A new take on my UBMQ

[jhitesma]

Some of you may remember about 2 years ago when twitchity and I first started working on a power board for his 250 acro frame I found myself with all the parts for a quad but no frame...I became inspired to make them fly anyway. That inspiration led to what I still refer to as the UBMQ or Ultra Budget Mini Quad. It was an attempt to get the most bang for my buck and centered on a $3 frame made from cheap ply and 1/2" square dowels.

I ran an afromini with baseflight and when it worked it worked really well:

But it had some annoying tendencies that would come up when the frame would twist. Which would happen fairly often I tried replacing the ply with acrylic and it only made the problem worse. Then I got a "real" frame and binned the budget frame. It had held up quite well, but did break arms fairly easy but since they were about $3 for 8 and fairly easy to drill and replace it wasn't THAT bad other than the twisting issue.

Well. Since I got it about 3 months ago (http://forum.flitetest.com/showthre...a-i3-Build-Log&p=251934&viewfull=1#post251934) I've been really obsessed with my new 3D printer. But mostly I've been printing tests and calibrations along with upgrades for the printer. I designed and printed a few things and shared those worth sharing on Thingiverse: http://www.thingiverse.com/jhitesma/designs

I've been wanting to do something for my RC habit though. I printed a FT Mini firewall/pod and a module case for my TX so I can finally protect a hubsan module I whipped up and am always scared to use because it's just loose wires and boards dangling off my radio right now. I even 3D printed a new button for my accucell. Heck I went so far as to 3d print a plane:


It kind of flew...but I could never get glue to hold the V tail together on this PETG and the nose warped a little when printing. I could probably do it better now.

Still. I wanted to design something of my own and make it fly.

I'm also trying to get better at openscad.

And a couple people on the dRonin IRC channel got to talking about a new CF frame with angled arms similar to the setup Eric Monroe did for FT just about a year ago. They were curious to see how it would affect autotune and cornering. I've been curious since before the FT episode. But I've yet to actually try it.

So I put it all together and came up with the UBMQ - Angled Arms OpenSCAD edition: https://github.com/jhitesma/jth_angle_quad

This is on github instead of thingiverse right now because I haven't test flown any of it yet. I've bolted enough bits together to know it should almost certainly work - but I haven't had time to tear down an old build and move the parts onto this yet.

I did print and test assemble (but again not fly) the motor mounts, and I got them far enough I was willing to post them: http://www.thingiverse.com/thing:1490002

It's my first parametric design and is used by the quad so seemed like a good first step. Being parametric means I could make a "customizer" out of it on Thingiverse that lets you specify things like the bolt pattern and size of the arm.

The details of the motor mount itself are a post or two all their own that I'll try to get to in the next few days if there's interest. Here's the short version:



Upper Left and lower right - original design based on modifications to this existing design: http://www.thingiverse.com/thing:1025343

Upper right - First "from scratch" mostly parametric design.
Lower Left - "Final" version of my mostly parametric design.

The design can still be improved. It's fairly easy to give values that make an unusable mount.

It's also a bit "heavy" for my tastes both at 11g and in looks. But 3D printed materials aren't particularly strong and I want it to hold up. I'm doing my test prints in PLA and will probably fly the PLA version. But I really want to print it in PETG I'm just low on PETG right now so I may have to make due with ABS which will be a real pain to get these parts to part well in.

The last time I built this quad up I just used 2 bolts and a bit of plywood laminated with hotel room key cards to clamp them on. It worked and was probably good for someone who crashed a lot because the motors just got pushed around on the shafts in a crash. It was however very annoying to have to constantly reposition them. These mounts are a bit heavier...but I'm also going to put 2204's instead of 1806's on here this time so I'm printing the motor mounts for strength with 4 perimeters and tops/bottom layers with 30% 3D honeycomb infill. The way this is designed most of the strength should come from the perimeters. So I'd cut back on the infill to save weight before I'd cut back on perimeters if I wanted to shave some weight.

There's also the quad itself to go with those mounts:


The idea is the brown bits are wood. The red bits are polycarbonate sheet or kydex or ply or CF. And the blue is 3D printed. So basically I'm adding a wedge between the plates which will hold the arms in the correct position and angle. Even without the angle this would solve most of the issues I had with the design originally. Eventually I'd like to machine the plates with my MPCNC when I finish it. But for now I'll hand cut them and may simplify them a bit.

It kind of works out like this:


You can see in the sideview that the arms are at 10 degrees.


I'm short some hardware to do final assembly but will pick it up in the morning. I was originally going to build this with 1806 motors again but mine are getting pretty beat up so putting this new set of YKS 2204 motors on seems like more fun. Plus those are setup with a Sparky2 which means it has a built in openLRS radio...if I can figure out a decent way to mount the antenna and keep it out of the props. But they're also setup with old slow emax "simon series" ESC's I do have 3 spare KISS 18a esc's on hand - since I've yet to burn one up I have all my spares. But using them seems like a surefire way to start having fires...so maybe I'll live with the slow ESC's on there for now. The original UBMQ had BS12's which weren't much better. Meh..we'll see what I feel like when I get time for it tomorrow.

Once I fly it I'll probably put it up on thingiverse...but the github repo will always be the latest (just not necessarily in the master branch)

I'm super open to ideas and suggestions. Better yet - I put it on github for a reason, anyone can fork it and make their own version then share back any changes.

I'm still working on making the main body more parametric and figuring out the best way to get it on Thingiverse as a customizer. I may do a full thing for the entire quad - but then upload the code for the angle block separately by itslef so it can be customized since the customizer won't work with included files.

Anyway I'm having a blast learning what OpenSCAD can do and would love to have some of you guys join in with experimenting. I started to figure out animations but didn't quite get it where I wanted yet:

 

[cranialrectosis]

Can't wait to see it fly!

Any idea about the final weight and are you planning to go 4S or 3S?

 

[jhitesma]

I only have 3s packs so won't be going 4s. Not really building this for all out performance anyway, just as a test platform to see how angled arms behave.

As for weight...I knew someone would be asking Right now without motors and only half the screws (but with a Naze32 and associated mounting hardware) it comes in at 153g



The emax Nighthawk 250 frame that I'm taking the parts off of to build this up was 167g fully assembled...but without a FC. So not too bad. The motor mounts are 11g each and the center section s 32g. (When printed in PLA, PETG will probably be a little heavier since it's denser) so that's about 76g heavier than the original version that didn't use any 3D printed parts. The original version was designed to mimic the dimensions of a warp quad and when fully built out the weight was very close to my friends warp. THat was using the afromini to cut a few g and ply which is lighter than arylic/polycarb. This could be done lighter - less infill on some of the parts, ply instead of acrylic, smaller lighter FC (though the Sparky2 is nice due to the built in radio.) But again I'm not going for all out performance here.

 

[airhawk]

that is amazing im looking forward to a flight video your original frame was ballistic it self

 

[jhitesma]

To be fair airhawk the frame wasn't the magic there...it was ridiculously high rates. I forget exactly how high they were but they were really stupid high for that

The UBMQ is coming together though. Stripped down the emax 250 today and started moving parts over...tossed the 250 frame on the scale for comparison:


Hey, that's the same as the UMBQ WITH FC

Let's see how the bits fit on it:


I found a minor mistake with the motor mounts. For most motors they'd probably be fine. But these YKS have extra "ears" on the circlips and need a bit more room. Since my printer has been busy all day with the last of the PETG parts for my MPCNC I just drilled these out. But I did update the design files and pushed them up to github. The "little_adjustments" branch is where I'm making minor tweaks as I assemble this thing for the first time.

This is a Sparky2.0 that Balu was kind enough to send me - which means it has a built in radio I can use with my OpenLRS. Not sure how I'll deal with the antenna on that yet but it will make it slightly easier to fit everything into this little frame. I kind of realized I forgot to leave room for a PDB or wiring. Oops. It still works though...the afro PDB that came with my emax 250 will just barely fit if I'm super careful with my wiring. I will have to replace the power wires on the ESC's but that's not a huge deal. May also shorten the motor wires while I'm at it. But if I have to do all that it's REALLY tempting to just put those spare KISS on here...except then I'd also have to dig up a 5v reg for the FC....

Anyway this is basically everything needed except props, battery strap and battery:


Not bad. Lower than I was expecting. Could shave some more weight swapping to aluminum hardware (there's a lot of fairly heavy m3 bolts in this the motors are mounted with 22mm bolts and the arms are held in with 2 30mm bolts each) And the KISS would be smaller/lighter than these emax ESC's. But I think this will do.

Starting to think I should maybe design a FPV pod for the top though.....Let's get it in the air and see how well it holds up to some crashes first

 

[airhawk]

cant wait to see it if i can find a cheap printing service i might do the middle section in a non clear material instead of the plexiglass like material you used just a preference thing

 

[jhitesma]

Polycarbonate/Lexan is nice becuase it's cheap, easy to work, and very strong.

But plywood is even cheaper, also easy to work, and plenty strong enough as well as lighter.

CF looks nice...but is heavier than Ply, a lot harder to work with...but very strong.

Acrylic (which is what I'm using until I cut new lexan) is somewhere between Poly and Ply for workabilty and weight but is really weak.

There really is no "perfect" choice.

Keep in mind though that most of the strength is meant to come from the top/bottom plate. The 3D printed center section just there to position the arms and keep things aligned. Without the plates it's pretty flexible and wouldn't work very well. Once the plates go on it gets VERY stiff.

My big concern is the motor mounts. Not sure they'll hold up well in crashes at all. Plus they seem bulky, and they don't attach to the arms very well. There's very little clamping force the way they're designed, and while it is a tight fit on the pocket it's not enough to keep them on . The screws all cut into the arms a bit but none of them go all the way through. So I may have to do some revisions on them still.

 

[jipp]

looks cool.. a 3d printer would come in handy i keep looking and finding things id like to try.

on the bonus i havea friend who will print anything for me at cost plus shipping.. i could not ask for anything better till i can designa nd print my own.

so would your hudson module work say ina taranis?

seems like that be a useful module for some.

im collecting parts for my next build.. i went with emax red bottom motors and all i can say is the quality is 110% above the regular emax motors i was use too. they sure upped there game. also found it odd they have a scratch off sticker to check for genuine emax motors. LOL

i guess ill always think of emax and good quality budget motors, not mongoose going after cobras for dinner. but yet they seem to have the logistic to support what they did make, and looks like willing to invest into better products.
i guess the industry learn how fast you can lose face with the sunnysky train wreck.

chris.

 

[mpbiv]

3D printers make hobbying so much more fun!

 

[jhitesma]

looks cool.. a 3d printer would come in handy i keep looking and finding things id like to try.

I've wanted one for over a decade. But could never justify it...too expensive to buy OR build and the quality of the printed parts was too low. But the past few years if it wasn't for my renewed interest in RC I'd finally have made it happen because the quality finally reached a point where I felt the prices were approaching reasonable.

The FolgerTech 2020 prusa i3 that I got is a killer value for the money - IF and I have to stress that IF...if you're willing and able to accept that it's not so much a "kit" as it is a "box of parts that can be used to build a 3D printer" and do enough research on your own to fill in the blanks. Thankfully due to it's low cost Folger has sold a bunch of these and there's a large community on the reprap forums, on Facebook, and to a lesser extent on other 3D printing forums/wikis/reddit/IRC who are will to help get noobs up to speed.

Since I've actually been collecting parts to build my own for close to 5 years and during that time doing a ton of research on 3D printing and how to build a 3D printer...that was no impediment to me. But I recognize that not everyone is in the same situation

I will say now that it's dialed in and printing reliably I absolutely love having it. I solve so many small issues around the house with a bit of CAD and a few hours on the printer that I'm pretty sure it's already paid for itself in the 4 months I've had it. (That isn't to say it's paid for all the upgrades I've done to it and all the filament I've put through it...but the $270 initial initial price has probably been recouped!)

on the bonus i havea friend who will print anything for me at cost plus shipping.. i could not ask for anything better till i can designa nd print my own.

I wish I had friends like that around here You need to finish building some of those quads on your desk though This is still VERY much a work in progress, I'm going to have to spray a cold hose in the general direction of the Phoenix area if you and airhawk keep it up

so would your hudson module work say ina taranis?

seems like that be a useful module for some.

To be fair it's not "mine" I just happened to build one. It's another design by midelic from rcgroups: http://www.rcgroups.com/forums/showthread.php?t=1954087

I just happened to have the parts on hand to build on and since I have my proto-x that uses the hubsan protocol made one up. Though honestly I've yet to even put it in the case I printed because the proto-x is more or less worn out at this point. It still flies..but not well. I've tried replacing the motors and replacing the battery...but it just isn't the same. Flies anemically and drains it battery way fast. Can't justify putting money into more batteries/motors for it since the last set didn't really help...at this point it's more or less a desk ornament

im collecting parts for my next build.. i went with emax red bottom motors and all i can say is the quality is 110% above the regular emax motors i was use too. they sure upped there game. also found it odd they have a scratch off sticker to check for genuine emax motors. LOL

The red's look pretty good and I've heard a lot of positive stuff about them. Haven't had a need for new motors though. To be honest I've mostly been looking at doing a <250g build due to FAA stuff. This 3D printed UMBQ is more or less just because I had an itch to scratch about seeing just how angled arms affect flight. I wish I could make it a <250 flyer...but just not going to happen with these materials.

i guess ill always think of emax and good quality budget motors, not mongoose going after cobras for dinner. but yet they seem to have the logistic to support what they did make, and looks like willing to invest into better products.
i guess the industry learn how fast you can lose face with the sunnysky train wreck.

The 1806 and 2204 emax I've got have both treated me well. I did finally kill one of the 1806, and had some issues with a replacement...but that wasn't emax's fault (poor amazon seller shipped me a used motor) (Side note: I'm rather happy in a schadenfreude kind of way that my post about the bad amazon seller has had more than double the views of my post about the YKS motors going on this quad which I had doubts about but have been working quite well so far!)

I'd love to pick up a set of the red's but my RC funds are basically on hold right now due to what this 3D printing habit has done to my hobby budget. Well, that and trying to keep both of our vehicles on the road for at least one more year now...was really planning on replacing the car before this summer but that's looking less and less likely for a laundry list of reasons I don't want to get into.

But yeah, I agree with that assessment of emax. They aren't going to achieve Cobra or SunnySky level of popularity...but they're a great bang for the buck that the average hobbyist would be doing themselves a disservice to ignore. May or may not be a good choice for a top level competitive flyer...but for the average Joe - I'd have a hard time recommending anything else.

 

[jhitesma]

3D printers make hobbying so much more fun!

They make so many things so much more fun I still want to try some of your designs...really starting to feel like a I need more than one printer. Not enough time to print everything I want with just one

Last night I ran out of PETG I've now used 3 full rolls of PETG. I've almost finished 2 rolls of PLA and a 3rd roll of PLA is getting pretty light...but PETG has definitely become my favorite material it seems.

I was trying to finish printing the last 2 feet for the MPCNC I'm building out of PETG but got a small jump in the X or Y axis on the 2nd to last foot and even after that didn't have enough filament left to do the final foot - so I'm going to have to order more PETG before I can finish the MPCNC now

On the upside I still had SOME filament left and decided to try and print two more motor mounts with a few minor changes (Slightly larger center hole for the motor shaft, slightly larger screw holes so I wouldn't have to drill them out as much, and shrank the overall height by 0.5mm to save a bit of material and give the motor screws a little bit more bite on the motor threads.) Figured I had just enough filament to print two of them...was more right than I expected:

Wound up with 220mm of filament left....that had me sweating at the end!

I also got the ESC's stripped down yesterday and have a plan for how I'm going to mount/wire them now. Hopefully I can do that tonight. Then just have to figure out a good way to mount the RX antenna and should be ready for some test flights.

 

[Snarls]

That's a close one right there. Looking forward to the rest of this build. I notice your extruder and bed tap the endstops twice before centering. On my printer there is just one tap followed by centering that usually ends with a grinding sound. Works fine, but it's kind of concerning.

 

[jhitesma]

That's a close one right there. Looking forward to the rest of this build. I notice your extruder and bed tap the endstops twice before centering. On my printer there is just one tap followed by centering that usually ends with a grinding sound. Works fine, but it's kind of concerning.

Yeah, that was a change in the newer version of Marlin I'm using. It can also home in two directions at once now which makes starting a print much quicker. The double tap is apparently standard for CNC machines to make sure they home correctly which is why it was added to Marlin.

Freaked me out the first time it did it after I upgraded, thought something was wrong at first! Then did some digging and learned it was expected behavior now

 

[Snarls]

Thanks, I haven't had the chance to upgrade my Marlin. I still need to clean up the wiring and think of some simple mods I can do to my printer without spending too much.

 

[Twitchity]

That's a nice looking 3D printer you have there, and the prints look great too. I'm really liking the PETG filament, and I've had great success with the eSun brand so far. For my past 3 spools this is the only filament I've used. Sticks great to the heated bed, doesn't warp, and is relatively strong. Right now I'm debating on redesigning my printer as there are a few aspects I'm not a fan on at the moment; one would be to remove the x-axis stepper motor off the carriage.

 

[jhitesma]

That's a nice looking 3D printer you have there, and the prints look great too. I'm really liking the PETG filament, and I've had great success with the eSun brand so far. For my past 3 spools this is the only filament I've used. Sticks great to the heated bed, doesn't warp, and is relatively strong. Right now I'm debating on redesigning my printer as there are a few aspects I'm not a fan on at the moment; one would be to remove the x-axis stepper motor off the carriage.

Thanks I'm really happy with it! For < $300 I'm really blown away by what this printer can do! Yeah, I put some money into upgrades...but not a ton. The leadscrews were like $25, the inductive sensor was $6, I've spend about $10 on fans and thermisters....and my big splurge was $80 on the e3d v6 all metal hot end so I can do higher temp stuff (Which I've barely taken advantage of yet other than the maker geeks PETG that like to run at 260c) Oh, and I spent $6 on a new stepper for the extruder so I'd have a spare since I was afraid I was going to burn out my original. By far my biggest "Expense" has been filament though so I can't complain

I really do like the PETG, I haven't tried eSuns and after the experience I've had with the roll of esun PLA I was sent (I ordered Hatchbox but Amazon sent me eSun with a sticker on it saying "Hatchbox") I'm not in a big hurry to try them.

The atomic filament PETG I've used I've been VERY happy with and get some very nice prints off of, the blue translucent I get amazing quality from - but it doesn't look that nice in photos because of the transparency. The gunmetal grey I didn't spend as much time dialing in my settings so I get good prints but not as good as I know I could from it. Only problem with the Atomic is that even with his 10% off code it's expensive

The Maker Geeks PETG has worked very well for me as well, and it's about 1/2 the price of the Atomic with the sale they have going on until the 24th. 2 rolls for $33 shipped. Can be PLA, ABS, PETG, HIPS - mix and match. Only catch...you get whatever color they send. (that's how I wound up with turquoise and fluorescent yellow.) I just ordered 2 more rolls today because I liked it so much. Kind of hoping I can get 2 rolls of PLA or maybe a roll of PLA and a roll of HIPS to try before the sale is over....at that price I have no problem not being picky about color And if I'm going to buy anymore PLA I kind of want to do it before summer so it doesn't melt together in the back of the UPS truck on the way here in our summer heat

Ok, enough procrastinating...guess I should go solder this thing up....

 

[C0d3M0nk3y]

Thanks I'm really happy with it! For < $300 I'm really blown away by what this printer can do!

How did you decide to go with the Folger Tech kit? It seems like there are about as many i3 kits out there as there are mini quad frames.

 

[jipp]

http://folgertech.com/products/folger-tech-kossel-2020-full-3d-printer-kit this is suppose to be a update to the i3s.. dunno, from the picture it looks like the one you built is bigger.

no doubt lots of kits..

chris.

 

[jhitesma]

How did you decide to go with the Folger Tech kit? It seems like there are about as many i3 kits out there as there are mini quad frames.

Markyoe's thread here: http://forum.flitetest.com/showthread.php?22307-Folger-Tech-2020-Prusa-i3-Build-Log played a VERY large roll in my decision.

I've wanted a 3D printer for a long time...When the reprap project first started in 2005 I followed it closely, but quickly decided that the cost of entry was too high for the quality of prints they were achieving.

Since then I've followed up a few times a year looking into how things have progressed. A few years ago (about the same time I got back into RC) I decided that the quality of prints was finally good enough to interest me - but the cost of kits was still too high and so was finding parts to build one from scratch.

But...I started collecting parts. In fact the Arduino Mega I used as the basis for my first flight controller was originally bought to go on a 3D printer - I figured I could start playing around with software first and collect hardware bits as I went.

Honestly I was never a big fan of the mendel designs. The i3 only moderately interested me. The big flat plates just didn't do it for me. I found designs like the cupcake/thing-o-matic/ultimaker far more viable with the box style construction. But I did like a lot of extrusion based designs I was seeing...just not the prices that went with them.

I hadn't heard about the Folger kit until Markyoe got his. I really liked what I saw. Specifically an extrusion based design that was priced <$300. Yeah, it's a Prusa i3 derivative (though some argue it has more in common with the i2 - I'd disagree) and the Prusa is a mendel derivative - but it's changed enough that the things I didn't like about the mendel/Prusa were gone.

There are cheaper kits - but they come from overseas with no support, I can't prove but strongly suspect that they have just as many (probably more) issues as the Folger kit. What the Folger does have is a US based company behind it that's very responsive to it's customers and a large user community - which add a TON of value to the kit. For example on my printer the hot end thermister was damaged out of the box. One e-mail and the next day Folger put a new hot end in the mail to me - not just a new thermister a whole hotend. I've seen quite a few people have problems and get full RAMPs/arduino combos as free replacements when they were suspect. My LCD didn't work, Folger asked me to do a little more debugging before sending a replacement - and I found that half the pins on the adapter board weren't soldered. While I'm sure Folger would have sent me a replacement it was quicker and easier for me to just solder the one I had myself.

So bottom line was for me the Folger kit hit the price/quality/support sweet spot dead on.

 

[jhitesma]

http://folgertech.com/products/folger-tech-kossel-2020-full-3d-printer-kit this is suppose to be a update to the i3s.. dunno, from the picture it looks like the one you built is bigger.

no doubt lots of kits..

chris.

Your link is for the 2020 kossel - that's a delta style printer and is actually considerably larger than mine. I would NOT suggest a delta as a first printer. They're considerably harder to get setup and working well and far less intuitive. Plus the Arduino/Ramps is barely capable of dealing with them as the calculations get a lot more complex.

This is the 2020 i3 that I have: http://folgertech.com/collections/3...ap-2020-prusa-i3-full-aluminum-3d-printer-kit

However their photos do make it look squished.

If you're interested in one the thread I posted above in the Tools section is full of great info. There's also a thread about them on RCGroups which is pretty good, and a thread in the reprap forums that honestly I haven't even managed to read ALL of yet. Oh, and I have a number of posts on my blog about the printer and my experiences with it: http://www.nevermindthesand.com/tag/3d-printing/ there's also another blog which played a large part in my decision to go with this printer: http://moosteria.blogspot.com/2015/05/3d-printer-project-part-1-decisions.html



Oh - I almost forgot. In my last post about why I picked the Folger. One other key element. I REALLY wanted a KIT not a ready to go printer. I've been following the development of 3D printers long enough that I really wanted to build my own as I knew that much like my RC Projects it would never really be "finished" and I would constantly be tweaking and upgrading it. The Folger isn't so much a "kit" as it is "a box of parts that can be used to make a 3D printer" which isn't appealing to a lot of people...but for me was VERY appealing.

You'll probably notice I've modified mine pretty extensively - mostly by using it to print parts for itself. All of the pink/blue/yellow/green parts on it were printed by it as upgrades. The only original printed parts still on it are the X carriage (I've printed an improved replacement but haven't installed it because it's not a big enough improvement and kind of a pain to tear things down far enough to install) and the bearing mounts under the bed for the Y axis - which I've also printed replacements for but not installed.

I actually have a major upgrade for the Y axis ready to be installed and may be doing so later this week - replacing the hardened chrome rods and linear bearings with a large extrusion and sled that run much quieter and are less fiddly to get aligned. But I've been trying to finish up the MPCNC before I tear things that far apart.