Batteries

[Motopreserve]

I don't know. Clearly there is no industry standard. Didn't one of the more expensive brands get caught labelling some batteries at 65C, but they were really only good for 5C? I'm not even entirely sure how they come up with mAH ratings. I mean, if draining past 80% of the battery damages it, aren't they all over-rated?

Ive been talking to Bart about whether there is a way to productively and safely test various batteries. Not sure there is a sure-fire way to provide valuable data on these Lipos - but I'm open to suggestions.

Ill send end through some data specs on some of the batteries I have. I've been using several models of the ReadyMadeRC 4S versions - been happy with them, and they're not on the list yet. Also have some Turnigy Nano-Tech 6S I could send as well.

 

[Quinton]

Watts = Power = current squared x resistance = Voltage x Current
E = I * R
i.e., Volts = current x resistance. Units are typically amps X ohms

Energy density should be considering only the output power when compared to total battery weight and should not take in to account the battery size.

Now I remember why I gave up physics 30 years ago... ..whoosh

 

[R_Lefebvre]

Watts (Power) = Voltage x Current. Cells x Volts/Cell x Amps

Watt-Hours (Energy) = Voltage x Current-Time.

Due to industry convention, LiPos are measured in milli-Amp-Hours. So divide by 1000 to get Amp-Hours.

Then:

Watt-Hours = Cells x Volts/Cell x Amp-Hours

Obviously my spreadsheet assumes the manufacturer is honest on their mAH rating, which, we have no idea yet. If we really wanted to do this, we'd get a load bank, and load the batteries at some rate. Maybe assume a 15 minutes discharge time, as that is pretty average. Then measure voltage as the battery drains to 3.3V/cell or something. This would check the OEM mAH rating, as well as capture the real power delivery of the batteries. Low-C, high density batteries aren't going to be as much of a good deal if the voltage sags more under load, as the actual delivered power will be lower.

 

[R_Lefebvre]

Ive been talking to Bart about whether there is a way to productively and safely test various batteries. Not sure there is a sure-fire way to provide valuable data on these Lipos - but I'm open to suggestions.

I've got some very large resistor banks I could get from work, designed as braking resistors for large machinery with variable speed drives. I've been thinking of getting one, and then setting it up with a large FET, and a basic control system. This actually isn't a difficult problem to solve, just takes a little time and money.

It would also be great to scientifically test efficiency ratings on motors. Set up a dyno cell, and take the propeller variable out of the equation.

 

[jfro]

If you guys want to throw out some input on your favorite batteries, I'll see if I can add them.

Just need:

Name, xS ymAH, C rating, weight (excluding shipping packaging, obviously), dimensions, and price. Though the last two are technically not required, nice for comparison.

Thanks for the offer...

Nanotech Hardcase $112
Capacity[/B]: 8000mAh
Voltage[/B]: 6S1P / 6 Cell / 22.2V
Discharge[/B]: 25C Constant / 50C Burst
Weight[/B]: 1111g (including wire, plug & case)
Dimensions[/B]: 195x50x55mm[/B]

Gensace $82
Capacity: 5000mAh
Voltage: 14.8V
Size: 138.05 * 45.81 * 49.18mm (L * W * H)
Weight: 562g
Discharge: 40C Continuous, 80C Burst

Quadcopter $255
Capacity: 10,000+ mAh 6S1P
Voltage: 22.2V (6S)
Size: 165 x 55 x 63 mm (L*W*H)
Weight: 1358g
Discharge rate: 25C continuous; 50C maximum

TAttu $269
Capacity: 10000mAh
- Voltage: 22.2V
- Max Continuous Discharge: 25C (250A)
- Max Burst Discharge: 50C (500A)
- Weight: 1382g
- Dimensions: 168* 64 *59mm

Tattu $229
- Capacity: 8000mAh
- Voltage: 22.2V
- Max Continuous Discharge: 25C (200A)
- Max Burst Discharge: 50C (4009A)
- Weight:1194g
- Dimensions: 167*65*52.8mm (L * W * H)

Tattu $379
Capacity: 16000mAh
- Voltage:22.2V
- Max Continuous Discharge:15C (240A)
- Max Burst Discharge:30C(480A)
- Weight: 1931.7g
- Dimensions: 180*74*65 mm

 

[jfro]

Asking again, as I'm confused.

Landed with cells in the 3.55 range. Put 7438 mah back in. Going to redo test today on another battery. Have 6 and they are all pretty new. Resistance is under 2ish on all cells.

I'm confused because, 3.5+ range shouldn't be hurting my battery. Battery was not hot at landing, just warm. However, if 80% rule of 8000 mah is 6400 mah, then didn't I go substantially over the 6400 by putting 7428 mah back in?

 

[Quinton]

Asking again, as I'm confused.

Landed with cells in the 3.55 range. Put 7438 mah back in. Going to redo test today on another battery. Have 6 and they are all pretty new. Resistance is under 2ish on all cells.

I'm confused because, 3.5+ range shouldn't be hurting my battery. Battery was not hot at landing, just warm. However, if 80% rule of 8000 mah is 6400 mah, then didn't I go substantially over the 6400 by putting 7428 mah back in?

I'm no expert, but when and how did you measure the cells.
Was it 3.5 under load using telemetry reading, or was it 3.5 after you landed?

 

[jfro]

I'm no expert, but when and how did you measure the cells.
Was it 3.5 under load using telemetry reading, or was it 3.5 after you landed?

Readings were from my lipo alarm right at landing, then used a 2nd lipo tester about 1-2 minutes later. Cells were in the (edit out 5.2- 5.4) 3.52 to 3.55 range if I remember when I landed.

 

[Motopreserve]

Cells were in the 5.2- 5.4 range if I remember when I landed

Jfro:

???? Is is 3.5 or 5.2-5.4?

R_L:

i am going to be doing the motor testing for Bart/MRF. He sent some motors today. I'm set up to test the 'standard' way with props covering draw, thrust etc.

Please shoot me a PM (so as not to clutter this thread OT) with details of the motor testing you're describing. I'd love to be as thorough as possible.

Thanks.

 

[Quinton]

Readings were from my lipo alarm right at landing, then used a 2nd lipo tester about 1-2 minutes later. Cells were in the 5.2- 5.4 range if I remember when I landed.

Well it also depends on how much extra weight you were carrying.
I am very conservative and I land when my telemetry says 22V, if I was to check each individual cell after landing they would have risen substantially to at least 22.6V/6 more if I was carrying a heavier load, maybe even upto 1V
So your 3.5V per cell really could be 3.4v or even 3.3V when you are in the air, which could be pushing it A BIT.

Its pretty important I think to actually have telemetry when you are flying so you can see exactly what is going on, as it will completely change when you are on the ground.

 

[gtranquilla]

As electrochemistry performance changes according to load changes, the load needs to remain constant.
I think constant current load is the best bet via a bank of conventional lights (current is self regulating) and ambient room temperature is less affected.... no thermal runaway.

Typical lab testing and measurement procedure is do it all at STP = Standard temperature and pressure. 25 degrees C? pressure not so relevant.
In the case of an open resistor load bank a fan is likely required and room temperature needs to remain constant within reason.

I've got some very large resistor banks I could get from work, designed as braking resistors for large machinery with variable speed drives. I've been thinking of getting one, and then setting it up with a large FET, and a basic control system. This actually isn't a difficult problem to solve, just takes a little time and money.

It would also be great to scientifically test efficiency ratings on motors. Set up a dyno cell, and take the propeller variable out of the equation.

 

[MombasaFlash]

Me? I'd use a helicopter.

Well I thoroughly enjoyed that little talk, thank you.

 

[gtranquilla]

R_Lefebvre,
Thanks for posting that. It helps to explain why my Cinestar 8 does not fly as efficiently as I was expecting relative to my previous Hexacopters.
What caught me off guard though was that an X Coaxial will not have more disc area than a regular Quad.

Me? I'd use a helicopter.

So for example, to achieve the 4.6 kg/m^2 disk loading my quad has, with with a 10kg AUW, you'd need 2.17m^2 of disk area. On a quad or X8, that would require 33" props. On an octo, 23" props. I don't have to tell you how big of a machine that would be. But that's what you'd need to do to be able to load it with enough batteries to fly for 30 minutes.

By comparison, a simple 700 heli, has 2.06 m^2 of area, and an 800 is 2.54m^2.

I know it may seem odd to focus so much on the disk area, but this is the key place that hovering efficiency is determined. It's a natural law. The disk loading gives you the basic hovering power. Then your drive system efficiency makes things less efficient from there. There's less to be gained in the drive system, than there is in the disk area.

This talk is too brief, but gives you some introduction to the topic:

 

[R_Lefebvre]

Yes, the added disks on a coaxial do not add disk area. You could have triple co-ax, and it would still have the same area. A co-ax is going to have many of the same characteristics of a quad with 4-bladed propellers. Basically, the solidity ratio goes up, which does increase the amount of power you can put into the air for a given disk area. But you don't get more disk area. The momentum theory is a very powerful predictor of efficiency. It predicts the basic efficiency of any machine very well, then propeller design, motor efficiency, etc, just takes away from that.

That's where helicopters suffer. They have large disks, low disk loading, but the rotors are much less efficient typically than a multirotor. This is because the multirotor propellers only ever spin exactly as fast as they need to, to generate the required thrust. Helicopters always run the rotors faster than needed, to hold back thrust potential in reserve, which can be accessed by increasing the pitch. The higher you run the RPM, with more power potential in reserve, the less efficient it is. However, they can be made fairly efficient. Particularly if you want to fly forward, not just hover. In this case, the much more streamlined shape of a helicopter frame helps a lot.

 

[Quinton]

Yes, the added disks on a coaxial do not add disk area. You could have triple co-ax, and it would still have the same area.

This is all very fascinating.
Not sure if you know the answer to this but if it does not add disk area, would it help having smaller props above or below on a co-ax?
Have you ever tested this theory out?
It would be great to get Barts thrust meter set up as a co-ax to do some tests.

 

[R_Lefebvre]

I don't know the answer to that. My guess is that, not matter what you do, you're not going to beat the efficiency of a single rotor with the same diameter as your largest propeller. Having a small propeller above or below that, could only be less efficient, I believe. Really, all that matters, is how fast that column of air is being accelerated downward. More weight with less area, means the column of air must be accelerated faster, and that's less efficient. No real way to get around that.

But, who knows, maybe some complicated aerodynamic effect would come into play. The momentum theory is a very basic concept.

There is a theory that using overlapping propellers, between 10-30% overlapping, could actually be up to 5% more efficient than the same two disks would be if separated. But it's never been proved out in reality. Even the guy who wrote a doctorate thesis on it, couldn't show it in lab testing. The possible 5% increase was lost in the measurement accuracy of his setup.

The only part that's clear to me, is that for a given frame wheelbase, a quad or X8 should have more area, and thus be more efficient, than an Octocopter. This seems to be well proven in practice as well. Nobody makes "duration special" octocopters when trying to set a record, they always use Octos. And again, the extreme example is a 100-copter. It would have a thin band of propellers around it, with almost no area at all.

 

[gtranquilla]

Two items come to mind in this discussion:
1) http://www.aerialtronics.com/ but there are others such as DraganFlyer etc.
2) https://www.youtube.com/watch?v=vfggwmhcvC0 Sikorsky X2

It seems that a X Coax offers a lot of benefits over a flat 8 from several perspectives except for:
a) some loss of efficiency (but relative to what? (a flat 8)
b) need to lower and/or rotate the camera down so that FOV does not allow view of the lower props.

 

[Motopreserve]

This is all very fascinating.
Not sure if you know the answer to this but if it does not add disk area, would it help having smaller props above or below on a co-ax?
Have you ever tested this theory out?
It would be great to get Barts thrust meter set up as a co-ax to do some tests.

Im going to be taking over for Bart for the motor testing and if he sends me some of his dual mounts I could do the tests with over/under props.

Should be up and running late this week (waiting on better bench power supply and the motors from Bart). I'd probably have to get an agreement on what size prop top and bottom would be best - I've never run anything but a flat MR, so lack the research and knowledge about this type of setup.

 

[R_Lefebvre]

Two items come to mind in this discussion:
1) http://www.aerialtronics.com/ but there are others such as DraganFlyer etc.
2) https://www.youtube.com/watch?v=vfggwmhcvC0 Sikorsky X2

It seems that a X Coax offers a lot of benefits over a flat 8 from several perspectives except for:
a) some loss of efficiency (but relative to what? (a flat 8)
b) need to lower and/or rotate the camera down so that FOV does not allow view of the lower props.

The Eurocopter X3 is actually faster than the Sikorsky X2. The coaxial rotor on the X2 presents a lot of extra drag in forward flight.

An X8 is not less efficient than flat Octo. That's the main point of the presentation. The only way you can say it's less efficient, is if you constrain the design to use the same size propellers. But why do that? The constraint makes more sense to be the total frame size (wheelbase) or possibly weight. In both these cases, an X8 can use larger propellers than a comparable flat octo, and those propellers will be more efficient.

The only disadvantage I see, is your second one. That it's harder to keep the props out of the frame.

 

[jdennings]

It would be interesting to know, from a practical point of view, how much extra propeller length is needed to get an X8 to the same level of efficiency as a flat 8. I suspect not much ... Depending on the type of propeller, a 16" X8 as efficient as a 15" flat 8?
I've been flying with an X8 with 15" and 17" props (soon to be 20"), and a flat 8 with 15" and 17", but have not experimented enough with the flat 8 to make meaningfull comparisons yet ...