• Welcome to the new Flite Test Forum home! Please note that the migration is still undergoing and certain content, user privileges or features may be missing or limited until the transition is complete. Thank you!

Ultra budget quad build.

jhitesma

Posted a thousand or more times
Mentor
#1
I'm cheap. Really really cheap. I kind of chuckle when I hear Josh or Chad say they're cheap because compared to me they're almost spendthrifts. But I'm also stupid cheap to the point I'll do things that cost more because I did them so cheap. I try to learn from my mistakes though and like to help others keep from repeating my mistakes. So I try to share a lot as a way to help others be cheap without making the mistakes I make.

Just as a taste of my cheapness...my first article on FT was about making a replacement button to fix my $20 battery charger: http://flitetest.com/articles/charger-button-repair

What I left out of that article is I also fixed a missing rubber foot on it with just a drop of hot glue :D


Growing up I always dreamed of flying RC planes. I built a number of balsa and tissue planes but couldn't afford RC gear so I was stuck with rubber power...and none of them got more than a half dozen or so flights because I got tired of spending more time repairing them than "flying" them. Then a few years ago I found out just how affordable the hobby had gotten and picked myself up a micro heli. I had fun with it but quickly learned that my house was too small to have much fun flying in and it was rare for me to find a day with light enough winds I could fly it outside. Thankfully just as I started to loose interest due to that I found Flite Test about the same time they first released their swappable nutball plan. Since then I've been having a blast building and flying cheap foam planes.

But I've really wanted a quad. Before I built my first foamie, before I bought that first heli...I wanted a quad. I did pick up a Syma X1 last year for about $50 and had a lot of fun with it - but it's just too small and too limited. I really want a "real" quad with a controller I can hack and parts I can easily and cheaply replace when they break. I just couldn't justify more than $100 in my budget and couldn't find a way to make one happen for that little. Now a few years later I have more parts on hand, more experience, and I think I can finally pull it off.

I should be working on it right now...but I feel like sharing more than building at this moment and I feel bad about hijacking a number of other threads talking about my in-progress project.

If I'm done I plan on writing up an article for the main site whether I'm successful or not. But for now I figured I should start this thread to share my progress and give others a chance to give me any advice (not that I'm likely to follow it even if I do appreciate it!)


I started the week before Halloween. After a stressful Saturday I did some browsing on Hobby King and started building an order. I had missed out on my second chance at grabbing a KK2.0 board that week when they were already back ordered again by the time I saw the e-mail announcing they were available. So I started reading up on the MultiWii. I had never had much interest in the MW since I hadn't heard much about it and the idea of taking parts from Wii controllers just seemed like it would cost more and give worse results than just buying a KK board. But as I read up on the MW I quickly learned that most of the MW boards have advanced to better sensors and don't actually use Wii parts anymore. But I also learned that the whole thing is based off Arduino and I just happened to have an unused Arduino Mega sitting within reach on my desk and a couple of wii controllers that haven't been used in months and I started thinking.

So I wired up the arduino and my Wii Motion Plus, loaded the MultiWii firmware...and within an hour had what seemed to be a functional (though not flight ready) flight controller sitting on my desk.

hbmw1.jpg

http://www.youtube.com/watch?v=v4I_DGs7rSk

Cutting the cost of a flight controller out of my order suddenly made it very viable to pull off a <$100 quad based on the FT RotorBones Knuckles with Blue Wonders......

I bit the bullet and ordered:
4 - 1300kv 24g motors $32.38
4 - 10A multistar ESC's $24.96 (These may have been a mistake...we'll see...10A is cutting it kind of close and it turns out that while they are supposedly flashable with SimonK they don't have pads to make it easy. I've read mixed reviews about their suitability for quads but the price was hard to pass up and if it comes down to it I'll break out my SMT iron and a magnifier and reflash them or use them for something else at some point.)
1 - XT60 to 4x3.5mm bullet breakout cable $2.94 (this WAS a mistake, the HK website listed the 10A multistars as having 3.5mm bullet connectors but they actually have 2mm. But the breakout cables with 2mm bullets only have JST connectors and since I plan on using my 1500 or 2200mah packs which have XT60's on them those don't help me much...so I'm going to end up doing my own thing instead, will save this for a bigger quad project in the future.)
2 - Sets of 8045 quad props $5.18 (I know I'll break props...probably should have ordered more than 2 sets...and really kicking myself for ordered both sets the same color...guess what I'm going to order this weekend :D )
1 - 8ch 9x RX $8.99 (I had planned on using one of my existing RX's...but my order came in low enough I was able to splurge..good thing too because the next day I lost one of my RX's and no longer had a spare!)
1 - MultiWii Bluetooth module $7.49 (this was a bit of a risk since I wasn't sure if it would work with my homebrewed controller...but I didn't see any reason it shouldn't and it was cheap enough to risk. Not to mention the idea of being able to program the controller from my phone was one of the things that REALLY impressed me about the MultiWii. Turns out it was worth the risk as it took me all of 5 minutes to set it up and it seems to work great!)

Even with shipping my order was under budget enough I was able to toss in a few spare props for some of my other planes and still keep the total under $95. Woo Hoo!

Of course then I realized I forgot wires to connect my RX to the controller. And I forgot prop adapters. And I started reading more about the MultiWii.....

So before my parts even arrived I had ordered off ebay:
1 - BMP085 baro sensor breakout board - $2.95 shipped (Still waiting on delivery, by mail from Hong Kong...I'm not holding my breath.)
1 - HMC5883L 3 axis magnetometer compass - $2.31 shipped (Still waiting as well...but like the baro I'm not in a hurry since it's not needed and is just for me to experiment with after I have this thing flying.)
1 - Arduino Pro Mini - $4.98 shipped (Figured for <$5 it would be nice to have on hand because I don't really want to devote my mega to this...though the more I read about MultiWii the more I start to think about picking up another Mega and sticking with the Mega since it can also run MegaPirateNG and ArduPilot giving me even more options down the road. But the Pro Mini will be nice to keep the weight as low as possible on the knuckle quad I figure and the price is right.)

I also ordered some male/male 10cm servo extensions. Though I messed up and misread the first listing I bought and didn't realized that unlike most sellers their price was per cable and not per pack of 10 :( So I ended up odering another pack of 10 from a second seller. ($5.80 for the 2 cables...$4.15 for the pack of 10 - still smacking myself over that especially since the seller I bought the 2 cables from signed me up on their spam list :( )

And I ended up placing one more HK order from their US warehouse for:
4 - 3mm prop adapter $4.72
2 - 10pc 2mm bullet connectors $4.86

As of this morning USPS was saying that order wouldn't be here until Tuesday so I wrote off my chances of building a quad this weekend....but while working on building a play house for my daughter this morning the mailman showed up and had 2 packages for me! My 10 pack of servo extensions came in and so did my HK order! Woo Hoo! I've got all the parts I need to build this thing!

So next trip to the hardware store for bits for the playhouse I also picked up a 100 pack of 6-32 bolts and nuts (none of the hardware stores in town had nylocs in any size small enough so I'm going to have to bust out the loctite for now :( ) I alrady picked up a few 1/2" square dowels on an earlier trip to the store. (Nuts, bolts and washers were about $10 for 100 which should be enough for a few quads...the dowels were about $1.20 each and I used 1 and a half to cut my frame.)

I'd love to take advantage of the ongoing 15% off FT sale...but well...this playhouse project is seriously depleting my hobby fund (but it's worth it because she's already having a blast with it even though it's just a frame at this point) and with all the other parts on hand I don't want to wait for shipping.

So last night I cut up some dowels and printed out the knuckle parts plans and this evening I started cutting them out...more on that in a moment.

When my first HK order came in earlier this week I promptly hooked up the BT module and as I already mentioned within 5 minutes I had it working wonderfully! So I went ahead and gutted my motion plus with the 2 servo connectors that had come in hooked up two more channels on my breadboarded controller:

1412841_10151687810416805_22615208_o.jpg

Seems to be working great! I still don't have yaw hooked up so I can't actually arm it yet...but on both my phone and the PC based config app things seem to be responding correctly!

With all of my parts on hand I laid everything out to get a feel for it all:

1453831_10151694782791805_606959994_o.jpg

Like I said I'm kicking myself for only ordering one color of prop...I do have some CW rotation 8045 props on hand of the same design so I may put one of those on for now and fly it plus style until I can get some other props...


Now the real reason I'm writing this up instead of building right now...I needed to take a break from making my knuckles.

I started by using some spray tack adhesive (the same Krylon spray I use to stick plans to redi-boad when building FT foamies) to attach the printed plans to a piece of ply I had laying around from when I made my own firewalls before getting hooked on the FT Firewalls. I hate this ply...it's the standard Monogram Revel ply from Michaels and it's horrible. Impossible to get clean cuts because it flakes so bad. Even when I cut it while sandwiched between other bits of wood or drill it on top of some scrap wood it splinters like crazy. But it's what I have on hand and it should be ok for now.

With the plans glued on I grabbed my center punch and started marking the holes:
1423148_10151694981761805_340826143_o.jpg

Then got out my 9/16" brad point drill (I LOVE brad point drills - so nice for drilling accurately!):
1453568_10151694981106805_1058864934_o.jpg

And drilling on top of some scrap to minimize (but due to the lousy ply not eliminate) tear out I proceeded to make all the holes in my knuckles:
1419566_10151694987196805_1889350552_o.jpg

With all the holes cut it was time to separate the knuckles...but how? I have a nice old 1940's Delta bandsaw that a friend gave me for free because it was a rusty mess on the of his house. I disassembled it, cleaned it, rebuilt it and it works...but I haven't found a new tension spring for it and as a result it tends to break blades and I'm out of silver solder to repair the blades right now. Plus it's not the best for straight accurate cuts. My table saw is overkill and has way too big of a kerf. My xacto razor saw has a nice thin (.015") kerf and would do it...but it's time to replace that blade as it has almost no teeth left so it would take forever to use it.

I decided to do it the hard way. And this is one of those points where I effectively spend more to do something the cheap way. $12 to order a set of knuckles from FT would be a lot cheaper than the time I'm spending cutting these by hand with this method (WAY cheaper if I was to charge myself the hourly rate I charge clients when doing my day job!) I got out a steel straight edge, some clamps, and one of my utility knives along with some new blades:
990642_10151694991361805_2015183430_o.jpg

A half hour later I had them cut out enough to test how accurate they are:
1467095_10151695149326805_1578560308_o.jpg

Looks pretty good! All of the bolts went it with almost no resistance except for the last two which needed just a bit of twisting from my fingers to get them through one or two tight spots.

And now is when I really wish my bandsaw had a blade in it (And no I can't just buy one locally...can't even get one on-line easily as not many people stock the length of blade needed for a 70 year old 10" bandsaw. Buying slightly oversized blades, cutting them and re-welding them is the only way I've found. Did I mention I'm "stupid cheap"? :D ) The bandsaw would let me quickly and easily finish these last four cuts in a few seconds.

Instead I just spent about 40 minutes typing this up while giving myself a break before making the last four cuts by tediously going over the line over and over again with my utility knife.

Oh well...here goes. Doubt I'll have any more updates tonight. Once these are cut I'll be drilling my frame and assembling it but my wife is due home from work in half an hour and I'll have to cook dinner so I doubt I'll have time to get any photos before we crash out for the night!

Tomorrow priority #1 is getting the floor and walls finished on my daughters playhouse...but my wife has the day off so I should have time for both that and getting more done on this quad!

Would love to get it in the air tomorrow but I don't think that's realistic at this point. Once the frame is built I still have to figure out what I'm going to do for the center (I'm out of ply and the knuckles center is nice but overkill for what I'm doing), get the motors mounted, solder bullets onto the motors, wire up the ESC's, finish cleaning up the connection between the WMP and the Arduino, make a power loom, mount it all.....

To be honest I'll be happy if I can power it up without props by mid-week so stay tuned!
 
Last edited:

jhitesma

Posted a thousand or more times
Mentor
#2
Halfway done with these last four cuts on my knuckles...and just realized I should have cut them with the grain of the wood going the other way.

Might be something nice to note on the plans. I know in the video they mention putting the grain one way on purpose for the center pieces to keep the battery mounts from ripping out. I totally spaced thinking about the grain when I glued the pattern to my ply for the knuckles though.

The way I did it (Grain running with the "short" side of the knuckle that only has 2 holes in it) is probably going to be prone to breaking right where the booms meet since the grain runs parallel to the joint between the center booms and the motor booms. Having the grain run across that joint would probably be stronger.

Of course part of the point of ply is that you shouldn't have to worry about grain since each ply should have the grain going in oposite directions. But this ply I have sure seems to have an "overall grain" to it and to break easier along the direction of the grain that matches the outer ply's.

How long does that 15% sale run again.....
 

jhitesma

Posted a thousand or more times
Mentor
#3
Ok, I did manage to finally get all of my knuckles cut out last night before crashing. For those who want to repeat this method (and I really don't advise it!) plan on using about 6 utility blades getting them all cut out. Using more would have made it go even quicker as the ply dulled my blades fast and fresh blades cut MUCH better. But my quick change knife was in another room (did I mention I'm also "stupid lazy" and do things the hard way to avoid what would actually be less work sometimes?) and these cheap blades I have right now don't hold an edge very long at all (Not buying the 100 pack of "Husky" brand blades again, will spend a few bucks more instead of buying the absolute cheapest I can find next time, 100 has lasted me over a year and I'm only half through but if these held their edge just a little longer I wouldn't have gone through this many already.)

First priority this morning is supposed to be working on the playhouse. But with my knuckles all cut I just had to put the frame together first.

I roughly used the method outlined in the official build video but since I don't have center plates yet let alone the little double pieces I made a few modifications to my methods.

I put my motor booms right on the printout of the center plates and marked the outer holes. I then drilled those holes and fearing that my knuckles aren't all quite the same used a bolt to hold a knuckle to that hole and then used the knuckle as a jig to drill the second hole:
1416284_10151695839241805_139142805_o.jpg

I then put a bolt through that hole and proceeded to do the same thing on the other side of the boom:
1468955_10151695840746805_2084387949_o.jpg

With two knuckles loosely in place on one end boom I then held a center boom in place using a square to make sure it was straight - and used the holes in the knuckle as a jig again. Then repeated this for the second center boom:
990646_10151695837656805_895621994_o.jpg

At this point I checked with my square and while there was a tiny bit of play (less than what it looks like in the official video despite my sloppy knuckles) everything was nice and square. So I added bottom plates and put some washers and nuts on finger tight. Without even aligning anything it was all square!
1468365_10151695853331805_894875164_o.jpg

So I repeated the same thing on the second side. I did run into one little issue with the 8th bolt on both sides not lining up with it's hole on the bottom plate. But a quick "adjustment" with the drill took care of that. With all four booms together I gave it a quick check with the square and it's good at all four corners! Wow, even with sloppy homemade knuckles cut with a utility knife this thing self-squared like a champ! I'm guessing accurate drilling of the holes is far more important than the shape/size of the actual knuckles for this part. I put it on the scale and came out at 150g for the basic frame. Note - I made my motor booms 2" longer than stock (12" overall) since I'm planning on mounting my motors directly to the boom since I don't have the rotor bones mounts/landing gear. I'm just going to use some scrap plastic and zip ties as landing gear, and drilled some divots into the tops of the booms to clear the shaft on my 24g motors. ( I figure I could just raise them up in their mounts a bit but the relief holes seemed like a good bit of insurance against the shaft rubbing.)
1419558_10151695851896805_954593898_o.jpg

I haven't decided just how I'm going to mount the motors yet. I'm not sure if I trust two servo screws per motor...and I'm not at all sure I just trust zip-ties. Going to do a few experiments later today to determine how I'll handle it.

It's not the prettiest thing...but it's looking pretty good to me and up in the air the sloppy knuckles won't be noticeable! Going to disassemble it and paint it a little later, then put it all back together with loctite.

For the next few hours though I'm on playhouse duty. Then soldering bullet connectors to my motors. Hopefully by this evening this thing will be looking quite a bit like a full quad and I can focus on cleaning up the wiring on my homebrewed control board. Starting to think there's a chance I might just possibly get a chance to try spinning up some motors late this evening!
 

jhitesma

Posted a thousand or more times
Mentor
#4
Off topic...other than it's why I didn't get to make any progress today. And it does explain where the scrap wood I'm cutting and drilling on came from :)

As I mentioned I'm building my daughter a play house. We had a HUGE ficus in our backyard. We loved the shade it gave us (a valuable commodity here in the desert) but it was getting into the sewer (It was planted right over the main sewer line, I had to dig up 20 feet of it two years ago when the roots crushed it) and the tree just made a horrible mess - so it had to go. February 2011 I had my tree guy come out and take it down to the roof line figuring I'd do the rest myself. [Side note - the weekend the tree came down was also the weekend I decided to start eating healthier I was 210 at the time it came down. I got down 155 about 6 months later and held it for almost a year...but have drifted back up to 165 the last few months...been falling back into old habits and working to stop that again.] Here the tree was a year after we had it taken to the roof line the morning of the day I had the tree guy come out to finish the job because the chainsaw I was going to use to finish it myself turned out to be a basket case:

IMG_20121229_103429.jpg

See that deck under it?

Turned out it was all redwood or cedar. So I saved the wood. Wasn't sure what to do with it but knew I'd find something to do with it.

Then a few weeks ago we went to a party at a friends house and they had a playhouse made out of cardboard boxes and my daughter LOVED it. So I figured it was time to do something with the wood. It was all 2x6 boards so I ripped it into 2x3's with my table saw to make it go further. Then framed out a playhouse:

1467054_10151696001301805_1300723158_o.jpg

I went kind of overboard...but the plan is to add a flat deck on top with a railing and a slide into our yard. So I wanted it to be able to support a full grown adult up there. The floor was old wood floor pieces the previous owner of our house left behind that had been sitting outside (and have continued to sit outside for another 6 years) so they were in kind of rough shape...but plenty nice for a kids playhouse.

She picked pink for the outside and she and my wife painted it this evening while I finished the "roof":
1464198_10151696684561805_1591733605_o.jpg

I still have to cut out a few more windows, add a door, finish the "second level" add trim around the windows, door and corners...and I may build a little deck between it and the sandbox I made her a few years ago. But she's already having a blast with it and it's now "done" enough that I can take a few weeks off from working on it and diverting my hobby budget into it. Despite 90% of the wood being recycled from that old deck I still spent way too much on this thing. The smile on her face is worth it though. And now she'll have somewhere safe to hide while I test fly my quads in the backyard. And I'll have an ultimate challenge for the smaller quads to try and fly through the windows :)

Sounds like she just finished her bath so I can take my shower and then it's daddy's quality time with his quad :D
 

jhitesma

Posted a thousand or more times
Mentor
#5
Didn't get much more done. Spent way too long looking for my zip ties which have suddenly mysteriously disappeared. Have a BUNCH of them and take them with me every time I go flying even though I've yet to use even one when I'm out in the field....but suddenly tonight the entire container is missing :( Not in my truck, not with my build supplies, not in my RV, not in my office, not in my shed....Grrr. And I almost bought a new pack yesterday because I was worried mine may be too old for mounting motors...but then put them back on the shelf because I knew I had so many.

Oh well. I did get bullets on all of my motors and I did get my frame painted. Still have to take each nut off one more time so I can put loctite on them. But it should be nice and visible now:

20131110_231718.jpg

20131110_233418.jpg

I actually labeled each knuckle so I would put it back together the same way...but even so the last two screws needed to have a drill bit run through their holes before they'd go in. Oh well, on the upside once again it squared itself up perfectly without me having to do anything. I forgot to paint the ends of the motor booms too...will just touch those up with a brush since it's not worth it to drag the airbrush back out. (Side note - painting these booms with the airbrush was a pain, they're so light they kept getting blown around!) Paint was applied really light (you can still see my pencil marks through it) but still added 3g - new weight on the bare frame is 153g.

Oh and the color choice was not influenced by my daughter. I just happened to see her little $.79 bottle of hot pink acrylic and figured she'd get a kick out of me painting the "new bee" as she's taken to calling it her favorite color. Besides, it's nice and visible (I've been using hot pink duct tape on my Versa in place of extreme tape for a bit of color) and to be honest when I was her age it was my favorite color too. Will have to get some better photos in the daylight though - light in my spare room doesn't do it justice.

Was really hoping to have the motors and esc's hooked up tonight but without those zip ties.....

Really wish I had tomorrow (well, today at this point I guess) off. Ah well, I'll try and pick some new zip ties up tomorrow I guess.
 

Craftydan

Hostage Taker of Quads
Moderator
Mentor
#6
um wow. That's pink all right! You shouldn't have any trouble picking it out from all the other quads parked on the street ;)

It should show up well against a blue or grey sky, but you might want to re-color one set of booms to be the "front". Symmetry is nice, right up to the point you loose orientation . . .

Either way, looking nice!
 

jhitesma

Posted a thousand or more times
Mentor
#8
um wow. That's pink all right! You shouldn't have any trouble picking it out from all the other quads parked on the street ;)
Yeah, when I first started building I was trying the Experimental airlines method and figured I'd try colored duct tape instead of extreme packaging tape since Ed said it was an acceptable alternative and I figured it would give some color to my planes. So I looked for a color that would stand out against the blue sky we normally have here...and something that my daughter would appreciate in hopes it would keep her interested in the planes. So hot pink won out and my default color scheme is now pink and black:

DSCN3891-L.jpg

20131014_160313.jpg

I figured if nothing else it will help keep people from wanting to steal my planes ;)


It should show up well against a blue or grey sky, but you might want to re-color one set of booms to be the "front". Symmetry is nice, right up to the point you loose orientation . . .
Yeah, I thought about that when I was painting it. But I'm not sure if I want to set it up H or I style yet and I'm planning on running two colors of props to help with orientation. To be honest though I've found that for me color doesn't really help much. Once they're a few hundred feet up/away I can't really tell top/bottom forward/back by color on much of anything. I fly my X1 without the bee body sometimes and just use the two colors of props for orientation - but even then I routinely fly high/far enough that it's just a dot and I can't see the prop colors. As long as I keep moving I don't have a problem loosing orientation - but if I stop to hover....I'm sunk.

I'm hoping to invest in some FPV gear in the next month or so and that's what I'm really hoping to use this quad for.

But I may add some tape on the booms to help once I decide if I'm going to fly it H or I. First few flights I may actually fly it + style since I do have black CW props that match the red props I have for it, so I could make the leading prop black flying that way.

Either way, looking nice!
Thanks. I'm blown away at how well it came out given how crude my methods were cutting the knuckles. I was really planning on doing an X configuration but when the knuckle stuff came out I started thinking H may be quicker and easier to build for a first scratch build - glad I went with it so far.

Wonder if I can find some hot pink props to match somewhere......
 

jhitesma

Posted a thousand or more times
Mentor
#11
(I typed this post up after lunch today but apparently forgot to hit "Post" on it. So here it is better late than never - another update to follow very shortly with some more good news.)

Can't wait to have video to share :)

RCTimer has the pink props...but they don't seem to be in stock. Ebay to the rescue! 2 sets of pink quad props on their way. Coming from China by post though so I'm not holding my breath on them getting here anytime soon :D

Hit harbor freight over lunch and picked up some new zip ties. Forgot it's a holiday (easy to forget since I didn't get today off) so the line there ate into my lunch more than I expected. By the time I got home, made my lunch, ate it, got the zip ties open and started test assembling....I was only able to get this far:

1457790_10151698269341805_1924541452_o.jpg

1469054_10151698269396805_1300787043_o.jpg


But progress is progress and an excuse to grab a daylight shot of the paint since I may not have time for one tonight before the sun goes down.

Been debating just how to do the power loom. Have had a few ideas, everything from just cutting off the bullets on the ESC's and soldering them all together, to building a distribution board out of some PC board material. But then last night I had an idea in a dream. Not going to divulge it just yet...want to see if it will work out and be clean first, but I have a feeling it will work out.

Will probably be able to finish mounting the motors/ESC's this evening but won't get to fire up the soldering iron until my daughter goes to sleep so major progress won't happen until after 9 tonight. Still, if my power wiring idea works out there's a decent chance I may get to hook a battery up tonight and see some motors spin.
 
Last edited:

jhitesma

Posted a thousand or more times
Mentor
#12
Well it's got power! Got home from work a little early because I had to meet a friend who was coming over to take a look at my beer fridge (I brew my own beer but the fridge I use to store it died at the start of summer and has had my last batch waiting in the secondary fermenter - thankfully it's a beer that actually benefits from a few extra months of sitting.) Got the 3rd motor mounted before he showed up - but then had to help him finish diagnosing the fridge.

Bad news. My beer fridge is dead. Compressor is shot. It's only a 7 year old fridge but life outside in the desert is hard. (Though I have friends with outside fridges that are over 13 years old and still going strong - guess they just don't make them like they used to :( )

Good news. After that I was able to finish mounting the fourth motor and ESC! And that was all I could do without firing up the soldering iron so things got put away until I got my daughter in bed.

More good news - she skipped her nap today and was exhausted...got her down a half hour early!

Good and bad news - FT got todays video out early so I burned half of my extra half hour watching that before digging back in :D

With visions of flying toasters still in my mind I set myself back to work. My idea for the battery harness was to take the four female bullet connectors and solder them to each other on the outside. My original vision was to do it in a 2x2 arrangement but finding a way to jig that up and solder it proved exceedingly difficult so I went with all four next to each other with two facing one direction and two facing the other (for each side.) Worked better than I expected. As I feared the mass of four connectors like that made it tough to avoid a cold solder joint with the tip I have on my iron (and I'm hesitant to change tips right now because the current tip is really tight on the heating element and I'm afraid I'll break the element getting the tip off - and my spare heating element was misplaced about 6 years ago.) I made up both sides of the harness, soldered it to some wire and an XT60, added some heatshrink...and seem to have a functional battery harness!

1467132_10151698904246805_152516493_o.jpg

I hooked up my ESC's one by one and confirmed that the BEC in just one ESC will power my RX, Arduino, bluetooth module and WMP - as well as checking that all four motors rotate the right direction. Oh yeah, that means I also finally decided how to make it fly. Going with an I config after all because that was how it was sitting when I started checking motor rotations.

899263_10151698904726805_544890065_o.jpg

You may or may not be able to tell in that fuzzy photo - but the center line mark on the right side is now a directional arrow pointing to what I decided will be the front.

I've now reached the point where if KK2's had been in stock or I had spent the money on a real MultiWii board I'd be balancing props or outside in my well lit driveway trying to hover.

But since I'm still trying to roll my own FC I'm now down to the bits I've been avoiding the past two weeks.

I have my homebrewed FC in a "functional" state but by no means in a "flyable" state. There are three things I need to work out before I can move forward:

1) Final wiring on the Wii Motion Plus - this isn't a big deal. Just a matter of finding some suitable wire in my junk box and making it happen.

2) How to mount the WMP board. One of the nice things about building like this is that I only have to worry about vibration dampening mounts on the little 20mm X 30mm WMP board (and the slightly larger nunchuck board when I get around to getting it out - but I plan on getting this airborne with just the gyros first then adding in the accels for stable mode.) I have some ideas about this and again just have to make it happen.

3) 5v power distribution. This one I've been putting off. I've been thinking about it...but haven't come up with anything I like. So I'm going to talk about this for a bit to think it out before I do any more work tonight:


I have four things that need 5v:
a) The Arduino board
b) My RX
c) The WMP
d) The Bluetooth module

In the future I'll also be adding the nunchuck - but it connects off the WMP so it's not a big deal. I'm also going to be adding a baro and a compass but they're both 3v and that will bring it's own issues - still the 3v regulator will need to be powered off of 5v so I need a way to hook at least 5 things off of my 5v rail.

The arduino actually has a built in regulator and can run off of 7-12v. I'm kind of cheating right now by feeding it 5v off it's 5v output. It works...but it's kind of dirty. I could just power it right off the battery and then use it's own 5v line to feed the WMP...but I'm not sure it's built in regulator would be happy running off a fully charged 3s pack and I'm not sure it's 5v rail would be happy powering the sensors RX and bluetooth.

I do have 4 BEC's...so I could theoretically use one for each need. But that just seems prone to causing all kinds of issues.

So my general plan is to use just one of the BEC's and tape off the other three. That one BEC I will use I then want to feed some kind of bus that I can connect the 5v loads to. The Bluetooth module is only connected for testing and tuning, so it needs to be easily removeable...but the others can be more permanent.

I've been staring at this board for two weeks and digging through my junk boxes looking for ideas. But nothing has really seemed "right" yet.

And unfortunately typing this all out hasn't given me any new insights either like I had hoped it would.

At this point I think the next step will be to take a small step backwards and take everything off the arduino, mount the arduino, then start hooking things back up and just see what I come up with.

Looks like I've got a bit of time before I need to start dinner so Guess I just need to get on with it now.....maybe I'll catch up on a few other threads first to see if I get any inspiration....
 

jhitesma

Posted a thousand or more times
Mentor
#13
It's alive! Well, aliveish.

I ate quick at lunch and then decided to try hooking a few things up and one thing led to another...next thing I knew I was ready to power it up and see what happens. Still nowhere near airworthy. But wanted to see if it could power up off a battery, connect to my phone for tuning, and spin up it's motors.

Big problem I realized was that the last time I flashed the arduino I messed up the multiwii config. Not bad, but kind of critically. I had accidentally commented out both arming options. No wonder I couldn't get it to arm! Re-enabled yaw to arm, re-flashed...and I could arm it finally!

Made a few more connections, some kind of temporary - I used one BEC to power the bluetooth module since it's not normally connected. I used a second BEC to power the arduino, and a 3rd to power the RX - the WMP is powered off the arduino's 5v line. I hooked the WMP back up temporarily (I still need to put proper wires on it) and gave it some power:

1464485_10151699932271805_875279623_o.jpg


It mostly worked! The front left motor wasn't doing anything...but the others were coming up when armed and responding to throttle and me moving the WMP board and TX sticks around!

Took a little look at it and realized I was off by one pin on my motor connection, I had the front right plugged in where the front left should go and the front left was plugged into an unused pin on the arduino. Doh! Moved them over one spot...and all four motors are up and reacting:


Front left was rotating the wrong direction, not sure why as it was going the right way last night...but 3 out of 4 ain't bad and a quick swap of two leads and it's going the right way now.

Haven't had a chance to actually confirm that they're reacting correctly to the sticks and gyros...but just hyped to see them reacting at this point.

Still a lot of work to get it airworthy but I'm feeling pretty confident that I may get to try a small test flight this evening. On my list:

1) Wire the WMP correctly. (I also gutted my nunchuck last night so I may go ahead and hook it up as well)

2) Um, mount everything.

3) Deal with the 5v power distribution correctly.

4) Cleanup some of the wiring.

5) Figure out a battery mount.

6) Balance motors.

7) Balance props.

But man oh man am I feeling hyped to have it this far! Hearing all four motors spin up and react gave me a charge I'm still hyped from. Really excited to get some time to keep working on it tonight now!

One big concern I have...I'm starting to really doubt my choice of the 10A multistar ESC's. They were noticeably warm after just a minute or two of idle testing. Not hot...but warm. Of course I'm used to running 30A ESC's on my blue wonders (because the 30A's were cheap and I needed them for my bigger motors and I didn't care about saving a few grams of weight) and they more or less stay ice cold. I'll have to put the IR thermometer on them after some tests later on. I did mount them with the FET's facing out so they'd get the most airflow. And with no props and no movement yet they're not exactly getting a lot of airflow yet.....

Fingers crossed that these will be ok, having to buy 4 more ESC's will cut into the funds I'm setting aside for a Mobius cam and a 5.8ghz video setup.....
 

jhitesma

Posted a thousand or more times
Mentor
#15
Well...she ain't very pretty but she's very close, here she is with all of her parts on the scale before I dug in on re-wiring the WMP:

1469006_10151700351081805_2042270245_o.jpg

611g Not sure if that's a good or bad weight for this kind of setup, seems reasonable to me though.

After that I dug in on cleaning up the wiring to the WMP, ditching the stock connector and adding some nice light ribbon cable (taken from an old IDE drive cable) with .100 pin headers so I can easily plug it into the arduino:

1460825_10151700565161805_1620546966_o.jpg

That actually took most of the evening, and was why I had been putting it off. Getting the stock connector off was a serious pain. I did it the hard way and actually removed the connector that was soldered to the board. Slowly nipping away at the plastic with some diagonal cutters then once the internal bits were exposed one by one desoldering them and removing them. But eventually I had the board clean. Then the fun of trying to solder on the IDE cables. The insulation on those things melts if you look at it funny so it took some quick work with the iron to get the wires on without melting through the insulation and shorting it all out. Eventually got it though. Plugged it back into the arduino, fired up MultiIWiiConf and confirmed it was still working! Woo hoo!

I then tried to do a bit of more solidly attaching things to the frame. For now I ended up just using some rubber bands to hold the arduino in place. And in a fit of "what the heck it's not going to fly tonight" judgement the WMP board as well:

1471729_10151700574216805_2086626284_o.jpg

Other than the totally inadequate gyro mounting it's looking close to flyable:

1453576_10151700574441805_1009767952_o.jpg

So I hooked up the conf program and confirmed that the correct motors spun up when tilted in various directions. The little virtual quad still isn't responding in the conf program and I'm still not sure why...but the graph is showing the expected responses and everything seems to be working correctly. I'm even seeing basically no vibration from the motors at idle - at about 3/4 throttle I do see a bit of vibration in the pitch reading, but that could easily be the way I have the gyro board "mounted".

Now. I still haven't loctited anything, and I already had one of the motor mount grub screws work itself loose letting one motor come loose. But I decided to do a little more testing. I put the props on. Battery still isn't mounted so I wouldn't be tempted to try and fly. (The only velcro straps I have on hand are too short, gotta pick up some longer ones in the next day or two.) Put them on the wrong way at first. Firing it up while holding it the quad tried to push itself into the ground as I gave it throttle. Swapped the props around though and fired it again....

And it seemed like it was doing what it should! Definitely seems like the default PID settings won't work though. Even with the throttle just off idle and it being handheld I could feel it oscillating pretty bad. Of course this could also be due to the wonky gyro mounting, or the only very roughly balanced props. But I have a feeling the default PID settings are expecting a somewhat larger setup and are going to be a bit too much even once I take care of those issues.

Still...getting very close and getting very excited. Still debating how I'm going to clean up the mounting. Kind of keeping my fingers crossed that the arduino pro mini I ordered comes in so I can swap to that which may make things a bit easier to deal with. The baro sensor I ordered did come in today but I'm no where near ready to start experimenting with it yet!
 

Craftydan

Hostage Taker of Quads
Moderator
Mentor
#16
Jh,

As long as you don't overprop, those ESCs will be fine. You aren't supposed to push a BW motor past 7.5A continuous, anyways, and the right prop should easily give you 600g thrust *EACH*, and stay under the 7.5A. You'll have no reason to stay at WOT for more than a second or two, unless you want this thing going orbital! For sanity, double check with a watt meter if you've got one.

On the whole, she should zip around just fine, once you've got it set up and taught the controller how to fly ;)

BTW, best advice I've received so far: It may not look like it, but it's a heli, and like any heli, it will always vibrate, so tighten and loctite anything down that you can, reduce vibe, and preflight EVERY time you fly.
 

jhitesma

Posted a thousand or more times
Mentor
#17
Yeah, the numbers on the ESC's mean it should all be good....but I'm still just a little nervous. I've been running the same 8045 props on my BW's in pods for about a year now and have tested them with my watt meter a few times, I don't remember exactly what they're pulling off the top of my head but I do remember being shocked at how little it was and that it was under 10A. Just a little nervous about these 10A multistars since when I opened one up to see if they'd be easy to reflash (possible but not simple, no easy to access pads) I was shocked to see absolutely no heatsink of any kind. But they're also the smallest ESC's I've used so that may be common on ESC's of this size. Worst case if I burn one out I use the other three in pods for planes and get a set of something that can easily be flashed with simonK, I've got plenty of ways to flash AVR's on hand.

FWIW - these BW's blew me away the first time I put power to one and they continue to. Still can't believe how much power is in those little things!

And yeah - last night when my wife got home from work I just had to show it off to her since I still had the props on it....and sure enough after I gave it just enough throttle to get out of idle and gave it a tilt forward it started oscillating again and revved itself up then the front right prop nut came loose and sent the prop flying. Thankfully I was holding it above us just in case something came loose so the prop and nut didn't hit anyone or anything important. But lesson learned to be more careful about snugging things down even before tests.

After that I sat down and did the loctite thing on all of the frame screws. About finished off my tube of blue, going to have to pick some more up at HF when I grab some velcro straps later today.

Think I had an idea on how to better mount the gyro too. Going to try cutting a piece of medium density foam that will fit snugly between the frame under the arduino and cut a pocket into it which I'll slide the gyro and accel boards into. Hopefully that will hold them securely but still isolate them from vibration.

Been putting off building a better prop balancer for a few months too, so that's on my todo list for tonight. I have one of the little finger held simple prop balancers from HK and a stack of rare earth magnets. Usually I just stick a magnet on the side of my dryer and rig something to hold another...but if I'm going to be doing 8 props at once (I want my spares ready to go) I figure it's time to setup something better.

I'm not sure just how I'm going to balance the motors. All of the demos I've seen basically use the graphs in MultiWiiConf to show the vibrations off the gyro/acc ... but the ways I've had my gyro mounted it's so issolated I don't see any vibration in it. Maybe once I have it mounted a bit better I'll get some better feedback....
 

jhitesma

Posted a thousand or more times
Mentor
#18
I guess it wasn't the holiday on Monday that made HF so crowded. Just that snowbird season is in full swing. Went in again at lunch today to pickup some velcro straps and loctite (and totally forgot the loctite!) and the place was even busier - but other than the workers I was the only one in there under 65. And the 3 workers were swamped with a hoard of snowbirds all demanding attention. Such is life here in the winter. Thankfully most of them are asleep by 6PM so if you can put off shopping until the evening they're easy to avoid :D

Did a bit more rigging on the gyro, still not 100% happy with how it's mounted but progress is being made. I should have made the wires for it an inch or so longer though, they're really limiting my options. Also did some testing while hooked to the configuration program and found that I had the gyro backwards and upside down. Or something...only the yaw was right. Took me about 10 minuts of fiddling to finally figure out that I was looking at it from the bottom and the diagram showing how it should be positioned was looking at it from the top. Doh! With that figured out it looked like things were responding 100% correctly!

I then tried to figure out how to adjust the PID values in the official conf program...and couldn't make heads or tails out of it. Grabbed
the windows GUI instead and found it to be MUCH easier to use. Went ahead and lowered the P values by 1 across the board. At that point holding it in my hand at low throttle with props it was feeling pretty stable. Since my wife had just a few minutes of her lunch break left...I couldn't resist...I headed out to the driveway.

The lack of landing gear made it tough...but I almost got it off the ground. Then I realized I forgot which way I had set it down and didn't know where the front was. Got that figured out, tried to get it to take off again...and it almost did. But the right rear motor cut out on me. Wasn't sure what was up with the motor, but if I disarmed and then rearmed it would come back on...but then cut out again.

Yep, turns out I left the ESC's in LiPo soft cut mode and had grabbed a lipo that was waiting to go on the charger <head smack>

Want to see it for yourself? I set the flip on the ground in cause I had success, but I guess I'll share my failure as well. Warning the video is horrible since my driveway is not very well lit:


(You can hear my wife and daughter both of whom really wanted to watch...but I made them stay in the house and watch through the screen door because I wasn't convinced things wouldn't go flying or the quad wouldn't take off on it's own.)

Oh well, knew it wasn't really ready to fly yet anyway. Didn't break anything. Did scrub the ends of the props against the driveway just a little...but not even enough to see on 2 of them - and I still have to give them a good balance anyway.

It's close though - real close. I may have to drop the P values a little bit more. But first I'm going to watch the new FT video and balance these props...then maybe fire up the soldering iron and hook up the accelerometer board since I want that hooked up before I make my final adjustment to the mounting of the gyro board.
 

jhitesma

Posted a thousand or more times
Mentor
#19
Too late. No photos. But....made some quick and dirty landing gear out of some thin acrylic sheet I had laying around. And got the nunchuck disassembled and wired up. Need to finish my final changes to the mounting of the sensors now but it's looking pretty good.

Need to spend a bit of time with my 9x tomorrow and figure out how to setup the AUX1/2 so I can change the mode and enable autolevel since the accels are working.

Wife has the next two days off (I still have to work) which means I'm not on daughter duty all afternoon/evening so I should have time to do the last few bits and get it flying - fingers crossed! Bed time for now finally though!
 

jhitesma

Posted a thousand or more times
Mentor
#20
Edited to update the video with a higher quality version (of the video, not the flight)

And over lunch today it flew. I should probably put flew in quotes though:


I didn't really think my landing gear would hold up, even so I had expected it to last more than one "landing". The acrylic I had was just a lot more brittle than I expected it to be. Will make some new ones out of something else.

I did have a second "Flight" as well and it's video is in my youtube but I'm not going to post it because I really shouldn't have attempted the flight and did so in a very unsafe manner. I'll admit to it...but I won't put the video here - if anyone wants to see me take unnecessary stupid risks they can look it up on my youtube channel directly.

The "IMU" is still not mounted very well, and I still need to balance the props better - but I suspect the bigger issue is that the P value is still too high for this small of a frame. After work I'll be adding new landing gear, doing a proper mounting of the "IMU" now that I have both the gyro and the accel working, lower the P a bit more...and if it's not totally dark out by then try again.

My magnetomter/compass arrived in the mail today too - so now I have baro and compass I can work on adding once I have basic flight working.

Still need to do a bit more research on flight modes and how to setup er9x to deal with them...I haven't updated the firmware on my TX in about a year though and it sounds like er9x has been surpassed by open9x and I've been putting off looking into whether I should change over or not...just haven't wanted to get sucked down that rabbit hole yet......hmmm....
 
Last edited: