Why do people quote 3.7V as nominal?

[Kingfish]

<soapbox: on>
The definition of nominal (and there are many) essentially means:

• In aerospace: performing or achieved within expected, acceptable limits; normal and satisfactory.
• With cost or price: being trifling in comparison with the actual value; minimal.
• In engineering: a value which is close but not exactly the same.

So are LiPo’s at 3.7V minimal in charge? Yes, that’s true: On my system they are damn close near discharge. However - it is certainly not a satisfactory condition; if I’m going cross-country then I am in a panic to find the next opportunistic charge (ala Pepsi machine).

Conversely for LiFePO4 chemistry that would be very satisfactory for the middle of the range, or at least in the fat of the pack. :wink:

The way I use the word “nominal” is similar to how NASA uses it: Close to normal, acceptable, within tolerance, conforming to expectations, median. 8)

The fat nominal part of my LiPo pack begins at about 4.07V/cell and ends at about 3.7V/cell. Above or below that range and there just isn’t a lot of depth or resiliency; if the power curve is represented by a lazy-S/ laying-on-the-side, then 4.07V+ is the end of the up-curve and 3.7V is the beginning of the down-curve, with the meat of the pack in-between. That is the part worth measuring, which has value and significance for calculating distance, duration, endurance, stamina, and throughput: Nominal.

I stopped using 3.7V for nominal measurements last summer; it’s a worthless false figure that is entrapping a false sense of security. On my system, when I reach 3.6V/cell I am about to become DitW (Dead in the Water), or ForD (Found on road DEAD): The cruelest of all lessons learned about power management. :x

Maybe 3.7V means something to the LiFePO4 people, but it sure sucks for LiPo and I just wish people would stop making blind references to it being the “nominal” cell voltage without qualifying the use of the word. That’s just plain wrong in engineering parlance.

Nominally-speaking, KF
<soapbox: off>

 

[rebelpilot]

so QTY 12 Zippy FlightMax (4s3p of 6S1P 5Ah 25C) would be 1.5kWh instead of 1.3 kWh?

 

[liveforphysics]

Nano-techs true nominal is over 4v.

It's just old conventions from gen1 Lipo that have stayed the same.

 

[TylerDurden]

A bit dated (3yrs, pre nanotech), but maybe still useful:

 

[Kingfish]

Rebel, here’s the way I look at it…

I run 15S LiPo: 3.7V * 15 = 55.5V which is about 1.5V away from being overdrawn. I think my LVC cuts out at 54.1V, and the available Wh will be very low ~ essentially unusable; presume Zero.

Typically we determine the system capacity by the potential energy.

• A rocket sitting on a launch pad has two types of potential energy: The total stored chemical energy normally used for propulsion, and the impact force (mass, height, & gravity) should it fall over. If it fell over and exploded the maximum potential energy would be realized (which btw is the answer to a trick question of an entrance exam I once took).

With batteries the potential energy is essentially electro-chemical in nature that is locked up as charge potential. The measurement of the maximum potential is taken when the battery is full, not when it is in the middle or end. You don’t gauge the distance that a new car will travel with half a tank do you?

Measuring the capacity of battery packs by nominal voltage is silly – no matter how the word is interpreted. Capacity is measured at the level that the user typically and traditionally sets as the top limit; what they determined to be full by their charging system.

Make sense?

TylerDurden, thanks for posting that! The use of nominal here is implied to mean the middle, the median, or the mean, and this is the correct use of the phrase. Nominal could also be used interchangeably as “optimum” or "preferred" voltage, especially when running system diagnostics.

Mollified, KF

 

[rebelpilot]

Re: 1 kWh LiPo club, who's in?

Postby icecube57 » Mon Jan 31, 2011 5:47 pm
I see your point and its kinda misleading. You can actually drain a little bit more than the calculated wh using the nominal voltage. Not alot but its enough to matter.

EX. I did a full discharge on a 12s 10AH pack. So thats 44v.4v x 10AH = 444wh but I actually discharged 456wh. Results with a SA CA with precalibrated shunt.

So in theory our 24s 3p Turnigy pack we have still deliver 1.3kwh or 1.332kwh but they deliver 1.368kwh to be exact.

 

[TylerDurden]

I've not expected the term "nominal" to be more than a reference to a category, prone to arcaneness.

Lumber sizes being a common example.

 

[number1cruncher]

I'm just a dumb auditor, but isn't nominal in this context the average voltage of a particular cell during a full discharge? Meaning that if you take the nominal voltage and multipy by the cell's capacity you should get to the cells usable watts. I've never tested this hypothesis, but have adopted it as my understanding. Simple is as simple does I guess...

 

[Kingfish]

Number1, you bring up a valid point and one I wanted to touch upon:

1. First issue has to do with voltage sag, and that will be unique to each user’s configuration, chemistry, environment, and operational behavior.
2. Second issue has to do with the true measurement of system capacity – which is likewise unique, and to do this, one needs to measure the high and low limits of the voltage, and amps or watts over time.

Example: Simplistically, to calculate Average Velocity => vave = (vi + vf) / 2

Applying this example to batteries,

• Vi = Maximum Voltage, hot off the charger
  Vf = Minimum Voltage; could be when LVC protection kicks in or some preset safe value
  Wi = Initial Wattage == 0; we haven’t used any energy.
  Alternately we could use Ai = Initial Amps == 0 for the same reasons.

Use the system until Vf is reached. Review the total Wattage or Amps consumed over Time.

Repeat the test on level ground at STP at least once more to instigate an average trend. Ultimately this will be the true measurement of your particular system at its’ maximum potential. The average voltage during discharge calculation then becomes very much like the average velocity equation: Useful for coffee table conversation, though not precisely accurate.

A better method of examining “nominal” voltage during discharge is to measure the rate of voltage drop during maximum throttle (WOT) over the full battery voltage spectrum. Note the rate of change in voltage drop at the beginning and ends in particular. Being a number cruncher I presume that the mean and derivation will become apparent and useful; the fruitful part of the curve is not always the middle

On my system, using 15S LiPo, my top voltage is typically 63V +/- 0.2. When I give capacity I often list my top-voltage for reference. The bottom limit is 42V, however due to system sage the battery doesn’t reach this value; it is only reached under extreme demand. In my experience the top volt burns off quickly regardless of the number of parallel cells; my present 6P commuter configuration drops 1/2V in the first mile or two. Likewise the bottom voltage drops even faster; once I hit 54V the system sages to 42V in < 5 miles under full load – and when you’re travelling over 130 miles on a single charge, that’s an important relative consideration.

In conclusion, the true limits of my system are calculated by using the fatter voltages and do not include the bottom end, though the top-end does factor in a wee bit as gravy for the goose & sause for the gander

Clear as mud? KF

 

[liveforphysics]

If you're looking at finding your real-world nominal for your particular application, you can simply back-out the math by using the values from your CA.


Charge to whatever you charge to. Run the bike until you call it discharged to whatever you let it discharge down to. Look on your CA for the watt-hours used, and the amp-hours used. Divide used watt-hours by used amp-hours, and now you have your real-world pack nominal voltage delivered to your controller. That takes into account the real-world sag in your wires/connections/etc.

I bet it will be something pretty close to 3.7v/cell (unless you wire up your bikes with 2awg and 4awg leads like I do)

 

[Doctorbass]

As i already explained on the forum, nominal voltage is explained by this:


This is the voltage a cell have when discharged at 1C when it approach half his ENERGY capacity. ( not half Ah)

... a kind of averaged voltage..over the discharge. but not exactly...

EX: a common full lithium full at 4.2V and empty at 3V will show 3.7V on the terminal when it is discharged at 1C rate under a load of 1C when it reach half his total Wh capacity

In other words.. you take a headway 38120 of 10Ah that is 32Wh ( 3.2V at 10Ah) and you should be able to measure close to 3.2V on teh terminal when it will have discharged 16Wh energy and is under 10A load.

that also revelate that if you measure the voltage of a lipo cell that is 3.7V, that this cell probably have less than than half his energy stored because it is voltage not under load.. and probably would measure few mV lower.. maybe 3.6 or 3.55 at 1C depending on the Ir of the cell...

The way LFP is explaning also good since it revelate the true nominal voltage of YOUR battery.. used or brand new.... The way i explaned that is coming from the industry method.

Doc

 

[liveforphysics]

As i already explained on the forum, nominal voltage is explained by this:


This is the voltage a cell have when discharged at 1C when it approach half his ENERGY capacity. ( not half Ah)

If you do this, a nano-tech cell has a nominal voltage >4v.

This is the point Kingfish is trying to make. The early formulas really were 3.7v nominal. The cutting edge HobbyKing stuff has a substantially higher nominal voltage when measured at 1C (or even 5C makes like a~1% difference).

 

[number1cruncher]

If you're looking at finding your real-world nominal for your particular application, you can simply back-out the math by using the values from your CA.


Charge to whatever you charge to. Run the bike until you call it discharged to whatever you let it discharge down to. Look on your CA for the watt-hours used, and the amp-hours used. Divide used watt-hours by used amp-hours, and now you have your real-world pack nominal voltage delivered to your controller. That takes into account the real-world sag in your wires/connections/etc.

I bet it will be something pretty close to 3.7v/cell (unless you wire up your bikes with 2awg and 4awg leads like I do)

Exactly, my thoughts...In addition, I would also venture to assume that usable watts should(all things being equal, yada, yada) come out similar in each discharge cycle. As KF and LFP have noted, where my rambling omitted, is the concept of voltage sag and its effect on average voltage. But as noted by LFP, the CA should take care of that. :wink:

Also, KF, I do agree about nominal not being quite the average, but the mode, as (regular, not nano) Lipo seems to spend most of its discharge time at ~3.7V relative to all other voltages. Good point!

 

[liveforphysics]

Also, KF, I do agree about nominal not being quite the average, but the mode, as Lipo seems to spend most of its discharge time at ~3.7V relative to all other voltages. Good point!

That's exactly the number he is trying to dispel. 3.7 was the average voltage, or the voltage at 50% of the energy discharged back in the day with older formulas. It no longer is the case, it's higher, and by a substantial amount.

I'm at my desk right now with a nano-tech pack, I will make a discharge curve at 1C and post it up.

 

[Doctorbass]

I'm at my desk right now with a nano-tech pack, I will make a discharge curve at 1C and post it up.

Excellent idea!.. Please dont use the stressed one of your High compression engine start attempt ! :wink:


Yeah.. I know.. you'll say.. it dont make any difference these cells are like god and are so much everything positive.. perfect etc.... ....


Doc

 

[Doctorbass]

Hey.. wait a minute...



nominal V of nanotech over 4V ??

That would mean their Wh is higher than the normal lipo for the same Ah ?? :|


3.7v to 4V = 0.3V.. around 7% more energy for the same Ah ??
Doc

 

[liveforphysics]

I'm at my desk right now with a nano-tech pack, I will make a discharge curve at 1C and post it up.

Excellent idea!.. Please dont use the stressed one of your High compression engine start attempt ! :wink:


Yeah.. I know.. you'll say.. it dont make any difference these cells are like god and are so much everything positive.. perfect etc.... ....


Doc

I'm using that very pack that I used for starting the engines, and then later used as a test pack to short-out and see if the tabs would pop. It got the 8awg wire so hot it melted the solder off the large bullet connector to open the circuit in about 3seconds. Still cycles exactly like the day it was new. ***EDIT*** Actually, I just looked at the graph it's making, and it looks like finally cell #1 is damaged. :'( It's falling off quite a bit sharper than the other 2 cells in the pack. :'( Poor nano-tech. I guess they can be damaged.

I'm discharging at 2c rather than 1c just because it's so slow to discharge at 1c, and 2c is more realistic for an ebike anyways.

 

[liveforphysics]

This particular pack is damaged it seems. (which is not it's fault, I did a number of extremely abusive things to it in tests)

The first part of the voltage curve fell off much faster than it normally does, so perhaps the first part to go when the cells are damaged is whatever effect it is that causes that odd voltage increasing effect.

And yet still, the weakest cell in this damaged 4Ah pack discharged past 3.1Ah of it's capacity before dropping to 3.70v under 2C discharge.


As soon as it finishes it's discharge cycle, I will post it up.

 

[liveforphysics]

Here it is:

And if the forum crops the width of the image for you (I HATE that!), here is a direct link:

http://img109.imageshack.us/img109/7663/4ahnanotechdischarge.png


Notice, even with that damaged cell dropping like a rock after 3.6v, it still makes 4.31Ah (I had it charged to 4.36v to start with though).

If you even just look at the time averaged voltage, it's over 3.8v even for the damaged cell. If you look at the watt-hours divided by capacity, it's over 3.8v, if you look at the percentage capacity used before reaching 3.7v, it's 79%.

There is no way or shape to work the numbers to get 3.7v nominal from modern LiPo, even a badly damaged pack like this. lol When you hit 3.7v, you had better not think you're at a half-tank, because you're actually about to be pushing your bike home.

 

[Hyena]

used as a test pack to short-out and see if the tabs would pop. It got the 8awg wire so hot it melted the solder off the large bullet connector to open the circuit in about 3seconds. Still cycles exactly like the day it was new. ***EDIT*** Actually, I just looked at the graph it's making, and it looks like finally cell #1 is damaged.

Who'd have thunk it!
Sweet test data man, good work.
Have you got a used but unabused nanotech you can also do some discharge tests on ? And take them lower ? Maybe down to 3.2 or 3.0v ?
It's good to see what the one you've beaten the snot out of can do but one that's only been used in a somewhat normal fashion would be more representative.

I agree, for normal ebike use the capacity measured at 2C is a good call

 

[liveforphysics]

used as a test pack to short-out and see if the tabs would pop. It got the 8awg wire so hot it melted the solder off the large bullet connector to open the circuit in about 3seconds. Still cycles exactly like the day it was new. ***EDIT*** Actually, I just looked at the graph it's making, and it looks like finally cell #1 is damaged.

Who'd have thunk it!
Sweet test data man, good work.
Have you got a used but unabused nanotech you can also do some discharge tests on ? And take them lower ? Maybe down to 3.2 or 3.0v ?
It's good to see what the one you've beaten the snot out of can do but one that's only been used in a somewhat normal fashion would be more representative.

I agree, for normal ebike use the capacity measured at 2C is a good call

I have my personal Nano-Tech pack, but it's all hooked together, and not sitting on my desk at work right now. I just used that one because I happened to have it with me at work.

 

[number1cruncher]

Also, KF, I do agree about nominal not being quite the average, but the mode, as Lipo seems to spend most of its discharge time at ~3.7V relative to all other voltages. Good point!

That's exactly the number he is trying to dispel. 3.7 was the average voltage, or the voltage at 50% of the energy discharged back in the day with older formulas. It no longer is the case, it's higher, and by a substantial amount.

I'm at my desk right now with a nano-tech pack, I will make a discharge curve at 1C and post it up.

I get that nano's nominal is above 3.7V, but what about "average" lipo, like 25C Flightmax or Turnigy? It seems to me through my very unscientific observations that 3.7V seems a reasonable nominal figure to accept for my 25C Flightmax pack. I would not put much effort in defending this however.

Back to nano. Why hasn't HK or whomever created this battery tech updated the nominal value to a more appropriate figure? I would think they would want to advertise a higher nominal value to attract customers. As Doc Bass says(I think), it contains more energy than average lipo.

 

[liveforphysics]

Back to nano. Why hasn't HK or whomever created this battery tech updated the nominal value to a more appropriate figure? I would think they would want to advertise a higher nominal value to attract customers. As Doc Bass says(I think), it contains more energy than average lipo.

Because the customers are stupid morons who all ready doubt that it really does 45c-90c (and they have no way to test that), when it can actually do 100c continuous and burst 200c. If they were to say, "by the way, it also has a higher nominal voltage", it would just create another 1000 posts by ignorant morons claiming that's impossible, all LiCoO2 has to be 3.7v nominal etc.

In practice, if you buy Nano-Techs for an RC helicopter or something, you generally get about 10-15% more run-time and better performance during the runtime than all other LiPo rated as the same voltage/capacity. This comes in the form of less voltage sag (than ANY other cells I've tested) under high discharge loading combined with a higher nominal voltage.

 

[number1cruncher]

Because the customers are stupid morons

Hey, I resemble that remark! :lol:

But yeah, you've got a point. Funny how life works, modesty is condemned to scrutiny. I just wish HK would stock some nano in their US warehouse!!!

 

[theRealFury]

I thought that 3.7v was classed as 'nominal' because of the early lipo batteries having a 4.2v full and a 3.2v discharged voltage so the middle ground was 3.7v. I do agree that its no longer relevant and its probably better to talk about 'rated application voltage'. So instead of saying my "3.7v nominal lipo" it would be "my 3.7v rated lipo"

Lets face it.. everything we use batteries in has a 'rated voltage' on it, a 36v rated motor will quite happily work on 32v or 43v (and if some of you guys are anything to go by... quite a bit higher than that as well lol :lol: ).