Efficiency of higher voltages?
- Thread starter thingone
- Start date
In reality there is no difference at all.
The very literal answer is that the highest voltage could save you a little weight in wire weight, but have the heaviest controller to cope with higher voltage. The battery would also be heavier to cope for the same capacity for the same reason.
The very literal answer is that the highest voltage could save you a little weight in wire weight, but have the heaviest controller to cope with higher voltage. The battery would also be heavier to cope for the same capacity for the same reason.
liveforphysics
100 TW
Voltage isn't a function related to power train efficiency.
A 5v system and a 500v system both are on equal ground with respect to capacity to be efficient. The 500v system brings with it lethal shock risks, major BMS expense and wiring and bulk to package. The 5v system brings with it a need for copper cables the size of your thumb (which in short runs are not expensive or very heavy), but then also needs a controller capable of many hundreds of phase amps (which adds cost).
From a pragmatic standpoint vs a physics standpoint, going for 20s pack voltage to enable using the outstanding power density of the 100v mosfet offerings is the pinnacle today with respect to total system performance for a given amount of controller size/cost etc.
However, if your aim is really to just do 20mph, going to a 12v or 24v system would be great with respect to a simple battery to build and manage.
A 5v system and a 500v system both are on equal ground with respect to capacity to be efficient. The 500v system brings with it lethal shock risks, major BMS expense and wiring and bulk to package. The 5v system brings with it a need for copper cables the size of your thumb (which in short runs are not expensive or very heavy), but then also needs a controller capable of many hundreds of phase amps (which adds cost).
From a pragmatic standpoint vs a physics standpoint, going for 20s pack voltage to enable using the outstanding power density of the 100v mosfet offerings is the pinnacle today with respect to total system performance for a given amount of controller size/cost etc.
However, if your aim is really to just do 20mph, going to a 12v or 24v system would be great with respect to a simple battery to build and manage.
dogman dan
1 PW
Bottom line, for that particular speed, 48v is all you need, even up large mountains. While one voltage could theoretically be more efficient, pulling only about 400 watts on the flat, you won't be heating up any wires that much regardless of the voltage chosen. ( assuming proper size wires)
Therefore, any difference will be so small it will be hard to measure. The resistance across a plug could be a whole lot bigger number. That brings you to, perhaps solder connections instead of plugs, for max efficiency. Or use multiple plugs, so one poor contact no longer matters.
I'd say go 48v, since there is such a wide variety of good stuff to choose from in 48v. Batteries, chargers, controllers, etc.
Therefore, any difference will be so small it will be hard to measure. The resistance across a plug could be a whole lot bigger number. That brings you to, perhaps solder connections instead of plugs, for max efficiency. Or use multiple plugs, so one poor contact no longer matters.
I'd say go 48v, since there is such a wide variety of good stuff to choose from in 48v. Batteries, chargers, controllers, etc.
Jonathan in Hiram
1 kW
- Joined
- Oct 6, 2009
- Messages
- 446
liveforphysics said:
That's just what I wanted to know for the next stage of my ebike evolution, thanks.
spinningmagnets
100 TW
There are steps in the efficiency and affordability of available components right now. 5 years ago, high-current batteries were expensive and hard to get. Now...the average ebiker really has a lot of choices as to what volts and max amps to use.
Common "48V" controllers have 63V capacitors. Subtract about 5V for voltage ripple and voltage spikes (I don't know what those are, but I've heard of them mentioned elsewhere), and a long lasting and reliable system can be devised using a voltage up to 58V.
Most new lithium-based ebike battery packs right now are usually NCM or NCA chemistry. For long life, I recommend charging to 4.1V per cell (Justin suggests 4.05V is even better for long life). Most 48V labeled packs are 13 cells in series (13S), so...13 X 4.1 = 53.3V when fully charged.
If you go to a 14S pack, the fully-charged voltage (using a common 4.2V per cell charger as a worst-case scenario) would be 58.8V
If you know for certain how many watts you would like to use on a build (around 1200W is pretty common around here for a commuter)...then the more volts you use, the less amps you can get away with using to get the same watts. Lower amps means a cooler controller and motor, plus lower amp-draw will allow your battery to provide you with a little more range (compared to drawing high amps from the battery).
Theres nothing wrong with finding a controller with higher-voltage capacitors so you can run a nominal 60V system (fully-charged to 4.2V per cell is 16 X 4.2 = 67.2V, so...it might be possible to get away with using the cool-running and efficient 3077 FET, which has been rated as capable of 73V (maybe with an unlisted safety margin of a few volts?).
I don't want to start a thread-drift, but...concerning safety, 60V DC is an over/under number for safety regs, as it pertains to voltage being able to penetrate dry human skin. I know that high current is very dangerous, even at only 36V, but a 60V border for recognizing an added zone of danger is reasonable.
em3ev.com and Luna Cycle are both retailing 14S packs right now (along with 13S / 48V packs). em3ev calls the 50V and Luna calls them 52V, but...they are both 14S. The German ebike "Spitzing" is using 14S, for what its worth...
edit: if you live where its relatively flat, I've heard that commuting on 36V is more affordable to build over a 48V system, and is very adequate for most ebiking needs...
Common "48V" controllers have 63V capacitors. Subtract about 5V for voltage ripple and voltage spikes (I don't know what those are, but I've heard of them mentioned elsewhere), and a long lasting and reliable system can be devised using a voltage up to 58V.
Most new lithium-based ebike battery packs right now are usually NCM or NCA chemistry. For long life, I recommend charging to 4.1V per cell (Justin suggests 4.05V is even better for long life). Most 48V labeled packs are 13 cells in series (13S), so...13 X 4.1 = 53.3V when fully charged.
If you go to a 14S pack, the fully-charged voltage (using a common 4.2V per cell charger as a worst-case scenario) would be 58.8V
If you know for certain how many watts you would like to use on a build (around 1200W is pretty common around here for a commuter)...then the more volts you use, the less amps you can get away with using to get the same watts. Lower amps means a cooler controller and motor, plus lower amp-draw will allow your battery to provide you with a little more range (compared to drawing high amps from the battery).
Theres nothing wrong with finding a controller with higher-voltage capacitors so you can run a nominal 60V system (fully-charged to 4.2V per cell is 16 X 4.2 = 67.2V, so...it might be possible to get away with using the cool-running and efficient 3077 FET, which has been rated as capable of 73V (maybe with an unlisted safety margin of a few volts?).
I don't want to start a thread-drift, but...concerning safety, 60V DC is an over/under number for safety regs, as it pertains to voltage being able to penetrate dry human skin. I know that high current is very dangerous, even at only 36V, but a 60V border for recognizing an added zone of danger is reasonable.
em3ev.com and Luna Cycle are both retailing 14S packs right now (along with 13S / 48V packs). em3ev calls the 50V and Luna calls them 52V, but...they are both 14S. The German ebike "Spitzing" is using 14S, for what its worth...
edit: if you live where its relatively flat, I've heard that commuting on 36V is more affordable to build over a 48V system, and is very adequate for most ebiking needs...
Jonathan in Hiram
1 kW
- Joined
- Oct 6, 2009
- Messages
- 446
If you play with the Grin simulator a lot you discover that hill climbing is considerably better at 72V than 48 or even 58V on a low speed wind DD motor which is what I have, the simulator does a good job of verifying my existing setup both for speed and range so I think it's probably reasonably accurate at enhanced levels as well.
Honestly I'm more afraid of texting-while-speeding drivers closing with me from behind on our curvy and hilly 60mph two lane roads while I crawl up a hill than I am of electrocution. We often have a gully or ravine at the side of the road, nowhere to go if you get forced off so I take my lane a lot, I'm more comfortable doing that at the highest possible speeds.
In a hill climbing situation with a low speed motor you have to have enough voltage to push the current through to make the torque you need for climbing.
On edit: I agree with Ron, if you are starting from scratch and live somewhere flattish then a 36V can do you fine at probably a lower overall cost but as always it's having the right combination of parts.
Honestly I'm more afraid of texting-while-speeding drivers closing with me from behind on our curvy and hilly 60mph two lane roads while I crawl up a hill than I am of electrocution. We often have a gully or ravine at the side of the road, nowhere to go if you get forced off so I take my lane a lot, I'm more comfortable doing that at the highest possible speeds.
In a hill climbing situation with a low speed motor you have to have enough voltage to push the current through to make the torque you need for climbing.
On edit: I agree with Ron, if you are starting from scratch and live somewhere flattish then a 36V can do you fine at probably a lower overall cost but as always it's having the right combination of parts.
Similar threads
