CVT RPM vs Efficiency
ChrisARitz
Does anyone have any data showing RPM vs efficiency??
I'm just trying to figure out, at what point does putting pipes on an engine,
vs hp increase, vs rpm increase does it make sense? If you raise the rpm for
peak hp and are only getting modest gains in hp, would the less efficiency slow
you down?
Of course, this is taking into account a change in gearing.
I'm excited about my new 800 Cat's low RPM hp peak. With the higher cvt
efficiency, I'm thinking it may get as much or more hp to the ground vs its 800
TRIPLE/TRIPLE cousins.
Just trying to generate some GOOD discussion around here again.
Chris
'01 ZL8 esr
'00 ZR6 efi
'99 ZRT6
Terminal doesn't mean the end.....
It's only the beginning.
Go FAST! Skydive.
Justbugg
The latest employee release notes show that GM is dabbling in CVT technology
for smaller cars in the near future.
Snow Etc. - http://www.SnowEtc.com
Now over 100,000 part numbers available
Member of...NPPA, MSA #29220, OSSA, SBA, OSSRA
1-877-623-SNOW (7669)
Brian 98 MXZ 670 Napoleon, Ohio
*discount Snowmobile, ATV, Trailer, PWC parts & acc.
ChrisARitz
>for smaller cars in the near future.
TRUE - The new rumored Saturn SUV will have a CVT as an option That's what I
hear ;-) I deal with Saturn on a daily basis , but due to confidentiality, I
cannot divulge any information. Just speculation and rumour. But, you can
email me if you really are interested.
Chris
David Courtney
rpm's increase?
TIGER
seventies showing an automotive applied CVT with clutch/belt drives for
each rear wheel. The article mentioned very high efficiency but low
power capability. Cars were quite heavy back then so I would imagine it
is more feasable today.
Also saw an article recently in Design News magazine showing a CVT which
used a kevlar belt and two variable size drums...same principle but more
power handling capability with longer belt life. This one may be a long
way off.
TIGER~
Justbugg wrote:
>
> Maybe they are getting better...
> The latest employee release notes show that GM is dabbling in CVT technology
> for smaller cars in the near future.
>
ChrisARitz
increases exponentially, creating more drag - hence, less efficiency. Also,
faster spinning causes more windage, creating drag. This loss is not linear,
but exponential. Also, faster spinning takes more power to hold the rotational
velocity.
Again, this increases exponentially, not linearly.
There a number of factors to make the CVT less efficient as the RPM increases.
The new CVT proposed by a certain car manufacturer is *from what I hear* using
a STEEL belt, and not a more traditional belt type used by a snowmobile.
Haven't actually seen it, but that's what I hear while at the proving grounds.
Back to my original post - anyone have some sort of graph showing efficiency
reduction vs RPM????
Doug Miller
wrote:
>As we all know, CVT's become less and less efficient as RPM increases.
we don't all know this. the above is basically untrue.
there is some argument that as belt speed goes up, efficiency goes
down.
But if this were true, you could simply gear down, or run a larger
diameter primary clutch.
>I'm just trying to figure out, at what point does putting pipes on an engine,
>vs hp increase, vs rpm increase does it make sense? If you raise the rpm for
>peak hp and are only getting modest gains in hp, would the less efficiency slow
>you down?
>Of course, this is taking into account a change in gearing.
no, it isn't. if you want to slow down belt speed, gear up!
-doug miller
ChrisARitz
>diameter primary clutch.
??? And slow down your engine speed to be off your peak of the hp curve???????
NO WAY!
The forces lost multiply EXPONENTIALLY, not linearly. For example, one of the
losses is aerodynamic, or windage.
Drag is a function of velocity cubed. Take 75mph^3/55mph^3. You'll see that
there is twice as much aerdynamic drag at 75mph as there is 55mph. There are a
lot of frictional forces lost in a faster spinning CVT, too. The tangent
forces, the inertia, belt friction on the sheaves, spinning mass of the belt,
etc... These forces all increase exponentially as a function of velocity.
For conversation purposes only - say I am making 100hp at 7500rpm. My CVt is
80% efficient. I am getting 80hp to the chaincase.
Now, I make some mods. I'm making 110hp at 9000rpm. With the higher CVT
speed, I'm only at 70% efficiency. My hp gains have been negated by my
decreased CVT efficiency. I'm now only getting 77hp to the chaincase.
Both cases - as stated before - would have appropriate gearing to compensate
for jackshaft speed vs HP.
With that example, I may actually slow my sled down!!!
So, does anyone have a graph showing efficiency vs cvt speed??
David Courtney
I think you're way off track here... Doug is right. The losses you
sight... even if increased exponentially are virtually insignificant.
Must of the research I've seen shows the CVT to be 95% efficient at
transferring power... with most of the loss occurring through heat generated
by belt friction.
"ChrisARitz" <chris...@aol.comSkydive> wrote in message
news:20000512085149...@ng-fk1.aol.com...
Wolfy
>Chris,
> I think you're way off track here... Doug is right. The losses you
>sight... even if increased exponentially are virtually insignificant.
Exponentally and insignificant (power losses) are a contradiction in terms.
The fact is, power losses do increase exponentially as rpms rise and that is
significant, not insignificant. And no way is a CVT transmission 95%
efficient. I don't remember the stat, but it's something in the 50-60% range I
believe for most sleds.
> Must of the research I've seen shows the CVT to be 95% efficient at
>transferring power... with most of the loss occurring through heat generated
>by belt friction.
Wrong. I'm with Chris on this one, and the power claims of the new 800 cat
motor is exactly why I'm pumped about them too. Big power, low rpms, better
clutch efficiency, more power to the track.
>
One of the faster bald guys.........
Matt Jensen
That figure is for the entire drivetrain of a snowmobile, not just the
CVT. The CVT is regarded as being a fairly efficient system, but other
parts of the snowmobile drivetrain are not. (That huge track dragging
along the slide rails, for one...) That's where you get the 50-60%
total efficiency numbers.
IIRC, the average snowmobile CVT is at least 90% efficient. The CVT
consists of the drive and driven clutch and the belt between them and
nothing else.
Wolfy wrote:
>
> significant, not insignificant. And no way is a CVT transmission 95%
> efficient. I don't remember the stat, but it's something in the 50-60% range I
> believe for most sleds.
--
Matt Jensen (my email address is "moc.noivbo@nesnejm" backwards)
'98 Ram 1500 QC 4x4 5.9L, '99 ZRT 600, '00 ZR 600 EFI
ChrisARitz
>motor is exactly why I'm pumped about them too. Big power, low rpms, better
>clutch efficiency, more power to the track.
AMEN
ChrisARitz
Back to my original post - Does anyone have any DATA to speak with???
The data can out this stuff to rest! (or probably start much more debate)
Michael Hart
"Doug Miller" <drmi...@cyberhighway.net.despam> wrote in message
news:391b8803....@news.cyberhighway.net...
> On 11 May 2000 15:26:18 GMT, chris...@aol.comSkydive (ChrisARitz)
> wrote:
>
> >As we all know, CVT's become less and less efficient as RPM increases.
>
> we don't all know this. the above is basically untrue.
> there is some argument that as belt speed goes up, efficiency goes
> down.
> diameter primary clutch.
They are already running Bearcat clutches in grass drags quite
successfully...
And BTW the Doo Race Manual has discussed the efficiency problem. Grip
force and heat cause more loss as the rpm climbs.
> >I'm just trying to figure out, at what point does putting pipes on an
engine,
> >vs hp increase, vs rpm increase does it make sense? If you raise the rpm
for
> >peak hp and are only getting modest gains in hp, would the less
efficiency slow
> >you down?
> >Of course, this is taking into account a change in gearing.
>
> no, it isn't. if you want to slow down belt speed, gear up!
> -doug miller
But a mod motor geared too tall might push your rpms too low to use the
added HP... Any wonder why some modded motors don't really turn much over
stock rpm?
BTW Doug, what's your 760 turn?
Michael Hart
"Matt Jensen" <n...@spam.com> wrote in message
news:391C3477...@spam.com...
>
> That figure is for the entire drivetrain of a snowmobile, not just the
> CVT. The CVT is regarded as being a fairly efficient system, but other
> parts of the snowmobile drivetrain are not. (That huge track dragging
> along the slide rails, for one...) That's where you get the 50-60%
> total efficiency numbers.
Those numbers have been pushed by a company whose reputation for inaccurate
inertia dyno HP numbers are well known... though I hear their latest
motorcycle dyno numbers are more realistic.
Barry Berg
quite understand it all. But from personal experience riding with you
...well lets just say to you...... as long as you are the lead sled
isn't that what matters??? You were happy with your 600 what made you go
to an 800?. Gee lets me see ...I bet I can guess????? That is if we
are still on speaking terms??????
Barry Berg
For those of you know me I am under my dads account now. That is why it
says Barry Berg. It should say Tina Yoder
Mike Baker
concepts. You can also mathematically figure engine torque from stated HP
and RPM figures. for example my engine is 104 HP@7800 RPM. SLP states a 16
HP gain at 8000 RPM for 120 HP total with their twin pipes on my engine. and
HP=torque X RPM. Without doing the math, HP and RPM will cancel out giving
you relative measures of torque at 7800 RPM and 8000 RPM. Not complete info
but I figured a 12% torque increase. This is a torque increase and so I
believe this would benefit performance in spite of increased RPM's of the
primary clutch. Also increased engine torque will help the secondary in
backshifting and should benefit driveability as well.
There is also a CVT clutch efficiency graph in the handbook. Amazingly the
CVT transmission is very efficient, approaching 90% at a 1:1 ratio. From
what I read, you will be very good to get 1/2 of your engine's horsepower to
the ground. Most of the losses look to be in the track, after the chaincase.
Skydive? I am afraid to come down most of the hills I justed climbed on my
600 RMK. CYA on the hill...MB
--
http://www.mcn.net/~bikemaker
bike...@mcn.net
ChrisARitz <chris...@aol.comSkydive> wrote in message
news:20000511112618...@ng-ch1.aol.com...
> As we all know, CVT's become less and less efficient as RPM increases.
> Does anyone have any data showing RPM vs efficiency??
>
> I'm just trying to figure out, at what point does putting pipes on an
engine,
> vs hp increase, vs rpm increase does it make sense? If you raise the rpm
for
> peak hp and are only getting modest gains in hp, would the less efficiency
slow
> you down?
> Of course, this is taking into account a change in gearing.
>
> I'm excited about my new 800 Cat's low RPM hp peak. With the higher cvt
> efficiency, I'm thinking it may get as much or more hp to the ground vs
its 800
> TRIPLE/TRIPLE cousins.
>
> Just trying to generate some GOOD discussion around here again.
>
> Chris
> '01 ZL8 esr
> '00 ZR6 efi
> '99 ZRT6
>
ChrisARitz
>successfully...
Isn't this because of the lower starting ratio of the Bearcat secondary???
Chris
David Courtney
Just for starters, consider this:
Two 100hp engines... one runs at 5,000 rpm and the other runs at 10,000
rpm. The engine running at 5,000 rpm has to make twice as much torque (and
transfer it through the belt) as the engine running at 10,000 rpm. This
means that the 5,000 rpm engine will have twice as much tension in the belt
as the 10,000 rpm engine. So, the 5,000 rpm clutch has to grip the belt
much harder in order to prevent the belt from slipping, than the 10,000 rpm
clutch.
Now, you have a belt that is being squeezed much harder from the sides
(by both the primary and the secondary clutches) and is being "stretched"
twice as hard by the load applied to it!
The harder you squeeze the belt and the more you stretch it... the more
hp you convert into heat! But, now you expect us to believe that your low
rpm, torque monster is going to be more efficient at transferring this power
to the ground than a higher rpm triple? Dream on!
Now, consider the fact that you will be applying two 50% bigger "shock
loads" to the belt compared to the triple's three smaller ones (even at the
same rpm or torque level) on every revolution... and you're going to have to
tighten the clutches even more to prevent the belt from slipping! More
heat, shorter belt life... less power to the track... again!
If you think that the "exponentially increasing" aerodynamic drag on a
small rotating body or the increased jackshaft bearing drag is significant
compared to these losses, then I've got bad news... you're probably not one
of the faster bald guys!
There are a lot of reasons to build a big twin... but CVT efficiency
definitely isn't one of them, in my humble opinion.
Dave
"Wolfy" <wolfi...@aol.comnojunk> wrote in message
news:20000512121700...@ng-fa1.aol.com...
>
> Exponentally and insignificant (power losses) are a contradiction in
terms.
> The fact is, power losses do increase exponentially as rpms rise and that
is
> significant, not insignificant. And no way is a CVT transmission 95%
> efficient. I don't remember the stat, but it's something in the 50-60%
range I
> believe for most sleds.
>
> Wrong. I'm with Chris on this one, and the power claims of the new 800
cat
> motor is exactly why I'm pumped about them too. Big power, low rpms,
better
> clutch efficiency, more power to the track.
> >
>
>
Larry
Chris throw a set of Bearcat clutches on your Cat and you'll love it. I
installed them on my 98 ZR5. The sled is geared down, .5 ratio, and the
speedo still reads 105mph. The holeshot is improved tremedously too.
Just thought this might interest you.
Ride Safe..........Larry
Out in Friendly Manitoba
timerak
around the pulleys twice as often. Also aerodynamic drag. Windage covers for
secondary clutches claim to save about 10 hp at 9000 rpm. Aerodynamic drag
goes up by the square of the speed (I believe)
David Courtney wrote in message
<391da51f$0$385$3936...@news.twtelecom.net>...
Michael Hart
"timerak" <ti...@ptialaska.net> wrote in message
news:shrtl1i...@corp.supernews.com...
> You are discounting the considerable power consumed in bending that belt
> around the pulleys twice as often. Also aerodynamic drag. Windage covers
for
> secondary clutches claim to save about 10 hp at 9000 rpm. Aerodynamic drag
> goes up by the square of the speed (I believe)
Also discounting cooling time between pulleys...
More heat + less cooling time = bad news.
ChrisARitz
>goes up by the square of the speed (I believe)
Actually, it is more - by the velocity cubed (to the power of 3)
ChrisARitz
>installed them on my 98 ZR5. The sled is geared down, .5 ratio, and the
>speedo still reads 105mph. The holeshot is improved tremedously too.
Is mid-range or top end shifting affected????
David Willis
rpm’s raise. There have already been reasons stated to argue either
way. But it is hard for me to believe that there is very much more
power loss due to higher rpm’s. If there is, why don’t they just gear
down before the clutch instead of after it?
> You are discounting the considerable power consumed in bending that
belt
> around the pulleys twice as often. Also aerodynamic drag. Windage
covers for
> secondary clutches claim to save about 10 hp at 9000 rpm. Aerodynamic
drag
> goes up by the square of the speed (I believe)
>
Sent via Deja.com http://www.deja.com/
Before you buy.
Larry
| Post Response | Return to Index | Read Prev Msg | Read Next Msg |
TECHNICAL TALK
Re: CVT rpm vs Efficiency
Posted By: saltydog <salt...@gvtel.com>
Date: Saturday, 13 May 2000, at 9:18 p.m.
In Response To: CVT rpm vs Efficiency (Larry)
Oh man the summer is too long. The question of RPM vs. efficiency can be
answered easily. If you are making more HP it is worth it. If you are
making 100 HP at 8000 with one engine and 100 HP at 9500 with another
engine then it will become an issue. Still a small issue. If you are
adding HP the trade off is worth it. Pro Stock engines prove this. It
all depends on what you are doing though. Be carefull raising RPM on a
long stroke engine like the 800 Twin Arctic engine. The piston speed at
7700 is probably higher than the piston in a 500 twin at 8600. The long
stroke accellerates and decellerates the piston at very high rates of
speed. This is not an efficiency question. Bottom line, quit worrying
about it. Its such a small issue you wont notice it. Aftermarket pipes
that raise RPM make more HP because of the increase in RPM. If you are
willing to run the engine at a higher speed and tune it right there are
big rewards if the pipes work. Spend that $$$$$$$$. "the dog"
-------------------snip----------------------
Chris, I posted your exact question on www.trailconditions.com in the
technical talk area. This fella that reponded is quite knowledgable on
clutching. Take it for what its worth.
Larry
Chris throw a set of Bearcat clutches on your Cat and you'll love it. I
installed them on my 98 ZR5. The sled is geared down, .5 ratio, and the
speedo still reads 105mph. The holeshot is improved tremedously too.
Is mid-range or top end shifting affected????
No, midrange and top end shifting didn't change. I run a digital tach
and can tell you nothing changed. I used a clutch kit from Ken Jaeck
and it performed identical using the ZR and the Bearcat clutches.
David Courtney
significant!
First of all, a savings of 10 hp @ 9,000 rpm from a windage cover is
absurd! 1 hp would be much more likely. The entire CVT on a well tuned
sled only loses about 5 hp @ 9,000 rpm. Windage is significant in high
speed, low torque machinery... like small turbine engines, turbochargers,
and high speed electric motors; when they run in the range of 50,000 to
100,000 rpm... but not in a 10,000 rpm CVT system.
Secondly, heat a drive belt up to 120 or 140 degrees F, then bend it
until you achieve about a 7" dia. loop... and see how much force it actually
takes. Almost none. I would be amazed if this even takes another 1 hp @
9,000 rpm.
I recommend http://www.gates.com/facts/index.cfm?show=Engineering , they
have 160 documents pertaining to belt drive systems. They state that in
general, properly designed and well maintained belt drive systems achieve
95% to 98% efficiency. Granted, this is not specifically referring to
snowmobile belts... but, all of the belts have to bend around a pair of
sheaves!
So, if we're 95% efficient and we've already allowed 1% for windage...
we only have 4% left to split between internal (bending) and external
(gripping) friction... now, you tell me which takes more power: bending the
belt around the pulley or keeping it from slipping on the sheave?
I firmly believe that you will lose much more power through external
friction at a lower rpm, higher torque than you will through internal
friction (and windage) at higher rpm and lower torque... but, maybe I'm not
one of the "faster bald guys" either!
Dave
"timerak" <ti...@ptialaska.net> wrote in message
news:shrtl1i...@corp.supernews.com...
Doug Miller
>Exponentally and insignificant (power losses) are a contradiction in terms.
No. I'll agree that they are exponential. But if you lose .1
horsepower, and you lose it exponentially as RPM increases, you still
have lost less horsepower then being off by 100 rpm in your clutching.
Lets try this another way.
Horsepower is a measure of work over time. That work over time gets
turned into one of two things. Either acceleration (including climbing
a hill) or friction. So if you are saying that the clutches are
inefficient, that means that they are making too much friction.
Friction means heat.
Anyone seen a frozen idler pully on the front of a car motor? The
motor can actually still idle, but the belt absolutely smokes within a
minute. So the car motor can't be making 10 horsepower at idle, but
the smoking belt won't kill the motor, but the frictional heat is
enough to smoke the belt really quickly.
My point is that it takes very little horsepower to turn into
humougous amounts of heat.
So if you really were losing 10 horsepower in your clutches, they
would turn into molten blobs of aluminum in about 7 seconds.
You are better off tuning your clutches to load the motor correctly at
all miles per hour at wide open throttle.
-Doug Miller
ChrisARitz
>inefficient, that means that they are making too much friction.
>Friction means heat.
Again, don't forget the windage friction. Anything that is constantly in shear
is considered a fluid - including air. The shear forces are friction.
I don't disagree with your points. However, like me, you don't have any hard
data to speak with (i.e. testing data). That was kind of the point of my
original post - I was inquiring if anyone has any data to ponder??? I want to
do some calculations and ponder a few mods......
As far as what you said about changing the gearing to control belt speed (or
cvt speed) - I don't ride in the mountains (like I think you do). However, us
flat-landers play with RPM by primarily with flyweight changes and combinations
of springs and helixes. Sure, gearing WILL affect RPM on top after the CVT's
are shifted out completely. But, if you're controlling all of your shifting by
gearing - that is the wrong approach. (like I said, unless mountains are
different)
About time we got some better discussion around here - the best since the 100%
jerk showed up!!!!!
Joe Peters
pain. In gods eyes we're all the same. Someday we'll all have perfect
wings..............
-100% Joe
David Courtney
situation; however, I'm not sure that I can properly explain why.
Aerodynamic drag... such as the "wind resistance" on a sled moving
through the air is very different from this "windage" that keeps coming up
here. Aerodynamic drag increases exponentially with velocity... however, I
believe that statement applies only to a body that has some frontal area.
(Cd* frontal area*V^3 where frontal area = 0?)
The horsepower dissipated by fans, propellers, etc. does increase
exponentially with rpm... but, only because the blades have some frontal
area... and even then, only within a fairly narrow rpm range. At some rpm
point, the air (or water) that is displaced by the first blade cannot return
in time to be caught by the next blade... and the blade cavitates. Once
this happens, increasing rpm actually decreases the load on the engine
(dissipating less horsepower). This is naturally overcome in a moving
vehicle such as a boat or plane because the forward motion of the vehicle
moves each blade forward into some "undisturbed" water or air... so, a boat
prop that is cavitating at 6,000@10 mph is probably working very efficiently
at 60 mph.
Aerodynamic (parasitic) drag has three components: form drag (caused by
the frontal area displacing air), skin friction (caused by air passing over
a surface), and interference drag (caused by airstreams intersecting). But,
since the secondary clutch has no frontal area (other than the tiny ribs on
the back side, which get covered by the windage cover), I don't think that
it can have any form drag... so, it acts like a propeller that is cavitating
at all speeds; and it's "drag" increases roughly linearly due to skin
friction only.
The second car in a close "train" of race cars does not experience the
same drag as the first car... because very little of the frontal area of his
car is exposed to the "fresh" airstream... so, mainly skin friction and
interference drag are acting on his car.
If the secondary clutch had blades or gaps in it's circumference, then
up to a certain rpm level, the "windage drag" would increase
exponentially... but, it doesn't.
Also, the actual "force" of friction should not increase at all due to
increasing rpm... in fact, it should be less because the there is less
tension on the belt in a lower torque, higher rpm situation. The power lost
due to the force of friction then should also increase linearly with rpm...
but, if you have significantly less friction force, you still lose less
power at higher rpms.
Anyway, the point of this whole discussion is that, while the losses may
not be large enough to worry about... it is very unlikely that your lower
rpm, higher torque motor will allow the CVT to run as efficiently as it will
on a higher rpm, lower torque motor.
Dave
"Doug Miller" <drmi...@cyberhighway.net.despam> wrote in message
news:39203666....@news.cyberhighway.net...
> On 12 May 2000 16:17:00 GMT, wolfi...@aol.comnojunk (Wolfy) wrote:
>
> >Exponentally and insignificant (power losses) are a contradiction in
terms.
>
> No. I'll agree that they are exponential. But if you lose .1
> horsepower, and you lose it exponentially as RPM increases, you still
> have lost less horsepower then being off by 100 rpm in your clutching.
> <snip>
> -Doug Miller
>
timerak
Joe Peters wrote in message
<28087-39...@storefull-266.iap.bryant.webtv.net>...
ChrisARitz
Yup - the frontal area IS hard to compute - but there is some. Without either
surface (primary or secondary) being smooth, there is surface area. What about
the towers? What about the fins? If these things were not significant, why
the windage plates on these? What about the skin friction of these surfaces
(although, probably not measuarble)
It would be neat to see these things put on a dyno and measured at various RPM.
> Anyway, the point of this whole discussion is that, while the losses may
>not be large enough to worry about... it is very unlikely that your lower
>rpm, higher torque motor will allow the CVT to run as efficiently as it will
>on a higher rpm, lower torque motor.
Don't you mean MORE efficiently??
Do you have any data supporting your theories? I must say - you seem to be
knowledgable.
I'm a 'former' engineer. I took some fluid dynamic classes, along with
vibrations, dynamics, combustion engines, etc.. along the way to my
engineering degrees. However, I have forgotten most of it. I'm now just a
dumb sales guy.
Still would be interested in seeing some data.
Doug Miller
<adv...@internetwis.com> wrote:
>here. Aerodynamic drag increases exponentially with velocity... however, I
>believe that statement applies only to a body that has some frontal area.
>(Cd* frontal area*V^3 where frontal area = 0?)
hmmm. i bet you're right. in which case, this probably doesn't apply.
> The horsepower dissipated by fans, propellers, etc. does increase
>exponentially with rpm... but, only because the blades have some frontal
>area... and even then, only within a fairly narrow rpm range. At some rpm
>point, the air (or water) that is displaced by the first blade cannot return
>in time to be caught by the next blade... and the blade cavitates. Once
>this happens, increasing rpm actually decreases the load on the engine
>(dissipating less horsepower). This is naturally overcome in a moving
Hmmm. Never thought of it this way, but I bet you are right. So if you
think of the clutches as "fans", then you should be able to compute
loss in the clutch at rpm by how much horsepower it takes to spin it.
So for instance, a balancer can easily spin the clutches up to speed.
And we aren't talking about the acceleration RPM on the clutches, only
the loss in maintaining the RPM. So for instance, if a 1/2 horsepower
electric motor can quickly spin my clutch up to balance it, then for
sure the horsepower required to spin it at 9000 rpm has got to be less
then 1/2 horsepower.
> The second car in a close "train" of race cars does not experience the
>same drag as the first car... because very little of the frontal area of his
>car is exposed to the "fresh" airstream... so, mainly skin friction and
>interference drag are acting on his car.
mmmmm. yeah, but the last car is still doing the work to maintain the
vacuum at the back of the pack. In other words, a "vaccuum" is being
created, which is only acting on the last car.
> Also, the actual "force" of friction should not increase at all due to
>increasing rpm... in fact, it should be less because the there is less
>tension on the belt in a lower torque, higher rpm situation. The power lost
>due to the force of friction then should also increase linearly with rpm...
>but, if you have significantly less friction force, you still lose less
>power at higher rpms.
never thought of this one either. hmmmm.
> Anyway, the point of this whole discussion is that, while the losses may
>not be large enough to worry about... it is very unlikely that your lower
>rpm, higher torque motor will allow the CVT to run as efficiently as it will
>on a higher rpm, lower torque motor.
makes sense to me.
And, when you are all done, it doesn't add up to very much loss.
So where does all that lost horsepower go??????? It ain't the shafts.
It ain't the gears. It ain't the chain. It ain't the
idlers............
One of my projects this summer is to take my sled motor out and put it
into a car. then run it on a land and sea dyno, then run it on a
chassis dyno.
-doug miller
David Courtney
I had not thought of the balancer! I think that you are 100% correct in
that analysis... if a 1/2 motor can spin it up to 9,000 rpm... it's got to
be using less than 1/2 hp at that speed.
Land & Sea has a water-brake engine dyno for sleds that bolts to the
crankshaft in place of the primary clutch... and uses an arm with a strain
gauge (to the jackshaft) to measure hp on the sled. We have access to one
of these, and it agrees very closely with the hp numbers off of a Superflow
for the same motor (within 2-3 hp)... and you don't have to take the motor
out of the sled!
My idea for a chassis dyno involves a "fake" skid-frame that has a
standard track drive axle at the back. A hydraulic pump will be mounted on
bearings in between the rails... and connected to the drive axle by a
standard snowmobile drive chain and sprockets. As you load the hydraulic
pump... the pump will try to rotate in its bearings, and you can measure the
force (a small hydraulic cylinder at the end of a torque arm) required to
resist the rotation. (Rpm of the dyno will be taken off of the "speedo
drive" on the dyno axle... using a digital speedo sender.)
We have a very accurate, small engine dyno that we built, which works on
the same principle. I just received (and conned a friend into assembling) a
data acquisition system (16 digital lines, 4 analog inputs) for our
home-made flow bench that will serve triple-duty: on the flow bench, the
Briggs & Stratton dyno, and the sled chassis dyno. (We only use one at a
time anyway!) However, I need to practice writing C++ code in order to use
it!
Anyway, my thought is that if you took the track off of the sled and
pressed on a cogged belt pulley to both the sled's driveshaft and the
chassis dyno's driveshaft... you would eliminate the track as a factor
(cogged or synchronous belts are 98% efficient). Then by comparing the
actual engine dyno numbers to the chassis dyno numbers... you could
calculate the efficiency of the "transmission" (CVT, jackshaft, chaincase,
driveshaft, plus 2% for the cogged belt) at any rpm or horsepower level you
wanted to.
Hopefully before this fall... right now it's all 20hp Raptor III go-kart
motors!
"Doug Miller" <drmi...@cyberhighway.net.despam> wrote in message
news:3920b1f8....@news.cyberhighway.net...
<Snip>
> Hmmm. Never thought of it this way, but I bet you are right. So if you
> think of the clutches as "fans", then you should be able to compute
> loss in the clutch at rpm by how much horsepower it takes to spin it.
> So for instance, a balancer can easily spin the clutches up to speed.
> And we aren't talking about the acceleration RPM on the clutches, only
> the loss in maintaining the RPM. So for instance, if a 1/2 horsepower
> electric motor can quickly spin my clutch up to balance it, then for
> sure the horsepower required to spin it at 9000 rpm has got to be less
> then 1/2 horsepower.
>
><snip>
> So where does all that lost horsepower go??????? It ain't the shafts.
> It ain't the gears. It ain't the chain. It ain't the
> idlers............
>
><snip>
David Courtney
I agree that the towers on the primary have some frontal area, and as a
"former engineer" myself, I have to admit that being away from school for 15
years has left me pretty much in the dark as to how to compute the drag on
them.
I was really referring to the secondary... and assuming (possibly
incorrectly) that all of the newer sleds have some sort of "windage cover"
on the secondary already... as Polaris has had for years. Honestly, I've
never really looked at any of the other brands to see.
Unfortunately, according to the info on belt drive efficiency on the
"Gates" site... measuring differences in belt drive efficiencies is nearly
impossible even under laboratory conditions with sophisticated equipment.
So, there isn't a lot of "hard data" available that is "trustworthy"...
I'm guessing that when you have a drive system that is from 95 to 98%
efficient, that you have reached the point of "diminishing returns" as far
as improvements in efficiency are concerned. 98% efficient is probably near
the practical limit for power transmission by any means.
I do still think that at a given horsepower level, the CVT will be less
efficient at a lower rpm and higher torque because of the increased internal
friction caused by stretching the belt more, and the increased external
friction caused by having to grip the belt more tightly.
Also, a snowmobile does not apply the power smoothly like a turbine or
an electric motor... so, I think that there is some slight advantage in
having three "smaller" power strokes instead of two "larger" ones.
If the belts slips even a fraction of an inch on each power stroke (300
times per second) the belt is actually "creeping" backwards with respect to
the sheave... which creates heat and reduces efficiency. I believe that in
order to prevent this; a motor with higher torque and/or fewer cylinders
will require higher side pressure on the belt at any given horsepower
level... and the CVT will be less efficient because of it.
This has been a very interesting thread, and in spite of the fact that
we didn't really turn up any "facts"... it has forced me to spend time
thinking about the question... which, I think, is the whole point of this
newsgroup.
So, thanks for starting the discussion. Dave
cc: ChrisARitz
----- Original Message -----
From: "ChrisARitz" <chris...@aol.com>
Newsgroups: rec.sport.snowmobiles
To: <adv...@internetwis.com>
Sent: Monday, May 15, 2000 8:32 PM
Subject: Re: CVT RPM vs Efficiency
"ChrisARitz" <chris...@aol.comSkydive> wrote in message
news:20000515213206...@ng-fj1.aol.com...
Wolfy
recall all too clearly what the SkiDoo racing manual said about this. There
are power losses, as you go up the rpm scale, due to inefficiencies in the
clutch design. That's not saying that high rpms will always be slower, not at
all. It seems this discussion has gotten off track, with wind resistance,
etc.. It seems simple, that if a 140 hp sled makes it's peak power at 7600
rpm, it will lose less hp through the clutch than a 140 hp snowmobile that
makes it's peak power at 8500, all other things being equal. Doesn't it?
Please resume this most enlightening discussion. No flames please.
Larry
<snip>
So where does all that lost horsepower go??????? It ain't the shafts. It
ain't the gears. It ain't the chain. It ain't the idlers............
<snip>
So it seems, maybe, the hp loss isn't a whole lot when we talk about CVT
rpm vs efficiency. If we put faith in track dyno's where do we lose the
most hp??
Obvioulsy the rotating mass of the drivetrain soaks up alot. Do we have
to accept the fact that unitl a more efficient drivetrain comes along we
are gonna lose about 50% of the hp at the track???
I've heard stories of drag sleds that have turning efforts of 12ft-lbs
or less at the rear ider. Maybe the whole setup wouldn't be ideal for a
trail sled, but what can be done to achieve similar results for the
"average" trail rider?
ExtremeDoo
dragged sleds measured time, taken the chain case apart and ensure the
gears are straight and square to each other and boom almost 1/2 a
second on the track, this is from a stock sled.
That being said, on I believe it was Gates web page, don't kill me if I
am wrong there, they talk about efficiency of chain vs. belt and
sprocket size as it relates to strength of the system. I think that
they claim high 90 % efficiency with cogged belt and low 90% efficiency
with chains of similar sprocket/gear size.
I don't know how accurate there figures are, but I know from up in the
mountains where we spend most of our time riding, run a larger set of
gears with the same ratios always amounts to better performance. Now
obviously there is a point where the diminishing returns meet up with
the added inertial mass issues of such a large chain.
Any thoughts
In article <8465-392...@storefull-134.iap.bryant.webtv.net>,
--
*Any thing I say is my opinion and my opinion only
and does not reflect the views of anyone else,
period, even though I say this someone is still
going to try to hold it over my head.
David Courtney
140hp @ 7,600 rpm is about 97 ft-lbs of torque.
140hp @ 8,500 rpm is about 87 ft-lbs of torque.
My contention is that in order to transfer this (12% more) torque
through the belt to the secondary, you will both stretch the belt more and
have to squeeze it tighter, in order to prevent it from slipping, and that
these two things cause the CVT to be slightly LESS efficient at transferring
power.
I do agree that there are power losses due to higher rpm... but, they
probably increase linearly with rpm... and are small in comparison.
This is all related to average torque, but on a "micro" level; the big
twin has a much higher "instant" torque during each power stroke. It is not
only producing more torque than the higher rpm triple... but, it is
concentrating all of that torque into two power strokes instead of dividing
it among three. So, we have 12% more torque divided by 2/3 as many
cylinders... for a 67% (it just worked out that way) increase in
"instantaneous" torque (belt tension) that has to be resisted by the side
friction of the belt against the sheave. And, at the risk of beating this
tired horse completely to death, that's why I don't think a big twin's CVT
will be as efficient as a higher rpm triple of the same horsepower.
Dave (Just one of the bald guys!)
"Wolfy" <wolfi...@aol.comnojunk> wrote in message
news:20000516125835...@ng-bd1.aol.com...
David Willis
efficient, does that only apply at specific ratios of the clutch? I am
asking this, because when I have looked at track dynoes I have noticed
that the maximum hp is usualy around 30 mph or so. If the cluch has
the same efficiency all throughout, wouldn't it have it's maximum hp at
start, then decreasing as friction increases in the track, and chain
case?
Now if the clutch is more efficient at different ratios, what
determindes this? Is it due to not enough surface area on the primary
at engagement? Is there any whay to change where your clutch is most
efficient? I think there must be, because it looking at the dyno
charts, they are not all in the same place.
Just some thoughts I had.
In article <3921655b$0$385$3936...@news.twtelecom.net>,
Doug Miller
<david...@hotmail.com> wrote:
>asking this, because when I have looked at track dynoes I have noticed
>that the maximum hp is usualy around 30 mph or so. If the cluch has
so the clutches fully engaged, motor turning at 8100 rpm or so, with
normal gears, turns out to be about 25-27 mph i'm told.......
which means to me that something is eating horsepower as it goes
faster.
and all clutching setups may not have the motor making max horsepower
at 30 mph like they should. OTOH, it is hard to find real world
conditions where you stay at 30 mph very long at WOT.
-doug miller
Jon
>Doug,
> I had not thought of the balancer! I think that you are 100% correct in
>that analysis... if a 1/2 motor can spin it up to 9,000 rpm... it's got to
>be using less than 1/2 hp at that speed.
There are problems even with this simple theory. A 1/2 HP electric motor is
rated for 1/2 hp output at some temperature for continuous duty. The motor
may be capable of putting out far more horsepower than its continuous duty
rating. I wouldn't be surprised if a 1/2 hp electric motor could put out over
10 hp. The rating has nothing to do with peak HP like is common with gas
motors.
> Land & Sea has a water-brake engine dyno for sleds that bolts to the
>crankshaft in place of the primary clutch... and uses an arm with a strain
>gauge (to the jackshaft) to measure hp on the sled. We have access to one
>of these, and it agrees very closely with the hp numbers off of a Superflow
>for the same motor (within 2-3 hp)... and you don't have to take the motor
>out of the sled!
That's good to know. I've been curious as to how the two types/brands would
compare.
> My idea for a chassis dyno involves a "fake" skid-frame that has a
>standard track drive axle at the back. <cool idea snipped>
> Anyway, my thought is that if you took the track off of the sled and
>pressed on a cogged belt pulley to both the sled's driveshaft and the
>chassis dyno's driveshaft... you would eliminate the track as a factor
>(cogged or synchronous belts are 98% efficient). Then by comparing the
>actual engine dyno numbers to the chassis dyno numbers... you could
>calculate the efficiency of the "transmission" (CVT, jackshaft, chaincase,
>driveshaft, plus 2% for the cogged belt) at any rpm or horsepower level you
>wanted to.
> Hopefully before this fall... right now it's all 20hp Raptor III go-kart
>motors!
Hey, more power to ya! Let us know how it works out!
David Courtney
the 1:1 shift ratio... it is worst at engagement and during the "clutching
phase" which normally lasts up to 25-30 mph... and also falls off as you go
into the "overdrive" area past 1:1.
I think what you're seeing with the chassis dyno is that:
A) The CVT is terribly inefficient up to 25-30 mph because the belt is
slipping during the "clutching phase". This is similar to "slipping the
clutch" in a motorcycle or manual transmission car when starting out... the
engine is turning too fast and the vehicle is moving too slow, for the
clutch to be fully engaged.
B) At 25-30 mph when the clutch fully engages... CVT efficiency
increases dramatically... but, the drag on the track (which has been
increasing linearly with the speed all along) now becomes the "big"
horsepower eater... and the overall drivetrain efficiency continues downward
as speed increases.
In other words, the drag on the track is probably the main "power
thief"... but at low speed, the clutching losses of the CVT are added in;
creating the "efficiency peak" at the point where the "clutching action"
stops.
That's just my theory based on what I've read... I've never seen any
actual chassis dyno sheets, so I don't know for sure. Hope it made some
sense... these things are hard to discuss without graphs & pictures!
"David Willis" <david...@hotmail.com> wrote in message
news:8fsldp$c08$1...@nnrp1.deja.com...
> This has just got me thinking a little. When you say the cvt it 95%
> efficient, does that only apply at specific ratios of the clutch? I am
> asking this, because when I have looked at track dynoes I have noticed
> that the maximum hp is usualy around 30 mph or so. If the cluch has
David Courtney
On the chassis dyno, the mph point where the horsepower "peaks", will
differ for different sleds, because they have different gearing and clutch
set-ups. It has nothing to do with the CVT efficiency.
If you have taller gears, you will be moving faster when the clutch
reaches full engagement rpm, than you would be with shorter gears... so the
horsepower "peak" will be at a higher mph.
Also, some clutches are set up to engage harder than others, so they
"force" full engagement earlier... but, if you do this, you either bog the
motor or get a "jerky" engagement (because the motor is still turning too
fast and the sled moving too slow for a smooth, complete engagement) unless
the track breaks loose (like on a grass drag sled).
Again, no proof. Just my theories. Dave
David Willis
In article <3922bd01$0$161$3936...@news.twtelecom.net>,
"David Courtney" <adv...@internetwis.com> wrote:
> Sorry... I missed the last part of your question.
> On the chassis dyno, the mph point where the horsepower "peaks",
will
> differ for different sleds, because they have different gearing and
clutch
> set-ups. It has nothing to do with the CVT efficiency.
> If you have taller gears, you will be moving faster when the
clutch
> reaches full engagement rpm, than you would be with shorter gears...
so the
> horsepower "peak" will be at a higher mph.
> Also, some clutches are set up to engage harder than others, so
they
> "force" full engagement earlier... but, if you do this, you either
bog the
> motor or get a "jerky" engagement (because the motor is still turning
too
> fast and the sled moving too slow for a smooth, complete engagement)
unless
> the track breaks loose (like on a grass drag sled).
> Again, no proof. Just my theories. Dave
>
>
DBarmes
the track. but the rest of it seems to make since. I wonder as the speed of
the track increases if the centrifigul forces on the trac would pull it away
from the slides and therfore make it more efficient??
R!
ChrisARitz wrote:
> >(Cd* frontal area*V^3 where frontal area = 0?)
Neat formula, what are the dimensions in the area to get a force as the result?
Typically, Drag = 1/2*rho*U^2*Cd, which results in kg*m/sec, or a Newton. Hence,
velocity squared is the quantity you are interested in.
>
>
> Yup - the frontal area IS hard to compute - but there is some.
Frontal area of the clutches is not the area in question, maybe the moving belt
area on the top (tension) side of the clutch is. Frontal area of a sled is
measured easily by shining a light on the front of the sled and marking out the
shadow behind it, assuming you have a reasonable light that emits parallel rays
(the sun?).
> Without either
> surface (primary or secondary) being smooth, there is surface area. What about
> the towers? What about the fins? If these things were not significant, why
> the windage plates on these? What about the skin friction of these surfaces
> (although, probably not measuarble)
Windage plates on Polarii came with the Storm and promoted higher top speeds in
some situations. Your argument about windage is valid but the quantity is pretty
small compared to other losses.
>
>
> It would be neat to see these things put on a dyno and measured at various RPM.
Been there, done that. The belt drive is as efficient as Gates and David state, in
almost all operating conditions. The belt does run hotter than 14-160 F though.
Big twins give the belt more time between torque pulses to cool and relax (could be
the incorrect assumption) than a triple. Typically, the 800 triple belt runs
hotter than the 700 twin, hence the cooling duct on the 800 XCR.
>
>
> > Anyway, the point of this whole discussion is that, while the losses may
> >not be large enough to worry about... it is very unlikely that your lower
> >rpm, higher torque motor will allow the CVT to run as efficiently as it will
> >on a higher rpm, lower torque motor.
>
> Don't you mean MORE efficiently??
>
> Do you have any data supporting your theories? I must say - you seem to be
> knowledgable.
Yes, but I can't share it here. You'll find more losses in track hysteresis than
anywhere else in the 'system'.
>
>
> I'm a 'former' engineer. I took some fluid dynamic classes, along with
> vibrations, dynamics, combustion engines, etc.. along the way to my
> engineering degrees. However, I have forgotten most of it. I'm now just a
Your shear arguments are valid but it is a pretty complex and unsteady flow pattern
in the clutch area at speed. Also, most frictions (belt, kinetic,etc.) work on the
velocity term and not on the acceleration term. The V^3 thing you state I believe
comes from a vehicle dynamics situation where there must be a characteristic length
added to the calculation to be dimensionally correct, unless a friction coefficient
with dimensions is used with more dimensional manipulation due to empirical
relationships establishing the overall result in a finite regime of operation.
David Courtney
Thanks for the info.
You wouldn't happen to have any data on the XCR 800 & XC 700 that would
show the peak horsepower@rpm & torque@rpm would you? I can't seem to find
much.
Dave
Doug Miller
wrote:
>> Yup - the frontal area IS hard to compute - but there is some.
>
>Frontal area of the clutches is not the area in question, maybe the moving belt
>area on the top (tension) side of the clutch is. Frontal area of a sled is
>measured easily by shining a light on the front of the sled and marking out the
>shadow behind it, assuming you have a reasonable light that emits parallel rays
>(the sun?).
So a tad bit of history. David and I were arguing that the clutches
are pretty efficient generally. The other guy was arguing using the
old Doo manuals that if he can get the belt speed down he's gonna find
100 horse or something.
David and I were trying to find the fiziks to prove that it ain't
gonna happen. Which you've given another great example of!!!!
>Windage plates on Polarii came with the Storm and promoted higher top speeds in
>some situations. Your argument about windage is valid but the quantity is pretty
>small compared to other losses.
We're agreeing.......
>Been there, done that. The belt drive is as efficient as Gates and David state, in
>almost all operating conditions. The belt does run hotter than 14-160 F though.
Only when the factory's ship wimpy springs and flakey helixes......
;-) ;-) ;-) (big smileys, just teasin). Seriously, some of
the local doo dudes can get the clutches consistently down below 120F.
They think it matters. My belt gets hot, but doesn't seem to get
eaten.....
>Big twins give the belt more time between torque pulses to cool and relax (could be
>the incorrect assumption) than a triple. Typically, the 800 triple belt runs
>hotter than the 700 twin, hence the cooling duct on the 800 XCR.
I never even thought of that one...... Could be also slip angles of
belts?? Which would also explain why soft belts are more efficent then
hot belts?????
> You'll find more losses in track hysteresis than
>anywhere else in the 'system'.
makes sense to me!!!!!!
>velocity term and not on the acceleration term. The V^3 thing you state I believe
>comes from a vehicle dynamics situation where there must be a characteristic length
could be. I know it also has to have some sort of minimum speed and
cross section. In other words, a small bar in the wind does not have
the vcubed effect, but a brick at 300 mph does.
This really is a COOL discussion. Thanks Mr. Kerner, whoever you
are.... ;-) ;-) ;-) (smileys!!!!)
doug miller
fiziks, math, cs back in my professional beer drinking days.
David Courtney
estimates, he told me that based on the info he has received from Polaris:
XC 700 makes about 122hp @ 8,000 rpm
XCR800 makes about 155hp @ 8,500 rpm
If these numbers are correct (or close) the XC is making 80 ft-lb of
torque, while the XCR is making 95 ft-lb of torque, when both are at their
peak hp rpm's... so, my uneducated guess is that the 20% extra torque being
transmitted through the belt is what causes the XCR800 belt to run hotter.
In our discussions, we were assuming two engines of equal horsepower...
running at different rpm's with a different number of cylinders. To make a
fair comparison, the XC 700 would need to make 155hp @ 8,000... which would
be about 102 ft-lb of torque! (28% more than it actually does.)
My guess is that then, the XC 700 belt would run slightly hotter than
the XCR800 for all the reasons we talked about earlier... but, I have no
proof yet!
"R!" <gke...@wiktel.NOSPAMcom> wrote in message
news:392A77BD...@wiktel.NOSPAMcom...
>
>
> ChrisARitz wrote:
>
> > >(Cd* frontal area*V^3 where frontal area = 0?)
>
> Neat formula, what are the dimensions in the area to get a force as the
result?
> Typically, Drag = 1/2*rho*U^2*Cd, which results in kg*m/sec, or a Newton.
Hence,
> velocity squared is the quantity you are interested in.
>
> >
> >
> > Yup - the frontal area IS hard to compute - but there is some.
>
> Frontal area of the clutches is not the area in question, maybe the moving
belt
> area on the top (tension) side of the clutch is. Frontal area of a sled
is
> measured easily by shining a light on the front of the sled and marking
out the
> shadow behind it, assuming you have a reasonable light that emits parallel
rays
> (the sun?).
>
>
> > Without either
> > surface (primary or secondary) being smooth, there is surface area.
What about
> > the towers? What about the fins? If these things were not significant,
why
> > the windage plates on these? What about the skin friction of these
surfaces
> > (although, probably not measuarble)
>
> Windage plates on Polarii came with the Storm and promoted higher top
speeds in
> some situations. Your argument about windage is valid but the quantity is
pretty
> small compared to other losses.
>
> >
> >
> > It would be neat to see these things put on a dyno and measured at
various RPM.
>
> Been there, done that. The belt drive is as efficient as Gates and David
state, in
> almost all operating conditions. The belt does run hotter than 14-160 F
though.
> Big twins give the belt more time between torque pulses to cool and relax
(could be
> the incorrect assumption) than a triple. Typically, the 800 triple belt
runs
> hotter than the 700 twin, hence the cooling duct on the 800 XCR.
>
> >
> >
> > > Anyway, the point of this whole discussion is that, while the losses
may
> > >not be large enough to worry about... it is very unlikely that your
lower
> > >rpm, higher torque motor will allow the CVT to run as efficiently as it
will
> > >on a higher rpm, lower torque motor.
> >
> > Don't you mean MORE efficiently??
> >
> > Do you have any data supporting your theories? I must say - you seem to
be
> > knowledgable.
>
> Yes, but I can't share it here. You'll find more losses in track
hysteresis than
> anywhere else in the 'system'.
>
> >
> >
> > I'm a 'former' engineer. I took some fluid dynamic classes, along with
> > vibrations, dynamics, combustion engines, etc.. along the way to my
> > engineering degrees. However, I have forgotten most of it. I'm now
just a
>
> Your shear arguments are valid but it is a pretty complex and unsteady
flow pattern
> in the clutch area at speed. Also, most frictions (belt, kinetic,etc.)
work on the
> velocity term and not on the acceleration term. The V^3 thing you state I
believe
> comes from a vehicle dynamics situation where there must be a
characteristic length