The new line of Multistar High Capacity Batteries hit the market last summer. I chose not to rush, because I didn’t want a pig in a poke and looked for user feedback, but that made the choice even more challenging. Eventually, after six months I did buy a Multistar High Capacity 4S 10000mAh Multi-Rotor Lipo Pack just for experiment: Multistar High Capacity 4S 10000mAh Multi-Rotor Lipo Pack
Product http://www.rcfair.com/en/product/1525105/


If you take a closer look at the battery’s appearance, you’ll see Turnigy written in a small font on the Multistar label. Multistar is the name of a product line, same as Nano-tech, A-SPEC, BOLT. They all share the brand name. According to the manufacturer, Multistar batteries are designed for multi-rotor copters. All unlabeled blue-color batteries are hereinafter referred to as Turnigy, and those belonging to the line - as Multistar.

The RCGroups is now actively discussing the product: http://www.rcgroups.com/forums/showthread.php?t=2207137; the RCDesign.ru is quieter, as the information is thinly spread across the threads.

There are few comments about this and other batteries at HobbyKing, and there is only one positive review from a guy owning a hexacopter with a 680 frame. The forums create a mixed impression of the Multistar line.

I’m intent on using accumulator batteries in multi-rotor copters and flying wings. I should say that I used Turnigy 4S 5000mAh 40C Turnigy 5000mAh 4S 40C Lipo Pack
Product http://www.rcfair.com/en/product/1082821/
throughout the last season, both singly and in pairs. As I tested my quadcopter, it would make a good 30 minutes in the air. To be honest, the price was total exhaustion and actually a tumble. Thanks to low altitude, there was no damage. In any event, the ultimate goal was reached, and that was thanks to blue Turnigy packs.

Wonder why I wanted this particular battery?

First, I chose it because of the beneficial weight/capacity ratio. With capacity reaching 10,000mAh, it weighs only 810g. For reference, two Turnigy 4S 5000 mAh 40C weigh more than 1,188g, I mean the Multistar is lighter by about a third. However, there has to be a sacrifice – this line boasts a discharge rate of 10C and a maximum of up to 20C. Consequently, a 10,000mAh battery handles 100A and 200A respectively.


The new accumulator battery makes me want to add a stage and increase flight time. That is quite possible, I guess. You may say that would be useless because capacity would remain the same, so it would be six of one and half a dozen of the other, all the more so because they have even lower parameters. However, I focus on lower weight. I mean, 400g off should contribute to longer flight time. Still, there is one question: will the battery have enough power to get my copter up and running?

I measured the current with a wattmeter as I was test-flying it on Turnigy 4S 5000mAh 40C batteries. It reached 16A at hover mode and about 45A at full throttle. As far as I’m concerned, the Chinese tend to overstate their products’ characteristics, so I guess I they can well be divided by two or even four. However, even that would be enough for unhurried flying.

I ordered the battery ex-stock in Russia during a promotion act when prices were lowered to the European level. Many people know that EMS delivery from Europe is quite expensive. In any event, it cost me about $5, and a Red Express messenger brought it to my doorstep. The delivery was quick, and it would have been even quicker but for the weekend. The invoice number was 101569303, and the tracking service http://www.redexpress.ru/ru/bonus/CargoInfo/ allows you to trace the route of your package. Although it is set up for large cities only, and in other cases ex-stock delivery is conducted by Russian Post via Red Express, it appears to be less costly and much faster.


Let’s do a bit of lab work.

First thing I did after getting the package was checking the battery. Voltage discrepancies between cells were minimal (3,887V, 3,903V, 3,894V, 3,885V), no mess.

Then I decided to run the batteries using a Hyperon EOS battery charger with 5.8 Firmware.

Charger control and log analysis was provided by Hyperion EOS Control & Data Suit 2012-10-24.


It took 3855mA of 10A current to shift the battery from storage mode to full-charge mode.

Next I discharge it until cutoff and then recharge it to see how it is going to react. For reference I parallel the two Turnigy 4S 5000mAh 40C batteries via Channel 1, and the Multistar 4S 10000mAh – via Channel 2.

Discharging batteries

I set the maximal discharge rate up to 5A for this charger.
It took a bit above four hours for them to discharge.
Turnigy gave in first, as Cell 1 dropped sharply. Although the rest of the battery did have a margin, the first cell was down way too soon.

The Multistar’s diagram looks pretty nice. The cells discharge smoothly, and all four cells have almost identical lines. Discrepancies are minimal and measure in thousands of 1V. All cells slid from 3.65V down to the cutoff 3.5V at the same rate.

To sum it up, the Turnigy showed 9810mA, and the Multistar – 10050mA. Given the battery’s stated capacity levels, the result is fully justified.

The maximum capacity diagrams are almost identical. The Turnigy has a 9% margin, and the Multistar – 6%. The reason why the Turnigy showed more was the drop of one of its cells to cutoff level.

A battery’s internal resistance depends on a number of factors and has a broad range.

High capacity batteries are known to be bigger in size than low capacity batteries. Because they feature bigger plates with larger work surfaces, they show lower internal resistance than low capacity batteries. Structural characteristics are also important, as are a battery’s capacity, charge state, operational current, temperature, etc.

While the latter are less important, structural characteristics are vital. No, I won’t take these apart, let’s just compare pack sizes: the two Turnigies (149 x 49 x 37mm) and the Multistar (160 x 65 x 36mm) have similar thickness, yet the Multistar has a smaller surface.

The total surface of the Turnigy plates: (149*49)*2=14602mm2 = 146sm2;
The total surface of the Multistar’s plates: 160*65=10400mm2 =104sm2.

As follows, the Multistar’s plate size is smaller by almost a third.

Charging batteries

On the next day I charged my batteries, since they’d had a good sleep and recovered from the stress of the previous day. I set charge current up to 10A.

Batteries are charged in the CC-CV fashion. In other words, initially constant current (CC) is applied with cell voltage increasing. By the time each cell got 4.20V, the total charge level reached about 95%, and the charger activates the second CV phase. The amperage decreased gradually, so that voltage per cell wouldn’t exceed 4.20V. This parameter is determined by the battery’s chemical characteristics. The permissible excess is 4.25V, and at 4.30V or more the battery can cause an explosion and fire.


The Turnigy took 9,711mA, and the Multistar - 9,906mA.



It took 1:17 and 1:19 for the batteries to charge respectively.

It should be noted that Multistars have a twice higher internal resistance than Turnigies

During the charging process, active mass builds up on the battery’s plates and increases the plates’ operational surface. Therefore, charged batteries have lower internal resistance than discharges ones.

Hyperion is a pulse charger, and it reduces current to 0 and raises it back up to the set value from time to time during the charging process. The diagram on the right reflects 0,2-0,25 voltage drops in the Multistar’s cells between pulses; in the Turnigy voltage drops by only 0,1-0,15V.


This explains the input voltage drops in Channel 2, to which the Multistar is connected.



The diagrams show that Multistar accumulates maximum capacity more smoothly. Turnigy reaches the 98% maximum capacity level faster, and after balancing out the cells it reaches 100%.


If you charge your battery up to 95% as recommended by the charger manufacturer (http://media.hyperion.hk/dn/EOS/EOS0720NET-MAN-EN12.pdf ), the Turnigy will be ahead of the Multistar.

Some may object that it is better to charge it up to 100%. However, Hyperion developers recommend that you charge it up to 90-95%, and there are two reasons why. First, you can increase your batteries’ life this way; second, it increases flight time during field sessions. Just compare the balancing time with the flight time, and you’ll see the difference. All you need to do is tune the TCS End Action option in your charger.

Field test

I was really looking forward to seeing the Multistar’s behavior in my copter.


The copter had the following components:

Tarot FY650 IRON MAN 650 Quad-Copter Carbon Kit TL65B01
Product http://www.rcfair.com/en/product/1469798/
T-Motor MN4010 475kv
Quanum Carbon Fiber T-Style Propeller 17x5.5 (CW/CCW) (2pcs)
Product http://www.rcfair.com/en/product/1567974/
ESC T-Motor 40A Opto
FC Pixhawk Arducopter v3.2
Ublox NEO-6M V2.3 GPS Module With EEPROM For Flight Control
Product http://www.rcfair.com/en/product/1691308/
3DR Telemetry 900MHz
3DR Power Module 90A
R/C Futaba 14SG+R7008SB

Actually, that was my first flight after winter. However, according to the calendar, it was still winter, but it had been thawing in St. Petersburg for several days, as it was +3 Centigrade at daytime. That was not an ideal type of weather for flying anyway, because +3 is the lowest admissible temperature for using LiPo packs, so I had to take that into account.

As it often happens, after a long and tedious preparation, I realized that I had bought the wrong type of props – they were too big and too loud. The Chinese know-how-to-fail.

Besides, I had not yet tuned the flight controller after reflashing it to 3.2, and the PIDs were all default. And the two GPS receivers, which I used, were conflicting.

The configuration was different, so I used a Turnigy 130A Watt meter and power analyzer Turnigy 130A Watt Meter and Power Analyzer
Product http://www.rcfair.com/en/product/1082333/
and measured the parameters during takeoff.

The peak amperage at this takeoff configuration was 29.67A, and the peak power was 445.3W
The first takeoff was powered by the two Turnigy 4s 5000mA batteries.




The diagrams show that the takeoff load caused the batteries to drop down to 15.6V.

The second flight was powered by the Multistar 4s 10000 mA.







The Multistar dropped down to 15.5 – a little lower than the Turnigies. With this battery the copter sheds 400g of weight, which is quite observable during takeoff, because I don’t need to push throttle that hard. In the hover mode, the throttle stick was around 50-55%.

By the universal time point, it was 60 for the Turnigy and 70 for the Multistar, because the flight preparation was a little too long. The Turnigy took 1500mA with a voltage of 14,8V; the Multistar took 1300mA with 15,0V respectively.

I decided not to put the green battery to the ordeal. It felt cold after the flight. I can say that the tests were fruitful, and I guess I’d better hold another one when it gets a little warmer.


Conclusions
The Multistar did well during the lab tests. Before it discharged to the cutoff 3.5V, it had given off 10050mA. Voltage discrepancies between cells wouldn’t exceed a hundredth of a Volt. When charging, the battery took 9,906mA. The Multistar has twice the internal resistance of the Turnigy, which is, most likely, explained by to structural characteristics.

The flight revealed a little deeper drop in the Multistar than in the Turnigy. That should be due to the multistar's discharge rate, because the peak current amounted to 29.67А during takeoff, which agrees with 3C. The diagrams based on the flight controller data are linear.

On the whole, the Multistar did very well. Now I’m going to wait until it gets warm enough for flight time evaluation.

My quadcopter flies well with the two Turnigies too, whose total capacity reaches 10,000mA and weight – 1,188g. Therefore, I can also consider a Multistar 4s 16000mAh, whose weight is 1,290g: Multistar High Capacity 4S 16000mAh Multi-Rotor Lipo Pack
Product http://www.rcfair.com/en/product/1523672/
.

I would really like to try a MaxAmps 4S 11000 mA 40С but I can’t do it right now, because it is inhumanely expensive ($290).

This is the translated version. You can read the original Russian article here.