Why 50V-60V is considered the optimum battery voltage?

Hi,
After being told my 72V system is not recommended and that I should work with no more than 50V-60V I wanted to discuss this.
Since all my components are compatible with up to 100V (CA, Phaserunner), and since with higher voltage I can expose all the wiring to less intimidating currents - then why not use it?
The biggest bottle neck on E-bike is the motor phase connector - which experience phase currents way over in excess of the 30A-45A those Andersons are rated for. Using a higher voltage system with a slower winding motor so you can keep the same top speed will solve this by requiring less phase currents for the same torque.
72V is also not yet a limiting factor with regard to insulation. For example, if I worked with 230V - I would have saved alot on copper weight (wiring only, not inside the motor!), but I would needed more heavy-duty connectors and insulations, not to talk about the danger of electrocution.

I really enjoy the fact I draw smaller currents to supply my 2000W hungry system, compared to if I fed it with 50V or less.
Any opinion about this?

What motor and controller setup ? I run 72v for the speed. But my next commuter build will be with a 52 volt just for size and weight. Use high quality cells for long lasting battery.

999zip999 said:

What motor and controller setup ? I run 72v for the speed. But my next commuter build will be with a 52 volt just for size and weight. Use high quality cells for long lasting battery.

Click to expand...

I have two motors - front 9C+ 2706 DD and rear ezee250rc. Each controlled with it's own Phaserunner. (Both together take 2000W but peaking at 3000-4000W where needed).
Especially because the phaserunner doesn't care at all about speed but on phase current (as long as it has enough voltage to counter the back-EMF), I started thinking about why to stay with lower voltages.

Size and weight? You know - with most modern controllers your range depends only on the Wh of your pack, so for a defined amount of cells you can aim for higher voltage or higher Ah. The range would be the same, but higher voltage will allow higher top speed and will induce less I2R losses on your wires. (except within the motor that stays the same)
In your case, with 72V you also benefit stressing your system with lower currents when you don't require your top speed.

Of course I will use high quality cells - I ask only from the voltage decision perspective.

I think it depends on the context that the recommendation was given. For some motors that's normally a bad idea. your 250w eZee, for example, would probably fail at higher voltages. it's not built for handling the kind of speed the higher voltage would give, and you'd either melt the winding s or destroy the gears.

Your 9C on the other hand, will rock at voltages above 72. IIRC someone a few years ago ran the calculations on the stator designs and lamination thicknesses of various motors. 9C and a few others peaked in efficiency near 100 volts, and weren't reaching saturation until someplace north of 120 volts. Other motors like the Crystalyte "The Crown" hit saturation around 100 volts and worked better if you kept them at a lower voltage, like 72.

The only caveat to running your 9C 2706 at high voltage is the wattage and heat. Unmodified, it can't take more than 2000w for long. Being a faster wind, it will get hot fast, too. Modified for cooling and with larger phase wires added, it's easily capable of 3000w or more.

Also in context, there is some safety concerns that will cause some people to not recommend higher voltages. 48v and below is considered "safe". Above that is pretty safe too, but the threshold has been set low. People who know don't care. people who don't know, stick to the guidelines.

High amps at any voltage can cause very serious burns. There are many examples of 12V car batteries that are shorted through a metal ring on a metal wrench and someones finger. There is no voltage that is safe when talking about high amps.

Once we accept that the amps are a danger that we must be conscious of, careful about, and plan for...we can turn our attention to the risks of higher voltages. In my research, I read that there is an international standard where 60V is the accepted limit for the electricity to penetrate dry human skin. If your hands are sweaty and salty, of course, it will be worse.

If a 14S Lithium battery if charged to 4.2V per cell, it is 58.8V when fully charged. If you charge to 4.1V per cell, its 57.4V...I know that is very close to 60V, but as soon as you ride even just a minute it usually drops to a fairly stable level below 56V.

4V isn't a huge safety margin, but if we are reasonably safe around 72V batteries, we are a lot safer at <56V. One of the driving factors in the popularity of the 52V batteries is that the BBSXX drives will run off of 13S and 14S batteries from the factory (Nominal 48V/52V).

Once any builder has decided to jump into the 60V-100V pool, it can be wonderful from a performance perspective. However, that can also force the battery to be much larger (harder to fit?), much heavier, and much more expensive.

edit: as Chalo mentions below, their are much more efficient components available for systems below 60V, such as 63V capacitors and 3077 FETs, etc

I like 72v. For me it came down to having a lot of 36v batteries and chargers, so it was an easy and cheap voltage for getting my motor to wind up a little higher with some cheap off the shelf stuff. But it is plenty sparky when things go wrong.. I got one of my plugs too close to my big aluminum battery rack and it jumped the air gap and dug a 2" long gouge in the metal while melting the plug. I got ones with more shrouding after that.

The usual posters here are accustomed to higher voltages and have apparently developed rides capable of handling higher amps and speed.

I think for most hobbiests and certainly beginners 14s is a good spot. Actually I'm finding more and more older adult path riders very happy with 10s, and moving to 13s and 14s simply due to the market offerings and push by a couple of the bigger vendors.

36v is chided and becoming less available. I'll be building my own packs to support to older 350w 36v mid drives due to dismal offerings. Guest bikes loved by everyone that rides them. Including busy path days by this couple. As a matter of fact 4 BBS01's are the least supported motors I ever installed for riders. But that's older riders, not daily commuters.

I've had a taste of speed with 14s and 33mph, but it took some work getting a good safe braking system and frankly it's just adequate. I won't make my due diligence speech, but I continue to worry over the direction and eventual problems as this xplodes into more power by brand new builders. No learning progression. Nearly instant speed. On not so great frames.

All the 72v riders I've followed here have been around and have the experience.

There are more efficient FETs available for nominal 52V operation than for higher voltages. That means less heat management problem solving to do, and of course more battery watts that make it all the way to the the wheel.

Drunkskunk said:

I think it depends on the context that the recommendation was given. For some motors that's normally a bad idea. your 250w eZee, for example, would probably fail at higher voltages. it's not built for handling the kind of speed the higher voltage would give, and you'd either melt the winding s or destroy the gears.

Click to expand...

I am talking about transferring the same torque at any speed. Why would the gears fail at high rotational speeds, if the torque they transfer stays the same? (It's more likely the torque would be fair less, since Power=Torque*Angular velocity)
And what is that speed limit (in rpm) you consider as an upper limit?

Your 9C on the other hand, will rock at voltages above 72. IIRC someone a few years ago ran the calculations on the stator designs and lamination thicknesses of various motors. 9C and a few others peaked in efficiency near 100 volts, and weren't reaching saturation until someplace north of 120 volts. Other motors like the Crystalyte "The Crown" hit saturation around 100 volts and worked better if you kept them at a lower voltage, like 72.

Click to expand...

That efficiency would be in effect only if they are let to be span that fast, and without the serious wind force load that is involved...
Even in the configuration I talk about, the motor would rarely face the "full" 72V phase-voltage from the controller, unless you go at really high speeds - which waste a lot of energy anyway.

The only caveat to running your 9C 2706 at high voltage is the wattage and heat. Unmodified, it can't take more than 2000w for long. Being a faster wind, it will get hot fast, too. Modified for cooling and with larger phase wires added, it's easily capable of 3000w or more.

Click to expand...

I was talking only about increasing the voltage while keeping the same phase-current within the motor. Higher input voltage would induce lower battery current drawn.
Of course if I use a standard controller that limits by battery-current only, then the power would increase, and since it has no phase-current limitation (At least a safe one), you can easily overheat your motor.
And last - once I added the rear ezee to my existing 2706, I solved ALL the heating issues of the 2706. I now limit it's phasecurrent (50A), and it shares the load with the ezee. Just by doing this, by battery consumption was cut by 40%! (much less battery power is converted to heat inside the 2706). I saved any need for drilling/cooling of any kind.

Chalo said:

There are more efficient FETs available for nominal 52V operation than for higher voltages. That means less heat management problem solving to do, and of course more battery watts that make it all the way to the the wheel.

Click to expand...

As I said, I use Justin's Phaserunner and CA, which are rated all the way up to 100V, so no problem here.
Of course if I already had lower voltage components, then this discussion would have became less relevant, as I would have needed to upgrade those as well.
I am in this dilema, since I need a longer range battery - and so I think which battery voltage to pick. I already know my total Wh required - the chosen "S" number would directly affect the "P" number of cells.

spinningmagnets said:

High amps at any voltage can cause very serious burns. There are many examples of 12V car batteries that are shorted through a metal ring on a metal wrench and someones finger. There is no voltage that is safe when talking about high amps.

Once we accept that the amps are a danger that we must be conscious of, careful about, and plan for...we can turn our attention to the risks of higher voltages. In my research, I read that there is an international standard where 60V is the accepted limit for the electricity to penetrate dry human skin. If your hands are sweaty and salty, of course, it will be worse.

If a 14S Lithium battery if charged to 4.2V per cell, it is 58.8V when fully charged. If you charge to 4.1V per cell, its 57.4V...I know that is very close to 60V, but as soon as you ride even just a minute it usually drops to a fairly stable level below 56V.

4V isn't a huge safety margin, but if we are reasonably safe around 72V batteries, we are a lot safer at <56V. One of the driving factors in the popularity of the 52V batteries is that the BBSXX drives will run off of 13S and 14S batteries from the factory (Nominal 48V/52V).

Once any builder has decided to jump into the 60V-100V pool, it can be wonderful from a performance perspective. However, that can also force the battery to be much larger (harder to fit?), much heavier, and much more expensive.

Click to expand...

High amps causing burn through electrocution or by touching shorted metal? High amp through your skin and you would be dead immediately.
In fact, 30mA are the threshold for the ability of a human being to still be able to use his/her twitching muscles to disconnect from electrocuting source.
If it's high amps through metal, then you are seriously doing something wrong. Actually with 12V or 24V, it's more likely to get heat-burned because the currents are low enough not to trigger any fuse - but to considerably warm and melt lots of wires and connectors until it's detected. With high voltage - the current would jump high enough to shut down the protection circuit or fuse, and therefore not having enough time to cause damage.

Regarding battery weight - of course if I go from 14S5P to 20S5P I would get a heavier battery, but it would also have a much higher capacity now. I talk about keeping the same Wh - raising the S number in favour of the P number. For example - going from 10S10P to 20S5P. A controller like Phaserunner or Grinfineon would effectively draw half the current with the 20S pack for the same given load. This is of course only relevant if you buy a new battery pack - which is my dilema now, and the reason for this thread

Optimum to do what?

Step one,, figure out what you want to do,, then we can start making suggestions.

If you want a max speed under 30 mph,, then most typical hub motors would reach 27-30 mph on 48v (54.6 v full). That's why its optimum, cuz you don't look like a hooligan going 25-30 mph and cops can still ignore you.

But if you want 50 mph out of most hub motors, then you are looking at something that charges close to that 100v limit, like 22 or 23 s. 24s if you undercharge slightly to 100v exactly.

dogman dan said:

Optimum to do what?

Step one,, figure out what you want to do,, then we can start making suggestions.

If you want a max speed under 30 mph,, then most typical hub motors would reach 27-30 mph on 48v (54.6 v full). That's why its optimum, cuz you don't look like a hooligan going 25-30 mph and cops can still ignore you.

But if you want 50 mph out of most hub motors, then you are looking at something that charges close to that 100v limit, like 22 or 23 s. 24s if you undercharge slightly to 100v exactly.

Click to expand...

I have two motors on a 26inch bike: 9C+ 2706 and ezee250rc.
I want to be able, temporary, to increase my cruising speed to 60km/h where I must merge with traffic. Otherwise, I don't go faster than 40-50km/h.
I also find it more safe, on main roads which are of high grade (10%) - to be able to climb them at a steady 50km/h (I can push a bit more if I want), thanks to the twin motor + 72V configuration, where otherwise I would have been capped at 40km/h.
72V is required for 60km/h on flat, but it can also be achieved with field weakening (although less efficient).
I find the 72V better overall, as it reduces the current over the wires and terminals, although motor side - all I2R losses are the same. It gives you opportunity, and it increases your system efficiency even if you don't go to the speeds it allows.

What do you think?

Yeah,, 72v needed to go that fast on that motor. There is your need that requires 72v. But 72v 40 amps can overheat those motors if run hard very long. So easy does it on the longer 10% hills,, I mean like if it will take longer than 15 min to climb it.

A 9c can easily take 72v 20 amps, for a very long time. 1500w,, and you will get about 35 mph out of that. But to get that fast takeoff in traffic, you need all 40 amps. I'm just saying, consider backing off the throttle the 9c motor on the really wicked hills, so it sees 1500w continuous if the hill is long and the weather hot. In winter, let er rip. I'd put a CA on that 9c, so you know what you are pulling at all times. Use that 3000w sparingly and you will never melt a motor.

The ezee, it will be in risk run above 1500w. so if it runs on a separate battery, 48v is plenty for it, used only when you are going slower anyway up that wicked hill. If the ezee runs on 72v,, limit amps on it somehow to 10 amps or so. you can clip a shunt on a 72v 30 amps controller for example, and lower the amps of the thing.

dogman dan said:

Yeah,, 72v needed to go that fast on that motor. There is your need that requires 72v. But 72v 40 amps can overheat those motors if run hard very long. So easy does it on the longer 10% hills,, I mean like if it will take longer than 15 min to climb it.

A 9c can easily take 72v 20 amps, for a very long time. 1500w,, and you will get about 35 mph out of that. But to get that fast takeoff in traffic, you need all 40 amps. I'm just saying, consider backing off the throttle the 9c motor on the really wicked hills, so it sees 1500w continuous if the hill is long and the weather hot. In winter, let er rip. I'd put a CA on that 9c, so you know what you are pulling at all times. Use that 3000w sparingly and you will never melt a motor.

The ezee, it will be in risk run above 1500w. so if it runs on a separate battery, 48v is plenty for it, used only when you are going slower anyway up that wicked hill. If the ezee runs on 72v,, limit amps on it somehow to 10 amps or so. you can clip a shunt on a 72v 30 amps controller for example, and lower the amps of the thing.

Click to expand...

Remember that in my setup the 2706 share the load with the ezee. Just adding the ezee reduced the battery consumption by half, since no more (almost) waste heat is generated inside the 2706.
I am using the phaserunner, and limit with it the 2706 to 50A of phasecurrent. Now it doesn't anymore what's the current on the battery side, but still I limit that value to 40A. That increased my range as well by considerable amount - since I prevent the motor from working in those high I2R losses zones.
I have the same phase current limit on the ezee - so it's also very happy with it. I wrote about this in another thread but I will repeat it here: Since I limit the phase current, I don't care about the battery input power anymore - regular controllers with no sane phase current limit (only battery current limit) need it - but the phase runner doesn't. It means that WOT at standstill will make the ezee draw few hundred watts (no back-EMF to counter), but at high uphills speeds, when it's helping the 2706 - it's drawing much more than 1000W - because of the high back-EWF the battery has to counter, and since I apply the same torque but at high speed. (laws of physics -> Power = Torque * Angular velocity).
The motor wouldn't overheat in neither scenarios - that's why I don't see why I should care at all on input power (if my battery can handle it), if I watch the true value - the motor phase current.

It also seems, that the CA and the Phaserunner has their share of problems when running on 72V. I run a flashlight via the CA DC port (Grin flashlight), and I happened to trip the protection circuit of the phase-runner more than once, and Grin told me it's probably because it's tripping due to the high voltage. (although by specs it's rated up to 100V with it's components).
Anyone else had hardware/protection circuit problems when running 100V components that make problem at 72V?

I'm also running dual PhaseRunners with 72V, and 9C 2810 rear and BMC V3/4TT front in 26". See the Bonanza2WD thread for more info on my machine and results.

The only tripping problem I've experienced has been with the PhaseRunner on the Gearmotor, and that required some adjustment to the PLL gain, as recommended by Grin. I also reduced the motor current, so that may have helped, it is now 45A.

High voltage from the battery does NOT change motor phase current. Motor phase current is determined by motor windings and the torque needed. Think ampere-turns on the motor as that determines the field which determines the torque. 9C motors hit the first magnetic saturation at about 480 amp-turns. So my 10 turn 9C motor requires 48 amps to reach this point, and your motor is probably 6 turns, if so it requires 80 amps to generate the same torque. The motor phase current flows from the PhaseRunner through the cables and connectors to the motor, regardless of the battery voltage. If you want to reduce motor current you'll need to rewind the motor, or get a higher turn count (lower Kv and higher Kt) model.

The only place currents are reduced from increasing battery voltage are in the battery and from there to the controller. The controller, similar to a buck type DC-DC converter, converts the extra voltage into extra current which flows through the FETS and out to the motor. So your motor lead PowerPoles are seeing the full current regardless of what you do with battery voltage, as long as there is enough battery voltage for the speed.

I don't know if your Gearmotor is a speed or a torque wind, your DD motor is a fairly middle of the road winding, certainly not a real "torque" winding such as an 8, 10 or 12 turn 9C. Both of my motors are "torque" winds with Kv's in the neighborhood of 5.3-5.5.

If I were making a vehicle with no budget constraints with the objective being ultimate performance (even 1000's of hp), I would run 20s (to use the excellent 100v MOSFETS), and run a poly-phase motor with as many phases as needed to achieve the power desired.

If I were making a commuter vehicle in the sub 5kW range, I would run ~24-36vdc, because every added series cell in the string is potential for hassles, where a few feet of cabling for 100-200A is easy and reliable compared to managing more cells.

If I were making a vehicle for sub 1kW, I would shoot for 4s-6s if I could find the right motor winding and controller.

That would be excellent, however the excellent and tiny PhaseRunner 6 FET limits the current a bit.

For my Bonanza, the 9C 10 turn was bought (from methods) in early 2011, and was the one I was riding when we met Luke and Eric and many others on that Marin peak, that fateful day the "Dangerous Bike Launch" video was made. So that 9C 10 turn motor (in the excellent JRH wheel build) drives my voltage requirement on this machine. The V3/4TT was chosen to have similar Kv, before when the plan was dual Lyen controllers. The PhaseRunner was a late change, albeit an excellent one, and it being a 6 FET 100 Volt model is better suited to 72V at lower currents. So it is a fairly well matched "system".

But if starting from scratch I'd probably choose the 52V system voltage (59V max) and have to find other controllers, none of which are as small and well suited for bicycle use as the PhaseRunners are at the moment, or use the PhaseRunners anyway and suffer about 30% torque loss due to lower voltage and higher current motors and the PhaseRunner current limitations.

The reason 52V is attractive is that so many other of my bikes run well on it, and that allows swapping battery packs around which is convenient and saves some funds for other things.

liveforphysics said:

If I were making a commuter vehicle in the sub 5kW range, I would run ~24-36vdc, because every added series cell in the string is potential for hassles, where a few feet of cabling for 100-200A is easy and reliable compared to managing more cells.

Click to expand...

Interesting. The new generation of high-capacity / high-discharge cells had started me thinking about the viability of smaller, lower voltage battery packs combined with more robust cabling for mid-range, mid-power bikes (commuters and grocery-getters) but I don't have the EE background to know if this was more than just wishful thinking. Sounds like it could be reasonable.

The new generation of high-capacity / high-discharge cells had started me thinking about the viability of smaller, lower voltage battery packs combined with more robust cabling for mid-range, mid-power bikes (commuters and grocery-getters) but I don't have the EE background to know if this was more than just wishful thinking

Click to expand...

Its easily doable. Back when there were very few options as far as ordering a specific kV of motor, the trend was to order a slower winding (out of two kV choices) and raise the volts to 72V (plus you had to make the 72V battery yourself). Anything can be ordered in bulk, but you have to find a wholesaler that has confidence that they can order 100 units of a specific type and they will sell fast enough to make it worthwhile.

Now, there are more choices. The MXUS 3000W and Edge 1500W (45mm wide stator and 35mm, respectively). Both have the desirable thin laminations, and the thick aluminum stator support that acts as a heat sponge, to absorb heat spikes from high temporary amps.

Both are available in fast windings that have low turn-counts, and fat wire that allows high amps. There are many that will insist that doing that is inefficient, because high amps means converting a lot of the battery watt-hours into heat.

A long time ago, doing that would limit range, but...now? its easy to find a lower voltage battery (10S, 12S, 13S) that will fit into a bicycle frame, and will have lots of Ah of range. Plus amps...LOTs of amps. Samsung 25R puts out over 20A per cell, so a reasonably-sized 5P pack can provide peaks of 100A, no sweat.

It is difficult to find small low cost BMS's over 50 amps, and controllers that handle really high current are physically large, like the Sabvoton. Circuit breakers for high current become difficult to source. Connectors for large gauge wire are often quite large. It is all do-able but the bulk of the parts is scooter-like, not bicycle-like.

Ron, Alan --
Thanks for insights on why low voltage / high amperage could work, and why it would be challenging in a light/inexpensive bike-like build. Food for thought. Last couple years I've been very happy with 48 and 52 V setups. But this hobby is habituating at minimum and I find myself thinking "what if..."
Like... What if I built a couple very simple 18650 1p batteries w/o BMS... Paralleled them on the bike to get sufficient wattage for target range of say 10 miles... Charged them separately... Could be light / fast / cheap / safe... Combine the traditional benefits of LiPo with current benefits of 18650 NCA or NCM chemistries.
Probably a pointless exercise (48 or 52 V 18650 batteries with BMS do just fine in my real world needs) , just interesting topic to think through.

Availability of controllers and motors that meet the torque, power and speed requirements is the thing to find first. Not much product outside of the 36/48V commodity gear.

Controller size is always an issue on a bike , if your chasing 8 to 10 kW 120v + Is needed as controllers get big above 100amps,