Siemens 1PV5138-4WS24

[Stiive]

The motor is rated at 300VDC max. You can get 10k RPM at full power with sag at that voltage

It's torque I'm interested in. Also, I never understood why they give an ACIM a DC voltage rating. That tells me nothing. Give me an AC voltage and a base frequency. I'm guessing they only intended it to be used with a (their) matched controller.

I disagree, its always the power your interested in - you then set the gear ratio to get the torque you want. This motor still has 90kW @ 8,000 RPM (30seconds) and 60kW+ continuous.

The constant torque is constant 300Nm to just over 3k RPM, which gives the peak of 100kW for 30secs.


*According to my simulations* With the standard 8.28:1 borgwarner direct drive for this motor, in a smallish un-aerodynamic 5door hatchback weighing 1500kg, your looking at;

- 0-100kmph of about 7.4sec and top speed of about ~145kph reached in ~22.5seconds. Interesting enough, the top speed is limited by RPM (10k) and not torque.

If you made the vehicle lighter, this would obviously improve the acceleration times significantly.

[woody]

Depending on what you are doing, a final drive of 8.28:1 or thereabouts may be hard to arrange in a car, with most final drive ratios around 3:1 and maybe 5:1 (Hilux/Hiace) being the upper limit of cheaply available, 6.5:1 the highest I've seen in aftermarket (Ford 9 inch).

Maybe a Hilux 5:1 behind a power-glide 1.73:1 is close? (8.52:1)

A thread on DIY electric car has a load of info on power glides.

[BigMouse]

The motor is rated at 300VDC max. You can get 10k RPM at full power with sag at that voltage

It's torque I'm interested in. Also, I never understood why they give an ACIM a DC voltage rating. That tells me nothing. Give me an AC voltage and a base frequency. I'm guessing they only intended it to be used with a (their) matched controller.

I disagree, its always the power your interested in - you then set the gear ratio to get the torque you want. This motor still has 90kW @ 8,000 RPM (30seconds) and 60kW+ continuous.

The constant torque is constant 300Nm to just over 3k RPM, which gives the peak of 100kW for 30secs.


*According to my simulations* With the standard 8.28:1 borgwarner direct drive for this motor, in a smallish un-aerodynamic 5door hatchback weighing 1500kg, your looking at;

- 0-100kmph of about 7.4sec and top speed of about ~145kph reached in ~22.5seconds. Interesting enough, the top speed is limited by RPM (10k) and not torque.

If you made the vehicle lighter, this would obviously improve the acceleration times significantly.

Yes, you're correct. But I want my torque to last, hence giving higher power. What's better for performance? 300Nm to 3k RPM? Or 250Nm to 7.8k RPM? *According to MY calculations* the latter should take my 1600kg BMW from 0-100 in 6 seconds.

250Nm at 7800rpm is ~200kw. My 48vac wound 4 pole 132 frame motor should be capable of (at least approaching) that. Some water cooling might be required, but I'll do everything I can to get it there. A 364v battery pack supplying a 48vac motor should allow constant V/F up to 265Hz with peak battery current of around 630A or so.

What voltage battery pack would I have to run to get the same performance from the Siemens motor? Approaching 700vdc I'd imagine, but it'd be a hell of a combination!

[Stiive]

250Nm at 7800rpm is ~200kw. My 48vac wound 4 pole 132 frame motor should be capable of (at least approaching) that. Some water cooling might be required, but I'll do everything I can to get it there. A 364v battery pack supplying a 48vac motor should allow constant V/F up to 265Hz with peak battery current of around 630A or so.

What voltage battery pack would I have to run to get the same performance from the Siemens motor? Approaching 700vdc I'd imagine, but it'd be a hell of a combination!

Yeh I'm not exactly sure what the limiting factor of voltage is - maybe someone here knows? I'm sure the insulation of the windings isn't the issue.

Perhaps its something to do with the time constant of the rotor - and switching fast enough to induce full magnitude of current in the rotor at high frequencies - obviously higher voltage allows for this, but maybe theres a saturation effect where increasing voltage past a certain point gives a diminishing return - can't think of why though - maybe switching losses in the IGBTs? Or dead time / propagation delay / timing inaccuracies? Perhaps increased harmonics and CEMF?

Otherwise, yeh, whats to stop you from running the 1PV5135 @ 650V like the 1PV5138 and producing 200kW+?
Surely with marketing these days if it could be safely done, they would have marketed it like that. The fact that they don't leaves me a bit skeptical about your dreams [Edit - hopes/goals/aspirations/etc - 'dreams' not meant in a derogatory way].

Again, only one (sure) way to find out... hurry up and FIRE HER UP!! (not literally)

[Richo]

whats to stop you from running the 1PV5135 @ 650V like the 1PV5138 and producing 200kW+?

Ah it'll be the bearings and balance.
If you can get 10krpm@300V@100kW then 20krpm@600V@200kW isn't realistic.
Doubling the current will just overheat it quicker.

I still haven't seen any high speed bearings for big machines (ie 132 frame and up)

[Stiive]

whats to stop you from running the 1PV5135 @ 650V like the 1PV5138 and producing 200kW+?

Ah it'll be the bearings and balance.
If you can get 10krpm@300V@100kW then 20krpm@600V@200kW isn't realistic.
Doubling the current will just overheat it quicker.

I still haven't seen any high speed bearings for big machines (ie 132 frame and up)

No, I mean increasing the 'knee point' where field weakening occurs from 3,200RPM, to 6,400RPM.
Theoretically this means the motor would retain peak flux and produce 300Nm up till 6,000RPM - doubling the power output.

[Johny]

No, I mean increasing the 'knee point' where field weakening occurs from 3,200RPM, to 6,400RPM.
Theoretically this means the motor would retain peak flux and produce 300Nm up till 6,000RPM - doubling the power output.

But if you double the speed at which torque starts to drop (lower voltage motor) you effectively double the current at lower speeds. There must be some point where the copper becomes a fuse.

[Stiive]

if you double the speed at which torque starts to drop (lower voltage motor) you effectively double the current at lower speeds. There must be some point where the copper becomes a fuse.

Sorry I don't follow what you mean?
The winding remains identical. I'm still applying just 300V at 3,200RPM.

[Johny]

The winding remains identical. I'm still applying just 300V at 3,200RPM.

Sorry - you're right - ignore me....

[BigMouse]

There's an excellent thread elsewhere on this forum about overvolting a motor for more power (same V/F to higher F). Many of you were participants in that thread, and it's the one that helped convince me to go this route.

If dead-time or switching losses become a problem at higher commutation frequencies (like you, I don't see why they would), then the switching frequency can be varied (reduced) as required. No readon for that though, as even at 400Hz, a switching frequency of 16khz would still have 40 steps per cycle. At higher frequencies, the resolution should be less important anyway due to the inductance of the windings. The current would still be "sinusoidal enough" at those frequencies. Timing inaccuracies even at 16khz should be negligible with a 32Mhz clock speed.

I think the main reasons this isn't done in practice it due to the high currents involved in low voltage machines. Specifically the added cost/weight of the larger IGBTs, wiring, and a battery pack that can output 200+kw.

Thankfully, IGBTs are getting cheaper, wiring doesn't add that much weight or cost that much more, and the CALB CA series cells have been shown to perform very well. I've seen numbers as high as 12C usable. 200kW on my planned pack would be less than 8C.

I feel like I'm hijacking this thread. Maybe I should make a build thread for my car.

[Stiive]

I feel like I'm hijacking this thread. Maybe I should make a build thread for my car.

No, this isn't a hijack because if there is no draw back from over voltaging the 5135 to 600V, why would OP bother spending double the amount of money on the 5138?

I've always wondered how much you can over-voltage the motor. I've done the 1/4voltage hack on an industrial IM, then ran at full voltage no problem - but this is slightly different to running a motor at double the voltage without changing the windings. 1/4 voltage hack essentially quarters the turns and quadruples the wires 'in hand' from the original design. The rotor remains unchanged though, so I guess this rules out the rotor time constant being an issue.

Anyone run their 415V 3-phase industrial motor unchanged @ 830V+?

[TooQik]

I feel like I'm hijacking this thread. Maybe I should make a build thread for my car.

No, this isn't a hijack because if there is no draw back from over voltaging the 5135 to 600V, why would OP bother spending double the amount of money on the 5138?

I've got no issue with the ongoing discussion in this thread.

I think it's probably worth pointing out that there already exists a 5135 that Siemens rate for 650 VDC, the 1PV5135-4WS28. If the question is why would I choose the 650 VDC 5138 over the 650 VDC 5135, then I'd answer because the 5138 produces 80 Nm more torque at the same current and my preference is not to be playing with drive ratios to get the torque figures I want.

[Richo]

Anyone run their 415V 3-phase industrial motor unchanged @ 830V+?

I don't think so.
Although I'll probably be able to do this hopefully at the end of the year.(fully charged pack with unmodified motor).
But the higher the voltage the more likely the insulation will break down in the motor or other parts in the system.
The point will be there are cheaper ways to get power rather than just upping traction voltage.
Given the common industry standard for 400/415V making this the voltage the cheaper option.

So buying a motor for 600V really warrants it to be rewound for more power.
Like you said before the 300V siemens motor may be better choice to start with.
As long as the Max RPM isn't exceeded.

[T2]

Too Qik wrote
my preference is not to be playing with drive ratios to get the torque figures I want.

This is the major mistake most amateur constructors make.

The better strategy is to think of the gear box as a way to make the outside world look easier to the powertrain.

Use of an additional single ratio torque multiplier mounted to the motor is technically the best way to go for generating torque as opposed to fitting a larger, heavier and more expensive motor.

Of course how that is to be accomplished is totally another thing.
I have seen all possible meshed gears removed from a standard transmission excepting those for first and second gear which are part of the input shaft itself. Sure it whined some. In hindsight we should have had new input and output shafts manufactured. Today finding someone to write G-code for a CMC may not be a problem.

[Stiive]

But the higher the voltage the more likely the insulation will break down in the motor or other parts in the system.

I dont think so... The mag wire insulation should be good for well over 1,000V. The 5135-4WS14 passes a megger test at 2.5kV

The point will be there are cheaper ways to get power rather than just upping traction voltage.
Given the common industry standard for 400/415V making this the voltage the cheaper option.

Not in this case, you cant get more power out of the 5135-4WS14 without increasing voltage further. Question is, what is the limiting factor?
The 5138 is always going to be more powerful though because it is physically bigger - just the lower voltage winding on the 5135 makes it more practical to overvoltage - I cant see anyone going a 1200V build to double voltage the 5138 .

I'm in the middle of making a 750V controller, since I have a spare 5135-4WS14, maybe ill give it a try - though i don't want to cause any damage

[TooQik]

Just thought I'd update this thread for anyone interested, I've ordered a 1PV5138-4WS24 along with a TI-9030 inverter from HEC and hope to have my hands on them in about 6 weeks.

[Jodie]

I'm interested,

Im up the road in Cockatoo.
I have a 5135 unpacked.
Interested to see how your inverter goes.

Jodie

[Richo]

Yep the 5138 should be good with oodles of continuous torque.
Here is the link to the motor specs
http://w3.usa.siemens.com/us/internet-d ... 0Sheet.pdf

The 5135-4WS24 passes a megger test at 2.5kV.

Sure but I said more likely.
Add heat cycling and voltage spikes and it won't pass that megger test for ever.

[TooQik]

Im up the road in Cockatoo.

A mere stone's throw away.

Will start a conversion thread once I actually start getting my hands dirty and I find my digital camera.

Yep the 5138 should be good with oodles of continuous torque.

I'm looking forward to taking the car to a hillclimb meet once completed and seeing how it goes.