diy solar

diy solar

LFP for EV's projected to be less than $56 per kWh within 6 months

MattiFin said:
Voltage makes no difference in motor size or winding size, just wire gauge and number of required turns.
Practical limit for lowest reasonable voltage would be 1-turn winding.
Click to expand...
So...
A 25hp 48V motor can be the same size as a 25hp 800V motor?
 
Rednecktek said:
Pretty much, yes. The 800v might be a little smaller.

I know our 240v and our 440v motors are about the same size physically on the ships.
Click to expand...
Hardly the best example. Want to know something cool my 120v motor can be wired for 240v.
 
All of this is apple to banannas comparison.

A 75+kW output motor would need impressive amounts of windings and conductor size at 48V compared to 800V.
The energy flow alone would be enormous...
1600A vs 100A...

Just imagine that for a minute...
 
1600A vs. 100A. Or maybe that is 3 phases of windings, 533A vs. 33A.

Simply 16x the wire cross section but 1/16th the number of turns. Same magnetic field strength.
perhaps easier to wind with 16 wires in parallel, rather than same total wire in series.

About the same as secondary vs. primary transformer windings if stepping down 800V to 48V.
 
Supervstech said:
A 75+kW output motor would need impressive amounts of windings and conductor size at 48V compared to 800V.
The energy flow alone would be enormous...
1600A vs 100A...
Click to expand...
The motor only cares about the magnetic field. If it takes 32 turns at 100 A or 2 turns at 1600 A, it is the same current density in the overall winding and thus the same magnetic field. A single turn motor is technically possible.

When leaving the motor, however, the 1600 A current is a huge challenge and the 100 A is not.

Mike C.
 
Hedges said:
Not pointless, just you may be OK with constant torque, constant thrust, same at low speed as at high. That is to say, horsepower linear with speed.
That could work for typical car use.

If you use 100 HP driving into the wind at 100 mph, are you OK only having 10 HP carrying a load uphill at 10 mph?

A 2-speed transmission would let you have much higher pulling power in low gear.
(even some winches are 2-speed.)
Click to expand...
Hmm. It's really engineering, but I was talking about cars where a linear curve and a generally static load it's kind of silly. Electric motors have a very wide rpm range. The reduction gear in my EV is obviously going to be optimized but once you get the RPM's up ever so slightly, you just keep pushing up to max rpm. Thus you can engineer for the worst case torque needs . . . To get up the hill in a headwind at 100MPH. Set your ratio, then just spin the RPM's up. This is what makes these darn things so crazy. There is no penalty from an energy use standpoint, gearing it up does not save any watt-hours, you keep applying more juice the motor just turns faster and faster.

My PHEV sonata had a 70HP electric, which was tied in behind the DCT along with the 200-odd HP ICE. Shower down the ICE turns on, driving normally the electric motor seemed fine even on hills never downshifted. YMMV. I think once you get rid of the bulky ICE it's simpler from an engineering standpoint to build a more powerful electric motor, which just isn't that much bigger . Dunno, I was actually surprised the EV did not have a gearbox of any kind.
 
Supervstech said:
All of this is apple to banannas comparison.

A 75+kW output motor would need impressive amounts of windings and conductor size at 48V compared to 800V.
The energy flow alone would be enormous...
1600A vs 100A...

Just imagine that for a minute...
Click to expand...
Should only need to be fatter at the feed point/bus bar. The amount of wire in the windings is roughly the same, your just going to tap it multiple times with lower voltages to get the field up. Same idea as a multi-tap dry transformer or mixed voltage motor. Wiring for 240 run the voltage end to end making one long winding, wiring 120 you have a center tap, and you flip-flop the 120 across the same wires split in the middle. The number of amps across each cross-sectional winding is the same. Obviously some engineering trade-offs busing, and optimizing the windings / wire sizes as you scale the voltages up and down.
 
Rednecktek said:
19 pages later, the only question I have is when are those prices going to hit us? Even now it's over $100/kwh for cells. 😥
Click to expand...
I purchased 280Ah EVE from Luyuan and with shipping included it came in at $102/kwh.

Close. Cells were $78 each. All tested over 304Ah so technically it is less than $100/kwh as that comes to over 93 Kwh instead of 86 Kwh.

It's the freight that cost $1338 that drives up the cost. With the current situation in the Middle East most shipping companies will be charging more. Sure, you might see cell prices lower but freight costs might make those prices irrelevant.
 
ksmithaz1 said:
Should only need to be fatter at the feed point/bus bar. The amount of wire in the windings is roughly the same, your just going to tap it multiple times with lower voltages to get the field up. Same idea as a multi-tap dry transformer or mixed voltage motor. Wiring for 240 run the voltage end to end making one long winding, wiring 120 you have a center tap, and you flip-flop the 120 across the same wires split in the middle. The number of amps across each cross-sectional winding is the same. Obviously some engineering trade-offs busing, and optimizing the windings / wire sizes as you scale the voltages up and down.
Click to expand...
Ahh.
Well, that makes no sense to me, but im just a wire jocky, i dont get into the appliances.
 
MattiFin said:
"Sweet spot" for mosfets or bulk dc-rail capacitor is currently somewhere around 400 volts. Superjunction mosfets totally changed the game and SIC is pushing the performance even further.

And 60v mosfet would be usable only for maximum 24v system unless you like to design your inverter for even less margin than Chinese do.
For 48V nominal system(max 60v input) 100v would be reasonable minimum, leaving 40v margin.
48v nominal 5kV inverter would need say 10x 100v 10mOhm mosfets in parallel for 10W conduction losses. Total gate charge to drive would be around 400 nC.
400V nominal system would handle the same power with single IPP60R065S with a total gate charge to drive only 51 nC. And it cost less than 10x low voltage mosfets.

On 400V system you have 10x lower part count, fraction of gate charge to drive, WAY simpler layout when you don't need to worry about paralleing 10x high current switches and lower component cost.

Like stated earlier, if 48v would make sense the electric cars would use it. Voltage required by motor is not any relevant argument, all EV motors are custom built and if it would make sense they would be wound for 48v system.
Click to expand...
Again, some points are valid.

But again I don't believe the argument of 48V being more common and cheaper (by Inverter/Batteries manufacturers) is for the DIY community ...

There must be another reason. And yes, it's your problem to buy the extra thick cables for such system.
 
ksmithaz1 said:
There is no penalty from an energy use standpoint, gearing it up does not save any watt-hours, you keep applying more juice the motor just turns faster and faster.
Click to expand...

To get higher torque from same motor, you have to drive more current. Above what it is designed for, that can only be done briefly, e.g. "ludicrous mode". Anything more would burn it up. Also more I^2R loss.

For maximum power and torque, want to run motor at max RPM, geared down to operating speed. Gearshift for some applications, although single speed is good enough for most EV. For a more quiet and pleasant ride, together with get-up-and go without shifting, you want a 400 cubic inch "V8" loafing along at low RPM, not a turbocharged 800CC screamer. Unless that is your thing.
 
Supervstech said:
All of this is apple to banannas comparison.

A 75+kW output motor would need impressive amounts of windings and conductor size at 48V compared to 800V.
The energy flow alone would be enormous...
1600A vs 100A...

Just imagine that for a minute...
Click to expand...
Well if you want to take it up a notch we can talk about 5MW+ Turbine Generators and Transformers.

I'm not saying all manufacturers do it, but in my field we are around 690VAC on the grid (LV of Transformer) and up to 1000VAC on the loaded generator. You get several kA in each converter (and reactor) ...

Reason being ? Sometimes cost, sometimes efficiency. Unless you go MMC or some other topology, there just isn't another way. And going from 1700V IGBT to 3300V IGBT isn't going to work loss-wise. 690VAC is the defacto standard for high power while still being LV (400V is for lower power). And manufacturers and users keep pushing the boundaries to even 1000VAC "plus tolerances" sometimes 🤣 .

It's NOT easy to insulate a MV/HV system. And it's a whole other beast to deal with. You might argue people get complacement and killed already in "LV" (<1000VAC/1500VDC). HV is to some extent safer and people respect it more ... Just thinking about NOT having to earth "LV" systems during maiontenance and the number of electrocution that happen as a result of that ... In HV it's MANDATORY to do it !

You start having all sort of extra issues (corona discharge, partial discharges, ...). And the clearances/creepages/insulation requirements go up exponentially (look for clearance for partial discharges as an example ...).

You can argue high voltage it's easier, but then you will have insulated stacks like in HVDC, oil filled transformers, etc.

1709572290568.png

EDIT: and sometimes you also have patents to deal with. So, at least in some fields, you cannot even use other de-facto standards with 3-level NPC/ANPC, etc, or you do not get access to second source supplier and other stuffs.
 
Last edited:
With an electric the torque is there from the get go. Typical ICE need to be around 2000 RPM to get to optimal. Torque is what snaps you off the line, horsepower is what gives you a shove when you are already moving fast. Rotary's dominated LeMans for a few years until the fuel restrictions took them out. Very linear steep horsepower curve at high RPM. Practically no torque for the displacement. As long as your RPM stays in range horsepower takes over at higher RPM's, as the torque drops off. I think the ICE curves here are somewhat generous, although modern ICE engines are pretty darn amazing.

The problem is with energy consumption. Gearing on an ICE vehicle is to hit the knuckle ~2000 RPM when the ICE is at it's most efficient. At high RPM a gas powered car starts drinking fuel at dramatically higher rates. I remember watching a NASCAR race, one of the drivers had a later pit, everyone else needed fuel. They calculated he could make the last few laps without additional fuel but he had to keep the RPM's way down, they did all the math on everyone else catching up a lap over the remaining time/laps, he pinned the RPM needle at the needed spot, and won the race.

Electric motors do not suffer from this. At 6000 RPM they use about three times the "fuel" they would at 2000. An ICE engine on the other hand will use 10-15 times the fuel over the same interval. This means from a practical standpoint you want to gear it down to get back to around 2000 RPM to save fuel. Start climbing a steep hill it just doesn't have the oomph at the higher gear ratio, so we have to get the RPM's up to optimize our horsepower at the expense of burning a much larger quantity of fuel/mile driven.

And that is why you generally don't need a gearbox on an electric. There really isn't an optimal "RPM" in terms of fuel consumption, because the consumption is linear. If you wanna go faster just spin the motor faster, you may lose a bit of torque as you really ramp it up, but it won't affect the efficiency.


1709572705605.png
 
ksmithaz1 said:
With an electric the torque is there from the get go. Typical ICE need to be around 2000 RPM to get to optimal. Torque is what snaps you off the line, horsepower is what gives you a shove when you are already moving fast. Rotary's dominated LeMans for a few years until the fuel restrictions took them out. Very linear steep horsepower curve at high RPM. Practically no torque for the displacement. As long as your RPM stays in range horsepower takes over at higher RPM's, as the torque drops off. I think the ICE curves here are somewhat generous, although modern ICE engines are pretty darn amazing.

The problem is with energy consumption. Gearing on an ICE vehicle is to hit the knuckle ~2000 RPM when the ICE is at it's most efficient. At high RPM a gas powered car starts drinking fuel at dramatically higher rates. I remember watching a NASCAR race, one of the drivers had a later pit, everyone else needed fuel. They calculated he could make the last few laps without additional fuel but he had to keep the RPM's way down, they did all the math on everyone else catching up a lap over the remaining time/laps, he pinned the RPM needle at the needed spot, and won the race.

Electric motors do not suffer from this. At 6000 RPM they use about three times the "fuel" they would at 2000. An ICE engine on the other hand will use 10-15 times the fuel over the same interval. This means from a practical standpoint you want to gear it down to get back to around 2000 RPM to save fuel. Start climbing a steep hill it just doesn't have the oomph at the higher gear ratio, so we have to get the RPM's up to optimize our horsepower at the expense of burning a much larger quantity of fuel/mile driven.

And that is why you generally don't need a gearbox on an electric. There really isn't an optimal "RPM" in terms of fuel consumption, because the consumption is linear. If you wanna go faster just spin the motor faster, you may lose a bit of torque as you really ramp it up, but it won't affect the efficiency.


View attachment 199980
Click to expand...
You are right for the reason you outlined. For a pure EV at least. But high-torque requires you to also have a sizeable power stack. A low-cost, lower current power stack could in principle be used if you are willing to sacrifice a bit your max torque.

Hybrid I think it's another story, depending on the kind (series/parallel). And depending on the power stack, motors and batteries. It's a system analysis after all.

But also electric motors and generators sometimes need a gearbox. Too high torque or too low rpm isn't good (Power Cycling of Semiconductors for instance, <16Hz isn't really "standard" anymore - 16.6666... Hz being what's used in Germany Railway for instance). You also have problem getting standardized industrial parts (16.666Hz is the lowest "standard" Breaker for instance here in Europe at least, anything lower typically requires you to buy a DC Breaker). Probably similar problems related to other components as well that are NOT qualified for such low operating electrical frequency.

50Hz/60Hz is just so much easier. Think about the size of 16.666 Hz Power Transformers. Yes, lower losses (potentially), but look at how big they need to be in order to handle the lower frequency and avoid saturation.

Your motor can handle "any speed", but what about your Power Stack ? You still need to be able to generate a "proper" voltage (and current) waveform the higher up you go and your control/measurement system needs to be able to follow that. Sure, everything can be designed based on the outcome of your system analysis. And of course this also influences the Diode/IGBT-MOSFET Loss Distribution, Thermal Cycling, Power Cycling, etc.

I learned quite quickly in my career that you cannot design an efficient, cost-effective power converter that does say 0-1000A and 0-1000V at the same time. You optimize for an operating point/range, and everything else is a bonus basically. Yes, you can do variable switching frequency and plenty of other stuff, but as usual, everything has compromises (losses/cooling, cost, complexity, reliability, ...). You won't have the same efficiency (and cost) if you over-design everything.
 
silverstone said:
You are right for the reason you outlined. For a pure EV at least. But high-torque requires you to also have a sizeable power stack. A low-cost, lower current power stack could in principle be used if you are willing to sacrifice a bit your max torque.

Hybrid I think it's another story, depending on the kind (series/parallel). And depending on the power stack, motors and batteries. It's a system analysis after all.
Click to expand...
I think this is key. I think for pure EV's the cost-benefit of downsizing the motor components is outweighed by the expense and power loss of a gearbox. With a hybrid you need the gearbox anyway. The difference in manufacturing cost of a 150HP electric motor vs a 75HP one has got to be fairly small when done at scale, and the size and weight differences are likely not that great. Factor in complexity / build costs (more components to bolt on more money to build it) it likely isn't worth it for a common EV application:

"The Tesla Semi has a single gear and uses four electric motors that operate independently to power its rear wheels."
 
You must log in or register to reply here.
Back
Top