(04-Nov-2023, 08:32 PM)Suros Wrote: I usually just go with 10ish volts over the max charge of what the battery can provide, then just choose a capacity based on physical dimensions that look sane. It's not very scientific, but my power has always been pretty darn clean. Helps to get them from a reputable brand too. Panasonic tends to be good. Just make sure they're proper low-ESR capacitors so they can effectively soak that voltage ripple.

These are the ones I have (still unpacked):
https://rotorvillage.ca/voltage-capacito...ow-esr-1pc

..and these are on their way (love Rubycons, they are 1000uF, but smaller than Panasonic, much easier to fit them inside the frame):
https://rotorvillage.ca/rubycon-capacito...-esr-10pc/

This is the test I did earlier today. I flew little more aggressively just to see what would happen. I managed to fly little over 3 minutes, but I still had to ignore Low Battery warnings. It was showing me 3.65 per cell at the end, but when I measured everything with my battery checker, it was 22.2V total, 3.71V-3.75V per cell. So I landed earlier than usual, I usually land at 3.6V per cell with my other quads. It looks like even after the proper Voltage Meter Scale calibration in Betaflight (settled at 113 at the end, and it was showing me in BF what I saw on my battery checker attached to the balance lead), it still looks like 1V is off in my OSD.

What could be the reason for that and how do I resolve this? I mean, I could keep ignoring these warnings, but it's annoying and it does not look right, especially because I know those warnings are lying to me. There is just enough juice left to finish my flights at 3.5, maybe even 4 minutes as I'm used to... but these warnings are screaming at me like battery is definitely dying on me, and like I'm abusing it...

Video:

Perhaps the hardware involved in reading your voltage is defective. Seems like your flights are going as expected, which suggests that the reading is incorrect, or every battery you stick on here sags terribly at just 20 amps. A serious amp draw that could sag you down that far would probably start a fire, so I'm doubting it's that.

Honestly, when I made my first post on this, it didn't even strike me as odd to see 4 volts right away as my only 6S model uses Li-Ion cells which immediately sink to 4.05 a cell at full charge. Just having the VTX powered is all it takes to make that happen.

For the voltage sensor calibration, yes I was referring to adjusting the BF settings. However you should take at least two different voltage points to check, usually I connect a fully charged and a fully discharged battery and sometimes you need to play with the divider setting as well. However you can also set the LOW/CRITICAL voltage levels in BF so you can move them lower for that setup if you do not want the warnings appearing so soon.

Not sure if you mentioned the specs of the battery you are using. But if the capacity or C rating is too low, then you will get the sag that you are seeing. Also you mentioned flying in high winds, that would draw more power from the battery as well. If your other setup has nearly the same AUW and motors, I suggest you fly them both back to back in the same conditions, then you have an accurate comparison to see whether or not there is an issue of excess current draw.

Somebody mentioned the C rating on the battery.  45 amps in your speed test might be taxing that battery too much.  The voltage sag might then be taking too long to recover.  It has recovered by the time you end the flight and take measurements.

I think I am observing this same thing with a couple 1s tiny whoops that draw 11 amps even in quiet level flight.

(05-Nov-2023, 02:56 PM)mstc Wrote: For the voltage sensor calibration, yes I was referring to adjusting the BF settings. However you should take at least two different voltage points to check, usually I connect a fully charged and a fully discharged battery and sometimes you need to play with the divider setting as well. However you can also set the LOW/CRITICAL voltage levels in BF so you can move them lower for that setup if you do not want the warnings appearing so soon.

Not sure if you mentioned the specs of the battery you are using. But if the capacity or C rating is too low, then you will get the sag that you are seeing. Also you mentioned flying in high winds, that would draw more power from the battery as well. If your other setup has nearly the same AUW and motors, I suggest you fly them both back to back in the same conditions, then you have an accurate comparison to see whether or not there is an issue of excess current draw.

Thank you! I will try to answer all the questions you've lined up...

Quote:Not sure if you mentioned the specs of the battery you are using. But if the capacity or C rating is too low, then you will get the sag that you are seeing.

This one was GAONENG GNB 1350 mAh100c/120c 6s. The thing is I fly with this battery with all my other quads and have absolutely no problems. One video was posted earlier, here is the link again:


...and that's with SpeedyBee F405 v3 stack inside. That was my 3rd custom build quad. This one I am having trouble at the moment, is my 5th built and it's using T-Motor Velox V50A, and Rush Blade F722 HD as my stack. If you can, please take a look at that video, I literally tried pulling its soul out, and had absolutely zero problems. Exactly the same motors, Xing-E Pro 2306 1700KV.
I have no idea where all these problems with T-Motor Velox V50A are coming. Could it be that my capacitor is dying/dead? 

Quote:Also you mentioned flying in high winds, that would draw more power from the battery as well.

Yes, that's usually the case where I live, but not every single day. That one video where I mentioned wind, I was flying while having 55km/h winds with wind gusts up to 70km/h (my local app that tells me details). But this last video I posted (TEST#2), I had zero wind while flying (rare, but happens around here). And yes, totally understand your point, winds, especially crazy like we get around here, would make motors work much harder, totally understandable. But I was flying in high winds before (tons of videos up on my channel, I am still a beginner by the way, trying to learn as much as possible), and I am super, super happy with how my other quads are flying.

Quote:If your other setup has nearly the same AUW and motors, I suggest you fly them both back to back in the same conditions, then you have an accurate comparison to see whether or not there is an issue of excess current draw.

Those are my exact observations. I've tried two other quads with similar or almost the same configuration, except one is F405 (GepRC MK5 with SpeedyBee F405 v3, and 2306 1700KV motors) and another one is F7 (QAV-S frame with Diatone Mamba inside, Xing-E Pro 2208 1800KV), using same batteries, and both fly with absolutely zero issues. I know, why I'm not flying those instead... I do, I enjoy them more than anything (have two more custom ones as well, and they both fly incredibly good, which can be seen from my other videos, I am not the best pilot, but these machines I've built are my babies and they perform amazing). But this last one, while it seems to fly superb, this is itching me so much, I can't figure out where this juice draw is coming from, and why am I dealing with this now. I want to solve it so bad, and have no idea where to look anymore.

Thank you so much for taking your precious time and trying to help me. It's greatly appreciated.

(05-Nov-2023, 05:27 PM)segler999 Wrote: Somebody mentioned the C rating on the battery.  45 amps in your speed test might be taxing that battery too much.  The voltage sag might then be taking too long to recover.  It has recovered by the time you end the flight and take measurements.

I think I am observing this same thing with a couple 1s tiny whoops that draw 11 amps even in quiet level flight.

I just replied to mstc above, please have a look. I have pretty much the same configurations, other 4 custom built quads, same batteries being used over and over again, and have zero issues in the same wind conditions. Last video link is me flying my 3rd custom built quad, and I posted it only because there are telemetry information presented on the screen, so you can watch what's happening with battery. It works perfectly fine, I pulled its soul out couple of times, and I see not significant sag at all. And that was with F405 in that quad... this last, 5th custom built, is with Velox V50A and Rush Blade F722 HD, so F7... and I'm getting nothing but trouble. OK, so those two are different, F405 vs F7... but I do have another quad with F7 inside and flying it with these same batteries with zero issues.

That's my very 1st custom built ever, and it flies like a rocket:


I really have no idea where to look anymore, other than to throw this Velox V50A in trashcan, and replace the whole stack with another SpeedyBee F405, v4 this time (local shop has amazing deal at the moment). That would be the easiest to do... but this itch what is going on and what is that that's drawing extra power, is killing me... 

Even though motors are rated with the same size/kv, there could be quite a large difference in performance. If you are using different motors on the two builds, it could very well account for some of the difference. The tune, FC, and ESC settings can also make a big difference. I am not familiar with these larger motors, but for smaller motors if you run 24khz vs 48khz vs 96khz can make quite a bit of difference on efficiency. Some motors might have stronger magnets and can suck more current faster, if you have a very tight tune or noisy gyro/setup or like a windy condition where quad is constantly being knocked, the FC could be excessively and unecessarily ramping up/down the motors. Maybe it is still flyable, but a lot of extra energy is wasted. Just some thoughts.

(05-Nov-2023, 06:51 PM)mstc Wrote: Even though motors are rated with the same size/kv, there could be quite a large difference in performance. If you are using different motors on the two builds, it could very well account for some of the difference. The tune, FC, and ESC settings can also make a big difference. I am not familiar with these larger motors, but for smaller motors if you run 24khz vs 48khz vs 96khz can make quite a bit of difference on efficiency. Some motors might have stronger magnets and can suck more current faster, if you have a very tight tune or noisy gyro/setup or like a windy condition where quad is constantly being knocked, the FC could be excessively and unecessarily ramping up/down the motors. Maybe it is still flyable, but a lot of extra energy is wasted. Just some thoughts.

Understandable. I do have identical motors on those two I was testing before (Xing-E Pro 2306 1700KV), but the only major difference is, one runs on F405, and another one on F722 (if that maybe affects things like power draw, etc.). For example, these couple of photos I will attach here, are from this 5th build that I'm having trouble with that weird power draw behavior, called MOPAX (that's how frame is called, designed by one of my local FPV shop owners who designed couple of other frames as well) with T-Motor Velox V50A/RushBlade F722 HD stack, and another one is QAV-s JohnnyFPV SE frame with Diatone Mamba F722 MK2 55A stack. I am comparing these two only because they both run F722, but they do have different motors (one is with 2306 1700KV, and the other one is 2208 1800KV)





...and this is how much they weight with the same battery, one is wearing RunCam Orange5 and another one Session5..





...and these are the screenshots from BLHeli_32

MOPAX with T-Motor Velox V50A/RushBlade F722 HD stack (PWM Frequency running at 48kHz)...



QAV-s JohnnyFPV SE frame with Diatone Mamba F722 MK2 55A stack (PWM Frequency running at 24kHz)...


I do have one more question... regarding the capacitor. Right now, one I have attached to the minus/plus pads for XT60 on the ESC, is 1000uF 35V LOW ESR. Even though it might not change anything, I've been planning to replace that one with brand new one, just to be sure my capacitor is in good condition. The problem is, I do have couple of brand new 35V ones, but they all are 470uF. Can I replace that one that's 1000uF with this, and would that cause any additional problems, or should I wait until the new ones (1000uF 35V) finally arrive in mail in couple of days? Just want to understand this little better, that difference between 1000uF and 470uF. For Voltage, I pretty much understand that it needs to be approximately 10V above, so 35V for my 6s setup should be OK if I am not mistaken.

Usually the right way to pick a capacitor is basing it on the size of your ripples. A smaller capacitor is actually better if you have smaller ripples as they tend to be able to adjust their level of charge more rapidly. That of course doesn't matter though if you get a ripple that's too big and ends up overwhelming the capacitor.

A little trick to get around the issue if a 470 doesn't cut it. Just add two in parallel instead of one. If your electronics aren't happy and you see wild stuff happening in your gyro data, just pop another on. I use motor wire to extend the legs and strap them to the back arms to keep the back opened up. Lots of small quads have no junk in the trunk.

(05-Nov-2023, 08:34 PM)Suros Wrote: Usually the right way to pick a capacitor is basing it on the size of your ripples. A smaller capacitor is actually better if you have smaller ripples as they tend to be able to adjust their level of charge more rapidly. That of course doesn't matter though if you get a ripple that's too big and ends up overwhelming the capacitor.

A little trick to get around the issue if a 470 doesn't cut it. Just add two in parallel instead of one. If your electronics aren't happy and you see wild stuff happening in your gyro data, just pop another on. I use motor wire to extend the legs and strap them to the back arms to keep the back opened up. Lots of small quads have no junk in the trunk.

I have one custom quad where I extended capacitor legs, and the wires I used are my motor wire leftovers, but they are little over 6 centimeters long (3.5"). I've read in many places that's not good (I haven't noticed any problems so far, but that does not mean problems don't exist only because I don't see them), and that the best thing to do is to place your capacitors closest to the ESC board with the shortest legs possible. Of course that's not always possible, I have that in one of my quads where Dji AIR Unit is sitting in the back, and you're guessing right... no space at all for anything, not even a grass blade, let alone 1000uF little piggy capacitor. So, I've extended it's legs, but planning to change that setup in the next couple of days. I've actually cancelled my Rubycon capacitors order, and place another one, and it will be delivered tomorrow (Amazon Prime). These are the ones I will be using in my quads now:
https://www.amazon.ca/dp/B073YR6LVK?psc=...ct_details

Little update... after extensive testing (still couldn't figure out how to fix it), I've come to conclusion that something... ESC, FC, BF is "lying" to me about the battery Voltage. Every single time I land, it's when OSD is alarming me Voltage is 3.6V, or even 3.5V since I am sure now, it's false information. I land, and immediately I've checked my battery levels with my little battery checker, as well as with both my chargers which are properly calibrated and showing me same values as that little battery checker.

BF is alarming me my end Voltage is 21.49V (approximately 3.58V per cell), and all three battery checkers are reading 22.6V (3.76V per cell). That is rather significant difference when it comes to these anyway short flights. If BF told me it's 3.7V per cell, I wouldn't panic nor even think to land until I hit steady 3.6V, maybe even 3.55V. Once battery recovers, it brings it back very close to storage charge. And I would get at least half a minute, maybe even a minute of flying (judging by all the other quads, and the difference between this reported end battery Voltage and what it really is).

Now, I know I can disable alarms, or even bring them down to lower values, so they don't scream at me while I'm enjoying flying... but I was thinking to rather do some kind of calibration and manually set it to some number that would be like the difference between the reported end Voltage and the real one. For example, when BF thinks it's hitting 3.6V, is it possible to make it read on screen as 3.7V somehow? I know this might sound silly, but I don't think I'll be able to fix this any other way. Quad flies like a rocket, stable motors sound, no oscillations at all (all my videos are always without a single second of stabilization, no hardware one, nor software one. This helps me iron out my flying skills, stick resolution feel, etc. The only thing I do in post, is color correction and lens distortion fix.

So, I'm thinking, if I have to fake the values to the ones I am 110% sure they are correct, and have my OSD (on this quad only of course) show me correct values, even though ESC/FC for some reason still think they are lower, I'd do that in a heartbeat. Something is definitely off. Batteries are not getting drained or destroyed, it's a simple miscalculated Voltage reading.

Here is the link with my 3rd test, wanted to fly little slower and smoother. At the end, I showed true battery Voltage (total and cells), right after I landed:



I was testing it with GNB 1350mAh 100c/120c 6s battery, of course.

It seems like you're using a 4in1 ESC that supports telemetry. Have you tried using the voltage reading from there instead of getting it from the flight controller?

(07-Nov-2023, 11:49 AM)Suros Wrote: It seems like you're using a 4in1 ESC that supports telemetry. Have you tried using the voltage reading from there instead of getting it from the flight controller?

Nope. To tell the truth, regardless of how embarrassing it might sound, I have never heard that's even possible. How do I do that? I'm all ears my friend! 

EDIT: Not sure if this version of T-Motor Velox V50A supports it (there are few other, improved versions released after that one), but this is the exact one I have:
https://rotorvillage.ca/t-motor-velox-v5...30x30-esc/

From the photos of the ESC, it looks like only onboard ADC is supported. So ESC telemetry will not give you voltage or current.