Powered Up motors: L vs. XL

[FoxOne]

I'm puzzled over very similar specs of Powered Up L and XL motors, according to Philo's research.

I believe I'm not the only one though )).

 

It looks like a good move to invest only in L motors, since they are lighter, smaller and just as powerful.

But what's the catch?

 

Are there any insights on the subject? Any additional field test results?

Looking forward, thanks!

 

Edited November 24, 2020 by FoxOne
Typos

[Zerobricks]

Good question indeed. I wish there was an official site where the differences could be better explained, I rememeber the PF components had quite detailed tech. specs.

The only difference I spotted in the Lego Shop description was the following:

L motor: Perfect for higher-load tasks.

XL motor: Maximum power for the most demanding needs.

Maybe it's a good question for the next designer interview.

Edited November 24, 2020 by Zerobricks

[kbalage]

I'm sure Philo's research was far more scientific than mine, but when I tested the motors after the release I measured the XL motor to have ~10% higher RPM and ~15% more torque than the L one. 

I'll try to come up with more comparative real-life tests when I get access again to my LEGO :) 

[rm8]

7 hours ago, Zerobricks said:

Maybe it's a good question for the next designer interview.

I've asked that qeustion)

They smiled only

[Zerobricks]

1 hour ago, rm8 said:

I've asked that qeustion)

They smiled only

That tells a lot... For now the only advantage of XL seem to be the additional mounting points if you want to use it as a structural part.

Edited November 24, 2020 by Zerobricks

[FoxOne]

Thanks for the info, gents!

@kbalage Look forward to seeing your new tests, Balazs! No hurry though, please take your time )).

[Milan]

23 hours ago, kbalage said:

XL motor to have ~10% higher RPM and ~15% more torque than the L one.

Even that doesn't justify the size difference, but it would be nice to have at least some reason to get the bigger one.
Interested to see what you will come up with.

[kbalage]

5 minutes ago, Milan said:

Even that doesn't justify the size difference, but it would be nice to have at least some reason to get the bigger one.
Interested to see what you will come up with.

Practically the torque difference is similar to the one that is between the PF L and XL motors when they are geared to a similar RPM. You can check out my first test here:

 

[FoxOne]

One totally non scientific observation to add --

 

Seems like XL is significantly less geared down than L. Found this while experimenting with steering control -- it was easy to rotate XL's shaft (unpowered) by "steering" front wheels; and nearly impossible to do it with L motor.

[2GodBDGlory]

On 11/24/2020 at 3:37 PM, rm8 said:

I've asked that qeustion)

They smiled only

Could it be designed as a performance motor to be used with a yet-to-be-released power supply?

It's a long shot at best, though. Do any of you know if the electric hardware inside could take significantly more power?

[Zerobricks]

2 hours ago, 2GodBDGlory said:

Could it be designed as a performance motor to be used with a yet-to-be-released power supply?

It's a long shot at best, though. Do any of you know if the electric hardware inside could take significantly more power?

No, both L and XL motors have the same polyfuse, I think 750mA.

[biasedlogic]

On 12/1/2020 at 8:19 PM, FoxOne said:

One totally non scientific observation to add --

Seems like XL is significantly less geared down than L. Found this while experimenting with steering control -- it was easy to rotate XL's shaft (unpowered) by "steering" front wheels; and nearly impossible to do it with L motor.

And I do think this is the key thing. Less gearing down means more internal motor torque. For the same power output means less motor speed (speed before gearing). Even at the same actual power available at shaft this should mean much longer life at full load.

It's a bit like the thing with engines in cars vs. boats. I have in my car a 2.0L diesel providing 100kW power, but if you want a marine diesel at 100kW it will be a 6.0L unit - three times the size/displacement. The difference is while my car diesel can provide the same 100kW, trying to take it continuously will wreck the engine in no hours at all, while the marine diesel can put its 100kW Monday to Sunday round the clock. 

[2GodBDGlory]

10 hours ago, biasedlogic said:

And I do think this is the key thing. Less gearing down means more internal motor torque. For the same power output means less motor speed (speed before gearing). Even at the same actual power available at shaft this should mean much longer life at full load.

It's a bit like the thing with engines in cars vs. boats. I have in my car a 2.0L diesel providing 100kW power, but if you want a marine diesel at 100kW it will be a 6.0L unit - three times the size/displacement. The difference is while my car diesel can provide the same 100kW, trying to take it continuously will wreck the engine in no hours at all, while the marine diesel can put its 100kW Monday to Sunday round the clock. 

Interesting reasoning; I never thought about that!

[RussianGuy]

On 12/1/2020 at 9:53 PM, 2GodBDGlory said:

It's a long shot at best, though.

Well, in the Power functions era it was 5 years between introduction of the four-wire cable and the servomotor which used it

[BrickBuildingFox]

Another difference I discovered is that the XL is much less noisy compared to L...

[New_User]

Not sure if the PU L is still using F260 core motor as the PF L motor and are both having the same reduction ratios (24:1). 

The PF XL had RF-500 core motor with 36:1 reduction ratio, does the PU XL still using the same architecture?

Changing the core motor with different motor turn number changes the characteristic of the output. 

[Andman]

My observations:

You need precision (steering, controlling pneumatic valve): C+ L motor (C+ XL tends to overshoot, thus is less precise)

You need inertia, don't need precision and space is not an issues: C+ XL

The reason is, what others already discovered, the different internal gearing.

[gyenesvi]

I had the same observation when trying to write custom control app for PU, the L motor was much more precise. As if the XL is not really designed for steering, only for crude positioning such as for LAs.

The internal gearing is also something that has puzzled me for a while, as it does not seem to be a good match for all the recent drivetrain parts, like the planetary hub’s massive down gearing. When the XL motor (and the hub) was first introduced in 42099, it had to be geared up to achieve proper speed, which looks a bit strange, given that inside the motor it is geared down, and all the down-up-down gearing probably loses significant power. I thought the reason was that in other models, like 42100, it needed higher torque, but that could have been achieved by gearing down outside the motor. Even the buggy used a bit of up-gearing (if I remember well), and the Zetros also has very slow speed at 1:1 gearing, that could only be made faster by up-gearing. It seems to me that the motor speed does not well match the rest of the parts in recent sets (planetary hubs and heavy duty diffs), and that with faster motors, a wider range of speeds could be achieved with less power loss (it’s kind of impossible to build fast models with PU and planetary hubs, as significant up-gearing of the motors looses all torque because of too much friction).

Does anyone have an explanation for this? Could this be intentional design, or just lack of compatibility between the design of PU motors and other drivetrain components? I think the PU system leaves significant power on the table because of this misalignment. Or is this the wrong way to think about it?

[FoxOne]

7 hours ago, gyenesvi said:

Does anyone have an explanation for this? Could this be intentional design, or just lack of compatibility between the design of PU motors and other drivetrain components? I think the PU system leaves significant power on the table because of this misalignment. Or is this the wrong way to think about it?

My guess is that motors' RPM ratio is relatively slow for at least one reason -- to be able to use them right out of the box in most cases, requiring no gearing down. Well, it's a toy set by design.

On the other hand, down-up-down method is totally justified: we don't want to have transmission with high torque in between (structural strength concerns). And yes, planetary hubs are for torque in a compact  shell, not speed.

 

On the whole "L vs. XL" quest, I do recommend to watch RacingBrick's subsequent videos, it seems he's cracked this puzzle.

[gyenesvi]

3 hours ago, FoxOne said:

to be able to use them right out of the box in most cases, requiring no gearing down

Well, sounds like a nice idea, but the very first set that used them (42099) is already an exception, as it needed up-gearing. Of course a single kind of motor will never be enough for all models without some gearing. That's why it seems odd that all PU motors have roughly the same speed.

3 hours ago, FoxOne said:

down-up-down method is totally justified: we don't want to have transmission with high torque in between

How is that justified? I understand what you say about less torque in between, but if you have a faster motor with less torque to start with, you'd have the same effect for the transmission but with less loss off efficiency due to the gearings that cancel each other and only add friction.

3 hours ago, FoxOne said:

I do recommend to watch RacingBrick's subsequent videos, it seems he's cracked this puzzle.

Sure, I have seen them long ago. As I remember, the conclusion there was that the XLs are a bit better for high torque applications. Well, the Zetros seems to kind of defy that argument, as it needs high torque, but uses two L motors. Even the Cat does not use linear XL motors, and as I have read it struggles with raising the blade (it is too slow as it is, and if geared up, it does not have the torque). 42100 uses XLs for drive, but with a lot of down gearing, and the technique has been kind of enhaced by the Cat's planetary drivetrain. Then there's 42114 that uses one XL for drive, which is a bit weak. So all in all, I don't see great usages of the XL motor.

[FoxOne]

20 minutes ago, gyenesvi said:

Well, sounds like a nice idea, but

I've mentioned "out of the box performance". This means you can just attach a wheel directly and expect some reasonable performance, as a sweet spot between torque and speed. Since it's an all purpose toy motor.

Don't see how 42099 is an exception here, because it has rather advanced drivetrain. Ultimately, if you want to kick it up a notch (with planetary hubs...), you always have an option to use buggy motors )).

Quote

 

Sure, I have seen them long ago. As I remember, the conclusion there was that the XLs are a bit better for high torque applications.

 

I believe there were 2 or 3 of them. Final conclusions were somewhat more sophisticated, regarding maximum current and cutoff levels.

[kbalage]

1 hour ago, gyenesvi said:

Sure, I have seen them long ago. As I remember, the conclusion there was that the XLs are a bit better for high torque applications. Well, the Zetros seems to kind of defy that argument, as it needs high torque, but uses two L motors.

It does not defy anything, as the Zetros already has way more torque than it requires. 

[gyenesvi]

2 hours ago, kbalage said:

It does not defy anything, as the Zetros already has way more torque than it requires. 

That's exactly my argument here. I have just rechecked your video about this, and your conclusion was that the XL motor has an advantage for big heavy and slow models (pretty much the description of the Zetros), and especially if higher voltage batteries are used. So if even the Zetros is not big and heavy enough to warrant the usage of XLs, then what is? The Cat? That might have been too heavy for it, as it uses the angular XLs that have 50% more torque (besides their form factor seems to be better for that build).

So the utility of XLs still seem too marginal, though the they could be useful for different applications (for example a single XL motor driving an LA to lift something), but I am struggling to find really good examples, where the XL is a clear winner over the L (for example, the Cat also uses an L for lifting).

3 hours ago, FoxOne said:

I've mentioned "out of the box performance". This means you can just attach a wheel directly and expect some reasonable performance, as a sweet spot between torque and speed. Since it's an all purpose toy motor.

Hmm, this is exactly my problem; if you build such a thing with the recent drivetrain components used nowadays for stock RC cars, the result is too slow. First, a wheel is rarely attached straight to the linear motors. Usually there is at least a diff and a wheel hub in between, and TLG seems to push the heavy-duty diffs and planetary wheel hubs / portal axles for good reason: with them things are less prone to break because the high torque only manifests itself at the end of the drivetrain. Second, the Zetros is an example of a straight to the wheel drivetrain in this sense; the motors go 1:1 to the axle diffs after coupling (the central diff is irrelevant). But the result is such a slow speed that is hardly playable and the torque is almost too much (for a trial truck that setup is acceptable, but for other kind of vehicles?). To get a sweet spot with these components, faster motors would be needed.

The other option is to build without planetary hubs. It is faster then, but then there's too much stress on the drivetrain that can break some joints or skip some gears (not to mention that steering angle is worse if the front is driven as well).

Also, about other RC functions like LAs, stock models are often criticised for being too slow to operate.

3 hours ago, FoxOne said:

Don't see how 42099 is an exception here, because it has rather advanced drivetrain.

I would definitely not call the drivertrain of 42099 advanced. It's one of the simplest possible, the motors go almost directly to the diff with some gearing that is needed for routing the driveshaft because of the high build of the chassis (and can be also used for adjusting speed/torque). No coupled motors, no gearbox, no floating axles, it's not even a realistic AWD (separate motors front and back). It's okay though for an RC car, I just don't really see it advanced.

3 hours ago, FoxOne said:

Ultimately, if you want to kick it up a notch (with planetary hubs...), you always have an option to use buggy motors )).

That's where it gets interesting. For one, the buggy motors (or Buwizz motors) are faster by 4x-6x, which is too much, it will damage the parts if you build a realistic drivetrain with it (not straight to the wheel). Second, its form factor makes it far from easy to use inside the chassis, (not inside the axles or in place of the diff in an independent suspension), typically, it would have to be put into the middle of the cockpit of a car to get the output where it should be. Actually, that is the motor that has been designed to put a wheel straight onto it, but that's also somewhat limited usage.

4 hours ago, FoxOne said:

Final conclusions were somewhat more sophisticated, regarding maximum current and cutoff levels.

How does that translate to practical usage? Isn't that just proportional to max stalled torque?

[FoxOne]

In no particular order:

1. I believe if one looks for a performance, he might want to switch to Hobby RC.
Lego is a toy set. Yeah, it's slow and not powerful -- personally I'm okay with it. Don't want some crazy 20,000 RPM motor to spin in my kid's hands, nor some serious torque to crash some tiny fingers.

2. 42099's drivetrain may not be advanced -- I agree, not the best choice of words. My point was drivetrain is long, with many gear couplings and ability to tune it by easy swapping.

3. Re Zetros -- wheel radius should be accounted for. 42099 has some monster size wheels to rotate.

4. Re practical usage. No, I believe we can't reduce it to max stall torque. As a complication, at least we have different power consumption and also should review cutoff levels. Yes, L can bring some serious torque, but it will swallow so much current it will trigger cutoff. Not the case with XL.

As a side note, it would be interesting to see your explanation, sir. We try to guess, true, since Lego has not provided any insights.
But I suppose TLG is neither stupid nor inexperienced, maybe it's just a lack of communications.

Edited December 28, 2021 by FoxOne

[Igor1]

I don’t understand at all the meaning of all the repeated discussions of PU L vs XL. There are technical characteristics of motors and they have been known for a long time: https://www.philohome.com/motors/motorcomp.htm.

The L motor is technically not much different from the XL. There is no need to look for some other hidden meaning, they are what they really are.

TL;DR PU XL sucks.

Edited December 28, 2021 by Igor1