Change of cranskshaft pulley to lightweight.

[texaxile]

Hi
I have a 2004 WRX Impreza, and I am considering changing the crank, A/C and PAS pulleys for one piece lightweight items from RCM, however, there seems to be some discussion that the Crankshaft pulley specifically acts as a harmonic balancer. The lightweight pulley is about 250 g versus stock pulley of 2.5 kg, gains are seen as being better throttle response and a claimed 10 - 15 bhp gain, with the very unlikely perk of it being a one piece item and not therefore prone to disassembling itself under load (very few recorded cases).

Now I'm not particularly techie, but this, along with a lightened and balanced flywheel is a relatively cheap route to power before entering the world of forged pistons, rods and divorce.

There is a wealth of qualified knowledge here, and there are also threads pertaining to this on scoobynet, but it ended up in a gentlemanly difference of opinion but no actual evidence to support or deny either way.

Can anyone here shed any light on whether this is a worthwhile idea or if it is a bad idea altogether.

TIA
Pete

[HJG]

Not completely familiar with these Subaru engines but a 2,5kg pulley is usually that mass for a reason and is likely a torsional vibration (TV) damper. Again, due to unfamiliarity of this engine I'm unsure if this TV damper is a rubber or viscous type, with the former tuned for a specific frequency and the latter effective over the engine speed range.

Eliminating a TV damper can increase the crankshaft's torsional response and tilt angle. The former can excite the crank to such an extent that the crank will fail completely and catastrophically. The latter will push the main bearings to potential failure.

Unsure also where the 10bhp power increase would come from by removing mass from the crank pulley?

[CrutyRammers]

10-15 bhp gain for swapping some pulleys? If only tuning was so easy.
As above, don't know this particular engine, but the crank pulley is certainly used as an external balance / damper on some engines.

[texaxile]

Hi HTG
Thanks for the reply, I'm also unsure why there are claims made of a power gain, I can only assume less rotational load on the engine, but would that give more power?.

I'm always sceptical when power gain figures are mentioned because they never translate in the real world and are often given in an "perfect" scenario.

However, there has not been a reported case of catastrophic failure (I choose my words carefully) on any on the forums, but this actually adds to my suspicion. I would ask why Subaru invest time and effort developing an engine to work reliably without making a mass production saving on a heavy crankshaft pulley.

That said, one only has to look at the massive cock up that was the EJ255 Engine to see that their credibility is not that great.

[texaxile]

CrutyRammers said:

10-15 bhp gain for swapping some pulleys? If only tuning was so easy.
As above, don't know this particular engine, but the crank pulley is certainly used as an external balance / damper on some engines.

LOL if only it was!.

Like me, you are probably wary when figures and percentages are mentioned with a seemingly simple fix.

That reminds me, I need to get some Slick 50

[HJG]

texaxile said:

Hi HTG
Thanks for the reply, I'm also unsure why there are claims made of a power gain, I can only assume less rotational load on the engine, but would that give more power?

No, it would only allow the engine to 'spin up' (and down) faster

[texaxile]

HJG said:

No, it would only allow the engine to 'spin up' (and down) faster

So there would be a noticeable amount of a reduction in engine braking (if coupled to a lightened flywheel)as well then?.

[HJG]

Yes, although the lightened flywheel would provide most of this difference.

The Subaru flat 4 should be quite a short and stiff crankshaft so torsional vibration isn't likely to be as serious of an issue as it would be on an inline 6 for example. However I still wouldn't want to be messing around with that damper. Flywheel is the safer option, providing it is of sufficient quality!

[stevieturbo]

Some people do use them with no ill effects.

Some people like nice coloured pulleys.

Yes some OEM pulleys have been known to fall apart.

But as for feeling any difference in response or any claims of more power.....show me proof.

I've used standard pulleys on many engines making 5, 6, 7 and 800+ with no ill effects. The standard pulleys when they're in good order just work. When something isnt broke...dont fix it.

But then I dont care about colourful pulleys.

[Little Pete]

The marginal gains to be had from not having accelerate the damper are not worth the risk of a broken crankshaft. I would fit a new OE damper and concetrate on getting the bottom end accurately balanced.

[MiggyA]

texaxile said:

So there would be a noticeable amount of a reduction in engine braking (if coupled to a lightened flywheel)as well then?.

HJG said:

Yes, although the lightened flywheel would provide most of this difference.

The Subaru flat 4 should be quite a short and stiff crankshaft so torsional vibration isn't likely to be as serious of an issue as it would be on an inline 6 for example. However I still wouldn't want to be messing around with that damper. Flywheel is the safer option, providing it is of sufficient quality!

While revving the engine you'd notice an increase in engine braking effect since there would be less rotational momentum to get rid of when you lift off the throttle.

[AdamIndy]

stevieturbo said:

Some people do use them with no ill effects.

Some people like nice coloured pulleys.

Yes some OEM pulleys have been known to fall apart.

But as for feeling any difference in response or any claims of more power.....show me proof.

I've used standard pulleys on many engines making 5, 6, 7 and 800+ with no ill effects. The standard pulleys when they're in good order just work. When something isnt broke...dont fix it.

But then I dont care about colourful pulleys.

I agree with all of this. I think it's more a cosmetic enhancement than for performance. Maybe a very slight gain with VERY being the operative word. I very much doubt it would be noticeable.

10-15bhp improvement? Not a chance.

A lightened flywheel can and will improve response(not so much power) providing a) it is properly balanced(including the clutch assembly) and b) not a lightened standard flywheel. I have heard a few stories of these being taken too far and breaking up.

There are plenty more things you can do with a scoob that WILL make a difference though I'm sure you know that already.

[227bhp]

You are attempting to measure the improvement incorrectly. It isn't about measurable BHP it's about accelerating up the road quicker. Taking weight from the rotating assembly will allow you to do this.
It isn't a simple calculation and is variable dependent on what gear, bhp, weight of car etc, but the interesting thing is it scales up. In certain circumstances 2kg lost from the FW can equal 20kg taken from the chassis in terms of acceleration from say 0 - 100.

[Stan Weiss]

2 kg weight on the outside of a ring which has a 152.4 mm diameter

The greater the acceleration rate the greater the torque needed

HP will be a function of RPM and calculated torque

Stan

_______________Road_Horse_Power_Prediction_Chart.
________These_numbers_will_be_similar_to_a_Chassis_Dyno.

_________________________Rear______Aero____Rolling___________Rear_W__Accele___Time__Rate
RPM______MPH___Velocity__Wheel____dynamic__Resist.__Elapsed___Horse__ration__Differ__RPM
________________ft/sec__Torque___Drag_-_HP___HP______Time_____Power__in_G's__ential__Sec

__0.0____.000_____.000____0.00______.000____.000____.0000______0.00__0.0000__0.0000__0.0
500.0__17.850___26.180____0.54______.000____.000___3.3333______0.05___.2441__3.3333150.0
_1000.035.700___52.360____0.54______.000____.000___6.6667______0.10___.2441__3.3333150.0
_1500.053.550___78.540____1.08______.000____.000___8.3333______0.31___.4882__1.6667300.0
_2000.071.400__104.720____1.08______.000____.000__10.0000______0.41___.4882__1.6667300.0
_2500.089.250__130.900____2.15______.000____.000__10.8333______1.02___.9764__0.8333600.0
_3000.0_107.100157.080____2.15______.000____.000__11.6667______1.23___.9764__0.8333600.0
_3500.0_124.950183.260____3.23______.000____.000__12.2222______2.15__1.4647__0.5556900.0
_4000.0_142.800209.440____3.23______.000____.000__12.7778______2.46__1.4647__0.5556900.0
_4500.0_160.650235.619____4.31______.000____.000__13.1944______3.69__1.95290.4167_1200.0
_5000.0_178.500261.799____4.31______.000____.000__13.6111______4.10__1.95290.4167_1200.0
_5500.0_196.350287.979____5.38______.000____.000__13.9444______5.64__2.44110.3333_1500.0
_6000.0_214.199314.159____5.38______.000____.000__14.2778______6.15__2.44110.3333_1500.0
_6500.0_232.049340.339____6.46______.000____.000__14.5556______7.99__2.92930.2778_1800.0
_7000.0_249.899366.519____6.46______.000____.000__14.8333______8.61__2.92930.2778_1800.0
_7500.0_267.749392.699____7.53______.000____.000__15.0714_____10.76__3.41750.2381_2100.0
_8000.0_285.599418.879____7.53______.000____.000__15.3095_____11.48__3.41750.2381_2100.0
_8500.0_303.449445.059____8.61______.000____.000__15.5179_____13.94__3.90570.2083_2400.0
_9000.0_321.299471.239____8.61______.000____.000__15.7262_____14.76__3.90570.2083_2400.0
_9500.0_339.149497.419___12.92______.000____.000__15.8651_____23.36__5.85860.1389_3600.0
10000.0_356.999523.599___12.92______.000____.000__16.0040_____24.59__5.85860.1389_3600.0
10500.0_374.849549.779___17.22______.000____.000__16.1081_____34.43__7.81150.1042_4800.0
11000.0_392.699575.959___17.22______.000____.000__16.2123_____36.07__7.81150.1042_4800.0
11500.0_410.549602.139___34.44______.000____.000__16.2644_____75.42_15.62300.0521_9600.0
12000.0_428.399628.319___34.44______.000____.000__16.3165_____78.70_15.62300.0521_9600.0

[Inline__engine]

people are forgetting a pulley size change can spin accessories slower and can reduce parasitic losses, not 10 bhp worth though but....

also parts with rotational inertia will show power gains because most chassis dyno accelerate at some rate they dont hold engine rpm constant to sample the torque. so the higher the acceleration rate rpm/sec of the engine used in the test the higher the bhp increase shown will be will be. most chassis dyno operators dont adjust for drivetrain inertia (too hard) etc so its never properly accounted for so it shows as the hp gain

Edited by Inline__engine on Wednesday 18th January 06:04

Edited by Inline__engine on Wednesday 18th January 06:06

[Mignon]

texaxile said:

and there are also threads pertaining to this on scoobynet, but it ended up in a gentlemanly difference of opinion but no actual evidence to support or deny either way.

So in the other words the general definition of the internet - fairly dim mathematically challenged people bickering pointlessly.

The mathematics of weight reduction of rotating components is hardly rocket science. It can be explained in a single page as here for example.

https://web.archive.org/web/20110918114916/http://...

It should be apparent after reading that that lightening something of such a small radius as a crank pulley is utterly pointless for the average car. Maybe of a little more use with a high revving bike engine.

Chucking in some average gearing and radius numbers, 1kg off a crank pulley would equate to about 3kg off the car's weight in 1st gear only and then utterly inconsequential for higher gears than that. If anyone thinks they can notice their car's weight changing by 3kg......

As for people who think that weight reduction of ANYTHING makes the engine produce more power. Grrrrr.

So here's my suggestion for that 3kg (or 6kg in your case if you can get a pulley 2 kg lighter). Go for a really good sh11te before getting in the car, take off your jacket and sweater first, drive for 10 miles in a fairly spirited fashion and you'll have burned off enough fuel and other weight to make the car as much faster as the pulley would have done.

[Stan Weiss]

Mignon said:

So in the other words the general definition of the internet - fairly dim mathematically challenged people bickering pointlessly.

The mathematics of weight reduction of rotating components is hardly rocket science. It can be explained in a single page as here for example.

https://web.archive.org/web/20110918114916/http://...

It should be apparent after reading that that lightening something of such a small radius as a crank pulley is utterly pointless for the average car. Maybe of a little more use with a high revving bike engine.

Chucking in some average gearing and radius numbers, 1kg off a crank pulley would equate to about 3kg off the car's weight in 1st gear only and then utterly inconsequential for higher gears than that. If anyone thinks they can notice their car's weight changing by 3kg......

As for people who think that weight reduction of ANYTHING makes the engine produce more power. Grrrrr.

So here's my suggestion for that 3kg (or 6kg in your case if you can get a pulley 2 kg lighter). Go for a really good sh11te before getting in the car, take off your jacket and sweater first, drive for 10 miles in a fairly spirited fashion and you'll have burned off enough fuel and other weight to make the car as much faster as the pulley would have done.

No the engine does not produce any more HP. But in any dyno test other than a step test it will show up as a HP increase. As I and others have said the faster the acceleration rate the more HP that will be shown. Since that HP not longer has to accelerate the removed mass / weight it will show up on the dyno test.

Stan

[GreenV8S]

I suspect the only significant effect would be that the engine revs up faster if the driver blips the throttle with the clutch disengaged. It will feel more powerful and more responsive to the driver, even if the actual performance impact is negligible.

[PeterBurgess]

Every little helps and we enter the 'aggregation of marginal improvements' philosophy which has always worked well in competition whether it be car racing or British Cycling Teams.
I attach a section of the superflow 901 manual which explains effects of varying acceleration rates on the dyno. The same obtains with lightening rotating and reciprocating components, the power 'gets to the wheels quicker' so the car accelerates faster.
A friend of mine in his 80s campaigned a standard race MGB to good effect in the late 80s and 90s. He always had more bhp at the wheels than my engines and I thought he must be cheating. One secret he shared was a crank scraper, 3 bhp more at the wheels but very iffy and wears the cam out as it is splash fed from the crank! When he retired he told me lots and lots of little goodies. The main ones were the scraper, using the old three compression ring pistons rather than the modern two ring, leave the second ring out ( it turns out the two ring version was same as three ring but top ring groove not machined)having the top ring higher gave 2 bhp more. Oil pressure run at 50 psi instead of 70-75, this gave 2 bhp more. So, from three simple ideas the aggregate was 7 bhp more at the wheels than I was getting. The guy ran our tweaked heads, same cam same CR same rolling roading just more knowledge from his mate who ran the test dynos at BL....aggregation of marginal improvements.

Peter

[Mignon]

Stan Weiss said:

No the engine does not produce any more HP. But in any dyno test other than a step test it will show up as a HP increase. Stan

It will show up as a "number" on a screen when a dyno is trying to do something that it's not been properly programmed for. To call that number horsepower is not even remotely correct though. Obviously when an engine is accelerating it is storing kinetic energy in its rotating and reciprocating components. That leaves less energy, or force, available to twist the torque arm on which torque is measured and in the absence of any correction for this stored energy will result in a smaller torque number compared to a steady state test.

However with any decent modern dyno software you should be able to correct for the moment of inertia of the engine component masses when doing an acceleration test. In fact there's a blindingly simple way of finding out what that correction factor should be even when you have no idea what the moments of inertia are.

Simply run the engine at two different acceleration rates. Adjust the inertia factor until both runs show the same gross torque numbers (measured torque plus inertia absorbed torque). That should bring you back very closely to what the engine will show at steady state and should of course now compensate properly for any other acceleration rate, even a variable acceleration rate.

The problem with trying to define any number showing on a dyno screen as being extra horsepower due to lighter components because of an accelerating run test is that this notional number will not show up at all when the engine is not accelerating. i.e. the car will not have a higher top speed which "proper" additional horsepower will clearly produce. It will not go any faster up a steep hill when acceleration rate is low or zero. It will not manifest much if at all in high gears.

All it will do is make the engine, and hence the car, faster accelerating, very briefly, in low gears. In fact if the vehicle is grip limited in low gears then it will make no difference whatsoever. By the time the vehicle is in a high enough gear to be able to take full throttle then the inertia effect will have dissipated anyway.