Originally Posted by JrcSeller

My iCharger has a 'storage' mode charge option. I use this immediately after doing a full ballanced charge and it brings the 2S pack voltage down to the recommended storage charge of 6.6 volts. In observing its discharge amounts, there appears to be only 40 mah removed to get it down to this voltage, I thus fly with it that way and so I don't need to ever worry about those peak spikes created during the charge and certainly don't miss the 40 mah for flight.

Ditto.

This is how I did when powering an electric 1/4 scale Cap232. The way I found to not fry a HV ESC using 18S A123. A little discharge after top off would keep the voltage in the "legal" zone (<60 DCV).
I've never thought to do the same in my DLGs. Once I have my wing with Hyperion D09, I may have to dischage the LiFe pack hot from charger, if I decide not to use regulator and live on the edge...
Tai

Originally Posted by Silent-AV8R

Both are true. So it is more correct to say that I am choosing from a small subset of the best performing servos out there AND I feel safe with exceeding manufacturers spec voltage limits.

I use Futaba S3150 and S3155 servos on unregulated 2S LiFePO4.

The funny thing is, one of these (or both?) are not even rated for more than FOUR NiXx cells!

If you want my opinion (or not; too bad!), RDS is normally draggy (and so are some folks conventional linkages), and 6.5-6.6V with a servo with zero deadband and a draggy linkage, is the problem....
Its just a theory of mine so far though. Likely always will be.

R,
Target

Originally Posted by target

The funny thing is, one of these (or both?) are not even rated for more than FOUR NiXx cells!

Neither is officially rated for 6 volts. But I have not told them that so they seem to work just fine. Stand behind me at launch and you won't hear a peep out of them. They generally do not buzz.


Your drag theory has merit. The mechanical load on the servo determines its current demand. The more load (drag, binding, etc.) you put on it in addition to the flight loads, the higher the current demand will be. In fact, I suspect that it is possible for the mechanical loads to exceed the flight loads. These little servos require great care since they lack the ability to dissipate excess heat efficiently due to their small size.

I'm planning on using this unit in a new SRTL build:

http://www.hkm-models.com/Emcotec%20Switches.html

The SRTL has no hatches, and you must tape/untape the wing joint to get to the innards. This is a combination voltage regulator and magnetic solid-state switch.

I have a similar, non VR unit in another SRTL and it works beautifully. It has a bright LED ON indicator, and you just place the magnet over the right spot, and the switch is on, the other side of the LED is off.

Mechanical switches are a common failure point that ruins your day all the way. Plus, I can tape the wing on and leave it there all day. Except for ballast changes...


WinchDoc

yes, 2 SRTLs...

While I understand the mindset of "it's worked so far so everything must be fine" I can't accept it. With typical manufacturing tolerances you may get lucky by hand-picking parts to work outside their ratings. But...when you run outside their design limits you're running with a system that has a calculable lower level of reliability.

I don't gamble with overvoltage in my planes. I add enough "reliability" stress of my own when I fly them.

Chris B.

Originally Posted by winchdoc

I'm planning on using this unit in a new SRTL build:

http://www.hkm-models.com/Emcotec%20Switches.html

I've been using these switches for a few years now and love them. Very cool, solid state, fail closed (never had one fail) and just flat out work. Emcotec makes a variety of great products and it is very nice that HKM-USA makes them easily available.

Originally Posted by SoaringDude

While I understand the mindset of "it's worked so far so everything must be fine" I can't accept it.

That is fine, but that is also not my mindset. I view it as I have no empirical evidence that shows me that the set up I am using has any issues. Until I have evidence to the contrary there is no reason not to keep doing hat I have been doing.

With typical manufacturing tolerances you may get lucky by hand-picking parts to work outside their ratings.

My equipment is not "hand picked". I place an order that comes from Tower and I get what they ship me, just like everyone else. It is "hand picked" only in the sense that some person in a warehouse grabs the boxes off the shelf.

But...when you run outside their design limits you're running with a system that has a calculable lower level of reliability.

Care to take a shot at calculating it then? I have no such evidence that it is a quantifiable failure rate, but if you do I would love to see it. I am a big fan of actual facts and data.

You know, I can understand all the buzz around LiFePo4 ease of use, by this I mean no need to really worry about self discharge great constant voltage under load.

But look at that, ok the battery is lighter, so that mean more lead in the nose of the plane to balance it no ?

Then look at all those risk of failure we add in the system ( regulator, more plug or soldering points).

I take care of my Nimh and I never had a failure in flight. With they're weight they help to balance the plane, simple system, less risks.

So why not just use some of those HV servos ( 378, hyperion has a great 15mm FMD 16 HV or even they 095 wing servo actually all hyperion except the 9 and 11mm are rated for direct use with LiFePo4) and just plug a LiFePo4 directly in the system.

I use LifePo4 in some of my big 4m electric planes so don't get me wrong I love them ( plane is Valenta Volcano and also in a e-Pike Superior) where I need to save weight and I have ton of room in the fuselage to even double the system but in small F3X fuselage... mmmm that's tight.

Thierry

Originally Posted by tiju000

You know, I can understand all the buzz around LiFePo4 ease of use, by this I mean no need to really worry about self discharge great constant voltage under load.

But look at that, ok the battery is lighter, so that mean more lead in the nose of the plane to balance it no ?

Then look at all those risk of failure we add in the system ( regulator, more plug or soldering points).

I take care of my Nimh and I never had a failure in flight. With they're weight they help to balance the plane, simple system, less risks.

No regulators here with LiFe and A123 packs, no switches, just a single Deans polarized micro plug with 18-20ga pack wiring. Compared to my prior system of five-cell NiMHs and a switch with 22ga wire, I feel there is even less risk of failure, plus the stable output of the lithium-based packs.

Even though I never experienced an in-flight NiMH issue, there were plenty of times catching false charger peaks or dealing with the self-discharge and cycling issues for packs that did not see continuous use. Total PITA. That is what drove me off of nickel-based flight packs.

I'm using the the JR HV servos wherever possible, but even the standard series rated for 6v have not been a problem. Lithium-based packs are lighter...but...so? Nickel packs still work, but for me the disadvantages are enough to dump them, given the other options now available.

Tom

I agree with Tom and others that the pro's outweigh the con's with regard to use of LiFe over NiMH, even if I decide I want a regulator in the system.

NiMH batteries are frequently tempramental.

R,
Target

Originally Posted by SoaringDude

While I understand the mindset of "it's worked so far so everything must be fine" I can't accept it. With typical manufacturing tolerances you may get lucky by hand-picking parts to work outside their ratings.

Chris, just for the record, I live in Lima, Ohio, not Champaign, IL. When I order servos or whatever, my stuff is pulled from stock just as your's would be, NO GROOMING THE ORDERS. I have been flying elevated voltage packs for a long time, 5-cell packs, 3S LiPo with CC Reg, and now 2S LiFe's. So, I guess that is a lot of hours turned on and no servo failures from non mechanical failures. I remember that when we started to use 5-cell packs, the same line was being used that we would burn up our servos and all devil would break out, it did not happen.

Chris, I would certainly say use what makes ya happy, but the myth that elevated voltages in at least Airtronics and JR servos (the ones I have used for a long time, especially JR specifically for five years) will lead to disaster is a myth, we do not use them that hard and that long in the great scheme of things.

Marc

Originally Posted by Silent-AV8R

Care to take a shot at calculating it then? I have no such evidence that it is a quantifiable failure rate, but if you do I would love to see it. I am a big fan of actual facts and data.

BTW, my "hand-picking" comment was misunderstood. When you buy a product I understand you get "what's there" and that you're not actually doing any testing beforehand. Onward to the main point.

The fact that failure rates increase when electronic components are subjected to overvoltage stress is a well-known mechanism. Google "component overvoltage stress effects" for many good references on the first page of search results. Even if your servos work now with excessive voltage, it is quite possible that they will fail much sooner than normal.

Here's one non-silicon example: tantalum capacitors. Most all RC radio equipment and servo electronics circuits use them. The Failure Rate of tantalum capacitors is given by the following equation :

F = FU x FT x FR x FB

where
FU = Applied Voltage / Rated Voltage
FT = Operating Temperature
FR = Series Resistance. FB = Basic Failure rate (by capacitor type)

Reference 1
Reference 2, page 13

The more you push the voltage the higher the failure rate. So if your voltage-stressed servos survived the infant mortality part of their life curve, then you still stand a chance that they will fail earlier than their normal lifetime. I could post many more graphs, charts, and online reliability calculators.

I do think the best overall reliability strategy is to choose the best voltage regulator/reducer and use it (or use HV servos). That's why I've offered to do stress tests on regulators that you folks send me.

Chris B.

The one regulator solution that still intrigues me is the Greentronics RegMax. Two separate single-cell inputs, voltage boost to ~5.5v, and two outputs. Works even if one cell fails. Seems like a good option if a regulator is indeed required.

http://www.greentronic.eu/en/regmax.html

Tom

In the manual, it states that your capacity is really double what each cell is, for an 1100 cell A123, that means you have 2200 MaH capacity.
Essentially the regulator somehow boosts the voltage by taking a hit in the 2P capacity rating (states the efficiency is @84%, IIRC).....
Very odd.
And it looks like the price has gone up substantially since I looked last.

Still a switching reg, so, not as radio silent as a linear reg.

R,
Target

Originally Posted by SoaringDude

The fact that failure rates increase when electronic components are subjected to overvoltage stress is a well-known mechanism.

What exactly constitutes a significant enough "over voltage" in a RC model servo. Your example is very nice in a theoretical sense, butin actual observed use it appears that its predictions are not consistent with observations.

The bottom line question is how much does say a 0.5 volt "over voltage" situation increase the failure rate of hobby RC servos? How much does it increase for a 1 volt or 1.25 volt "over" voltage situation?

So far, based on my direct observations and that of many others, there does not appear to be a significant failure rate increase for Airtronics, JR, or Futaba servos. MKS appears to be the one more prone to issues with small amounts of over voltage.

I am reminded of the saying that a good engineer will never let observed reality get in the way of a good theorty.

Having said all that, people need to look at the reports of the numerous users who have had success and decide for themselves what their comfort level is. That is what makes this hobby so fun, we can each do what we need to do to make ourselves happy