Thread Tools
Jan 15, 2003, 02:54 PM
Registered User
Thread OP

Advantages of Higher Voltage & Lower Current


The concept of feeding higher voltage and lower current to e-flight motors has been discussed in many different threads recently. Proponents say one advantage is extended brush life, and perhaps overall motor life as well. I thought this subject deserved its own thread, instead of being lost in the middle of discussions about different aircraft or cell types.

Actually, the fast rise in the popularity of lithium cells is probably the motivating factor in the increased interest in +V/-A discussions. With LiPo cells rated at 3.7V each, combined with the fact that they are generally not as tolerant of high current draws as NiCd and NiMH, the concept of moving up from 2- to 3-cell LiPo packs is being tried by a number of people. With the light weight of the LiPo packs, they represent a smaller percentage of all-up weight than typical NiCd and NiMH packs, so the addition of a 3rd cell does not represent as great a weight penalty overall. But what is the best way to compensate for the increased voltage to keep current levels down?

A discussion about mixed results in using 3-cell LiPo packs with GWS IPS applications led me to run a number of hypothetical Speed 280 and 400 combinations through P-Calc. Starting with a 6-cell NiMH pack, I used a typical gearing/prop combo for a 280 or 400 to set a baseline. Figuring that most people would want to keep the same prop size for a given aircraft, I upped the cell count from 6 to 9 for a 50% voltage increase and added 50% to the gearing, i.e. 4:1 on 6 cells and 6:1 on 9 cells. Using the same size prop, current loading stayed about the same, but rpm and thrust were substantially up. Then I tried increasing the gearing by 100%, i.e. 4:1 on 6 cells and 8:1 on 9 cells. Again, using the same size prop, current was substantially reduced with rpm and thrust remaining roughly the same as the 6-cell configuration.

So, one potential advantage with 50% more gearing is to get an improvement in thrust at about the same current and only adding the weight of 1 cell. With 100% more gearing and current cut in half, a potential advantage might be to use 3 smaller cells in place of 2 larger ones for a small overall weight reduction with roughly the same amount of thrust. Of course all of this is hypothetical based on an imperfect *Calc program, and may or may not be useful to many people.

What I think might be of interest to many people would be extending the lives of their brushed motors, especially the smaller ones such as the GWS IPS motor, which typically has a pretty short lifespan. I don't have the answer to this, I'm not even sure I understand the theory of why higher voltage and lower current are easier on brushes. Since amps times volts equals watts, and watts in both configurations remain about the same, what exactly is it about increasing voltage and reducing current that results in less stress on the brushes?

I hope this thread will serve as a catalyst for discussion on this subject, as well as a place to post results of any experiments that anyone may try.
Sign up now
to remove ads between posts
Jan 15, 2003, 04:06 PM
Registered User
Steve McBride's Avatar
I'm just waiting for someone to come out with a speed control and motor line capable of 4-6 or more LiPoly cells and a BEC capable of handeling it and 4 -5 servos.

No doubt we need to start thinking higher voltage/lower current to take full advantage of the new cell technology.

Steve
Jan 15, 2003, 04:26 PM
Registered User
Andy W's Avatar
Indeed. But I wonder, can you safely charge many LiPo cells in series as we do with NiCd/NiMH? Charging may become an issue..
..a
Jan 15, 2003, 04:39 PM
Registered User
Steve McBride's Avatar
Good point - for the low end stuff (low current) we could look at a balanced charger (I am not sure of that name) where there is some sort of tap between each cell pair so that the cells are each peaked properly. Additionally, again for the small stuff, there are some nice small charge monitors which could be attached to each cell - still may require some funky wiring to acheive a good balance on the pack.

My head hurts - going back to work......

Steve
Jan 15, 2003, 05:57 PM
Fixed Wing Fanatic
Jim Walker's Avatar
As I was reading your post, (with great interest I might add), I was thinking the exact thought with which you finished the post. To wit, watts in is watts in whether it's 10 amps @ 10 volts or 1 amp @ 100volts. Thinking about it theoretically, an amp is a representation of a quantity of electrons. So if you get your watts by less amps with more voltage, you're moving less electrons faster. To use my example; 1 amp is 1/10 the electrons moving 10 times faster at the 100 volts. Could it be that less electrons having to jump the infitesimal gap between brush and commutator at an icreased speed causes less arcing which is one of the major causes of power loss and component deterioration?

But wait, 1 ampere is also a measure of time. It measures how many electrons pass a given point during a given time period so we're back to watts is watts. Is it documented and proven that more volts and less amps for the same watts causes less arcing? This interests me for another reason that is related. Why do higher turn motors get better efficiency than their low turn, hotter wound brethren? I've never understood that even though it is well documented by many e-flight experts. The relationship to this subject is that lower turn motors need more cells/voltage to get in their happy spot.

Great subject! More input please as I would like to understand these relationships better!
Jan 15, 2003, 07:17 PM
Registered User
Thread OP
Honestly, I don't think we have to worry too much about chargers. The speed with which new chargers are being designed, produced and brought to market for the newly booming LiPo business is simply amazing. As far as anything over 3-cell packs, I think the electronics to support that won't be too far off in the future. Electroman at FMA Direct is already hard at work on this, and you know that he's not alone. What we can learn from comparing optimized 2- and 3-cell setups will give us a clue of what we can achieve with packs containing 4 or more cells in series.

Jim, you ask some good questions that are way beyond me. I'm hoping that some of the EE types will discover this thread and give us some sound engineering input on this.
Jan 15, 2003, 08:19 PM
Old Desert Rat
Arizona Chuck's Avatar
I don't think there isn't any question of going to a higher voltage as we find motor that work good.
We went up from the 7 cell trap to get more power, I don't see why that doesn't apply to the LiPo's. Now I can go to 37 volts with only 10 cells compared to 37 NiMH cells.
10/1080 at C5= 5 amps at 37 volts =185 watts for 7.7 oz of batteries. The NiMH would weight maybe 10 oz at 1/3 the capacity.
Time will tell about charging in series, Troy and Fred have done a lot of test and say it is OK. I wish we could charge in parallel and fly in series.
Jan 15, 2003, 09:12 PM
DNA
DNA
registered user
DNA's Avatar
If a motor is designed for high voltage and low current then that
is one thing. If you try "feeding higher voltage and lower current
to e-flight motors" that aren't designed for high voltage, you will
naturally increase the rpms on the motor. Higher rpms work better
with ball bearings.

If someone would tell me how you can get "extended brush life"
when a faster spinning comm causes more friction on the brushes,
I would be interested in knowing how to do that. Maybe I could
use that technique when sanding balsa. If I sand balsa at 60 strokes
per minute, it would take longer to sand than if I used 120 strokes
per minute. So the 120 spm will wear down the balsa faster than
60 spm with the same amount of applied friction.

Increasing the number of cells and increasing the gear ratio so you
can use the same prop doesn't decrease the current much, but
it Does increases the rpms. The motor will wear out sooner, not later.
Jan 15, 2003, 09:17 PM
Motors beat engines!
Ok, a couple of things:

Amps passing through a wire creates heat. ( electrical friction).

Volts alone have no movement and create no heat.

The same is true for amps jumping a gap. ( brushes).

Think about arc welders. To increase the heat for welding thicker objects, amps are increased, not volts. ( although they often follow along with the amps.)

Passing more amps through the motors brushes is literaly trying to arc weld them to the comm, creating pitting and wearing them out fast. Just increasing volts does not do this.


A low wind high kv motor creates less torque than a low kv high wind does because less winds means less magnetic field so therefore less torque.

The low wind motor compensates by spinning faster, making its weaker torque pulses more often for a potential increase in power.

The advantage to this method is that you can use higher amps to compensate for lower volts ( cell count), allowing lighter weight battery paks. These motors would also work well at higher cell counts till you hit the motors physical rpm limit. ( boom!)

Another downside of high volts low wind is that you could get into some very big gear reductions to get to normal prop speeds. Much easier to just use a high wind motor. A bit more efficient too. ( more torque again plus less friction from rpms)

The advantages to higher voltage lower amps:

1) less current means longer flight times

2) less current means potentially even smaller, lighter batteries

3) less current means less brush wear.

4) less current means motors running closer to their most eff. amperage, ( instead of being pushed to high amps for more power), so cooler motor and better output watts for the input watts.

I think the biggest short term uses are with 7.2 volt motors on 3 cells. This is essentially the same as running them on 10 nicad cells but with less weight and better duration.

An example I mentioned in a thread here recently was my buddy Randy's homemade cub clone. About 39 span and 21 oz running a gws 300 at 5.33:1 with 10x8. Pulls 6.0 amps on his 2s3p 950 surplus LI's, motor stays cool, power is good, and flight times are around 25 minutes. He has quite a few flights on it so far, and its holding up well.

One I'm trying out now, also posted elsewhere is my pico f Aileron, was running two qualcom cells, ips A with 10x8 for 2.4 amps. I just changed out to ips C and added a third cell with the same prop. Amps dropped to 2.3, motor seems to run about the same temp as before, but power is amazing, and flight times are actually up despite the huge infusion of power.

Time will tell if this setup will last, but since I'm still only doing about 30k rpm on the motor, and temps are not higher, I think it should hold up. We will see.


Dean in Milwaukee
Jan 15, 2003, 09:24 PM
Motors beat engines!
DNA, obviously your right that a faster moving motor will wear it brushes faster than a slower moving one, but don't forget that most of the wear actually occurs from arcing and pitting, not from friction between com and brush.
The brushes are eroded ( burned) away, not worn away.

Think about the lowly sp400 motor in its industrial applications.

Voltage, amperage and load are all low, but hours are very high. ( thousands of hours?) and they keep right on running. Thats one heck of a lot of total revolutions compared to anything e-flight use will ever put on them.


Dean in Milwaukee
Jan 15, 2003, 09:38 PM
Registered User
I have been thinking about this exact subject for a while, especially since the Lipo batteries have arrived. There are several different subjects already mentioned and I have a couple to add.

Starting with charging a lot of Lipo in series. Someone mentioned tapping out each cell junction in the stack. Starting with that idea there are two possibilities. One is to use individual 5V isolated power supplies, either DC/DC converters or AC/DC supplies feeding any one of the charger IC's for single cells. There are a lot of them available. The other idea is to feed the entire stack with a fixed charge current and place a precision shunt regulator around each cell. The shunt regulator would take more of the fixed current as the cell approached full charge until all of the current went thru the shunt at full charge. I have sketched some circuits for both and neither is very complex.

About the efficiency of higher turn motors. Every text I have ever read about motors points out that a lot of turns with finer wire packs the limited winding space better than a few turns of coarse wire. This gives a relatively lower winding resistance and better performance.

One real obivous advantage of higher voltage and lower amps is lighter wiring and smaller connectors.

The regulator to obtain 5V from high voltage has already been done by the folks that make the Ultimate BEC. The circuit is a standard buck switcher and the parts are made by many companies.

Another advantage is the control FET's in speed controls. It now takes several FET's in parallel to obtain low enough resistance. Changing the FET to a unit with 40V or greater rating raises the on resistance slightly but the current is lower and the power loss which is the current squared times the resistance is low for a single FET. This would lead to smaller, cheaper controls, especially brushless controls which have three control channels.

The upper limit on voltage is somewhere in the 40V to 60V area for safety reasons. Above this level the voltages are starting to become a health risk. The automobile manufacturers have settled on 42V for future car electrical systems which is compromise between safety and moving more power on less copper.

Interesting stuff all!
Lynn
Jan 15, 2003, 10:13 PM
Registered User
Thread OP
DNA, I can understand that more rpm would tend to produce some additional wear. However, I'm not sure if this might not be counter-balanced by lower loading on the motor due to the lower gearing. For instance, I'm assuming that the little IPS motors spin at higher rpm in the EDF-50 configuration than when spinning a prop. In the EDF-50, GWS recommends running up to 10.8V with the 7.2V motor, which has to produce a lot of rpm. Also, as Dean mentions, it is the arcing on the brushes from high current that I thought did more damage than just the friction.

Dean, that's a lot of good information that you posted. Not only that, but I can actually understand a lot of it. Your experiment with the IPS-C and 10x8 prop @ 11.1V will be very interesting. I was thinking of trying something a little more conservative than that -- an IPS-D with 10x4.7 prop @11.1V on a Mini IFOish. If we can gather enough data in one place, it will make it easier for everyone to decide which way to go.

Lynn, thanks for your contribution. I hadn't thought about the motor winding itself playing a role in the efficiency.
Jan 15, 2003, 10:13 PM
DNA
DNA
registered user
DNA's Avatar
"but don't forget that most of the wear actually occurs from arcing and pitting,"

And don't forget that the faster the comm spins, the more sparks that must
be delivered by the brushes. Where does the pitting occur...on the brushes
or the comm?

"Think about the lowly sp400 motor in its industrial applications.
Voltage, amperage and load are all low, but hours are very high."

Of course, and at low voltage, current and load, so are the rpms,
so the motor should last a very long time.

"running a gws 300 at 5.33:1 with 10x8. Pulls 6.0 amps on his
2s3p 950 surplus LI's, and its holding up well. "

I don't see how this qualifies for a high voltage, low current application.
The motor is only running on 7 volts. What's the big deal about that?
Jan 15, 2003, 11:38 PM
Motors beat engines!
"Dean, that's a lot of good information that you posted. Not only that, but I can actually understand a lot of it."


Thanks, it took me a while to get it written in an intelligable format.


DNA: Your right about industrial sp400's turning low rpm due to the factors mentioned, but bear in mind that they also will run about 8800 hours each year. Even at 2 or 3 volts, that is one heck of a lot of rpms, and still they keep right on running.

For me, thats proof that most of the brush wear is from arcing.

"And don't forget that the faster the comm spins, the more sparks that must
be delivered by the brushes. Where does the pitting occur...on the brushes
or the comm?


The amount of sparking is directly related both to the rpm, but its intensity is related to the amps. Run a motor once under load, observing the sparking, and then start to throttle back. The sparking practically dissapears as the load and amps are reduced.


Dean in Milwaukee
Jan 15, 2003, 11:42 PM
Motors beat engines!
""running a gws 300 at 5.33:1 with 10x8. Pulls 6.0 amps on his
2s3p 950 surplus LI's, and its holding up well. "

I don't see how this qualifies for a high voltage, low current application.
The motor is only running on 7 volts. What's the big deal about that?"



Oops, sorry, 3s2p. About 60 watts into that sp300, and it really seems to be doing well.


Dean in Milwaukee


Quick Reply
Message:

Thread Tools