Prepping My 3-Inch Freestyle Kestrel for Long-Range Flights

| Comments

Let’s start this blog post out with a summary of what I’m excited about. My 3” Kestrel with its little 1306 motors has just barely enough room to squeeze on a set of 4” biblade props, and using the Outcast Droneworks 3250 mAh 2S Lithium-Ion pack, it is getting some amazing flight times. So far, I’ve had a 19-minute flight while cruising at a pretty good pace, and a 12-minute flight that covered just over 6 miles on a day with 21-MPH winds!

In its current long-range configuration, my 3” Kestrel weighs in at 248.5 grams with a Caddx Turtle, TBS Crossfire, GPS, and Zoe FPV’s 3250 mAh battery.

This is all exciting, but at this point I have more questions than answers. I can’t give you a definitive guide to building a sub-250 gram long-range quad. I can tell you what I’ve learned so far. I can tell you about what I’m hoping to learn over my next three or four flights. I can also tell you about how I got to this point.

I’m going to be repeating a lot of information from [my previous post on this topic][e4o]. With any luck, I’ll only repeat the bits that are directly relevant to my new findings!

My Kestrel is not designed to be a long-range quad!

I did not set out to build a long-range micro quad. From the moment I began designing the Kestrel frame, I knew that I wanted an amazing HD freestyle quad. The goal was to keep the build under 250 grams, but if I went a little over, that wasn’t a big deal.

Just because a quad is good at one task doesn’t mean that it can’t pull double or triple duty. The front end of the Kestrel is set up in such a way that it is easy to use a battery strap to mount a GoPro on top. I’ve done short flights on both my 3” and 4” Kestrel builds with a GoPro Session.

You aren’t going to fly like Mr. Steele with all that extra weight, but either of my builds could do the job of a Cinewhoop. In fact, I think my 4” Kestrel with 1606 motors would do an amazing job at getting that buttery Cinestyle footage. Maybe I should design some prop guards?

I was hoping my 4” Kestrel build would be a good long-range quad

I hoped that 4” props would add enough efficiency that it would be easy to get a 10-minute cruising flight out of my 4” Kestrel, but it didn’t work out that way. The bottom-end power and the general feel of my 4” build is better for the type of freestyle I fly, but if anything, it wound up being less efficient than my 3” Kestrel.

I put heavy, overpowered VTX modules on both my Kestrel builds in the hope that they would one day make a long-range flight.

If the 4” was going to cruise for 10 minutes, it was going to need a bigger battery. A bigger battery would push it way over 250 grams, though, and that was going to be a bummer. This encouraged me to shelve the idea of going long range.

The Outcast Droneworks 3250 mAh High-Discharge Lithium-Ion battery pack

Zoe FPV’s amazing 2-cell battery changed everything. I quickly learned that my quads will fly on 2S. They didn’t fly smooth at first, but it didn’t take much to tune out most of the wobbles. My tuning isn’t quite done. I noticed that as the battery drains enough, when it gets close to 6 volts, the Kestrel doesn’t fly as smoothly. I’m sure I can compensate for that.

WARNING: The video above is the entire 19-minute flight.

Zoe’s battery weighs in at almost exactly 100 grams. My 650 mAh 4S packs are 85 grams, and the extra weight very nearly pushed my 3” Kestrel over the 250-gram limit. For testing purposes, I wasn’t too worried about the weight. I was just glad it was in the right ballpark.

My 3” Kestrel is forced to land when the Outcast Droneworks battery starts showing about 2.8 volts per cell in my OSD. When the load is removed, it bounces back to 3.2 volts per cell. This is nice, because it means I’m not ruining the battery when I take these long flight.

The new HQ 4x2.5 biblade T-mount props are awesome

These props are amazing. They’re quite gentle. They weigh less than the HQ 3x2.5x3 props. On my 3” Kestrel, the 4” props managed to fly almost 20% longer than the 3”. I think that’s impressive.

They actually feel great for freestyle on 4S, but my little 1306 motors just don’t have enough torque to manage propwash with such a long propeller.

The combination of Zoe’s battery and the new HQ props is a game changer for me.

I lucked out. I’ve designed three different arms for the Kestrel. There are 4” arms with a ton of clearance. There are compact 3” arms, which have just barely enough room to keep 3” props from rubbing on the fuselage. The arms I fly on my 3” build are the long 3” arms.

Kestrel with HQ 4x2.5x2 biblade props

I wanted the Kestrel to be light, but it doesn’t need to be tiny. The farther apart you can get the props, the better a quad flies. The extra distance helps with leverage, and getting the rear props farther away means they have less turbulent air to work with.

I lucked out. My long 3” arms are just barely long enough to fit a 4” prop. The arms on the Kestrel have a tiny bit of play because each arm is only held on by a single screw. I had to pivot each arm to keep the 4” props from rubbing on the frame. If I crash, they may pivot the other way and get stuck!

If all this experimenting works out, I will need to adjust the design of the arms to eliminate this problem!

Let’s talk about the 19-minute flight

I don’t have a lot of data on the 19-minute flight. At the time, I didn’t have a GPS module on my 3” Kestrel. I do have one on the 4” Kestrel, so I used the 4” to calibrate my brain. I flew for five minutes to get a feel for how fast 15 MPH is.

My plan was to fly the 3” Kestrel with the 4” props in roughly the same manner. Was I really flying at 15 MPH the entire time? Probably not. In fact, I’m starting to think I was at 20 MPH or more a lot of the time. For now, I’m going to assume that my average speed was indeed 15 MPH.

If that’s true, that first 19-minute flight should have covered about 5.75 miles.

I had a 16-minute flight on HQ 3x2.5x3 props

There isn’t a lot to say here. I flew at the same pace on the 3” props as I did on the 4” props. I didn’t have GPS to verify, though.

This flight was 3 minutes short of the previous flight. That means the 19-minute flight covered nearly an extra mile of distance.

I’ll be repeating both these flights with GPS.

Adding GPS and getting back under 250 grams was hard work!

I’m using a 5-gram GPS module from Banggood. It is a fantastic little piece of hardware, but I was right at the 250-gram limit. This pushed my to 253 grams, and shaving 3 grams was really difficult!

I wanted to save a bit more than 3 grams, though. I wanted to be far enough under that an error on my scale wouldn’t be making a liar out of me!

I removed extra zip ties. I removed two screws from each motor. I took off my gummy battery pad. I even used a pair of scissors to make my battery strap more narrow. The last bit got me under 250 grams, but only by 0.04 grams. That didn’t feel right, and I had another problem. The ND filter doesn’t weigh much, but it definitely pushed me over 250!

I got my old PH145 out of the closet, and I borrowed four aluminum M3 screws. Swapping out my steel screws for aluminum brought me down to 248.5 grams.

That’ll do it, right?!

Flying slow might be the wrong idea

To have this discussion correctly, I need to watch the amp-draw readout in the OSD. I haven’t done that. I only have estimates for throttle position. Throttle position doesn’t scale directly with power consumption, but I’m going to pretend that it does. It is close enough to help illustrate my point.

When I think about efficiency, I assume slower is better. That’s why I decided that 15 MPH would be a good cruising speed. While it is true that you’ll get a longer flight at a slower speed, I’m not so sure that’s the point of this exercise.

If my goal is to fly up the side of a mountain, I don’t want my flight to last as long as possible. I want to cover as much distance as possible!

Here’s what I’m thinking. With my long-range setup, it takes about 30% throttle to hover in place. Then it takes a total of about 45% throttle to cruise at 15 MPH.

That means about 2/3 of my energy is being used just to keep the Kestrel in the air, and only 1/3 of the energy is being used to propel the craft forward. If the goal is to cover as much distance as possible, this doesn’t seem like a good plan!

What if I cruise at 60% throttle? That would mean half my energy would be propelling the craft forward.

Comparing throttle position isn’t accurate here. I need to compare amps at different speeds. In any case, I think the concept is sound.

My first GPS flight with the big battery

I was hoping to make several test flights over the weekend. It was way too windy to get good data, so I only took one flight.

I was cruising at about 62% throttle for most of the flight. When I flew south, I was going at 15 to 20 MPH. When I turned around to fly north, I was flying at 50 to 60 MPH! The wind was pretty constant, but there were some gusts. I’m hopeful that the directions of my flight canceled out the wind assists, but I’m not all that confident.

This needs to be repeated on a calm day, but I’m going to tell you how things went anyway.

The flight lasted 12:41, and I covered 6.01 miles according to the GPS. If I had known that this was 9.67 kilometers, I might have tried to push it past 10 kilometers!

Should you build a long-range micro quad?

I’m not sure. Even with the Kestrel’s extremely long fuselage, it has been difficult to get enough separation between all the various radio components.

When I put my first Kestrel together, I mounted the Crossfire Nano module under the VTX board. Even though the antennas weren’t close to each other, this was still horrible for my video signal. I’ve since moved the Crossfire Nano to the opposite end of the quad, and that helped tremendously, but I’m running into new problems.

My GPS wouldn’t even pick up a single satellite when it was mounted near my Crossfire antenna, so I moved it to the front of the quad. Now it is directly above my Caddx Turtle board, and I have to be careful to mount my battery farther back to keep it from blocking the GPS module.

A friend of mine told me that he can’t mount a GPS module too close to his Caddx Tarsier, because the Tarsier is noisy, and prevents the GPS from locking on. I might be in trouble if I upgrade!

Getting enough separation between these three radio modules and antennas is difficult enough on my rather large micro quad frame. Things will be even harder on a smaller, lighter build, and I’m not even sure how well this will all work if I actually managed to fly two or three miles away!

Why use a micro quad for long range?

I always tell people that my Kestrels are my hold-my-beer quads. The last time I priced one out, either of my Kestrel build comes in at around $250 including props and the battery. The addition of the GPS module and the more expensive 3250 mAh battery probably pushes that number up to $280 or $290.

My pricing spreadsheet includes a whopping $45 for the cost of the Kestrel frame. If I ever manage to put them up for sale, I sure hope I don’t have to charge that much!

This is where I usually say that the entire Kestrel costs less than a GoPro. While this is true, I’m still flying with a Session 5, so it isn’t quite true for me.

Why send my $500 5” miniquad that carries a $150 to $400 camera on a risky mission when my $250 Kestrel could get the job done just fine? That risky mission could be crossing a stretch of ocean to reach an island, flying freestyle in a sketchy part of a bando that you could never climb up to yourself, or maybe even just scoping out a new spot when you’re uncomfortable.

I’ve been examining all sorts of compromises

I’ve been ignoring the Runcam Hybrid. I don’t want to spend $90, take a Kestrel apart, solder in the new camera, then realize that I’m almost as disappointed in the HD footage from the new camera as I was in the old camera!

My newfound long-range excitement has me rethinking this. The limited field of view of the Turtle looks awful when flying freestyle. When going on a slow, steady, long-range cruise, it doesn’t look bad at all! The new problem is that the Runcam Hybrid is 5 to 10 grams heavier than the Caddx Turtle.

Where do I save another 10 grams? My heaviest component is the massive RaceDayQuads 1,000 mW VTX and the MMCX antenna. The lightest replacements wouldn’t be suitable for long range. A light VTX with reasonable output power for long-range flights will be something pricier like a TBS Unify of some sort, and that bumps up the cost of my hold-my-beer quad!

I’m sure I can cut a frame that is a few grams lighter. There are four M3 screws left that can be swapped out for aluminum. I’m sure I can puzzle out a way to get a Runcam Hybrid on here and still manage to get under 250 grams.

I don’t really have to stay under 250 grams, but I think the build is more interesting if I do. I’d like to have a good, reasonably priced parts list. If you live in a country with a 250-gram weight limit, and you want to replicate my success, I would like to make that easy for you to do. That is assuming that this long-range Kestrel stuff is truly successful!

What’s next for the long-range Kestrel testing?

I need to take more flights! We have some nice-looking days coming up, and they should be free of wind.

I need to replicate that 19-minute flight, but this time I need to do it with GPS. Most of that flight happened at around 45% throttle.

I’ve done a flight at around 62% throttle. I’d like to do another at around 75% throttle.

I need to do more tuning. The Kestrel is pretty solid now when the 2S battery is full, but it has a big case of the wobbles towards the end of the battery. For most of the flight, the voltage stays above 7 volts. The last mile or so is at 6 volts or less, and that’s where it gets problematic.

These test flights are long and boring, and I have to wait for a battery to charge between flights. With any luck, I can get the 45%– and 75%-throttle flights in during a single session.

I also need to burn through an entire Outcast Droneworks pack on my real 4” 1606 Kestrel. She comes in at about 275 grams with the 2S pack. She flies fine on 2S, but the voltage dropped quickly enough on my 5-minute test flight that I can see that it won’t compete with the 248-gram 1306 Kestrel.

Update: List of flights, times, and distances

I’m not sure this is the best place to keep track of this data, but it is definitely the best place for now!

Flights using Outcast Droneworks 3250 mAh 2S Lithium-Ion battery:

  • 11/23: 19:01, unknown distance, 45% throttle
  • 12/09: 12:41, 6.01 miles, 60% throttle
  • 12/14: 10:06, 6.05 miles, 75% throttle
  • 12/15: 16:00, 5.5 to 5.75 miles, 45% throttle (forgot to DVR!)

Flights on 650 mAh 4S:

  • 12/15: 4:29, 2.58 miles, 25% throttle

My 4S 650 mAh packs are old and tired, but they save me about 25 grams of weight. These are very different batteries, but I figured I should include the data.

Conclusion

I have absolutely no idea where I’d fly some sort of long-range mission. Maybe flying from one end of our abandoned golf course to the other and back would qualify. Google Maps says it is about 4,800 feet, but I’d have to cross a street at about 3,000 feet.

It is possible to fly under the bridge, but that’s a long way to fly so close to the ground, and there’s no way for me to get a height advantage on my end. I’m a bit bummed out that this is the most exciting and interesting long-range flight I can imagine taking.

If that works out, maybe I’ll drive out to a nearby lake and attempt to fly across!

What do you think? Am I on the right track? Is sub-250 gram long range interesting? Do you think the Outcast Droneworks battery is as awesome as I think it is? Do you think I’m going to have any use for a long-range micro quad, or should I just stick to freestyle? Let me know in the comments, or stop by the Butter, What?! Discord server to chat with me about it!

Comments