT
These are the target and actual PWM values achieved using 7 resistors. (6) 1.5K and (1) 768 ohms
Pos 0: <= 1230 = 1165 Target - 1175 Actual
Pos 1: 1230 - 1360 = 1295 Target - 1300 Actual
Pos 2: 1361 - 1490 = 1426 Target - 1431 Actual
Pos 3: 1491 - 1620 = 1556 Target - 1555 Actual
Pos 4: 1621 - 1750 = 1686 Target - 1684 Actual
Pos 5: >= 1750 = 1815 Target - 1811 Actual
That's about as perfect as it can get.
ADDITIONAL NOTES:
A few things worth mentioning during the "build."
1) Before you try to solder the wires going to the switch, take a pair of large cutters and trim off the bump on the switch. If you look at the top of the switch, you'll see a keyed piece of metal that would be used to keep the knob from rotating, if the plastic housing was designed for it. If you try to clip it after the fact, you may run the risk of breaking off a resistor like I did.
2) Take an x-acto knife and ream out the hole a little bit. The threads on the switch won't go through otherwise.
3) The radio case will be a bit tight to get back together in the corner where you installed your switch. I forced mine ever so slightly....but I could have done a better job cutting the metal tab off my swtich.
Comments
I now did a similar (but different) thing for my Spektrum DX7 in two steps.
First step: I Mixed two switching channels (the 3 position flap channel and one of the 2 position knobs) and fiddled a little bit with the setup until I had the PWM values required by the APM
Second Step: Following the comment from Mark Colwell I did not try to use the "resistor" approach. I just replaced
the 2 swiches with one that has 6 positions. That works fine, with the disadvantage that I need 2 channels.
And the advantage of keeping my hands out of this extremely complicated resistor math :)
I did some more math....since AR's min and max values are different than mine, the ideal plan would be to hit the middle of our two ranges.
That turns out to be 1.3K, 1.5K, 1.5K, 1.5K, 1.5K, 1.5K, 1.0K.
The best 5 resistor solution would be: 2.8K, 1.5K, 1.5K, 1.5K, 2.5K
Why yes AR Projects, my posts from last night are after soldering actual resistors on to an actual switch and installing it in my actual Turnigy 9X and reading the actual values of the actual PWM's thereby validating my CALCULATIONS. I was not off by more than 10 PWM in a range of 65. That's off by less than 16%.
Then, I can CALCULATE how far off your values would be. Using your posted numbers, You're off by 92 in your worst case which is off by 141%
Basically, what I'm saying is you have picked the wrong resistor values and your trial and error process has netted you very poor results. Logic and calculations instead of guessing have led to far better results. So you're charging a premium for something people can do at home for 1/2 the price and have FAR better results in the end. I guess you don't always get what you pay for.
that's funny, because you seem to ignore my technical explanations, for finding anything could be against my posts.
Have you TRIED ACTUALLY my resistor values ? not CALCULATED,
but actually have you BOUGHT 1% resistors of these values and really TRIED them ?
get back to my post, READ it. im explaining why you cant calculate the values,
and why you CANT 'scale' by multiples of correct values.
Also as a side note, I did the math on AR Projects; resistor values (or what he claimed). Using his values, position 1 does not fall within the required range.
Pos 0: <= 1230 = 1165 Target - 1047 Actual
Pos 1: 1230 - 1360 = 1295 Target - 1371 Calculated
Pos 2: 1361 - 1490 = 1426 Target - 1479 Calculated
Pos 3: 1491 - 1620 = 1556 Target - 1595 Calculated
Pos 4: 1621 - 1750 = 1686 Target - 1703 Calculated
Pos 5: >= 1750 = 1815 Target - 1811 Actual
This didn't make any sense to me, so I went ahead and did the math using his measured values for min and max and reversed everything (from the calcaultions above....the resistors can be reversed and get different results and going the same direction as above every single value was outside the range)...these results are worse!
Pos 0: <= 1230 = 1165 Target - 1085 Actual
Pos 1: 1230 - 1360 = 1295 Target - 1268 Calculated
Pos 2: 1361 - 1490 = 1426 Target - 1377 Calculated
Pos 3: 1491 - 1620 = 1556 Target - 1485 Calculated
Pos 4: 1621 - 1750 = 1686 Target - 1594 Calculated
Pos 5: >= 1750 = 1815 Target - 1920 Actual
There's no way I'd use his resistor values.
For those of you watching this thread, I figured I'd give you an update on my build.
These are the target and actual PWM values achieved using 7 resistors. (6) 1.5K and (1) 768 ohn
Pos 0: <= 1230 = 1165 Target - 1175 Actual
Pos 1: 1230 - 1360 = 1295 Target - 1300 Actual
Pos 2: 1361 - 1490 = 1426 Target - 1431 Actual
Pos 3: 1491 - 1620 = 1556 Target - 1555 Actual
Pos 4: 1621 - 1750 = 1686 Target - 1684 Actual
Pos 5: >= 1750 = 1815 Target - 1811 Actual
That's about as perfect as it can get.
I couldn't wait for Mouser to finally get the switch in stock so I spent $9.40 including shipping to get them from eBay.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=140481713934...
I will update the OP to relect my results.
<50> buffer in the code ?
-
1
-
2
-
3
-
4
-
5
of 5 Next