Through spring and summer of 2016 we have performed a plethora of long tests while running the onboard electronics from the rocket’s internal batteries. For that reason we have also spent a lot of time – far too much annoying time – charging the internal batteries. Although we have dedicated multichannel chargers, it has turned out that fast charging has been challenging, to say the least.
The Nexø rockets are equipped with five internal battery packs. Each pack consists of 10 NiMH cells in series, with 4500 mAh capacity each, giving a nominal voltage of 12 volts. Four of the packs are series couplec two by two, resulting in two independent 24 V power rails for the various electronics boxes onboard the rocket. The fifth battery pack is dedicated the video camera and transmitter sitting next to the parachute bucket.
The long charging time is not caused by the battery packs. The nice and competent people at ACTEC A/S has produced a set of excellent pattery packs, just as expected. The cause of the arduous charging process is the very long cable connection between battery packs and charger units. As we would very much like to avoid having to climb the launch tower toting the charger unit flightcase, the cable length between carger units and the charging connector on the rocket was determined by the distance from the launch platform deck to the charging connector, some five meters above the deck.
As it turns out, the chargers couldn’t agree on the voltage drops developed over the long cable run. This confused the charging algorithms and caused the charging to be terminated prematurely. So despite a large number of switches, installed to facilitate simultaneous charging of several battery packs, we ended up having to charge one pack at a time. And this even at a charging rate much lower than first planned.
We made an extra, very short charging cable, for use in the workshop, where the rocket was placed horizontally on its trolley. This improved things a bit, but charging rate was still well below planned nominal.
To avoid going through this hassle once more, Scott has designed a new and very different charging circuit for the Nexø II rocket. The electronics team has also persuaded the structural team to allow us to place the charging connector at the bottom end of the port cable conduit, much closer to the platform deck. (The starboard cable conduit is actually a fuel pipe conduit with no cables at all.)
Scott has based his charging and control circuit system on a Microchip PIC16F1769 processor, which is designed for applications like this. The system is made up of two different units, a supply unit placed on the platform, which provides regulated DC voltages to the rocket, and a battery unit placed in the rocket, which monitors and controls charging and operation of each battery pack.
These two units communicate through a one-wire open collector link. Each unit also has a serial link, one for remote control (supply unit), and one for telemetri data (battery unit).
The supply unit is placed on the launch platform Sputnik, and provides two regulated DC voltages of 14 V and 28 V to the battery units. The two DC voltages is powered from Sputniks DC generator, converted by two remotely controlled boost converters.
The supply unit communicates with the battery units onboard the rocket, and regulate the DC voltages to fit the need of the battery units.
Remote control is performed through an ethernet interface from Mission Control onboard Vostok, and a local control panel allow personel on Sputnik to monitor and operate the units.
Battery unit schematic, charging circuit. Illustration: Peter Scott.
The battery unit, each unit is assigned to two series coupled battery packs, contains two charging circuits and two switched 24 V outputs, controlled by the abovementioned PIC processor. The processor also communicates with the supply unit on Sputnik, and with the TM&TC unit (TeleMetry and TeleCommand) responsible for communication between rocket and Mission Control. Thus there’ll be no ignition key for the Nexø II rocket, this function will be replaced by remote control from Vostok.
Battery unit schematic: internal regulator, output switches and communication circuit. Illustration: Peter Scott.
At the moment we are finalizing the schematic capture and preparing circuit board layout. Especially the battery unit will be a challenge, as the allocated space in the avionics bay is the same as on Nexø I, and we will have to fit three boards where we had the much simpler single board circuit before.
5 Comments
BRETHIOT François · 18th December 2016 at 3:44 pm
Hello,
50 pcs 4,5Ah cells is very heavy when we do rocket, even this testing one is not optimized to get highest performance but check that all systems works perfect.
Why not use Lipo batteries, much more capacity for lower weight and fast charging ability.
Best rgds
François
Bo Braendstrup · 18th December 2016 at 5:38 pm
There’s room for improvement in the weight budget, but I honestly don’t remember the rationale behind our choice of battery technology for the Nexø rockets.
When choosing technology we also have to consider performance in low ambient pressure and mechanical robustness, and in the future we may choose lithium based batteries.
/Bo.
Titi Kinn · 22nd December 2016 at 1:40 am
NiMh are safer and it makes sense using them for that purpose, but if you can have the batteries in a secure case, i believe the best batteries today are the Li-ion 18600 batteries. you would need a battery management system which you can get COTS systems for fairly easily. They batteries are not that much more expensive and have the highest energy density batteries currently.
Great work otherwise, and your answer above answers why you decided to go against lithium based batteries.
Thanks
Alaba Baju · 14th May 2017 at 8:08 pm
Hi!,
Great project, I love what you guys are doing. I know you may already have reasons for these, but I just want to contribute some points. Firstly, did you consider using thicker cables? That would have been the first thing I would consider when I suffer from voltage loss that I feel is too high due to the resistance of a long cable. Since it’s on a launch pad I guess weight is not a factor. Is it cost? I guess you have reasons, it would be great to put it in the article so curious readers can find out. Secondly, when you revisit the battery chemistry choice, if you haven’t before, try considering LiFePO4. They are safe and have better energy density than NiMH. I have some and coincidentally the PIC16F1769 too 🙂 . Good luck and God bless!
Bo Braendstrup · 16th May 2017 at 7:51 pm
We doubled the wire gauge, with only marginally better results. The problems listed prevailed. Having separate ground returns would probably have solved most problems, but the connector didn’t have enough pins for this.
Any lithium chemistry is a strong contender for our future choice, but we have no preference at the moment.
Thank you for your well wishes.
Comments are closed.