I wrote a huge post on this, but it's been "lost" somehow so all I'll say is this:

Many people swear by the eneloop pack for RX duties, but I prefer A123 packs for heavy duty:

http://www.rclipos.com/A123_Packs.htm

1) MUCH higher C rating
2) 5C charge rate
3) no regulator required for (normal) digital servos
4) shelf life as good as any lithium battery (heaps)
5) cheaper than eneloop packs
6) huge cycle life
7) low fire risk

I'll still use eneloop for ignition though as 6.6v is too much for most

micahel.

My close reading of the AR9100 showed that the battery voltage is passed directly to the Servo buss after arbitration, but without any regulation, so although the LiPo solution gave the lowest internal resistance a 2S LiPo was putting up to 8v on the servos and Hitec says this may reduce servo life by 20% so I went back to Eneloop NiMH 5 cell solution and serial regulators would increase the component count and the probability of system failure, defeating the simplicity of the AR9100 solution.

A123 sounds good.

Lonz...

I use 2S LiPo with a 3 amp UBEC from Hobby King regulating the voltage to 5v for RCEXCL ignition and Opto isolated switch from Model Aviation - Coffs Harbour <$20.

Lonz...

My UBECs from Hobby King went all into the bin.

Had a 5A UBEC with 2s LiPo in a glider with 4 servos (2 digital, 2 analog) and stirring the sticks caused the voltage to drop enough for a receiver reboot. Measured peak current was 3.2A - way below the rated 5A!

Luckily I discovered this on the bench before the maiden flight but this made me very suspicious of the cheap Chinese BECs and I have switched to better quality BECs and more thorough tests including measurement of current draw on the bench and sometimes during flight.

Have looked at using A123 but the 6.6V are still on the high side for most of my cheaper servos. But will give it a try next time.

P.

Agree.

My Hitech digital servos each stall at about 20 amps.

I don't think a UBEC is even in the ball park for that sort of load. A large model may have digitals 2*aileron, 2*elevator, 2*flaps, and n*for rudder, plus normal on throttle and retracts.

So a Spectrum AR9100 with its 50amp servo bus and EC3 connectors is good as the current does not go thru the switch, but the choice of good batteries is still the issue without a regulator etc......so if Eneloops fold with these higher currents then some 2000-4000mah Ni-cads seem to be a wise choice ( good voltage range, lowest internal resistance )

Also the max stall current of Hitec servos I could find is just over 5A it can still add up to a lot (too much) power sucked through the receiver.
There are BECs capable of handling this but you won't find them on the Hobby City web site.

The Reactor series from http://www.scottgrayrc.com/ReactorX.html or Bec Pro from Castle Creations do deliver serious power.
The AR9100 is good in the power department but hampered by the Spectrum RF system.

For really serious equipment check out this http://www.weatronic.com/en/shop?page=s ... egory_id=1

Always thought Ni-Cads are at their limit long before that but I'm not using Ni-Cads and don't know their capabilities well enough to judge this.

P.

This should be a discussion based in engineering fact, published numbers and measurement.

If you don't know then please find out and look it up. Political opinion is better paced on other posts I think ( eg Flying Giants etc ). This electronic engineering information has been around long before we put men on the moon in the 60's

Fact: the internal resistance of NiMH is greater than that of an equivalent mah Ni-cad.

Measurement: if you put a clamp current meter ( non insertion ) on the plus wire of a system with a very good battery and NO switch or other circuit resistance then as you turn the output arm of a good Hitech digital servo then you can see the static current increase to over 20 amps.

Fact: any BEC or hobby voltage regulator has a higher series resistance and slower response time than a good Ni-cad.

Conclucion....our current breed of servos have been designed for about 6v max, you cannot reduce a 2+cell LiPo to 6v without significant series resistance = heat = voltage drop = Rx reset = crash in the larger models.

The lowest internal resistance and nominal 6v supply is only available from 5 cell Ni-cad packs.

......."but hampered by the Spectrum RF system" ?????

Ask Jim why his FAAST system crashed 4 large expensive models before he was able to get the supplier to send it back to Japan when THEY modified the hardware and software after THEY spent months of R&D on his equipment.

Who do YOU know who has had this sort of miserable performance from a " but hampered by the Spectrum RF system" ????

Lonz
=========================================

PS: electronic reliability is proportional to component count...the fewer the better...

Lonz wrote:This should be a discussion based in engineering fact, published numbers and measurement.

If you don't know then please find out and look it up. Political opinion is better paced on other posts I think ( eg Flying Giants etc ). This electronic engineering information has been around long before we put men on the moon in the 60's

Wow, haven't seen this part in the forum rules.

Lonz wrote:Fact: the internal resistance of NiMH is greater than that of an equivalent mah Ni-cad.

Measurement: if you put a clamp current meter ( non insertion ) on the plus wire of a system with a very good battery and NO switch or other circuit resistance then as you turn the output arm of a good Hitech digital servo then you can see the static current increase to over 20 amps.

I actually did look up the stall current of Hitech servos. As you don't specify what a "good Hitech digital servo" is (not following your own rules??) I went trough all the highest specced Hitec servos. While the stall current is not always published, it is available for some of them and the max value I could find was 5.4A.
But if you give me one of these unnamed "good Hitech digital servos" I will be happy to measure it and confirm your numbers.

Lonz wrote:Fact: any BEC or hobby voltage regulator has a higher series resistance and slower response time than a good Ni-cad.

Conclucion....our current breed of servos have been designed for about 6v max, you cannot reduce a 2+cell LiPo to 6v without significant series resistance = heat = voltage drop = Rx reset = crash in the larger models.

The lowest internal resistance and nominal 6v supply is only available from 5 cell Ni-cad packs.

Yes, NiCad can obviously deliver high currents. That was something I was not aware. Thanks for telling me that.

Lonz wrote:......."but hampered by the Spectrum RF system" ?????

Hampered may be a bit too strong but because you like facts....

Spektrum/JR DSM2 is the oldest of the 2.4GHz systems. It selects 2 channels at power up and stays on them until you turn it off again.

All other 2.4GHz mainstream systems employ frequency hopping. They use anywhere from 3 channels (cheap Chinese versions) up to 80 channels for the more sophisticated systems.

There is some obvious advantage from transmitting on 80 channels vs 2 channels. This doesn't make Spektrum a bad system (I use it myself) but it does mean that there is less headroom in unfavourable conditions for DMS2 compared to most other systems. There are quite a few well documented cases where Spektrum users were over-proportionally plagued by lockouts at big events where many RC systems are active at the same time. You can easily find the reports over on RCGroups.
There is also a good documentation of a DSM2 weakness here http://rcmodelreviews.com/dsm2flaw.shtml

We do have a rather "clean" (measured myself!) RF environment at Belrose and and I have not experienced any issues with Spektrum but the fact is, that DSM2 currently is technically the weakest solution.
Another fact is that JR is now abandoning DSM2 and switching over to a new frequency hopping system. I'm sure they do have some very real reasons for that.

Faults with certain modules or receivers can happen with any system. Every single one of the big manufacturers had also their fair share of issues. For example Futaba had their overheating problem and Spektrum/JR in their early time had brownout/rebooting issues and there are currently still issues with AR500 receivers not working properly.

Fact is that today most any system is "good enough" for average modellers and certainly much more reliable than anything from the pre-2.4Ghz area. Still, if I would start again today and with my current knowledge I would probably not choose a Spektrum DSM2 system.

Lonz wrote:PS: electronic reliability is proportional to component count...the fewer the better...

Which must make the AR9100 a very unreliable system with 3 additional remote receivers compared to a standard single Futaba receiver



Another fact is that I don't like the way this discussion is going. I like participating in friendly discussions and I'm happy if I can learn something on the way. Getting attacked immediately because I dare not blindly accepting all your statements is not what I'm looking for.
So I think it's best to consider this "discussion" closed. (At least from my end, you are certainly free to continue as long as you want )