Diatone ET 150 Micro Quad "Marcie"All the features of a 450 class quad on a 150 frame. Take off weight / AUW = 230.7 grams.
8-20-16: More outdoor testing today, now using the new video system. Video wasn't perfect, but flew very well. Should have used crosshair antenna. Flew 4 batteries in "2d no hold" mode, LOS (line of sight) because the city has criminalized FPV. Two other guys were there early flying their Phantoms FPV, but left when people started arriving. Very tame gains, punch out was like an elevator. Think she's ready for some distance flights / FPV.
7-10-16: Installed rebuilt video module. Works beautifully. Weight down to 230.7 grams.
7-10-16: Test-flew "Marcie" at the field today. Flew two batteries (awaiting more XT30 connectors) without flight issues. Gains seem a little weak, punch-out was a joke. But no surprises or fainting. Drank the 450mw/hr batteries in 3-4 minutes, left them very warm at 20% remaining.
Video looks terrible, top half of picture missing. Rebuilt video system using a VT200 in stock. Instead of the system sitting atop the VTX (which gets reallly hot), it now stands on end using the aluminum stand as a heat sink. I didn't want to use any glue or tape at the contact point, which might insulate heat transfer or fracture from the thermal stress, so I used a nylon tie for structural support. There is some hot glue on the back to eliminate slipping. The VT200 has 32 selectable channels, although the camera blocks the LEDs. There's a pushbutton usable from the side. It also came with a SMX(?) connector which I removed. For me they come off easier by using a Dremmel to cut the legs off, then the individual legs come off without any stress to the PCB.
I made a new camera holder from some a plastic box, glued to the aluminum base with E6000. It will stay clamped as shown for 24 hours. As before, a loopback cable lets me test the video system without the Vector plugged in. Normally video runs into the Vector, where it overlays the OSD info, then back to the video system for transmission. The initial test looked great, powered by a bench power supply. Will get connected direct to battery, this VTX can take up to 3S directly, and outputs 4 volts for the camera. Under the (same old) aluminum stand is some 3M Dual Lock to attach it securely to the frame.
7-9-16: Reassembled "Marcie" and test-hovered OK. Ran the battery down to 3% and shut down, so the Vector's safety must be off. Replaced JST with an XT30 power connector. Failed to consider current, which is why she was failing. The motors need 24.7 amps (that's 6.17A each) at full throttle, the JST can only pass maybe 10 at best. The video system takes 5 volts from the PDB, then runs it through a Pololou regulator down to 3.3 volts for the camera and VTX. There are connectors now so the video system can be removed and tested separately. She looks the same externally, just some wiring fixes.
5-19-16: Video components tested OK. Ordered 20v to 5v regulator for the camera. With both devices on separate regulators the video should be clean and nothing should overload. In the pictures below you can see the sandwich I'm building, with plenty of air getting in for cooling. It's not done yet. A video in/out cable goes to the Vector, and a single power cable goes to the battery with a JST connector to carry a couple of amps. The regulators are right next to the video system so they can't pick up as much noise. So what did I miss?
5-16-16: Video system's a mess. Plastic camera bracket melted into transmitter and camera. Going to provide separate regulators for the camera and vtx, sitting on aluminum mounting plates made of coke can aluminum. Drilled holes in the mounting plate to save weight. Mounting camera on another aluminum bracket made of Pepsi can. All this will shed heat faster and provide cleaner power. Have to test everything first, probably order more parts. Have to find out if epoxy can tolerate the high heat these parts generate. Going to sandwich the regulators between the VTX underneath and the camera bracket above. All separated by thin aluminum. Will all be one piece for durability.
4-10-2016: OK so she's back together and still flies. The Vector's PSU powers the FC and VTX. The Diatone's PDB powers the camera and radio. The OSD shows about a 1 amp draw, which is the PSU's limit. No more video battery, but now there's noise in the video, but it's not too bad. Working on the wiring diagram so you can see what I did.
Youtube video of today's test hop.
4-8-2016: Still had a power problem, because the Vector PSU only allows 1 amp total output. Something's gotta come directly from the battery. So I ripped out the generic PDB I just put in, and put the Diatone one back in. This gives me five volts for the camera direct off the battery. The PSU provides 12 volts that I regulate down to 3.3 for the VTX, five volts for the RX, powers the Vector, and feeds the OSD with battery info. I could feed the radio with the other five volt regulator on the PDB if needed. The pics below show the Diatone PDB going back in.
4-3-2016: Trying to button "Marcie" up, but problems keep appearing. Found the video module is pulling 1.3 amps, keeps tripping the shutdown in the Vector PSU which is only good for one amp. So running the 12v regulator direct from the 3S battery. First this blew a regulator when the cable turned out to be reversed. The next one got very hot, because it's only good for 12 volts, not 12.7. Hope I didn't blow up the video camera / tx. Connected "Marcie" to the config software and changed the OSD around. Shortened cables to save weight, GPS still works. Did a test hop indoors to see if she can at least hover. Would like to find a 4S to 3 volt regulator that can provide 2 amps. Other than that I think she's back together.
4-3-2016: About to install a 3 volt regulator so the video module can run on the 12 volt filtered line from the PSU. Found a pinout for the Vector video harness, have not checked it yet. Had removed the pins that power the Vector to remove the video harness, and power it (and everything else) from the Diatone 5 volt regulator. Powered the Walkera RX1002 radio from the PSU, tested OK. Uses only .04 amps, not the 1.3 amps the manual states. Mounted the RX on some thin plastic insulator and hot-glued to frame. Was painted with liquid tape and strapped before with a nylon tie. Thought the vibration might wear the liquid tape off and short the RX to the frame. Re-mounted the antenna wire to a smaller nylon tie, with strips of shrink instead of a single long piece to cut weight. Added a little hot glue into the 3M Dual Lock that holds the PSU and FC so they can't shift or fall off. Should also paint some liquid tape on the exposed PSU board.
3-31-2016: Installed generic PDB, trying to keep wiring short. Reinstalled the FC deck as before, then the Vector. Put the Vector's PSU unit where the video battery went. Have to shorten the wires, and install a 12v to 3v regulator for the video system. The wiring for the GPS module gets shortened also.Trying to save lotsa grams.Took the paint off the ESC and FC chips for cooling. No wonder I was having problems, the Walkera RX-1002 receiver alone pulls 1.3 amps, and only works on 5 volts.
3-28-2016: On every outing so far, "Marcie" has fainted in flight and fallen out of the sky. Think I found out why: The Diatone P.ower D.istribution B.oard has 2 five volt regulators, which I used to power everything, including the Vector flight controller. Sadly these have a thermal protection circuit that shuts them down, and a one amp limit. To make matters worse, I buried the PDB under the Vector FC, which also generates lotsa heat. To fix this, I'm replacing the Diatone PDB with a plain one without regulators, and installling the Vector P.ower S.ensor U.nit, which provides filtered DC for accessories and powers the FC. This also gives the FC current and voltage data for a safer landing, and adds that to the OSD so I can see what's left. Because it's filtered, at least I can remove the video battery, and it weighs less so it should keep the weight under the FAA registration limit. After ripping her apart I found paint in bad places, which I'm removing. I should have used dayglow tape or shrink, the paint was a mistake. To remove the XT60 connectors I clipped the pin headers, cut off and desoldered the connectors, and soldered the sub-board back on the shorter headers. It was tested with the battery shown, and is putting out the right voltages, so I didn't destroy it yet.
3-28-2016: Removed paint in stripes to lower weight. New weight 224.3. Remaining paint adds 1.5 grams.
3-27-2016: Flew 4 batteries at Apollo field. Orange paint helped find her in long weeds, but makes her a little sluggish. On one flight she shut down and fell about 20 feet. Without the current sensor the battery can run down without triggering the safety mode. No damage from the session, including several hand catches. Think I may remove paint in strips like a zebra to lighten the load.
3-23-2016: Fixed video issue, connector not fully seated in flight controller. Removed paint from motor clip pockets and screw heads.
3-21-2016: Flies OK with the paint, but looks like a video problem developed. PM: Video connector to Vector was backing out. Pushed back in to fix video. Also cleaned out motor clip pockets and all screws.
3-19-2016: After walking right by "Marcie" in some grass during an hour-long search I decided to enhance all my girl's visibility. Ugly? You bet! But the objective is not to lose another quadcopter. At first I hoped to do this without removing anything or taping, but it required several coats, so I ended up removing the rotors and taping up the motors and lenses, after cleaning up a lot of overspray. The paint thinner didn't damage anything, and I hope the paint isn't conductive. I really should have done one first to see if it's fatal, but painted all three at the same time.
2-13-2016: Crashed at end of first battery. Found the GPS module was loose when recovered (about an hour later). The bluetooth locator was lost, kinda ironic. Found some grass around the motors, may have wound in from launch, or the GPS module may have fallen off. Added two nylon ties to hold GPS down, and will use a launching pad from now on. When home she flew fine again, no damage. Added the zip tie antenna mast seen online, removed the antenna tube mast, which was heavier.
Youtube video of today/s flight.
1-16-2016: Finally got a good hop out of "Marcie". Has just been sitting in her cigar box waiting to fly. Just wanted to see if she could do a little distance, and get to cruising altitude. Also pushed the speed up to about 30 mph, and did a rapid dive at the end. Some instability during rapid descent, but never lost control. Auto-detected landing and disarmed herself.
Another great capability of the Vector flight controller is it's built-in logging of both GPS and system functions. The config software reads the logs and provides this map overlay.
Youtube video of today/s flight.
11-15-2015: Several great test flights today. Took my time, calibrated compass whenever advised and waited for a good home fix before taking off. Flew three batteries, about five minutes each. Realized that the OSD (receiver) voltage monitor says nothing about remaining flight time, have to add a timer. In the video below I tried a "punch out" to test thrust, she climbs to 80 feet in a second. I'm only flying in 2D mode (no hold) to avoid any flyaway potential. May add gear to protect battery on landing.
Youtube video of third flight today
Learned that the motor shafts are so long that they extend into the spinner "windows", meaning that if your tool takes up the entire hole the spinner can't tighten completely. After removing the tool, the spinner is loose and the rotor is coming off. Now using a large cotter pin (opened so one side flat) so the spinner can completely tighten.
11-14-2015: "Marcie" flew outdoors today, and flew pretty well. There isn't any video because the cold seems to affect the FPV video system. Since I couldn't calibrate the compass, or see if any GPS sats were found, I flew in 2d no-hold mode, just Line of Sight for about five minutes. She got a little further away, then made a dash for a wire fence. I didn't notice at the time, but the crash ripped off one of the surface mount antenna connector on the Walkera radio. A prop broke also. Not sure if it was just loss of orientation (pilot error) or something mechanical. When I got her home I noticed the props were on upside down. I decided to reinstall the motor screws with thread sealant, and tighten everything else up. Wind was calm, and temp was about 45. In warmer temps the video system works fine. Also flew "Lucy", see her page. PM: Turned the gains down to 50% on pitch and roll, and 75 on yaw. Left stabilizer responsiveness at 2, but tempted to turn it down to 1. Turned off the autoland at low volts, forgot that I'm not using the Vector's power adapter, so it doesn't know what voltage it's got. I'm using the "receiver" voltage to watch the 5 volt supply.
11-3-2015: Always used nylocks or self-tightening rotors on larger quads, and I read how much others hate using spinners. So when I couldn't get the nylocks on last weekend I decided to wait. On Sunday I spoke with Chris, who knows his stuff, and said he uses both just fine. He suggested opening up the nylocks with a regular bolt, so they would screw on easier, or just crank the spinners down hard. I did both. I used the spinners for now, and "broke in" the nylocks for later. I still want a wrench of some kind to hold the motors while I screw on the nylocks. The Phantom comes with such a wrench, and I really need one with these tiny motors. I also trimmed the radio antennas down and used some antenna tube to keep them away from the props and angle them 90 degrees from each other. If she flies well I'll invest in a Dragon Link Copter micro receiver. Also considering landing gear. Currently lands on battery. As for modes, she's set to 2d no hold, 2d with hold, and RTH. That's the only switch being used. The Vector allows a second mode switch and a knob to adjust one gain in flight. The Vector's stick menus can also be used in flight, but not by me. So still waiting for a first real flight, maybe next weekend.
10-27-2015: Still have to figure out what to do about those antenna wires. Otherwise mechanical assembly done. Started looking at gains and the OSD tonight. Turned vertical down to 25%. Turned idle speed down too far, now motors don't spin at arm. Turned safety back on and calibrated the compass. Have to re-learn the stick menu system. Added a battery strap, which unfortunately covers the USB port on the Vector. Currently set to only 2D no hold and RTH on the mode switch. Have to figure out how I did the 3 position mode switch in Deviation again.
10-26-2015: Completed the video module, then installed it and GPS module for some testing. Everything's working fine. I'm must reconfigure the OSD to remove items that won't work on "Marcie", like the voltage and RSSI functions. I don't need the artificial horizon either. Maybe there are some I can add. A very short video shows the system aquiring satellites, with the gps module mounted where you see the video battery in these pics. That little battery gets stuck under the nose in flight. I couldn't test the video module alone, because it routes the video into the Vector and back to the transmitter. You can see here now nicely the module fits in the nose. Have to trim those long Walkera radio antennas.
Youtube video of GPS test
10-25-2015: Today I started to work on the video system. I wanted a removable module containing the camera, transmitter and antenna that would fit in the "nose cone" of the ET 150. An aluminum bracket forms both a base and a mast for the antenna. The transmitter is CA glued (for heat transfer) to the base, and the camera support bracket (plastic scrap) is glued to the transmitter. The camera is glued to it's bracket, which has a slightly adjustable angle. The antenna is attached to the mast with a couple of nylon ties, ensuring that no mechanical stresses get to the antenna connection. The wiring is next. I've removed all the unnecessary pins from the Vector video harness, which will solder to the camera and transmitter, allowing the video system and the Vector to be removable. Once this stage is complete I can determine if the GPS will work right atop the aircraft, or if I must use the mast. A little 3M Dual Lock holds the module in place. The whole subassembly weighs 11.1 grams.The whole video system comes from "Banggood.com", and costs about $35 bucks. It's exactly the same design I used on "Patty" and then "Lucy", because it works so well.
10-24-2015: Things moving faster now.Soldered the Vector radio harness to the Walkera radio, painted the bottom with liquid tape and attached to the frame with 2mm nylon ties. Found the easiest / lightest material to mount the Vector over the deck screws was a piece of cardboard. Some 3M Dual Lock makes it removable. It has to be removed to remove the deck, and access the power dist board and receiver. Had to pull it apart because I forgot to bind the radio first. Reassembled and hooked Vector to the computer, updated firmware, then ran the radio analysis program. Radio working fine, have to adjust throws, but channels correct. Radio is not powering the flight controller as hoped. Will have to run another power line. Have not been able to arm motors yet, and won't even start FPV work till she flies. Have to re-learn what I knew of the Vector.
Click HERE for today's first test hop video on Youtube.
10-23-2015: Today the EMAX Nano 12 amp ESC's arrived. They look great and fit perfectly as you can see below. I removed and trimmed the wiring, painted the back with liquid tape, and used auto trim tape to stick them on the arms. This lifts them up to get some air under them. Then I added a 2mm nylon tie for mechanical support.It took about three hours to get them in place, then I crossed my fingers and plugged the battery in. These are opto-isolated escs, so the tester had to be powered from the bench power supply. They spun up very smoothly, but of course two were spinning in the wrong direction. A few minutes later, four wires swapped, and it's all good. I'm using the connectors in case I have to switch flight controllers. I really have no idea if this thing will fly with the Vector. After watching these motors fly 220's on Youtube, I'm sure weight is not an issue. I also bought some 3030 props, 3S batteries and nylon locking prop nuts from the local shop. Maybe this weekend I can get the radio and flight controller wired up and we're ready to go. OOPS! Just realized that the pins come out the BACK of the Vector. Had to remove and reinstall all the ESC's. Retested OK. Now I'm looking at the Walkera radio, which goes under the flight controller deck. Think I'll use heat-shrink to insulate it. If I can get this flying OK with the Walkera, I'll invest in another Dragon Link "copter" receiver for longer flights.
10-22-2015: The motors arrived from MYRCMART.COM, and I managed to install them without destroying them in the process. This time I was a little smarter. I used the LC meter on them first, to make sure they were at least all similar to start with. Then I test-ran each one on the test jig. They sounded great! Then I installed one on the frame, and tested it again to see if I fried it. Once I was sure the screws weren't too long I installed the rest, and tested them all again on the frame. Gotta know what I can trust. They are beautifully made, sound silky smooth, and mount on the ET 150 frame without any modifications. The long wires will be trimmed short, as the ESC's will be on the arms. You can also see the flight controller deck, made out of thin plastic from a snack container. Same stuff I made the upper deck on Lucy from.
The LC readings look like this, all in uH:Severe differnces mean shorted windings, or opens. The LC meter came from Ebay.
10-15-2015: Ordered four RCX H1306 3100KV motors from MYRCMART.COM for $35.96 total, four Emax Nano series 12 amp 3S to 4S ESC's from Emaxmodel.com for $51 total, and three more Micro MiniOSD's from an Ebay seller for $27 bucks total. Found a good Youtube video on rewinding brushless motors, so maybe I can use the 1104's on something else. It will probably take a month for these parts to arrive, and I haven't even flown "Lucy" yet, so no rush. The video antennas, tiny cameras, and props for the 1104 motors arrived yesterday, along with a better ESC tester that shows the pulse rate.At Harbor Freight I picked up a nice assortment of heat shrink for a few bucks, but they didn't have the little 1.5mm drill bit I need for "Lucy"'s props. And from Amazon I bought the Dremmel circle cutter (for trimming props down to 3") and some 1.45mm drill bits for "Lucy". None of the big hardware stores stock the circle cutter, which surprised me. The power distribution board is a pain. It doesn't sit flat, so it creates mounting problems for the big Vector flight controller. The screws stick out the bottom, making the battery mount difficult. I can make something better. One of the ESC's can power the Vector and video stuff.
PM: Starting to consider how I'm gonna fit things together. Think I'm gonna need a deck to mount the Vector flight controller on, so I'm mocking it up with a business card. This is a naked Eagle Tree Vector and a naked Walkera RX1002 fighting each other for space. Half the Walkera receiver is under the deck, next to the power board, with the bind button sticking out for easy access. The ESC's will have to stay on the arms. The Vector goes on a removable deck and I'll use the pins and connectors so it can be disconnected. Maybe I'll power the Vector with the five volts from the power distribution board, but that may be too noisy. Thinking the video system should all be clustered in the front together, with the video transmitter standing on edge. For clean video I want to use exactly the same setup as "Patty" and "Lucy". The Vector includes an OSD. If she actually flies I can use the Eagle Eyes antenna tracker (I already paid for) with the ground station to aim a directional antenna and get a better picture at distance.
10-7-2015: Got a reply from LIFTRC.COM today about the RotorX 1104 motors. Another expensive lesson for me, it was entirely my fault. Even though the motors turned freely, the long screws cut into the coils, shorting and cutting them. Note to myself: measure everything with a caliper. Nuts. It wasn't even necessary to pull the bell off the motor, you can see the damage through the vent holes. The plan is to use 1306 motors now anyway, need "more power, captain". Wonder if they fried the ESC's also. Can't describe how much I hate wasting parts like this. It's not the money, but that they never got to work. I'm struggling now with how to mount the Vector flight controller, while keeping the power distribution board. I don't want to use the Vector's power sensor either. With the GPS and a video battery on top, the battery has to hang underneath.
10-5-2015: I ordered the "Round Shape Bluetooth 4.0 Anti-lost Key Finder Camera Self-timer Remote Control Smart Tracker For iPhone Samsung" (SKU241569) a while ago, and it finally arrived. There are some problems, but I hope to stick it on "Marcie", to track her with the cellphone if lost. It's bluetooth, so doesn't depend on the cell network at all. That also means it has limited range, maybe 75 yards. It came with a dead battery and no instructions, at least in English. The app they suggest has been removed from the Google Play store (I have a low-end Android phone), so I had to find something else. An app called IFinder works just fine for location, but not for other features like the remote camera shutter or beeper. The tag is always on, for at least a few months. It weight about 43 grams with the plastic case. The case is necessary to hold the switch and battery in place. Maybe I can shrink wrap it. It was about seven bucks. I was thinking of using a 433 mhz beacon, or the one built into the Dragonlink Copter model, but this is cheap and easy for now. It will fit under the camera under the nose.
10-4-2015: After testing the motors with the LC meter and a multi-meter, I've found three are open. I've removed all the motors and ESC's from the frame, and replaced the power input cable with a JST for a smaller 3S battery. After removing the motors I checked the screws with a loupe, looking for contact points, and found none. Where did I go wrong? I wrote to the vendor, maybe it's a bad batch? Think I'll use 20 amp Afro Mini ESC's and 1306 motors instead. Have to rig up a test for the Turnigy Plush ESC's to see if they're fried also. The frustrating part is that I didn't learn anything from (yet another) expensive disaster. Next time I'm going to test the motors and ESC's separately before installing anything on the frame.
9-28-2015: Well that's disappointing. Since I've run out of work I can do on "Lucy", I decided to mount some ESC's on "Marcie" and test the motors. I used the Turnigy Plush 6 amp that I wanted to use on "Lucy". They're for 1S or 2S, or 3S with no gurantee. The motors are the RotorX 1104's I ordered specially for "Marcie". They're for 1 to 3S and being tested with 4S. Lately I think I should use 1604's instead. I soldered everything to the power distribution board that comes with the Diatone ET 150, plugged in a big 3S battery (the one that came with "Christine") and plugged in the servo tester, just like I did with "Lucy". One by one the motors smoked, sparked and ran intermittently. At full power they stop. Everything gets hot. Tried on 1s and 2s, got nothing. So did I blow up the motors or the ESC's or both? Think I"ll try the motors on my 40 amp esc, and the power supply instead of the battery. I have some Afro 20a mini esc's that I actually intended to use with "Marcie", but the're a little to big to mount on the arms. Maybe I'll put them on a mid-deck above the power board. These motors look too tiny to me. This was supposed to be the easy part.I'm not interested in an overweight 120, I want a monster that can carry my Eagle Tree Vector, left over from "Gidget". It has a flight controller, a color OSD, and a tiny GPS module with return to home and auto-land. It also includes a data logger. But it's heavy for a micro-quad, so I want an overpowered Clydesdale that can the mail.
9-17-2015: The 1104 motors arrived, so I stuck them on the frame. They look much smaller than I thought they would. I really should have got some 1306 motors. The 1104's make 110 grams thrust, the 1306's make 230 plus. Below I parked "Marcie" next to "Lucy", also in construction. I'm not interested in acrobatics, but "Marcie" is going to be heavy.
9-11-2015: First entry: Have ordered all the parts needed to build yet another quadcopter. I'm getting very good at sourcing parts, and spending a lot less per unit than I used to, which is something. However I'm running out of Peanuts characters to name them after. "Marcie" is large enough to carry the stuff that wouldn't fit on "Patty" or Lucy. She will be a 3S unit with larger motors. I'm thinking of using the Eagle Tree Vector (flight controller / GPS / OSD / data logger) to get all the functions I want. I used it on Gidget, but wasn't happy with how it flew. Maybe I can figure it out this time. The frame was ordered first (on an impulse) for $20 and FPV parts from BANGGOOD.COM, the RotorX RX1104 motors are from LIFTRC.COM.From HOBBYKING.COM I ordered the OrangeRx R615X DSM2/DSMX Compatible (works with my Devo 10 transmitter with Deviation) 6Ch 2.4GHz Receiver w/CPPM (2) for $21.74, and the Afro Race Spec Mini 20Amp Multi-Rotor Speed Controller with BEC (5) for $64.95. I think these ESC's will fit on the arms, which is necessary to use the Vector flight controller. The upper deck will have the GPS module, minus the mast I hope. The battery will hang underneath. The objective is a true FPV aircraft able to cover some distance, like a tiny TBS Discovery.