do you know if there’s a way to add a fourth uart on this board simply for a one wire current sensor that I will install on the Lipo lead? I have the finished bind and fly version of this quad and there is no current sensing which is kind of ridiculous on a long range quad

No there is not, even the 3rd soft serial UART was a pain to activate. It's connected to the Vista and has issues in that some people are experiencing OSD drop outs. I would recommend you go with the 4'' 20x20 frame that is offered by CNCMadness in Canada and use a better FC.

yes u r correct! osd wigs out (numbers go all over the screen) 2-3 times on a 10 min flight and i even had the osd show "disarmed" several times and then restarts the flight timer and once the caddx completely rebooted during flight, that was scary

Wait - the Caddx actually rebooted? Now THAT IS SCARY!! I hope it won't happen to me - if it does I may get myself a 20x20 adapter and swap out the FC altogether.

yes! i never lost video but it went thru the whole boot up process that puts it in low power mode then switches frame rates etc and restarted timer again, luckily i was coming in and was only a couple hundred yards away, i was freaking out waiting to hit rescue mode for the first time. The explorer frame has 20x20 mount holes too so i started looking for another possible 20x20 fc with 4 uarts ;)

i have 7 dji quads, 2 air units and 5 caddx, 4 are prebuilt and 3 i built and in hours and hours of flights ive never had a reboot

until this one ugh!

oh i forgot to mention i was on the 1200mw hack and flying slow so that time i think the caddx overheated so i lowered to 1000, this is the only quad ive tried on 1200

I had a thought you let me know what you think if you could help me out or not I noticed the OSD on this goes a lot more wacky when I upped the pid loop to 4K so I put it back down to 2K and then I was considering overclocking the processor normally overclocking doesn’t help much unless I’m close to 50% CPU load and when at 2K Pit Loop, I’m only around 20% but when I was at 4K I was at approximately 44% and having crazy bad OSD issues I also figured out when the OSD is going crazy land the drone reboot just the goggles and it seems to hold a lot better. I also found an article from flywoo explaining the problem with this and obviously it is because of the soft serial uart and they claim they are working with caddx for a software fix to help the situation so let’s hope. Let me know what you think about overclocking the CPU processor I know there are three variables on processor speed to overclock and I would only do the lowest one

i think instead of overclocking im just going to turn black box logging off, that will save some cpu usage

That's not a bad idea. Black box is helpful but not if you keep losing OSD feed ;-) Some guy I'm chatting with on the JB forum also complains about difficulties acquiring GPS, so this build definitely has its problems. He's working on an extended mount that contains a copper or aluminum plate for insulation. Please let me know how you fare. I won't have much time to test my quad before my vacation next Wednesday and it would be good to get this sorted as I definitely plan to do some LR cruising in the Pyrenees.

I’m glad we’re talking I think we’re helping each other out a lot I have noticed that the CPU is totally being maxed out I’ve been going to the CLI tab and typing in "tasks" and if you do that you’ll notice that serial is taking the majority of the CPU usage so I’m doing whatever I can to lower total usage RPM filtering takes a fair amount but I don’t want to give that up 2k seems to be helping on Pid loop unfortunately 4.2 beta flight forces 8K on the Gyro loop I really wish I could take it to 4K in 4K but it won’t let me I did a flight last night for nine minutes and the OSD freaked out on me within the first 30 seconds and then stayed stable the entire flight which is definitely an improvement because when I had it on 4K pid loop and black box it would freak out about every minute or two so I think we are onto something that might be a possible partial fix

And I think flywoo knew it was CPU usage because I’ve never seen any firmware presets with a pid loop at 2K so on their Facebook page when they say they’re trying to have caddx with a solution is just a bogus excuse because the main issue is CPU overload at least that’s my best guess right now

I have no idea why a triple posted my last comment

As far as acquiring GPS the best thing to do is plug in the power pack into the USB port have GPS telemetry through your crossfire on your radio wait for the coordinates to pop up and then fly it does take about 3 to 5 minutes which is a long time the other cool thing is if you get the GPS set up on telemetry on your radio if you crash lose signal battery ejects or whatever it will save the last known GPS coordinates on the screen of your radio type those into your phone on Maps and walk straight to your quad. The second flight seems to get GPS signal considerably faster. Or if someone is not using crossfire or doesn’t want to do it that way plug into the laptop or the Speedy Bee app and click on the satellite tab and wait for the satellites to populate

Okay so I talked to Joshua Bardwell about this and he suggests dropping down to 2k - that was before I was reading all of your new comments. Based on what you're explaining dropping BB to 2k helps but you still experienced a drop out. In that case I think the only remaining option would be turning off BB altogether.

oh and the best part I no longer get that goofy OSD message saying "disarmed" that one scares the crap out of me it never really disarmed but it showed that on the screen and I have not had any air unit reboots either

actually what I’m talking about is dropping the pid loop to 2K and turning off black box all together 2K is the default on PID loop but I had changed it to 4K so now I’m back down to 2K with no blackbox logging at all I also considered overclocking the processor but I have not done that yet

Also It is natural that the SERIAL task shows unreasonable values, as CLI output blocks the task and force it to remain in the run state for prolonged period of time.

yes I’ve noticed that weird percentage that’s in the thousands but I just keep hitting the task command and every once in a while it will show a realistic number and make sure you have your radio on when you type in the task command that seems to help also to get a semi-accurate reading

do you have any luck or suggestions or advice on overclocking the CPU if I did that I would only do it 192

NEVER tried that so cannot help on that end, sorry.

And yes - I actually ordered an external USB battery the other day just for the purpose of priming the GPS ahead of the flight.

keep me up-to-date on any progress on your end and I will do the same, Oscar did a review on this quad and never had any issues with the OSD but he kept the pid loop at 4k and he did not use RPM filtering

I found a stack that could be a good replacement it’s a 16 x 16 F7 processor 32 bit ESC with current sensing two full uarts and a VTX uart that you could possibly use the camera pad for the other half of the uart if not you could use the LED pad for a serial uart

look up Tiny TALON Stack 16x16 F7 Stack & BLHELI_32 ESC

just a thought in case we can’t get this current stack working the best plus the Talon has current sensing and an F7 processor so it should work perfect

One quick question why did you and Flywoo both use the soft serial uart for the Caddx Vista instead of the GPS on the soft serial? I would think Crossfire and video would need the most stable uart connection and then GPS fall back onto the soft serial uart

I talked to several experienced people on JB's private forum about that and the consensus was that GPS had priority to the OSD. If you don't have GPS or if it gets interrupted RTH won't work.

exactly I had that thought after I posted this I was worried that the home point would be deleted

check out that stack (talon tiny stack) I mentioned in the previous post I even called a guy who helped design it and I think it will work perfect if the current stack in this set up doesn’t work out for us especially since it is an F7 chip and a huge plus is current sensing

I checked out that stack and it looks great. However I don't see a pad for the buzzer, did I miss that? Buzzers are an absolute must for anything LR.

im glad u mentioned that ill have to check more into it

crap ur right i didnt notice that i have used this VIFLY Beacon Wireless Self-Powered Drone Buzzer and it works great any sudden g force it beeps loud! and had a super bright led but u cant trigger it with radio or get low voltage alerts its also cool cuz u just zip tie it anywhere so now im not sure im stoll getting osd mess ups 2-3 per flight and today im trying the overclocking at 108mhz

well overclocking didn’t help I did notice that when I did some freestyle moves it seem to freak out more on the OSD so I bet you rpm filtering has most of the CPU usage and causing the OSD to mess up either way I’m not gonna worry about it anymore I’m just gonna fly bunch I did go for 4250 feet away from home and it did fine OSD messed up a couple times but not too bad and not for very long

I did a lot of testing with this and I found that Betaflight 4.1.1 doesn’t glitch the OSD at all.

For real? You should report this to the BF team. I wonder why that would be the case.

Did you try to turn off bidirectional d-shot in BF 4.2?

i gave up and scraped the stack, went to 20x20 with full uarts

Yes I reported it to betaflight but they told me that softserial is not supposed to be used with MSP so they will not try to fix it unless we can finds the code that broke it.

I tried 4.2 bare bones and still had the glitching.

Can you guys check 4.1.1 to see if the problem goes away? I'm using the BetaFPV V3 board with MSP over softserial as well. I want to know if it's not board-specific.

I've got a busy week ahead of me and won't have time to swap BF on my quad. Also would be worried about screwing up the settings as I'm going on a trip soon. But I will pass this on over on the DaveC FPV Facebook group. Perhaps some people can give this a shit and report back.

u did it!!! Flywoo is posting about the osd fix with 4.1.1 on their IG page

Heheheee - I told Joshua Bardwell about it. Also posted it in the DaveC group. He must have pulled some strings. I'll post your name so that you get the credit.

this hobby is so tuff when things are unknown but great when people pull together to figure stuff out. kudos to u both

I'm posting your name everywhere so that people know it was You and not them who spent the time figuring this out. Was just about to email Ally over at Flywoo to give you credit - at minimum they should send you a free quad or something!

I would’ve tried 4.1.1 but about a week ago i had a vibration and I noticed my stack was loose and when I went to tighten the stack bolts (one full turn) the gummies squished down too much and grounded out the flight controller on the lipo solder pad so it fried the flight controller and this is why I ultimately gave up on the flight controller and had to buy a new one and switch to the 20 x 20 stack. Even when I first received it almost every nut and bolt on the whole quad was loose. I normally build everything I fly but I was eager to get this one quickly since it was already set up and ready to go

So how did you accomodate the 20x20 on that frame? Mine from CNCMadness would not accomodate it.

well im still waiting on the stack its delayed in transit but my frame has both 16x16 and 20x20 mounting holes and i have the same 20x20 stack in another build so i have double checked all the measurements so im fairly certain it should fit fine. i have a f7 dual gyro iflight board coming