Recently I've been messing around with eCalc (https://www.ecalc.ch/xcoptercalc.php) to check some build ideas. I didn't know where to post this, but since my primary goal is to optimize efficiency I figured I'd post it here. 

I first ran some calculations using my Cinerat, which I was trying to convert to a LR quad. Putting in the specs as-is, I get the following result which seems reasonable given my experience:
   

I know this looks complicated at first, so I'll walk through it to the best of my understanding. Please let me know if you spot any mistakes!

1. First you specify the model weight. You can either specify the AUW, the weight without battery, or the weight without the entire drive system (motors, props, ESCs, battery). For my purposes, I specify the weight w/o battery so I can play around with various battery options (ie LiPo vs LiIon) and it will update the AUW automatically. Note that if you specify the weight without drive, the way it calculates the AUW is a little unintuitive. For a quad it's: 4 x motor weight + 4 x ESC weight (it assumes individual ESCs) + battery weight x cells (again, it assumes you're entering the weight of an individual cell, not the whole pack), and then it adds 10% on top of this for things like props. For a LR build, this is likely way too high, so I prefer either specifying the AUW or the dry weight for better accuracy.

2. The frame size is only taken into account to make sure your propellers won't hit each other and in the drag calculation later on for range estimation. 

3. Next you enter the battery specs. Here I'm using my GNB 6s 1100mAh pack. Note that the weight is per cell, so for a 6s pack you divide the total weight by 6. It has a drop down selection for a range of generic batteries and Li-Ion cells. Again, for a LR build you'll probably be using LiPo's that are much lighter than the defaults here, but having the quick Li-Ion selections are handy.
   

4. Here you specify the ESC parameters. Note that it assumes individual ESCs, so take that into account if you're using a 4 in 1. I just make sure the current limits are correct, and set the weight equal to my stack weight / 4.

5. Even though the other presets seem quite out of date, the motor drop down contains a lot of the newer efficient motors you'd use in a LR build. You can also define a custom motor if you have the specs. I couldn't find the specs for the Ethix Flat Rats on my Cinerat, so I'm using the closest I can find which are the GEPRC 1507 2800kv. 

6. Here you define the prop. Most manufacturers are represented in the presets, and they change the properties of the props slightly (I assume due to different airfoils/blade shapes/thicknesses etc). For example, the Gemfan props seem to be slightly more efficient than the HQ props but produce slightly less thrust. 

7. Here you can define the weight and current draw of any other accessories on your build. I've entered the approximate amp draw of the Caddx Vista, but since the weight is already accounted for in the dry weight I left that at zero. 

Then you hit calculate and the magic happens. It will give you estimated current draws at max power, max efficiency, and hover, along with things like estimate range, rate of climb, thrust to weight ratio, throttle at hover, etc. It will also graph range vs speed estimates and motor characteristics at full throttle. 
   
   


Results
For my Cinerat, it predicts a mixed flight time of 4.1 mins and a hover time of 7.9 mins. My flight times were just a touch over 6 mins with some easy cruising, so this is in line with reality. I also have some 3x4x3 props so I reran the calculations for those as well, along with seeing how removing the prop guards would affect things. 

Configuration - Mixed Flight Time - Hover Time - Power:Weight - Specific Thrust (at hover)
3x3x3 with prop guards - 4.1 mins - 7.9 mins - 3.8:1 - 2.88 g/W
3x4x3 with prop guards - 3.9 mins - 8.1 mins - 4.3:1 - 2.96 g/W
3x3x3 w/o prop guards - 4.4 mins - 9.3 mins - 4.3:1 - 3.03 g/W
3x4x3 w/o prop guards - 4.2 mins - 9.5 mins - 4.8:1 - 3.11 g/W

So the higher pitch props produce more thrust and allow you to hover at a lower throttle setting, but are less efficient and result in a slightly reduced mixed flight time. 

Next, I wanted to see if my idea of dropping the prop guards and using a 4s 3000mAh Li-Ion pack would give me the flight times I was looking for. Short answer: No. The 1507 2800kv motors produce plenty of thrust but are not particularly efficient, especially when restricted to 3" props by the frame. I'll post up that analysis in a bit, along with an analysis of some of the LR 4" builds posted here to see how accurate eCalc is when it comes to max efficiency builds. After that, I'll post my theoretical calculations for a ~100g dry weight 2.5" build with a DJI digital system and a predicted flight time of 20 mins!

Ecalc is a great tool, and reasonably accurate.

Next I wanted to see what would happen if I tried to use a 3000 mAh 4s Li-Ion battery instead of the 6s LiPo. Here are the results:

Configuration - Mixed Flight Time - Hover Time - Power:Weight - Specific Thrust (at hover)
3x3x3 with prop guards - 9.1 mins - 11.3 mins - 1.4:1 - 2.72 g/W
3x4x3 with prop guards - 9.0 mins - 11.6 mins - 1.6:1 - 2.79 g/W
3x3x3 w/o prop guards - 10.1 mins - 13.1 mins - 1.6:1 - 2.84 g/W
3x4x3 w/o prop guards - 9.9 mins - 13.5 mins - 1.8:1 - 2.92 g/W

4s just isn't enough voltage to turn the 1507 2800kv motors efficiently, and it takes 60-70% throttle just to hover. Power to weight ratios are also quite low, which means performance would be severely lacking. 3000 mAh would only provide an additional 4 mins of flight time at a significant hit in performance.

   
   

So what if I change the motors? The Flywoo NIN 1404-4850kv's were the best option I was able to find so far for a heavy 3" quad like this. On 4s Li-ion's, with a 3020 biblade (for efficiency), the results look like this:
3x2x2 w/o prop guards - 10.1 mins - 15.6 mins - 2.2:1 - 3.17 g/W

Slightly better, and at least it would be decently flyable, but not worth spending $60 on new motors. So, I think this idea is a non-starter. I'd need a much lighter frame and/or larger props to get a decent flying LR rig.

In general, for flight time, assuming you have "decent quality" components
tune/try in order of having greatest effect :

1) Prop size
2) Prop pitch/ #blades
3) reduce AUW
4) motors

Example1
2" props will never get anywhere near the efficiency of 4" or bigger. Doesn't matter what motor you try or PID tune. Its just physics.

Example2
A steep 3052*3 prop will suck battery fast regardless of motor, ESC, or anything else. Physics.

Example3.
Motor + prop have a efficiency peak curve. Typically 20-40% throttle. If you extremely reduce weight, an ultralight, you can be left of this curve, and get worse flight time. On the curve, and hover vs cruising will see big diff in flight time. Far to the right - like 90% of quads - pretty much anything over 150g AUW on 3". And doesn't matter if you are hovering or cruising.. only get a few minutes.

Yes, this is a good summary and has all rapidly become apparent when trying to work out my build. My goal is a micro quad with 10km range that can carry a DJI digital system and is equipped with GPS.
I was originally aiming for a 2.5"/65mm build, but as romangpro said, efficiency is hugely dependent on prop size. There is a huge drop in efficiency going from 3" to 2.5" props, and a huge jump going from 3" to 4" props. A 4" long range build is trivial at this point, and I think 3" is obtainable. Going down to 2.5" might be possible, but with a significant reduction in performance.

I have more analysis to post tonight, but here are the few remaining things I'm trying to figure out:

1. What sort of power to weight ratio should I be targeting? 2:1 seems like it would be flyable, but very boring and not very agile. 3:1 would be better, and 4:1 seems like it would be ideal for a smooth cruiser. Do these numbers sound right or am I way off?

2. One of the main reasons for going small is stealth. What factors affect the noise level? I'm guessing disc loading is the primary factor, with RPM and prop pitch (both related) impacting noise as well. Can someone point me to some data or references on this? Anything I'm missing?

3. According to eCalc, the most efficient motors by far for my planned 3" build are the SunnySky R1104-5500kv.
https://sunnyskyusa.com/products/r1104
Does this sound right? I'm skeptical such small motors will work well on a 3" quad with a 200+g AUW, but that's what eCalc is telling me. Also, I don't see SunnySky motors used often, and the few reviews I can find for their 1106 motors aren't positive. Are they a reputable brand? Has anyone used the R1104 motors?

10km with DJI on a 2-3 inch?  That would be allstar performance there.

When you say 10km....thats one way right with a return to home?

Sorry, I mean 10km total distance covered (ie going for 5k badge). Although you could probably do 10km one way... once.

Has anyone done a back of the envelope calculation on what is possible with current battery technology in a sub 250g drone?

After realizing the Cinerat wouldn't cut it as a long range quad, I wanted to see what was possible at the 3" (or less) scale.

Since everything is going to be a function of weight, I started putting together a build list for a test quad so I could calculate an AUW. Here's what I came up with:


I ended up setting my dry weight to 111g in eCalc to add a little bit of cushion. I also wanted to explore some different motor options, but didn't like how eCalc adds 10% to the calculated drive weight when you specify a weight w/o drive, as it made things more difficult to compare between motors. 

So with the dry weight and prop diameter fixed, I went about playing with the various options in eCalc to try to find the optimum combination I was looking for. 
Here are the results, with several different motor/prop/battery combinations, sorted in order of max range.

Key:
Thrust Ratio - Thrust to weight ratio at max throttle. Note that due to the current limitation of the Li-Ion batteries, this may not always be attainable. It gives a good idea of what the throttle % required to hover and the maneuverability/performance in flight. 
Hover Time (min) - Expected flight time based on all-up weight when hovering only.
Flight Time (min) - Expected flight time based on all-up weight when moving (current draw halfway between hover and max).
Range (m) - Max range considering standard drag.

Here is a link to eCalc with the "winning" combination if you want to play around and see if you can find something better: https://www.ecalc.ch/xcoptercalc.php?eca...sLongRange

Takeaways:
You definitely get better efficiency with a bi-blade prop over a tri-blade. Also, I was surprised to see that higher pitch props seem to give you better efficiency.

So with a 3" 111g dry weight quad and a 3s 3000mAh Li-Ion pack, a 10km round trip flight or 27 mins of airtime should be possible. If you wring all 3000mAh out of the batteries, those numbers go up to 11574 m range and 30.7 mins of hover time! Performance seems reasonable as well, with a thrust-to-weight ratio of ~3 with the heavy LiIon pack, and ~5 with the big LiPo.

Thoughts? Am I missing anything? Or do I just build it and see what happens?

On my 3inch mid range tadpole build I found one issue that is a challenge. You need to fly at a high altitude for LR and the wind is a real problem. Smaller props fare poorly in moderate to high winds and you will loose your efficiency (aka batt capacity) quickly. Just my arguably uninformed two cents.

---

Oh and build it to see what happens! that’s what I did and it’s been a real insight.

(04-Jan-2021, 11:53 AM)ph2t Wrote: On my 3inch mid range tadpole build I found one issue that is a challenge. You need to fly at a high altitude for LR and the wind is a real problem. Smaller props fare poorly in moderate to high winds and you will loose your efficiency (aka batt capacity) quickly. Just my arguably uninformed two cents.

---

Oh and build it to see what happens! that’s what I did and it’s been a real insight.

Good point, lightweight and wind don't go well together. I should probably add a column for max speed to my spreadsheet, and use that as a proxy for how well it will be able to handle the wind.

I wonder if the higher density of a 250g 3" will help it in the wind compared to a 250g 4" (like an Explorer) or 250g 4.7" (Mavic Mini 2)?

I have most of the parts either in hand or ordered, the last thing to decide on is the FC/ESC. I was having trouble finding a lightweight 3" frame that would fit a 19x19 camera, so I wasn't sure what my mounting options would be.

I ended up custom ordering a frame, so I have plenty of options for mounting (16x16, 20x20, 25.5x25.5). I like how low profile the Whoop AIO boards are, but the Goku 16x16 stack is a few grams lighter. Any thoughts or suggestions?

Hi V-22,

Be sure to post a BUILD thread.

It will be interesting to see how this goes. My builds always end up a little heavier than I calculated. 

From my experience, larger props are more efficient and more powerful. 
A 4 inch handles the wind better than a 3 inch; I believe the thrust of the larger props combats wind better than weight or density.

I will be curious to see how closely your actual build performs compared to the eCalc estimates. 




Also, take a look at the actual data from my Phantom-X {see attached graphic}   

The 2s 18650 3000 mAh pack out performed the 3s 18650 3000 mAh pack.

I have been gathering info for my build thread and will get that up soon!

I'm looking forward to seeing how accurate eCalc is as well. It seems reasonably close with the numbers for my Cinerat (even though I didn't have the exact motor specs), and I did a quick analysis of your 4" LR build (I will post in that thread) that gives numbers pretty close to your reported flight times.

Unfortunately I can't find any specs for that FPVCycle 1303 motor, or the new Lumenier motors above. I have reached out to some of the companies to see if they can provide specs but haven't heard back yet. The critical parameters for efficiency seem to be motor resistance and Kv. Motor resistance seems to be the primary determinant of efficiency for these long range builds, as it determines how much power is wasted to heat (P ∝ I²R). The optimal Kv depends on voltage (2S, 3S, 4S, etc) and prop size/pitch. I just bought a new benchtop DC power supply so I can do some milliohm resolution measurements of these motors should I not be able to get the data otherwise. Unfortunately shipping is still taking forever right now, so the project timeline is at the mercy of the USPS 

Hi V-22,

The various manufacturers are cranking out a wide assortment of new motors so fast that it is hard to keep up. In fact, I am willing to guess that they don't even know {or bother to care enough} how they actually perform.  I just watched a Drone Mesh YouTube video and he mentioned exactly that. In my opinion, the new FAA regulations have just cranked up the heat in the sub 250 gram market...so... there will likely still need to be quite a bit of actual trial and error testing to be done by builder/pilots.
I am actually considering a 1402 motor as an alternate for the 1303; just a thought 

Since each 18650 3000 mAh cell weighs roughly 46-48 grams, anything over 2S 18650 is going be pretty heavy, that doesn't leave much room for the quad's dry weight and allow motors big enough to pull it. 

@V-22, You are having way to much fun with that eCalc 

(07-Jan-2021, 10:22 PM)iFly4rotors Wrote: I am actually considering a 1402 motor as an alternate for the 1303; just a thought 

Depending on where you are in the efficiency curve, that might work. Note that the 1303 motor has a stator volume about 30% larger than the 1402.