LiFe packs

[Martin McIntosh]

I have put stuff about these cells on other forum threads but they still mystify me a bit.

I just cannot understand why a model using these apparently draws much less from a pack than an equivalent NiMh pack.

I have replaced two Tx packs with these and they just seem to last forever.

I also replaced a twin pack via a failover switch on a model with 2x 1800 LiFe cells and they drew just 60mA/hr each as against about 180mA/hr on the original NiMh`s.

Today I did some work on a simple Magna model which had been switched on and flown occasionally for over an hour but on topping up the 700 pack it only took 10mA/hr!

A discharge test at 300mA gave 640mA/hr left in it!

This seems too good to be true so any ideas out there?

[Denis Watkins]

A mystery and impossible Martin, as you suspect. There are many boffs on this site who will offer up some explanation, but I do know that digital input output readout equipment that we use is "thereabouts" accurate.

I always view pouring in and pouring out current from storage as into, and from a bucket.

A small charger has difficulty pressurising the current needed to fill up a large pack. E.g. a 60 ma wall charger, charging a 5000ma 3s, would be an extreme example, it would never fully charge.

I feel differing chemistries would respond "differently" to being "pumped up"

I think that this is what is happening here

[Phil 9]

in addition to the OP may I ask how to check lremaining charge at the field

[Shaunie]

The energy conversion efficiency of Nimhs is much lower than Lithium chemistries. For instance a 2000 mAh Nimh charging cycle would be 200 mA for 14-16 hours which calculates out to 2800-3200 mAh. With Lithium chemistries you get back out almost all the energy you put in, not sure how much you lose, I've not done any tests but it would be fairly easy to set up. I suspect much of this is down to the internal resistance.

Shaunie.

[Peter Jenkins]

Could it be due to a higher internal resistance for the NiMhs and therefore a higher internal losses within the cells. I know that as LiPos age their IR goes up and that's when they start to puff and heat up or overheat.

[Bob Cotsford]

I put it down in part to the fact that LiFe cells don't self-discharge to anywhere near the same extent that Nixx technologies do so you are seeing what the radio took out, not what the radio took out plus a large self-discharge value. I've found 75ma per pack pretty much average for medium/large model radio packs going up to 350 when also driving electric retracts, but I still try to stick to 2000mAh packs for simple models just because of their ability to supply a higher current. 3000mAh seems good with retracts.

[Martin McIntosh]

Thanks for the replies chaps.

It would seem that internal resistance is playing a part here but not self discharge since the tests were conducted back at home just after flying. The NiMh`s in each case are Eneloop 1900 Mignon type which are the best I have found and used in everything until I started to convert to LiFe`s.

Eneloops do have a full charge when new but after a charge or two they seem to self discharge just as much as any other hydride.

This brings me to another point which even a charger manufacturer could not answer. It is generally accepted that you need to put in about 1.4 times what you get out, but all of my chargers show the charge/discharge to be roughly the same.

The 700 pack in question above, by the way, refuses to take more than 585mA/hr when charged at 400mA despite having given out 640mA/hr. I shall cycle it again, this time with a discharge at 300mA and a recharge at 1C. I`ll post the results.

[Martin Harris - Moderator]

I've certainly noticed this phenomena with LiFe cells and have done discharge tests to reassure myself that the small mAh figures put back into the cells after a flying session have actually recharged the batteries fully!

Perceived wisdom from the choice of using 5 nickel cells over 4 in order to speed up servo response has always been that if you increase pack voltage you increase current and therefore the rate of discharge goes up but perhaps there's an alternative way to look at this? If you need a certain amount of energy to deflect a control surface and it's supplied at a higher voltage then the current required will drop assuming the same power requirement. Therefore the significantly higher voltage of LiFe cells may well give a noticeable effect on battery Ah recharge amounts.

There may be other variables at play here - the use of digital servos for instance - but energy density seems to be a possible explanation.

Edited By Martin Harris on 10/09/2016 11:33:43

[Steve Balaam]

I recently fitted out a 30CC plane with 5 Digital Servo's and 2 analogue servo's . I used 2 X 2300MaH LiFe batteries through a power distribution box the box also supplies output to the ignition unit.

Whilst testing I set the distribution box to alarm if the input voltage fell below 6 Volts.

When stirring the sticks the voltage alarm was constantly alarming which I wasn't happy with and have gone to 2 X LiPo with no problems. I tried 3 different pairs of Life's with exactly the same results. The total max current being drawn was 5 Amps so I guess the Life's couldn't hold their voltage under 5A load. I tried the same LiFe packs in a smaller plane with 4 X Mini size digitals with no problem at all.

So after this experience I'm not confident using LiFe batteries in the larger more expensive planes.

Steve B

[Martin Harris - Moderator]

Does the power box actually monitor the voltage at the input or output? It might be that it utilises silicon diodes which will give a 0.6 - 0.7 volt drop which would mean the output would be very close to the alarm threshold when delivering power at the nominal voltage.

Edited By Martin Harris on 10/09/2016 12:43:52

[Martin McIntosh]

The 700 pack gave out 644mA/hr despite only 585 going in according to an Overlander charger. It then took 606 at 1C. The mind boggles even more. I suspect that the chargers are telling porkies.

M.H., Increasing the voltage to a servo will certainly also increase the speed and current draw, not to be confused with electric propulsion when the drain can be decreased for the same output power but only if the load is reduced, i.e. a smaller prop. at higher revs.

S.B., A LiPo at a nominal 7.4V is very unlikely to drop below the 6V alarm setting, but drop it will. I would have thought 5.5V would maybe be better but I do not use a power box. 5A unloaded sounds rather on the high side.

I do however have a turbine with a similar unit by Jeti running on 2x 2S lipos. Don`t know if it alarmed or not because I did not build this.

[Steve Balaam]

The box regulates the Rx output at 5V. The CDI output can be set to the users requirements. Whatever voltage is presented at the input goes straight to the servo's.. The instructions say you can freely set the alarm input voltage, I have a circuit block diagram but it's not possible to tell if there are any diode's in circuit. As a precaution I've now gone to 2 X LiPo's with regulator on one airframe and 2 X 5 Cell NiMH with regulator on another.

Steve B

[Martin Harris - Moderator]

Posted by Martin McIntosh on 10/09/2016 15:53:05:

M.H., Increasing the voltage to a servo will certainly also increase the speed and current draw, not to be confused with electric propulsion when the drain can be decreased for the same output power but only if the load is reduced, i.e. a smaller prop. at higher revs.

Agreed, that is the generally accepted theory but I'm looking for an explanation for what seems to be a fairly widely experienced anomaly.

So...thinking out loud:

If we are moving the sticks constantly then yes, current draw is higher with a higher supply voltage but is that the reality? My flying style, even when doing fairly vigorous aerobatics, will involve periods of constant stick position or very minor adjustments and the servos will reach the commanded positions faster at a higher voltage so the moving duration will be shorter. In the meantime, we have a power source which, at say 700mAh, contains a nominal 4620 mWh of energy (700mAh x 6.6V) as opposed to a 700mAh 4 cell NiXX with a nominal 3360 mWh. If we use a similar amount of energy during a flight, the LiFe battery will use a smaller percentage of its capacity - therefore requiring fewer mAh to recharge it.

As I've said, this goes against conventional wisdom but it might be worth considering as an explanation as to why several people have all found that so little capacity is being taken out of their LiFe batteries...

[Bob Cotsford]

I think Martin has a valid point as most of us are not getting anywhere near to loading down servos in flight, hence the amount of work being done is the same whether powered by NI or Li batteries. Maybe in a high performance 3D model being thrashed within an inch of it's life the extra power available would be called on, but that's not the case for most of us.

I've also noticed that LiFe packs do show a drop to 6.3 or even 6.2v under load but as I'm using HK packs with quite light wiring I wonder whether the drop is from wiring resistance as much as anything? It's certainly not enough to drop the regulated output (where I'm using one) even with 5 or 6 10Kg servos.