|
I don't have a know working setup, im adding FPV to my first scratch build. And yeah the cap's off 
|
#
?
Jul 12, 2017 23:31
|
|
|
|
| # ? May 27, 2024 17:03 |
|
|
I had a similar issue on my first power on. It was because I had braided the power/gnd/video wires for neatness. Once I unbraided them it was fine.
|
|
#
?
Jul 13, 2017 00:26
|
|
|
I had similar issue and the ground pin on the camera was loose.
|
|
#
?
Jul 13, 2017 00:30
|
|
|
I really wish it'd be that simple on this build :/
|
|
#
?
Jul 13, 2017 06:06
|
|
|
I'd you're getting solid black then it means that the vtx and vrx are working fine, else you'd get static. The problem is either your camera or the connection from your camera to the vtx, that's where I'd start investigating. Do you have another camera you can try?
|
|
#
?
Jul 13, 2017 07:53
|
|
|
Yeah I have both a mini and a micro, same result. If the black screen means the vtx is good then it's just a matter of waiting for another camera to get here. I've tested the harness extensively, and eliminated the JST on the camera side by hooking the metal crimp-ons straight to the camera pins. Thanks for the help fellas, now I get to go through the RMA fun times
|
|
#
?
Jul 13, 2017 11:31
|
|
|
Does anyone know if this is even reasonably acceptable positioning for a FrSky L9R's antennas in relation to the receiver itself? That second box is a vector GPS unit. This position is pretty much the best place on the plane as far as separation between the 5.8ghz video transmitter and the RC receiver, but I really didn't want to stuff the RC receiver inside the fuselage since it's so tiny in there I'm probably not going to be able to secure it (everything in that picture's held on with 3M VHB tape while I get the layout figured out.)
|
|
#
?
Jul 14, 2017 09:23
|
|
|
BabelFish posted:Does anyone know if this is even reasonably acceptable positioning for a FrSky L9R's antennas in relation to the receiver itself? That second box is a vector GPS unit. Should be ok. The L9R doesn't have telemetry so it won't be interfering with the GPS reciever. I have similar separation between by L9R aerials and Vector GPS. I would try and get the reciever inside the fuselage though. Can't you just velcro it somewhere?
|
|
#
?
Jul 14, 2017 09:28
|
|
|
Slash posted:Should be ok. The L9R doesn't have telemetry so it won't be interfering with the GPS reciever. I have similar separation between by L9R aerials and Vector GPS. I can probably velcro it to the inside of the fuselage, and just struggle with it if I ever need to re-bind. As an aside I finally got my 900mhz crossfire diversity receiver up and running (took an update to the beta firmware for some reason.) I assume a 900mhz telemetry receiver in the same spot will probably cause the GPS issues? Edit: Rough Mock-up 
BabelFish fucked around with this message at 10:19 on Jul 14, 2017 |
|
#
?
Jul 14, 2017 10:02
|
|
|
BabelFish posted:I can probably velcro it to the inside of the fuselage, and just struggle with it if I ever need to re-bind. I believe that may cause issues, I've not tried crossfire myself. I believe there is something called a notch filter(sp?) which you can use to mitigate impact of the 900 MHz signal on the GPS though.
|
|
#
?
Jul 14, 2017 10:22
|
|
|
Runcam update: a rep got back to me promptly and offered to send me replacement main boards, so that's nice.
|
|
#
?
Jul 15, 2017 00:42
|
|
|
Anyone else at flitefest? I'm doin an overnight this time and it's pretty sweet. Love the night flyers. Ended up buying a bnf radian xl with a flight or two in it for $160. Gonna build a simple cub in the morning and debate buying five more for a build night.
|
|
#
?
Jul 15, 2017 03:54
|
|
|
moron izzard posted:Anyone else at flitefest? I'm doin an overnight this time and it's pretty sweet. Love the night flyers. What's the porta potty/washroom facilities like there? Also the power situation, is there like a central BATTERY CHARGE section or does every site have a hookup? Can you even camp in a tent or do you need an RV? I'd really like to go next year but I don't fly planes at all, would I still enjoy it? I figure they have a ton of quad stuff too from the videos I've seen but they also like to focus on the 1000x planes in the air.
|
|
#
?
Jul 15, 2017 04:11
|
|
|
From their vids it seems like you can just drop off your packs to be charged, and you can tent-camp for sure
|
|
#
?
Jul 15, 2017 06:08
|
|
|
Ahahahaha remember that weird FC with the separate gyro on a bit of foam I was making fun of a month or so ago? https://www.youtube.com/watch?v=wwOwja_jOzM Turns out that was Mr. Bardwell's idea
|
|
#
?
Jul 15, 2017 08:17
|
|
|
I find that foam padding iffy. While it might probably be able to filter out very high frequent noise, the mid frequent range might be able to cause some resonance that'll feed into your poo poo. Not sure how that helps. Plus mechanical delays, depending how soft the foam is. I eventually want to learn KiCAD and design a board with four gyros (secondary ones offset 22.5° from each other) and average their poo poo. Each doubling of gyros should reduce overall sampling noise by 3dB. That should certainly prevent poo poo getting fed into P and D terms and/or allowing to raise the filter cutoff and slope for delay reductions. That project might probably end up in a dud though, I can feel it coming Combat Pretzel fucked around with this message at 13:02 on Jul 15, 2017 |
|
#
?
Jul 15, 2017 13:00
|
|
|
You'd spend a lot longer talking to all those gyros, add in any latency due to filtering or averaging ==> effective sampling rate would go way down. Not saying that's bad, mind you, but in the drone world "more samples per second" is usually equated with "more better".... noise be damned.
|
|
#
?
Jul 15, 2017 16:19
|
|
|
Nah, reading out a single BMI160 is like 20us with DMA, running two SPI buses, the whole readout part should probably take maybe 50us total. The math stuff is just a couple of multiplications to rotate the data on two axes and then average the poo poo. Should be negligible. I'm foremost about killing noise instead of cutting delays. Latter can be considered if for some reason the result ends up being overwhelmingly good. This is mostly an experiment first. I'm not into the Betaflight looptime craze. We're currently running the BMI160 at 1600hz and it performs just fine. Assuming perfect clock, that's 625us between updates, delaying ESC updates by 30us of delay wouldn't matter much. I'm currently running the lowpass filter at 105hz second order. That's about ~1.5ms delay, dropping the order to one (PT1 on BF) that'd cut it down to ~0.8ms. If averaging multiple gyros results in a similar noise profile after a first order filter as for a single gyro after a second order filter, then it is all a net win. Because I loving hate noise. Makes everything inefficient. Burns energy and makes the D-term whacky in general.
|
|
#
?
Jul 15, 2017 16:54
|
|
|
Oh sweet, that's really fast. I guess I did not realize you had 2 busses to work with and the reads were so quick. Are you actually hacking up the BF firmware, or..?
|
|
#
?
Jul 15, 2017 16:59
|
|
|
The F4 (and F7) controller has actually six SPI buses. Altho I think you can bundle four into this QuadSPI thing. Nope, working with dRonin. I have this whole new sensor framework in the commit queue, that's theoretically capable of multisensor out of the box. Kind of need a test board. I guess I could stick each IMU on its own bus, but two per bus would let me validate some other code. This all needs to get past the sticking-tiny-things-onto-a-pcb-with-tweezers phase without me killing myself, though.
|
|
#
?
Jul 15, 2017 17:06
|
|
|
That's cool as poo poo. I didn't realize we had a developer in our midst.
|
|
#
?
Jul 15, 2017 17:24
|
|
|
Nah, just dabbling. There's a bunch of things I want to try to improve the performance, so I fiddle with the code. If it works well, it goes into the codebase. The dRonin codebase, at least flight-side, is pretty well structured and readable. I find Betaflight to be an annoying mess. IMO, anyway. However, as things usually go, amount of developers scales with the audience, and everything is a popularity contest, which Betaflight won (one of the dRonin main devs doesn't actually mind, because he hates doing support). So respectively the project moves only so fast. It's mostly just about board support, though, since the flight performance "improvements" over on Betaflight is a lot of woo-woo or kludges to fix bad tuning. The core of it all are still PID controllers, so there's not much you can do. Out of the box, quads running BF fly better though, since they can ship with a decent set of default PIDs, whereas our defaults have to work with autotune, which doesn't like twitchy PIDs, and autotune can fail, too. Granted, the multi-gyro thing is going a little overboard, but it'd be cool to have my own custom flight controller, if the effort works out well (even if it'll be rather barebones).
|
|
#
?
Jul 16, 2017 01:47
|
|
|
You could maybe support redundant IMU's on whatever UART's are pinned out, so when everyone moves to F7 you'll only have to get IMU boards, not full on custom flight controllers.
|
|
#
?
Jul 16, 2017 17:03
|
|
|
bring back old gbs posted:What's the porta potty/washroom facilities like there? Also the power situation, is there like a central BATTERY CHARGE section or does every site have a hookup? Can you even camp in a tent or do you need an RV? I'd really like to go next year but I don't fly planes at all, would I still enjoy it? I figure they have a ton of quad stuff too from the videos I've seen but they also like to focus on the 1000x planes in the air. they had showers, toilets were fine. Last year there was definitely a battery station, wasn't sure if anyone was sponsoring it thisyear. a decent amount of quad / fpv stuff there also. I walked by joshua bardwell's tent and I think he was about to review this https://www.youtube.com/watch?v=EJDAeQ0UtS4 I ended up buyin a bnf radian XL off a guy for $160, flown like twice. He had to emergency buy one at joe nall and had covered it in LEDs (it was very, very visible in the night fly. Also a simple cub kit, shirt (it looked good), and like 50 sheets of wr foam board.
|
|
#
?
Jul 16, 2017 17:18
|
|
|
HK just put about 320 items on sale (that I track anyway). It mostly seems to be spread over batteries, motors, and ESCs. I really need to add airplanes and other stuff to my tracker. Here's the link to my tracker if you want to peruse, and also reminder that I don't make money from it.
|
|
#
?
Jul 16, 2017 20:59
|
|
|
My Rhythmic Crotch posted:HK just put about 320 items on sale (that I track anyway). It mostly seems to be spread over batteries, motors, and ESCs. I really need to add airplanes and other stuff to my tracker. Here's the link to my tracker if you want to peruse, and also reminder that I don't make money from it. this is dope as gently caress, do you mind if I share it with others?
|
|
#
?
Jul 16, 2017 21:18
|
|
|
Be my guest. I am open to feature requests, like other vendors to scrape, email alerts, etc. Not sure when I will really get time to work on it though.
|
|
#
?
Jul 16, 2017 22:10
|
|
|
My Rhythmic Crotch posted:Be my guest. I am open to feature requests, like other vendors to scrape, email alerts, etc. Not sure when I will really get time to work on it though. Would it be much work to scrape the EU and UK warehouses too?
|
|
#
?
Jul 16, 2017 23:03
|
|
|
TTerrible posted:Would it be much work to scrape the EU and UK warehouses too?
|
|
#
?
Jul 16, 2017 23:05
|
|
|
It's possible, but will require a rewrite of the scraper portion  I will put it on my list though. At the very least I'd like to see if I can get that working without killing the performance. It already takes like 30-40 minutes to do one scrape.
|
|
#
?
Jul 16, 2017 23:10
|
|
|
whats the verdict on buying used batteries? there is a local guy selling a heap in my spec and I figure 50% of retail price is a fair start for something that hes 'flown once'
|
|
#
?
Jul 18, 2017 02:35
|
|
|
Laserface posted:whats the verdict on buying used batteries? Bring your charger, and do a resistance test on them. If they're 15-20 maybe stay away but if they're all 10-15ish then go for it. I have 8 tattu 650mah 75c 4s packs that are all 18-22ish and they still fly alright but come down very warm its just that around that number the cells are starting to lose their capacity I think.
|
|
#
?
Jul 18, 2017 02:43
|
|
|
Anyone got experience with higher capacity Tattus? Specifically 10.000mAh 4S. My Multistars sag over 1 volt on a 2C load (20A), which brings me towards my (self-imposed) 14v limit quicker than I would like.
|
|
#
?
Jul 18, 2017 03:27
|
|
|
bring back old gbs posted:Bring your charger, and do a resistance test on them. If they're 15-20 maybe stay away but if they're all 10-15ish then go for it.
|
|
#
?
Jul 18, 2017 12:14
|
|
|
FrSky have a new transmitter: http://www.multirotorguide.com/news/frsky-horus-x10-lighter-horus-x12s-or-advanced-q-x7/  X12 display, M9 gimbals and X7 case shape, basically. 330 euros, so I'm sticking to my taranis :-]
|
|
#
?
Jul 18, 2017 15:27
|
|
|
Oh, if anyone is considering using a Storm32 gimbal controller, just be aware that it spams enough noise to severely limit 433mhz UHF control signals and completely blank out 915mhz telemetry. Developer doesn't give a poo poo and isn't even going to put that little tidbit in the documentation.
|
|
#
?
Jul 19, 2017 05:29
|
|
|
ImplicitAssembler posted:Oh, if anyone is considering using a Storm32 gimbal controller, just be aware that it spams enough noise to severely limit 433mhz UHF control signals and completely blank out 915mhz telemetry. Developer doesn't give a poo poo and isn't even going to put that little tidbit in the documentation. In other news, am I crazy for wanting to put an inav board + miniquad motor/ESC in my first fixed wing build? I just want the autoleveling poo poo, nothing fancy, and I'd love to not have to stock a whole new set of powertrain components. That way I'd only buy the board, and use my current spare RX/ESC/motors.
|
|
#
?
Jul 19, 2017 10:49
|
|
|
Why is there not more frames that utilise a tilting camera? It seemingly makes sense to have a camera that tilts in relation to pitch but I would imagine it's only really feasible on larger frames and then they aren't as nimble as a 250. All the software is there in betaflight to run a servo, isn't it?
|
|
#
?
Jul 19, 2017 11:17
|
|
|
So Betaflight has no strict I-term windup limit. It only throttles I-term integration depending how much adjacent motor-specific throttle values are spread apart. If you do high throttle maneuvers without airmode enabled, the I-term may wind up and your quad gets hard to control until it comes down again. As a remedy, airmode gets now force-enabled beyond 50% throttle regardless you switched it on or not. From how I understand their code, with airmode, at high throttle, it gets reduced based on motor throttle spread, too. Which implies that airmode needs to give the system some headroom to allow the P-term some more authority to keep the quad from doing stupid stuff, while crossing fingers the I-term settles. Control systems in the whole drat world uses windup limits for a reason.
|
|
#
?
Jul 19, 2017 12:11
|
|
|
|
| # ? May 27, 2024 17:03 |
|
|
 Airbot's new F4 board does the same thing as bardy's board, with an added shell for physical protection (who gives a poo poo when it's in the middle of your build really) but also FFC connectors instead of soldered cables. Laserface posted:Why is there not more frames that utilise a tilting camera? It seemingly makes sense to have a camera that tilts in relation to pitch but I would imagine it's only really feasible on larger frames and then they aren't as nimble as a 250. Combat Pretzel posted:So Betaflight has no strict I-term windup limit. It only throttles I-term integration depending how much adjacent motor-specific throttle values are spread apart. If you do high throttle maneuvers without airmode enabled, the I-term may wind up and your quad gets hard to control until it comes down again. As a remedy, airmode gets now force-enabled beyond 50% throttle regardless you switched it on or not. From how I understand their code, with airmode, at high throttle, it gets reduced based on motor throttle spread, too. Which implies that airmode needs to give the system some headroom to allow the P-term some more authority to keep the quad from doing stupid stuff, while crossing fingers the I-term settles. evil_bunnY fucked around with this message at 12:46 on Jul 19, 2017 |
|
#
?
Jul 19, 2017 12:36
|
|