CAMSHAFT 101

Discussion pertaining to positive pressure E28s.
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

CAMSHAFT 101

Post by paul burke »

Its time to put the whole camshaft thing on the table so to speak. I made an adapter for my linear gauge to accurately measure BMW valve event/timing where it counts, at the valve. NO if ands or buts, this is the most accurate method.

Image
Image


Paul
gabo1
Posts: 107
Joined: May 01, 2008 1:40 AM
Location: northern california

Post by gabo1 »

WOW!
your awesome.
next question:
how much :D
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

gabo1, I wasn't planning on marketing a tool here, unless you have a shaft encoder and a software program such as Cam Pro Plus or a Cam Doctor?The only thing the adapter could do is measure lift with a dial indicator.

I am hoping that this may start a discussion on BMW camshaft design/application and what really works.


Paul
M. Holtmeier
Posts: 3024
Joined: Mar 11, 2007 3:06 AM
Location: Dallas, Texas

Post by M. Holtmeier »

:popcorn:

Paul, I am still interested in picking up an adjustable cam gear from you for my car. I won't be able to hold a conversation with you on cam design, but I'm curious to see what gains could be had in my turbo setup over the b35 cam that's in it now.
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Hey Mike, cam math is cam math and nothing more. You have a one up already since you installed and checked your B35 cam timing. Do you remember what your numbers were? These intake numbers should look familiar.

B35 intake lobe
Measured @ .012 lash .406 lift

@ .020 262 Dur

@.050 218 Dur

@.200 150 Dur


CNC grind intake lobe
Measured @ .012 lash .448 lift

@.020 252 Dur

@.050 229 Dur

@.200 164 Dur



The CNC grind has less duration (smaller) @ .020 than the B35.
Has less seat to seat timing (less overlap) than the B35.
Has earlier closing than B35 with less return to intake when piston is on its way back up(ABDC).
Is longer at .050 (11degrees).
Is longer at .200 (14 degrees) where cylinder filling is most optimum.
Is .448 lift @ valve

With a later opening and an earlier closing centers can be tightened up, resulting in higher torque numbers.

E-mail me about the cam gear if you still have the address, if not PM me


Paul
Last edited by paul burke on Jan 17, 2009 8:45 PM, edited 2 times in total.
Jeremy
Beamter
Beamter
Posts: 15841
Joined: Feb 12, 2006 12:00 PM
Location: Connecticut

Post by Jeremy »

Higher torque numbers, but at what rpm? Compared to the b35 cam specifically, what does the CNC cam do for the torque plateau? Is it any wider? Higher or lower in the rpm range?

Do you have any dyno sheets handy? I ask for the sheet not necessarily in a "prove it" sort of way, but I'm a visual person and seeing the shape of the torque curve would do wonders for me.

Jeremy
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Jeremy, I understand visual aids can help. There are several shops, some you are familiar with, doing builds using my camshafts. I am sure they will be happy to post dyno sheets when done. Tightening centers increases torque mostly in the middle of the rpm range in the B34/35 (2800 to 4500).BMW used a 108 center on there smaller (earlier) grinds. To clean up the Idle on the B35 they went to a 112 center. With less seat to seat timing you can tighten up the centers and advance the intake while leaving the exhaust in the same position and not adversely affect the Idle.

Paul
Entropy_Engineering
Posts: 84
Joined: Dec 19, 2008 10:48 PM

Post by Entropy_Engineering »

I wonder if the actual lift and duration would be a tiny bit more because the rocker riding on a thin oil film.
Coldswede
Posts: 6859
Joined: Oct 10, 2008 1:48 PM
Location: Back U.P. North,. Where the water's blue, the wind is free and seasons four.

Post by Coldswede »

I wonder if the actual lift and duration would be a tiny bit more because the rocker riding on a thin oil film.
Umm, is this one of those Quantum Mechanics questions? I think the difference would unobservable. :argue:

The oil film on two unpressurized surfaces would be strictly limited to things such as viscosity and molecular attractions, it would be very small.
Duke
Posts: 9986
Joined: Feb 12, 2006 12:00 PM
Location: Ormond Beach, FL

Post by Duke »

So....how would you build the perfect FI cam for the B35 engine?

How is it different than what is out there....the Alpina "turbo" cams for example.
Good & Tight
Posts: 461
Joined: Oct 29, 2007 11:16 AM

Post by Good & Tight »

paul burke wrote:Jeremy, I understand visual aids can help. There are several shops, some you are familiar with, doing builds using my camshafts. I am sure they will be happy to post dyno sheets when done. Tightening centers increases torque mostly in the middle of the rpm range in the B34/35 (2800 to 4500).BMW used a 108 center on there smaller (earlier) grinds. To clean up the Idle on the B35 they went to a 112 center. With less seat to seat timing you can tighten up the centers and advance the intake while leaving the exhaust in the same position and not adversely affect the Idle.

Paul
Paul
I want to see a dyno sheet on how much more torque or power your cam makes. You need to be a bit more specific on how much more torque it makes, 10 20 30lbs..???? is it just more peak torque with less area under the curve? Also is the cam power increase based specific to a stock or ported head?

Thanks
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

I started this thread not to sell or promote one particular profile (hence the title). I want to try to convey what makes one cam different from another, what to look for when selecting a cam, what the numbers really mean.

Good & Tight,
What did you experience when you advanced your camshaft?

Paul
Brian D
Posts: 324
Joined: Feb 12, 2006 12:00 PM

Post by Brian D »

paul burke wrote:The CNC grind has less duration (smaller) @ .020 than the B35.
Has less seat to seat timing (less overlap) than the B35.
Has earlier closing than B35 with less return to intake when piston is on its way back up(ABDC).
Is longer at .050 (11degrees).
Is longer at .200 (14 degrees) where cylinder filling is most optimum.
Is .448 lift @ valve

Paul
Paul, from looking at the numbers I would guess that the CNC cam is a very "streetable" stock replacement camshaft, or am I off? Would this cam work best in a stock head, or one with some ( or a lot of?) porting done to it?
iXer
Posts: 96
Joined: Feb 12, 2006 12:00 PM
Location: Sweden
Contact:

Post by iXer »

Duke wrote:So....how would you build the perfect FI cam for the B35 engine?

How is it different than what is out there....the Alpina "turbo" cams for example.
There is no optimal/perfect turbo cam. It depends on many hard factors and soft ones like driver preference and skill.

Gearing and weight of the car?
Manual or autotragic?
Where is your powerband ?
Drag-race, auto-x or heavy duty trailer pulling?
Full tubular pulse tuned manifold or cast log exhaust manifold?
Stock or a custom tuned intake?
Pressure differentials?
Stock ports or lots of custom port work?

Usually there is a tendency to cut down on the exhaust lobe to make the engine less sensitive to engineering compromises.
Good & Tight
Posts: 461
Joined: Oct 29, 2007 11:16 AM

Post by Good & Tight »

paul burke wrote:I started this thread not to sell or promote one particular profile (hence the title). I want to try to convey what makes one cam different from another, what to look for when selecting a cam, what the numbers really mean.

Good & Tight,
What did you experience when you advanced your camshaft?

Paul
Paul
The car feels like it makes more power, but I need to re-dyno again to see were it made more power.

Otis
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Otis, If you repeat your previous dyno day (same dyno, same operator and such) you should see a marked gain in torque, earlier spool, lower IATs and more. At this point it would be nice to know how much you actually advanced it. The torque may flatten a little earlier and it may drop some hp at the top, but acceleration is what everybody's after anyway, and that is what your feeling. When you dyno it please post both sheets, then will go through the cam timing changes as they relate to the power changes.

Paul
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Q-ship, this intake profile is completely streetable when combined with a suitable exhaust profile. Will Idle like a stone in a B35 with 8-to 1 static comp. Because of the increase in valve motion the valve train dynamic noise will go up slightly (not ticking noise, motion noise). It is there on stock M30s anyway and usually goes unnoticed.

Paul
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Mattias, I have profiles generated that rival roller cam ramp speeds.

Measured @ .000 lash .475 lift (12.07 mm)

@.015 lift 292 Dur

@ .050 lift 252 Dur

@.200 lift 182 Dur

If you have some Cat numbers, would you post them for comparison.

Here are the numbers for my second smallest M30 turbo grind measured at valve with .010 lash. I have increased the lift slightly since this printout. Intake .450 exhaust .420
Image
Paul
iXer
Posts: 96
Joined: Feb 12, 2006 12:00 PM
Location: Sweden
Contact:

Post by iXer »

Catcams have a cam with 299 degrees advertised duration:
Measured @ .000 lash 11.15mm valve lift
@0.35 mm lift 299 duration
@1.25 mm lift 258 duration

Of course they have profiles that lift more, the one with 12.00 mm valve lift is cited as 314 + 272 duration at the above lifts. It would be nice with the additional measurement at 0.2" to reveal how the ramp accelerates, I think you're being way more honest about that than anyone I've seen and it's appreciated and good for this discussion.

Just for comparisons sake, it might say something to someone, a friend of mine installed a very aggressive Catcams profile in his M10. I think it's more than a problem you can solve by throwing money at it, you can't fit anything with more area under the curve without major hardware changes. This is a test-fit of the 277 @ 0.05" cam:

Image

When we plotted valve lift against his old Schrick 304 we could see that the cams were quite identical until the Catcams just continued to lift (12.75 vs 10.50? mm).
At the moment these numbers are on paper in a garage, I will try to get them digitized in the coming weeks.

Anyw, this leads up to a couple of questions.
Your cam has significantly faster ramps than either of these two cams from the looks of it. What can be said of the longevity and service life of this approach?
Is the ramp speed made possible through the CNC production method or has it been avoided by the big name cam manufacturers due to problems with valve train wear?
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Mattias, the problem arises when trying to keep the rocker foot geometry. I have the 314 Cat in my possession. It is on a .950 base circle. I also have the 310-306, also on a .950 base circle. When I measured these cams at the valve, they were less lift than advertised. I do not know how they measure there stuff, but in the U.S. there is a recurring problem, the grinders fail to take into account the ratio change the foot/lobe contact creates during its motion. I have modeled this extensively and along with the CNC grinder and motion analysis have come up with several profiles. The clearance ramps on these profiles are adequate and will save the valve job and hard parts.

Lets keep this going. There are a lot of myths that have followed the M10/M30 valve train for years, like to clear some of them up.

Have you swiped the bottom of the rocker foot (in the picture) to measure lobe/foot contact area?

Paul
Last edited by paul burke on Jan 18, 2009 12:28 AM, edited 1 time in total.
iXer
Posts: 96
Joined: Feb 12, 2006 12:00 PM
Location: Sweden
Contact:

Post by iXer »

You know what? I remember that we were questioning the Catcams numbers once the cam arrived. This is one of the reasons we set up a dial indicator and started turning the engine around to see what was going on. I guess I'll have to find out what the conclusion was.

I'll report back in a few days, maybe sooner.

Do note that we used steel rockers made in Germany, apparently the same that were used way back in the days of racing of these machines. I believe VAC or IE sell these at hilarious prices (we paid half, if not less).
I'm not sure how these are different in shape than the stock items, but they use the same method of adjusting clearance with an eccentric. They have a very wide foot, nearly the width of the cam lobe. There is no "wear pad".
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

I am familiar with those rockers. They were used on the mid engine stuff raced in the '60s, like the Chevron B16. I have wanted to get my hands on them but do not know their origin.
Are they still being manufactured?

Paul
Last edited by paul burke on Jan 18, 2009 11:32 AM, edited 1 time in total.
turbodan
Posts: 9153
Joined: Jan 09, 2007 10:19 PM

Post by turbodan »

Those are definitely some serious looking rocker arms.
paul burke
Posts: 844
Joined: Sep 08, 2008 4:51 PM
Contact:

Post by paul burke »

Mattias, I plotted the Cat # 1302145 314 degree intake lobe.

Measured @ .000 lash .488 lift @ valve

@.010 336 degrees

@.020 303 degrees

@.050 276 degrees

@.200 206 degrees

Measured @ .012 lash .476 lift

@.010 300 degrees

@.020 288 degrees

@.050 268 degrees

@.200 201 degrees


This is with the eccentric in the stock position.

This profile is extremely fast at the top and relatively slow at the bottom and would be best suited in a 14 to 1 compression (or higher) NA motor. Like you said its an extension of the Shrick stuff with a little more lift. Most of these profiles are to long @.050 and give up to much cylinder pressure requiring large static comp. numbers to build torque.

Paul
iXer
Posts: 96
Joined: Feb 12, 2006 12:00 PM
Location: Sweden
Contact:

Post by iXer »

I appreciate the numbers Paul, these will be checked against our own measurements at a later time. What you're saying does confirm how the cam "feels" and there is no real increase in valve train noise and from the first hours of use there doesn't seem to be any trouble on the horizon.

I should explain that this M10 with the cam you just profiled is using that profile exclusively on the intake, it has a 114 degree lobe separation to the most aggressive exhaust profile they had for the M10: No. 1302146, with @0.05" 281 dur and supposedly 12.75 mm valve lift. I can discuss why this choice was made more in-depth privately, I have to check and discuss some simulations and practical results from yesteryear.

The engine will see 2.5 bar boost in the intake, perhaps 1.5 bar back pressure measured in the turbine collector of the exhaust manifold. With E85 fuel and around 7.5:1 static c/r we're looking at a mean M10 later this year.
It was started up with this cam and tuned for 1 bar boost the past autumn and it really surprised us at how relatively nice the idle and low load was. A 1200-1300 rpm idle rpm helps of course, but there was no hesitation for just driving around town off-boost but I wouldn't call it a grocery getter.
Powerband is from 4000-4500 rpm (full boost) and it pulls beyond 8000 rpm if you let it. This is the consideration that made me ask the question about longevity of the valve train, considering it has a powerband up to 8000+ rpm, with relatively soft ramps compared to yours.

Not many engines discussed here are like this, it is far from what a typical John Doe BMW owner considers a turbo engine and definately not very common on the M30 since it doesn't favor 6500+ rpm easily due to bottom-end issues.
Many things must be considered to make the system as a whole work together and it pays off big-time.. I'm not sure that the intention of starting this thread was discussing the cam and valve train requirements of type of extreme engine.
Post Reply