Jump to content

Peak power’s peak duration?


mrelwood
 Share

Recommended Posts

When choosing to buy an EUC, seems that many of us organize the candidates by the nominal power.

Say one starts to accelerate hard into a steep long uphill. How long can an EUC keep up with the peak power? At what point does it fall back to deliver only nominal power?

Link to comment
Share on other sites

2 hours ago, mrelwood said:

When choosing to buy an EUC, seems that many of us organize the candidates by the nominal power.

Say one starts to accelerate hard into a steep long uphill. How long can an EUC keep up with the peak power? At what point does it fall back to deliver only nominal power?

If you look at 7918cNC.jpg

(Taken from my http://forum.electricunicycle.org/topic/7855-anatomy-of-an-overlean/ topic) for an KS16C.

There are the dark red (800W), medium red(1200W) and pink lines(2200") which show constant power output. 

The grey areas are (depending on the battery voltage) the not reachable areas - The motor cannot provide this much output power of example the pink 2200W output power line.

Along the blueish limit lines (light blue - highest battery voltage till dark blue - lowest battery voltage) one could imagine the corresponding constant power lines, which are reachable within this limit. The more the battery voltage drops (by discharge and temporarily by "internal chemical processes" but not by normal voltage sag over the internal resistance - this voltage sag by internal battery resistance is already regarded in the limit lines) the lower this limit line and by this the maximum reachable output power gets.

So in this example the KS16C will stay "forever"(unitl EUC shuts of in because of low battery) in the range above nominal output power of 800W if one stays in the speed range above approx 14 km/h. At lower ranges the current is quite high and could/should lead to overtemp situations while asking continously for this amount of output power.

Edit: PS.: please ignore values for the violett lines for now (current for incline xx and acceleration of xx) for now - i am not sure about their scale!

Edit: The limit lines are determined by

- The "no load" battery voltage which specifies the max no load speed at one end and

- coil resistance and internal battery resistance at the other end. coil resistance can vary by something around 20% depending on the temperature of the copper in the coils and also battery resistance can vary depending on load/age/temperature/etc...

- the 80 A limit is choosen arbitrarily - don't have any idea of the real value, but fits in the diagramm and somewhere around up there should/could be some limiting...

Edited by Chriull
  • Like 2
  • Upvote 1
Link to comment
Share on other sites

Thank you, Chriull! For some reason I thought the ”peak” value is limited by time, perhaps since it is always such in the music/audio world. But if I now understood correctly, it is limited by the speed/current curve instead. Therefore:

  • Nominal power can be delivered or exceeded at any speed.
  • Peak power can be reached only at a certain speed, usually around 2/3 of max speed.

When I got my KS-16S (1200/3000W) I watched the power usage from the KingSong app while slowly going up a very steep incline. I reached a maximum of 1300W, and thought that there is tons more power available. So I decided to accelerate and go as fast as I dared to get a feel of the available power. I got up to 1500W (bumpy dirt road, 19degree incline, 91kg rider) and thought that I will never reach the max power of the 16S. But since the speed was still quite slow, I might have been actually quite close to the maximum.

Link to comment
Share on other sites

13 hours ago, mrelwood said:

When choosing to buy an EUC, seems that many of us organize the candidates by the nominal power.

What is usually provided is the nominal and peak power of the motor, not of the wheel. So it is mostly for selling purpose and I would say not very useable.

Definition is : The nominal power in kW (Pn) of a motor indicates its rated equivalent mechanical power output.

That would be the mechanical output that the motor could provide indefinitely, without heating/over heating... But what it you can cool the motor ? Moreover the efficiency of the motor vary from almost 0 to 80% depending on the rotational speed. If you have 800W mechanical output at 30% efficiency, that would require 2400W of electrical power coming from the mainboard... and some cooling if you want to keep it continuously !

Peak would be the maximum mechanical power, like the horsepower of a car motor which you usually don't have because they provide the real horsepower of the car measured on bench test instead (some Russian guys do it for the wheel).

For brushless DC motor I think max power is usually provided at half max empty rotational speed under heavy load... While for a combustion motor, max horsepower is delivered at high rotational speed with load. 

For a motor car to deliver the max power, it requires a big fuel rail and air intake. For your wheel it is battery power and mainboard power but we do not have informations about these... (also knowing the cell type and battery pack configuration can give half of the information)

Edited by jbwheel
  • Like 2
Link to comment
Share on other sites

4 hours ago, jbwheel said:

@mrelwood here is the website providing real power output test.

A test of the GT-16 858Wh

http://ecodrift.ru/wiki-article/rockwheel-gt16-review/

"Conclusions

EUC is very questionable and controversial. Such a unicycle is 100% contraindicated for the beginners. There are no factory user restrictions. This will all lead to a fall. Therefore, it is necessary to use good protection. If you plan to go above 35 km/h, then you need enhanced protection, like a motorcyclist. And necessarily — a good helmet. It is desirable that it was a full-face.

As for the design of the EUC, we were convinced that everything was done only for a beautiful picture. One should not wais for the practicality and benefits from such a solution.

Buy this device only if you understand 100% what you need it for, and where and how you will use it."

Ouch. Its beauty is only skin deep. [I have come close to buying one anyway several times :blink:]

I realize now that this article was posted at the beginning of the year when the wheel first debuted, so nothing new here. 

edit: see italics

Edited by litewave
Link to comment
Share on other sites

11 hours ago, litewave said:

I realize now that this article was posted at the beginning of the year when the wheel first debuted, so nothing new here. 

You forgot to pick this :

Rockwheel GT16 is a very powerful electric unicycle. For now, it is the leader of our tests among EUCs with a diameter of 16 inches.

...

As a comparison, among the 16″ unicycles are presented custom solutions based on Gotway ACM and Inmotion V8, which were about one third more powerful than their regular versions. And nevertheless, GT16 confidently bypassed them in power, leaving them far behind.

There has been some improvement, the mainboard seem to be tropicalized now. Also the bigger battery version would provide  more than +50% continuous power if NCR18650GA is used instead of NCR18650PF. 

  • Like 1
Link to comment
Share on other sites

14 hours ago, mrelwood said:

Thank you, Chriull! For some reason I thought the ”peak” value is limited by time, perhaps since it is always such in the music/audio world. But if I now understood correctly, it is limited by the speed/current curve instead. Therefore:

  • Nominal power can be delivered or exceeded at any speed.
  • Peak power can be reached only at a certain speed, usually around 2/3 of max speed.

I think your original idea of the terminology is the common one. Nominal power (Watt) is the maximal power that can be delivered for "unlimited" time. Electric motors (and other motors too) can deliver their nominal power essentially only at one particular speed (for electric motors about half of their maximal no-load speed). Peak power would need to be defined by the time span over which it can be delivered and is a rather volatile notion I would never trust without digging deeper.

The above graphs are extremely useful, but they show torque=current, not power (ie Watt). Power is the product of speed and current, so it could be computed from the graphs.

Edited by Mono
Link to comment
Share on other sites

40 minutes ago, jbwheel said:

There has been some improvement, the mainboard seem to be tropicalized now. Also the bigger battery version would provide  more than +50% continuous power if NCR18650GA is used instead of NCR18650PF. 

 

Both of the mentioned cells are rated at 10Amp continuous....The GA does NOT deliver more continuous Amps than its competitors from Pansonic or LG.... it is quite easy: in the 3500mah class NCR/NCA chemistry there is NO cell existant which can provide more than 10,5 Amp continuous. If you ever see a 3500mah cell stating to provide more than 10,5 Amp...its just fake announcement.

Only in the 3000mah range you find higher continuous values...some examples:

LG HG2 20 Amp

Sony VTC6 20 Amp

Samsung 30Q 15 Amp

Unfortunatly these cells are nearly double as expensive then our NCR 10Amp cells

Edited by KingSong69
Link to comment
Share on other sites

21 hours ago, mrelwood said:

Thank you, Chriull! For some reason I thought the ”peak” value is limited by time, perhaps since it is always such in the music/audio world. But if I now understood correctly, it is limited by the speed/current curve instead. Therefore:

  • Nominal power can be delivered or exceeded at any speed.

Mostly. At low speeds by current limitation nominal output power cannot be reached and/or can/will lead to overheating.

At higher speeds once the limit curve is reached nominal output power also cannot be reached anymore while accelerating... (for the short time while faceplanting)

21 hours ago, mrelwood said:
  • Peak power can be reached only at a certain speed, usually around 2/3 of max speed.

Peak power can be reached at exactly the speed where one of the constant power lines just "touches" the limit line. As the slope of the lines is determined by some factors and the constant power lines are also "distorted" by some factor the ratio between max speed and the speed at which peak power can be reached should imho not be some universal constant (once i get into the formulas again i could double check this - maybe it is some constant? Both factors could depend on each other?). It will depend on the wheel design and also the wheels operational state (copper temperature, battery state, etc). But in the above example it seems to be around 50% max no load speed.

21 hours ago, mrelwood said:

When I got my KS-16S (1200/3000W) I watched the power usage from the KingSong app while slowly going up a very steep incline. I reached a maximum of 1300W, and thought that there is tons more power available. So I decided to accelerate and go as fast as I dared to get a feel of the available power. I got up to 1500W (bumpy dirt road, 19degree incline, 91kg rider) and thought that I will never reach the max power of the 16S. But since the speed was still quite slow, I might have been actually quite close to the maximum.

The power shown by the Kingsong App (and all other apps) is just some number - the slower one drives, the more the value is off (too high) (http://forum.electricunicycle.org/topic/7549-current-demand-versus-battery-voltage/)

 

6 hours ago, Mono said:

I think your original idea of the terminology is the common one. Nominal power (Watt) is the maximal power that can be delivered for "unlimited" time. Electric motors (and other motors too) can deliver their nominal power essentially only at one particular speed (for electric motors about half of their maximal no-load speed). Peak power would need to be defined by the time span over which it can be delivered and is a rather volatile notion I would never trust without digging deeper.

I don't have any idea how the real nominal and peak output powers for the used motors are. Nominal Output power could be "reused" for the EUC nominal power by the manufacturers (would make sense to me).

Peak output power for an EUC is determined by the (current)torque over speed limit for each actual battery state and not by the motors maximum output power (at least i'd hope that the motors max output power is not really reachable within the torque-speed limit of the EUC - there should be some safety margin for a sane design)

6 hours ago, Mono said:

The above graphs are extremely useful, but they show torque=current, not power (ie Watt). Power is the product of speed and current, so it could be computed from the graphs.

Power = torque/radius * speed = constant * current * speed

So the constant power lines (reddish lines) are current * speed = constant (1/x lines)

Link to comment
Share on other sites

15 hours ago, Chriull said:

At low speeds by current limitation nominal output power cannot be reached and/or can/will lead to overheating.

You seem to suggest that nominal output power is defined as a speed dependent function rather than a single number.

Link to comment
Share on other sites

40 minutes ago, Mono said:
  15 hours ago, Chriull said:
At low speeds by current limitation nominal output power cannot be reached and/or can/will lead to overheating.

You seem to suggest that nominal output power is defined as a speed dependent function rather than a single number.

Yes. By the current limiting (if it is really implemented - what i strongly assume) most output power values cannot be achieved by this at lower speeds.

I assume that this "limited" power values at low speeds by current limitation also cannot be achieved for "longer" periods, since with this still quite high currents overheating should occur... (most dissipated power is "related" to the square of the motor current...)

Without current limitation the nominal power output can not be reached throughout the whole speed range, too - the "1/x" constant output power line has to have two (real) intersections with the current over speed limit line, which defines the physical possibilities of the wheel.

Edit: PS.: Additionally an output power of 800W at 5 km/h would mean for 100kg going up an incline of ~35° (~70%) which already sounds quite steep...

Edited by Chriull
Link to comment
Share on other sites

 

17 hours ago, Chriull said:

The power shown by the Kingsong App (and all other apps) is just some number - the slower one drives, the more the value is off (too high) (http://forum.electricunicycle.org/topic/7549-current-demand-versus-battery-voltage/)

 

Totally with you with all you wrote here.....just one question for the quoted statement:

The Thread there is only about reporting to much amps for a Gotway V3 (which i can agree with...much to high). But actually my KS18S is giving very good and sense making numbers for the Amperes and Watts. For example on normal cruising 20-30kmh about 4-7 Amps 250-420 Watt or at  acceleration or uphill 13-18amps.

Only when very hefty accelerating on a hill i can get the amps to Peak very shortly  at 25-30Amps(1500-1800Watt)......and that only for a nanosecond Long.

So i would say this numbers are making sense and seam accurate?

Link to comment
Share on other sites

Sidequestion, after my crash of last week. If I were to get a KS wheel that also goes 30km/h but has a motor that's a lot more watts than my V8 I will be able to ride over bigger "obstacles" without the motor cutting out at the same speed?

Link to comment
Share on other sites

15 minutes ago, ir_fuel said:

Sidequestion, after my crash of last week. If I were to get a KS wheel that also goes 30km/h but has a motor that's a lot more watts than my V8 I will be able to ride over bigger "obstacles" without the motor cutting out at the same speed?

I think I'm the least educated on this motor technology in this thread, but my understanding is that while it depends on several aspects, the answer is "most probably yes". Battery capacity, battery state, motor power curve, pedal softness, and firmware specifics all play a role.

Edited by mrelwood
  • Like 1
Link to comment
Share on other sites

1 hour ago, mrelwood said:

I think I'm the least educated on this motor technology in this thread, but my understanding is that while it depends on several aspects, the answer is "most probably yes". Battery capacity, battery state, motor power curve, pedal softness, and firmware specifics all play a role.

Of course I am not talking about going vmax - 5 km/h with a battery that's at 30% :)

Link to comment
Share on other sites

3 hours ago, KingSong69 said:

The Thread there is only about reporting to much amps for a Gotway V3 (which i can agree with...much to high). But actually my KS18S is giving very good and sense making numbers for the Amperes and Watts. For example on normal cruising 20-30kmh about 4-7 Amps 250-420 Watt or at  acceleration or uphill 13-18amps.

Gotways reporting too much ampere is an quite "old" story - i do not know if this relates to the same reasons or is something "extra".

I also had the impression that lately reported KS18S values seem more reasonable, but that would mean that the wheel now reports battery current instead of motor current - if so the values are now correct, if not they are as wrong as they ever were. Did not hear anywhere anything about this - one would have to lock closely at one log with quite some speed and incline combinations to come to an real conclusion.

If the new KS firmware now reports battery current there would still be to consider that this would mean power supplied by the battery and since the efficiency of electric motors can be quite bad at lower speeds the achieved output power would still be quite lower than this shown power value.

 

2 hours ago, ir_fuel said:

Sidequestion, after my crash of last week. If I were to get a KS wheel that also goes 30km/h but has a motor that's a lot more watts than my V8 I will be able to ride over bigger "obstacles" without the motor cutting out at the same speed?

The KS16B/C (which goes 30 km/h) has the same nominal power (800W) as the inmotion V8. The advertised maximum power of the KS16B/C of 3000W is just wrong and not reachable by this wheel. @EcoDriftmeasured about 2,2kW which is about the maximum power he measured for the Inmotion V8. So beside firmware (algorithm) and electronics there should be no real difference powerwise. Just the bigger batteries on (some of) the KS16B/C.

The KS16S has now 1200W nominal and imho still 3000W max power advertised. Unfortionately there are no new dynamometer tests performed till now, and also no testdummy rider published wheellog logs of an overlean with this wheel... :ph34r:

But driving the KS16S compared to the KS16C gives me the gut feeling, that it will take bigger obstacles much much more easily at the same speed.

And theoreticly a more powerfull wheel can take such bigger obstacles easier than a not so powerfull machine, but there could be limitations by electric design (how fast how much power can be supplied) and also by the firmware implementation (responsivness of the used control algorithm).

So comparing both of the kingsongs should be not only a feeling but for quite sure also reality - but i've never ridden an inmotion v8 to have any opinion, but there is quite a chance that changing from 800W nominal to 1200W nominal (+ more batteries in parallel) gives more safety margin. Alone the higher lift cut-off speed (almost max no load speed) is a very good clue in this direction.

If you meant in comparison one of the >800W nominal KS18 than you have the additional advantage of the greater wheel diameter which also helps to overcome obstacles easier.

  • Like 1
  • Upvote 2
Link to comment
Share on other sites

1 hour ago, Chriull said:

If the new KS firmware now reports battery current there would still be to consider that this would mean power supplied by the battery and since the efficiency of electric motors can be quite bad at lower speeds the achieved output power would still be quite lower than this shown power value.

Yes, sure...it is still battery current, and not motor, thats no question.

I can only say that the values i watched now, are really, really making sense. When at all, they are more on the to low side then on the to high side. The values now showing go also pretty good with battery consumption and range.

I know from the developer of darkness Bot, that he is reading the amp values “raw”, without doing any maths on them, so it seams KS has done some improvements on value reporting.....

Link to comment
Share on other sites

On 10/22/2017 at 1:17 PM, Mono said:

The above graphs are extremely useful, but they show torque=current, not power (ie Watt). Power is the product of speed and current, so it could be computed from the graphs.

Unit displayed is kW, mechanical output power  measured on a dynamometer test bench.

To get the electric power you need to divide by maybe 0,8 or less depending on the speed (brushless motor efficiency).

And for ncr18650GA, there are measures of an independent tester. It has been able to provide 2000-2750mAh, continuous 15A current. The tester provide the  same type of data for all 18650 lithium references.

http://lygte-info.dk/review/batteries2012/Sanyo NCR18650GA 3500mAh (Red) UK.html

On 10/22/2017 at 1:28 PM, KingSong69 said:

If you ever see a 3500mah cell stating to provide more than 10,5 Amp...its just fake announcement.

Link to comment
Share on other sites

2 hours ago, jbwheel said:

Unit displayed is kW, mechanical output power  measured on a dynamometer test bench.

To get the electric power you need to divide by maybe 0,8 or less depending on the speed (brushless motor efficiency).

And for ncr18650GA, there are measures of an independent tester. It has been able to provide 2000-2750mAh, continuous 15A current. The tester provide the  same type of data for all 18650 lithium references.

http://lygte-info.dk/review/batteries2012/Sanyo NCR18650GA 3500mAh (Red) UK.html

Perhaps you missed this part of the review:

The cells are rated for 10A and at 15A discharge the batteries gets very hot, on the first cell I terminated discharge at 75°C and the second cell I terminated at 85°C (The maximum allowed cell temperature is 70°C).

All cells CAN go over their rated discharge....BUT going over the rated discharge rate, puts not only enourmos stress to the cells...it also on middle and longterm view the heat distroys the chemical inside of the cell! That can not be seen from outside, but can lead in the worst case to venting when used again... The Discharge rate is not set for fun by the producer...if the cell would do that continuously without problems the producer for sure would use that for marketing.

So this test means: Yes, This cell is still living after 15Amp discharge...but it got seriously hot, so the test must been terminated,.and the best thing is now to throw the cell away after this test ;-)

Edited by KingSong69
  • Upvote 1
Link to comment
Share on other sites

3 hours ago, KingSong69 said:

So this test means: Yes, This cell is still living after 15Amp discharge...but it got seriously hot, so the test must been terminated,

So you do agree that it is possible and not a fake announcement. The point is it is almost  impossible to use a wheel like that. And for quick accelerations and even a long time, NCR18650GA will provide >50% power compared to previous PD or PF series. So you just need to let it cool (and maybe  it will naturally since it will be outside with some air, maybe with a cold weather and not just on a bench). It is also why the Mten3 is so powerfull and so responsive. The review is just telling us not to use the Mten3 for a long climb at max speed because the wheel WILL NOT limit the power output to 20A... Back to what I was saying about the GT 16, if the biggest battery pack use GA instead of NCR18650PF/PD it WILL provide >50% more power if necessary, because the mainboard will not limit the output power to 40A. 

And OK stressing the battery is accelerating aging, this is not some breaking news.

Edited by jbwheel
Link to comment
Share on other sites

1 hour ago, jbwheel said:

So you do agree that it is possible and not a fake announcement. The point is it is almost  impossible to use a wheel like that. And for quick accelerations and even a long time, NCR18650GA will provide >50% power compared to previous PD or PF series. So you just need to let it cool (and maybe  it will naturally since it will be outside with some air, maybe with a cold weather and not just on a bench). It is also why the Mten3 is so powerfull and so responsive. The review is just telling us not to use the Mten3 for a long climb at max speed because the wheel WILL NOT limit the power output to 20A... Back to what I was saying about the GT 16, if the biggest battery pack use GA instead of NCR18650PF/PD it WILL provide >50% more power if necessary, because the mainboard will not limit the output power to 40A. 

And OK stressing the battery is accelerating aging, this is not some breaking news.

NO, i dont agree...

I have quite a good knowledge about 18650 cells, as i am using them in my vaping devices since 4 years, and so got real deep into this.

ALL other cells with 3500mah and 10Amp can ALSO deliver peek performances over the cont. 10Amp rating....perhaps even 20Amps “peak”

Each cell on the world can deliver a peak over its rated max.....but first no reliable cell producer gives an amount/number to this “peak” as it is not defined, unreliable, unsafe and chemistry destroying....and second that doesnt mean you can use a failed continuous test as “proved” that the cell is stronger.....and third: No, the cells will not work better “outside in cold weather instead than on a bench”...best argument ever!

And nope the GA is just not better or more powerfull cell than other 10Amp cells. It has the same chemistry..in all test comparisons it can not do anything better. So to say it delivers 50% more power is just ....Sorry, completly nonsense. Yes, the GA IS one of the better cells in this class...some tests go that far and say it might deliver 11-12 Amp continuous, but to be honest...all NCA cells are better not used over 5amps, as there voltage sag gets that high over 10Amp, that it directly sags to a voltage that our wheels would declare as near empty, so that this high draw is only usable when the cell is fully loaded.

And no, stressing the batteries over their rated max is not “aging” them...Using them over their rated max, and overheating them is destroying their chemistry inside and making them unreliable and giving the probability to venting/blowing much further. Unfortunatly this is a process that can not be seen from outside,but that doesn’t make it less true. “Aging” would be to use them longer AT their rated max....but not over it. 

Even your tester -which has a good name in flashlight and vaping batterie testing- i guess edid this 15Amp test for the hole purpose of showing that it is better NOT meant to use over 10Amp...as he had to stop the test because of to high temperatures. 

But i give it up now, as this was the 3rd of 4th try to make it clear there is no much “better” cell in the NCA chemistry class.

For all others, can all be read in/on dampfakkus.de, mooch’s blog, batterie university.

 

 

  • Like 3
Link to comment
Share on other sites

11 hours ago, KingSong69 said:

And nope the GA is just not better or more powerfull cell than other 10Amp cells

 

11 hours ago, KingSong69 said:

there voltage sag gets that high over 10Amp, that it directly sags to a voltage that our wheels would declare as near empty

11 hours ago, KingSong69 said:

 

Lots of words but till wrong . Here are the facts.

Voltage sag of GA at 15A is about the same as NCR18650PD at 10A. So, +50% continuous power. Just keep it COOL and dont empty it in one run at max power. Which we NEVER do while vaping or riding. Peak power increase is probably much better but not measured (peak duration  has a definition in electronics/electric system), it is definitely below 10s and while riding it is useful to pass a pothole, accelerate etc.

LG HG2 voltage sag at 20A is same as NCR18650GA at 15A (but less capacity !). +33% max power.

Mten3 is not empty until ~3,3V / cell at 0A (power beep). I don't think I get a power beep at 3,3V ; 30A (=15A x 2).

The key is to have a design to keep the themperature (motor, mainboard, wires and battery) at reasonable levels but there might be a small weakness for the Mten3 depending how it is used...

Screenshot_20171028-121452-01.thumb.jpeg.b63afa94e9d5512812a918b7a1684ad8.jpeg

Screenshot_20171028-122424-01.thumb.jpeg.8622099273a7b371b39fe768efe27fa4.jpeg

Link to comment
Share on other sites

20 hours ago, jbwheel said:

Unit displayed is kW, mechanical output power  measured on a dynamometer test bench.

Though not in the above displayed graphs, AFAICS.

 

Link to comment
Share on other sites

7 hours ago, Mono said:

Though not in the above displayed graphs, AFAICS.

Sorry, but I thought you were talking  about these graphs from the ecodrift  measures from the review I posted, also because you were talking about torque and this is something you can measure on a dyno bench.

RW-GT16.gif

 http://ecodrift.ru/wiki-article/rockwheel-gt16-review/

Edited by jbwheel
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...