H920 replacement battery

[PatR]

More Turnigy 8000mA Graphene Battery
Part 1

I'll get through the boring stuff first but weather was good so the first flight on the Turnigy 8000mA 15C battery was done today.

Back to the boring stuff; I made up a rather long twin power adapter lead yesterday. By long I mean ~335mm. Because of the 10awg multi-strand wires used for the primary battery lead I felt it would be difficult, if not impossible, to make a short split power adapter that when combined with the bulk of the battery lead would allow the battery hatch to be closed and remain closed. Since the battery leaves a little room on both sides when fully installed in the battery bay I figured a longer adapter could be pushed between the battery and the outer battery frame to double back to the power plugs. The balance lead is long enough to place it on the opposite side of the battery bay between the battery and the inner frame, keeping the balance plug clear of the primary power wires. The end result is that it works as intended. If I was to do it again the adapter would be about 15mm shorter that what it is now. If anyone is interested I used 12awg multi-strand wire for all the adapter wires and had to use a pair of 120W soldering irons to heat up both sides of the wire junction at the same time to obtain a good solder joint. I tried a mini torch but those are somewhat of a pain to properly moderate wire temps and solder flow, even when using paste flux. The main power connector is an XT90 to avoid cutting the battery power lead to adapt a different connector.

The good stuff follows in Part 2

Edit: The mention of 8awg wire as the battery primary lead was a "senior moment" error. The actual wire size is 10awg.

 

[PatR]

Part 2, Turnigy 8000mA, 15C Graphene Battery Test

For those obsessed with numbers and tons of detail this post might be satisfying.

Charging
The battery as received had a charge state of ~22.6V. After a little fussing around with it a bit it was at 22.46V when I placed it on a Hitec X2-400 DC multi-charger using a 110V-220V, 30A power supply at a 1C charge rate of 8A. Note, although the HobbyKing website makes mention of higher charge rates the info sheet that came with the battery states not to charge at a rate higher than 1C unless the battery is labeled for a higher rate. There are no annotations anywhere in the package or on the battery that suggests a rate greater than 1C is permissible. Room temperature at the time of charging was 67*F, or 19.44*C. I turned my back on the charger for a little while and did not notice the time, but at 23.97V the charger terminated the charge cycle with an "over capacity limit" warning. Measuring the battery verified the state of charge so the charger was adjusted to a 1/2C rate of 4A to continue the charge. After another 2200mA or so the battery achieved a full charge state of 25.2V to 25.22V, depending on which device I used to measure it.

IR
Allowing the battery to cool off for an hour IR was measured, providing IR values of one Milli-Ohm for 5 of the 6 cells. The 6th cell measured 2 Milli-Ohms. I don't know what good IR values for a graphene cell are but 7 Milli-Ohms for a 6 cell pack can't be all that bad.

Flight Test Conditions
A Kestrel 4000 weather station was used to obtain weather data.

• Wind: 0-1mph
• Temperature: 67.6*F (19.77*C)
• Humidity: 54.3%
• Density Altitude: 1087' MSL
• Dew Point: Who cares, but ~50*F for those that might
• Visibility: >20 miles w/scattered clouds at ~2/10

Flight Test
Normally I define flight time as the time from motor arm to motor disarm. I do not normally record any time other than "flight time". As that does not represent 100% of the time a flight battery is in use I recorded battery time in use for specific segments of the power up and flight process. Each time is specific to the function and is "in addition" to actual flight time.

For roughly 3/4 of the test flight the 920 was flown normally, climbing to a max altitude of 300'agl and descending to 127'agl, flown out to a lateral distance of ~1500, and kept maneuvering during the first 3/4 of the flight. The gear was cycled twice, the camera was on and recording 4k video for the entire flight. For 1/3 of the flight the camera was placed in Global mode and kept in a continuous slow pan. In ways I worked the 920 harder than I normally do. For the last 1/4 of the flight the 920 was kept close in a hover with periodic climbs and descents from 25'agl to 50'agl as the battery level dropped from 21.7V to the 21.4V when it was landed.

Aircraft Boot Up, inc camera connectivity: 0:01:09
Post boot settings adjustment, includes boot time: 0:03:20
Time to first voltage warning @ 21.8V: 0:13:38
Second voltage warning at 21.6v: 0:19:00
ST-16 Vibration warnings initiated at 21.5V
Time at landing: 0:21:12 @21.4V, 3.56V/cell
Battery temp: Slightly warm, not at all hot, checked immediately after landing

You can add the 03:30 of boot and set up time to the flight time, or not, as you feel appropriate.

With only one flight thus far I think the battery works great. If anything, Yuneec could provide a firmware update that reduced the multiple and massively annoying ST-16 low voltage pop up warnings. They could actually do a lot more but minimizing the pop up warnings would provide a lot more functional camera time to the owners. Sure, you can over ride them by pressing the "Back" icon twice for each warning but the 21.8V warning is extremely persistent and requires many over rides. A maximum of two at the first and second warnings should be enough since the ST-16 goes into a vibrating cycle at 21.5V. There are no pop up warnings between 21.7 and 21.6V, which is quite a bit of flight time. At 21.5V the ST-16 initiates a vibration cycle for the voltage warning with only one pop up to go with it. I haven't flown below 21.4V with any battery so I don't have a clue when the gear would auto drop and go into emergency landing mode.

Opinion: It appears to be a viable battery choice for those that can make or obtain a power adapter lead.

The flight time obtained during this test was not much short of the longest flight time I've obtained (~25 minutes) from a flight using three new Yuneec 4000mA batteries. The Yuneec batteries sell in the U.S. for $279.99/pair, making a 3 battery installation pricing out at U.S. $419.98. The Turnigy 8000mA, 15C Graphene battery is priced at $68.52 at HobbyKing and provides almost the same flight time as three Yuneec batteries do. You can do the math. The only question is how long will the Turnigy battery hold up over time?

Edit, after posting the above I returned to the HobbyKing site and found they've raised the price of the Turnigy Graphene battery to $112.89....

 

[Dougcjohn]

Outstanding flight and Empirical data! Part 1, nice split lead set, looks store bought (hope that’s a compliment). With the lead wrap back going forward then back would it be any effective gain inserting lead end first, and bringing Lf & Rt leads on each side to rear plugs?

Bird’s flight equivalent time of a 3 x 4K pak (12k) pak with a 8k pak, electrically that’s introduced some impressive data. You feel that’s primarily the 8k size of cell and higher discharge rate / resistance or more towards the newer Graphene LiPo chemistry being more efficient in discharge and retention? It’d be a good comparison with a stnd 8k LiPo with similar discharge C value.

Congrats on a successful & productive experiment.

 

[PatR]

Thank you very much! Unfortunately it appears they raised the price in the few days since I bought one; Turnigy Graphene Professional 8000mAh 6S 15C LiPo Pack w/XT90 but even at that we're money in the bank. We could still buy a pair of 8k mA graphene for less than a pair of Yuneec batteries and end up with almost as much flight time as when flying with three Yuneec batteries. A bit of disparity in all that.

As for the possible reasons, I believe there are several. One is the burst discharge ability, which means it can also better sustain a higher current load. It's rated at 15C sustained and 30C burst discharge. The duration of a burst discharge is typically very short and measured as a few seconds or less.

I'm certain the very low IR contributes to better performance. Of the 7 brand new Yuneec batteries I obtained with the 920 all of them had IR values usually associated with old batteries. Pack totals with an IR of 40 to 67 was the norm. That may well explain why the Yuneec batteries are very hot after a flight while the graphene was warm but cool enough to hold against your cheek after the flight. I wish I had a temp gun, or better still, a data logger with a thermocouple that could be placed in the battery bay during flight. Alas, I do not.

The larger gauge power lead will certainly permit a more efficient energy transfer. I don't know that having a graphene chemistry makes it better or not. What I've read suggests improved energy density and transfer efficiency but I don't have the equipment to check that, or the money to buy a bunch of batteries to compare. The cell size has me on the fence. I don't know enough to make an informed decision in this. I know two Yuneec batteries weigh less than a single 8k graphene but the two Yuneec batteries provide less flight time than the single graphene.

I'm still of the opinion that all the Yuneec batteries being sold are from an old production run, perhaps even the original production run for that battery. As Yuneec did not sell a large number of 920's there's room to think they may not have scheduled a follow on battery order, using up what's been sitting on the shelves. The IR values with all my new batteries strongly suggest either poor cell quality or old age. Two years is getting long in the tooth for a lipo and two years ago is when they came out with the 920. Coincidence? I don't usually believe in them. I did learn one more thing today; I'll not be buying any more Yuneec batteries for the 920. If I can depend on the graphene for relatively consistent flight times under similar conditions I don't mind sacrificing a couple minutes flight time for consistency and reliability, especially when that costs less. There's also having to deal with only a single battery instead of two or three to conduct a flight. Less charge time, fewer chargers, less data tracking required. I like simple whenever such is possible.

My next go 'round with batteries will be focused more on locating a 5000mA or greater battery of a size that will permit carrying two of them in the battery bay. I figure the 8000mA graphene is a done deal so from this point it's all about providing more capacity for more flight time and at what cost.

As for feeding the power lead from front to back, I don't know that would be ideal. It would work but there are about 4 large JST connectors plugged into a board at the front of the battery bay and the possibility of having the battery power wires getting in there and applying any pressure to the board, wiring, or JST's is not something that gives me any "feel good".

 

[seabee]

Great research and documentation. Looks like a real viable alternative to the Yuneec battery even with the price increase. This is something those of us who are operating the H920 desperately need. Thanks Pat.

 

[Dougcjohn]

Pat, the discharge and burst does make sense, also age of cells coupled with IR. Is there reference of the OEM burst C? The burst of the 4K’s recently located are 25C - 50C burst. It’ll be interesting how they pan out in testing 2 & 3.

When you introduced the Graphene, I looked for a 4K and 5k setup. Didn’t find workable 4K, need to relook on 2 x 5k, I was attempting a 3 x 5k for run time.

At the non-sale price it’s still a major improvement!
The same will occur with the 4K, I asked if price was sale or standard, they stated sale, but sales run often.

 

[Dougcjohn]

Battery Alternative...
On another discussion; brought up teathering for unlimited time at a job site, sporting event, news, etc.

The H920 was one of the first they experimented with. Currently several packages are offered for h9290, h520, h480.
https://www.dronefans.com/tethered-drone-safe-t-yuneec-h920/

Tethered Yuneec Drones

Safe-T | Smart Tethered Drone Station

The Tethered Drones | UAV Expert News

 

[PatR]

From my perspective any event that would benefit from a tethered multirotor would gain much additional benefit from a much less maintenance intensive tethered balloon or aerostat. They can host much larger and better payloads and are silent. We left them up for a week at a time in Afghanistan. The only down side were the outside snipers when they had to be serviced, often induced by the same people. They are pretty hard to beat where persistent surveillance over a large area is needed

As for OEM burst ratings, I have not come across any references for that.

 

[Dougcjohn]

From my perspective any event that would benefit from a tethered multirotor would gain much additional benefit from a much less maintenance intensive tethered balloon or aerostat. They can host much larger and better payloads and are silent. We left them up for a week at a time in Afghanistan. The only down side were the outside snipers when they had to be serviced, often induced by the same people. They are pretty hard to beat where persistent surveillance over a large area is needed

As for OEM burst ratings, I have not come across any references for that.

On positioned observation posts, or sporting events, I’d agree better to utilize other options... great examples by the way! Where the balloons & Aerostar equipped with any method to reposition or all relied on camera rotation once deployed?

A few quick thoughts on where a teathered platform might be useful; On building / bridge inspections, being able to stay up and focus on completing inspection, or arriving to a disaster/weather incident being able to deploy quickly, maneuvering around (minimal) for position and assess damages from an aerial position that may provide a clear perspective of event, or single crew covering multiple positions of a looping course event (ie: off-road course, etc) performing multiple aerial deploys for short intervals at each location but longer than a battery pak.

Current offerings are pricey, I’d be curious to see the design: weight, electrical, mods to aerial platform, safety features?

 

[PatR]

Graphene Follow Up

I elected to let the Turnigy battery sit over night before recharging. Checking pack voltage this afternoon prior to charging provided a voltage level of 21.92V, only 0.52V higher than indicated pack voltage when battery use was terminated. I find this different from normal li-poly batteries where voltage after a flight will rebound by a full volt or more after the battery cools. i don't know what this difference in rebound voltage is telling me but I'm certain there's something in it that could be important where flight operations are concerned.

Since the first charge was interrupted by a capacity over limit on the Hitec X2-400 charger I used a new Yuneec A-10/Hitec X2 AC-DC charger for the second charge, using a 1C, 8A charge rate. Interestingly, this charger also interrupted the charge cycle at ~23-6V with the same capacity over limit indication in less than an hour into the charge cycle. I do not know the reason this is occurring. Both chargers had no problem completing the charge after starting the cycle again. During the first charge the charge rate was reduced to 4A, or 1/2C for the second portion of the charge while the next charge was left at 8A, or 1C for the balance of the charge cycle. In both cases there were no further interruptions of the charge cycle. I had wanted to obtain the milliamp count for the second charge to establish what percentage of capacity was consumed during the time of flight but as the charge cycle was interrupted this was not possible and will require at least one more test flight to determine.

Yuneec 920 Battery Notes

As chargers and cables were in use a thought occurred to check the IR values of the Yuneec 4000mA batteries again. The results were distressing. All 7 batteries, obtained last November, had been subjected to between 4 and 6 cycles, and none discharged below 70% of capacity and none stored fully charged, being cycled to storage levels for much of December. Of the 7 batteries the lowest IR was 68 with the highest 87. I'll not bother with individual cell values as at these IR counts the batteries are all indicating they are deteriorating rapidly, worn out or soon to be worn out, and should not be viewed as a dependable power source. They will still power the aircraft but may provide a little less flight time with each use, or fail completely in flight. How low do you want to go? I'll suggest anyone using the Yuneec batteries for the 920 check the IR as soon as possible, preferably before the next flight. The increasing IR values adds a little more fuel to my theory all the batteries currently being sold by Yuneec are old stock, and have passed their functional age. If such is the case, which would be very troubling, we may be dealing with a business decision that does not consider safety of flight.

 

[PatR]

"On positioned observation posts, or sporting events, I’d agree better to utilize other options... great examples by the way! Where the balloons & Aerostar equipped with any method to reposition or all relied on camera rotation once deployed?"

Although a bit more sophisticated camera gimbals and lenses or often remotely controllable through the use of a rather neat joystick. Pan, tilt, zoom, and a few other tactical features are always in use.

"A few quick thoughts on where a tethered platform might be useful; On building / bridge inspections, being able to stay up and focus on completing inspection, or arriving to a disaster/weather incident being able to deploy quickly, maneuvering around (minimal) for position and assess damages from an aerial position that may provide a clear perspective of event, or single crew covering multiple positions of a looping course event (ie: off-road course, etc) performing multiple aerial deploys for short intervals at each location but longer than a battery pack."

Works for me. Governments all over the world are doing the same to monitor their populations. You can't hardly go anywhere in New York and not be on camera, and London is so saturated with surveillance cameras they could prolly find any loose change that was dropped if someone brought it to their attention. Some cities in China are even worse.

"Current offerings are pricey, I’d be curious to see the design: weight, electrical, mods to aerial platform, safety features?"

The price is why so few are in use. Most were intended for the military establishment, had to go through a government procurement process, and therefore 100 times more expensive than need be because of the wasteful, inefficient bureaucratic processes involved. Buy a hammer yourself at a hardware store for $15.00. Have the same hammer provided as part of a government/military cost plus contract and the end result is well over $100.00. Like a VAT tax...

Two great safety features are the use of stranded cable, not rope, and a remotely operated emergency deflation device. However, both have failed in the past, as demonstrated by a wayward military radar balloon that got loose on the East Coast a couple years ago.

 

[PatR]

Capacity Over Limit Cut Off

After the previous post where I mentioned a repeat of "capacity over limit" termination of the charging cycle Seabee sent me a note that included the ops manual for the Yuneec A-10 charger (I did not have it) that suggested the charger was factory set with a 5000mA charge limit. He was 100% correct. On a side note the Yuneec A-10 charger is a re-branded Hitec X2 AC/DC charger and a much better instruction manual for the X2 is available for download at the Hitec RCD site and attached in this post.

As the 5000mA limit was present in the A-10 I also checked my Hitec X-2/400 charger and found the same 5000mA limit. Those of us using "after market" chargers and batteries with >5000mA capacities should probably go into our charger System Setting to check the cut off limits for both time and capacities and adjust them accordingly. The flow charts provided in both the A-10 and X-2 charger instructions are a lot more complicated than they need to be and make entering charger system settings appear to be a daunting task. It's not and really pretty simple. For the Hitec X2 type chargers it's only a matter of turning them on and scrolling through battery types until reaching the System Settings menu, pressing the start button to enter the System menu, then scrolling through that menu with the "+" button until reaching the function desired and making mA adjustments.

Thanks Seabee!

 

[PatR]

New battery test coming soon!

HobbyKing now has the Zippy Compact 6s 25C 5000mA batteries back in stock so I ordered a pair for testing. Combined weight is a bit over 1400 grams and they are shorter than the Turnigy Graphene which will leave a wee bit more room for adapter cables. Since they come with an XT60 connector the adapter cables will be less expensive to make than the XT90 version. Two of these are only $7.00 more than a single Turnigy but provide 2000mA more capacity with 10C greater burst discharge capability.

Who knows, we might see a 25 minute flight time again with these. That would be nice.

 

[PatR]

More Graphene Battery

Just did another test flight with the Turnigy Graphene battery with the results pretty much the same as the first. Total flight time 0:21:14, landing with a pack voltage of 21.2v as displayed on the ST-16. The battery had been charged and left sitting for a few days because of fog so starting voltage had dropped from 25.2 to 25.15. Not all that much of a difference but it was not a fresh charge. Measured battery voltage after removal from the 920 was 21.48v.

Interesting, but the 920 appears to self land at 21.2v. I say appears because I was descending to land but stopped the descent to hover for a moment and the 920 took over and landed itself. This is kind of a pain as it cuts a little useful flight time from a battery but it's also a fair safety measure when using a graphene battery as voltage drops very quickly when graphene cell voltage drops below 3.5v, and once achieving 3.5v/cell it's a good time to be looking at and being on your way to your landing spot. If the 920 auto lands at 21.2V I doubt you could fly the battery downbelow 3.5v per cell so be looking to land at 21.4V. 21.2v comes quickly after that. Cell voltages after landing were 3.517, 3.596, 3.60, 3.610, 3.581, 3.581.

Prior to flight the cell voltages were 4.178, 4.198, 4.197, 4.197, 4.192. IR was still a pack total of 7, same as it was prior to the first flight. Voltage alarms again very persistent from 21.8V through 21.7v, which occurred at 0:13:40 from the time motors were armed. Next voltage alarms trigger at 21.5v, but are not as persistent as the 1st warning, but are associated with a constant vibration cycle at the ST-16. First voltage warning initiates a single red flashing light at the aft LED, second initiates a double flash of red at the aft LED. At auto land the aft LED goes into a steady red flashing light mode.

Conditions of Flight
Temp: 63.3F
Density Altitude: 814' MSL
Humidity: 70.5%
Dew Point: 53.2F
Wind: 1-3 MPH

Only a couple flights on this battery but they have performed similarly.

 

[matt750]

More Graphene Battery

Just did another test flight with the Turnigy Graphene battery with the results pretty much the same as the first. Total flight time 0:21:14, landing with a pack voltage of 21.2v as displayed on the ST-16. The battery had been charged and left sitting for a few days because of fog so starting voltage had dropped from 25.2 to 25.15. Not all that much of a difference but it was not a fresh charge. Measured battery voltage after removal from the 920 was 21.48v.

Interesting, but the 920 seems to self land at 21.2v. This is kind of a pain as it cuts a little useful flight time from a battery but it's also a fair safety measure when using a graphene battery as voltage drops very quickly when graphene cell voltage drops below 3.5, and once achieving 3.5v/cell it's a good time to be looking at and being on your way to your landing spot. With the 920 I doubt you can fly the battery down to 3.5v per cell so be looking to land at 21.4V. 21.2 comes quickly after that. Cell voltages after landing were 3.517, 3.596, 3.60, 3.610, 3.581, 3.581.

Prior to flight the cell voltages were 4.178, 4.198, 4.197, 4.197, 4.192. IR was still a pack total of 7, same as it was prior to the first flight. Voltage alarms again very persistent from 21.8V through 21.7v, which occurred at 0:13:40 from the time motors were armed. Next voltage alarms trigger at 21.5v, but are not as persistent as the 1st warning, but are associated with a constant vibration cycle at the ST-16. First voltage warning initiates a single red flashing light at the aft LED, second initiates a double flash of red at the aft LED. At auto land the aft LED goes into a steady red flashing light mode.

Conditions of Flight
Temp: 63.3F
Density Altitude: 814'MSL
Humidity: 70.5%
Dew Point: 53.2F
Wind: 1-3 MPH

Only a couple flights on this battery but they have performed similarly.

 

[matt750]

On these test flights, are you flying around, using the camera, operating the landing gear, or just hovering?

 

[PatR]

All of the above. Hovering was minimal. Camera recording 4k for the entire flight. Flying it like I’m shootings a job. Pure hovering to test a battery provides nothing useful for functional battery duration.

 

[Dougcjohn]

New battery test coming soon!

HobbyKing now has the Zippy Compact 6s 25C 5000mA batteries back in stock so I ordered a pair for testing. Combined weight is a bit over 1400 grams and they are shorter than the Turnigy Graphene which will leave a wee bit more room for adapter cables. Since they come with an XT60 connector the adapter cables will be less expensive to make than the XT90 version. Two of these are only $7.00 more than a single Turnigy but provide 2000mA more capacity with 10C greater burst discharge capability.

Who knows, we might see a 25 minute flight time again with these. That would be nice.

Will be interesting to see your outcome data with 2 5K’s. The only ones of this spec I could find have either 5.5 Bullets or XT90.

Looks like a great fit in battery bay too! With dimensions 157x45x46 for 5K compared to 170x69x48 for 8K, the width is very workable for 2 x 5k; in comparison 4K @ 39 (4K desired spec for 3) total 117, the 5k @ 45 at 90, 8k @ 69 (138 too wide for 2). With a bit more height clearance of 2mm too.

Looks like a nice fit all around for 10,000mA. Great Find!

 

[PatR]

Thanks.

Don’t know how I did it but I’m pretty sure I saw ST60 when I first looked at the battery specs. After placing the order I looked again and saw they were XT90. My error. I have enough 90’s on hand to make adapters but it would have been nice to have been using 60’s as they are less bulky. Perhaps it was just wishful thinking...

In case anyone is interested the dimensions of the Zippy Compact 4000mA and 4500mA 6s are almost the same as the 5000mA and should allow a pair to fit in a 920.

What will be more interesting is learning what a 25C discharge rating will do, or not do, for flight duration. The Yuneec 4000mA batteries are rated at 8C and on a good day a pair will allow about 17 minutes of flight. The single 8000mA 15C graphene is providing 21 minutes of flight, only 4 minutes less than the best time I’ve obtained when using 3 Yuneec batteries. It may be the much lower IR of the graphene is providing the increase in flight time, or it could be IR combined with battery quality and increased C rating. In any case, the after market batteries are a heck of a lot more cost effective. Using the graphene as an example, we can obtain 3 of them, with each one providing power for a single flight longer in duration than 2 Yuneec batteries will provide for less than what a set of 3 Yuneec cost that would power a single flight only 4 minutes longer in duration than the graphene. I’ll trade 4 minutes of flight to save $300.00 any day of the week. A pair of Zippy’s (flight time TBD) allows purchase of 6 batteries (3 flights) for much less than 3 Yuneec batteries. There’s no longer any reason to buy the lower quality Yuneec batteries.

 

[Dougcjohn]

Another possible on the dual pak that exceeds a tri-pak.
The Zippy 6200mA 25C/35C is 157x45x60, dual pak (12,400mA) @ 120mm width compared to 3 x 4K (12,000mA) at 117 width. May be too tight in width, but looks like a possible fit. Peal off outer wrap cuts 3mm difference in half.
ZIPPY Compact 6200mAh 6s 40c Lipo Pack w/ XT90 @ $78 / $156 set