Jump to content

3.0L Roller Rocker Arms - WARNING New Found Info!!!


kev
 Share

Recommended Posts

I was in the process of degreeing in my hydraulic roller camshaft using the 3.0L Mitsubishi roller rockers, when I ran into an issue. This is pretty major and will affect anyone using these rocker arms, regardless on the type of camshaft.

 

Here is my setup:

Cam: Monsta'Banshee Hydraulic Roller Camshaft (purchased over a decade ago)

INT Duration @.050: .207° Seat-To-Seat: 265°

EXH Duration @.050: .196° Seat-To-Seat: 262°

INT Lift: .480

EXH Lift: .459

Lobe Center Separation: 111.5°

Power Range: 2500-6500RPM

INT lobe center: 107°

Valve degrees at .050" lobe lift:

IVO BTDC: -3.5

IVC ABDC: 30.5

EVO BBDC: 34.0

EVC ATDC: -18.0

 

Rockers: MD195450 3.0L Mitsbuishi Roller Rockers, 1.6 ratio

Cylinder Head: New AMC Marnal NJV casting

Valves: 1mm OS stainless valves

Springs: Schneider HD

Cam Gear: Shawn Silva Adjustable

 

 

I'll give a background and lead up to my conclusion:

 

Degree Setup:

 

I installed a summit racing degree wheel on the crank with a bent coat hangar as a pointer. Using a piston stop, I found true TDC and positioned the wheel/pointer in the correct position:

 

full

 

Then I took a part an old lifter, removed the spring and ball to make it fully collapse. I installed the intake rocker arm assembly using feeler gauges to take up all lash. Used that magnet as a nice flat base for my indicator

 

full

 

full

 

full

 

 

Intake Centerline:

 

I used the intake centerline method (well documented on the web) to locate the intake centerline.

 

I took the cam to the peak lift on the intake lobe, zero'd out the indicator, and took measurements .080" (1.6 RR X .050") from each side of the peak. I was surprised how far that I had to go from the 12 o'clock position. At 12 o'clock my intake centerline was 119.5 degrees! For the heck of it, I did the same at .050" of each side of the peak...same centerline, as it should be.

 

I got the centerline at 107 degrees and then advanced it 3 degrees per Shelby's recommendation....so it is at 104 degrees. I've checked this at least 6 times even with taking the head off and literally starting over....it is consistent.

 

 

 

full

 

full

 

To hit that 104 deg centerline, here is how far I had to advance my cam gear:

 

full

 

full

 

 

I pondered this for a long time. Why would I have to advance this sucker so far. The Monsta Banshee has the exact same intake centerline as a stock cam!!! Machining tolerances wouldn't result in this much cam advance.

 

I decided to 'understand' the camshaft some more and went further to take measurements on the valve open/closing:

 

The IVO/IVC/EVO/EVC numbers for the cam above are at .050" of lobe lift. To get valve lift, I multiplied these numbers by my rocker ratio (1.6) to get .080". I took measurements of the intake valve open and closing at .080" of valve lift:

 

IVO @ .080 valve lift: 8 deg ATDC

IVC @ .080 valve lift; 30.5 deg ABDC

 

I kind of expected these not to be spot on but I expected the duration to be. With my 3 degrees of advance, I wanted 0.5 deg ATDC and 27.5 deg ABDC. So it is opening 7 degrees late and closing 3 degrees late. I expected these to be proportional. I should be within 1 degree consistently. But at least this lobe is int he ballpark...unlike what I found next:

 

I flipped the indicator over to the exhaust valve and took the EVO/EVC measurements;

 

EVO @ .080 valve lift; 63 deg BBDC

EVC @ .080 valve lift; 50 deg BTDC

 

No where even close!!!!

 

I scratched my head over and over, pulled it down and started from scratch 3 times! Studied the deg wheel as well, etc. Why is the exhaust valve opening and closing about 30 crank degrees too early? I pondered this for a week, swapped out rockers using some brand new OEM ones, some new aftermarket ones, and several used oem ones that I had laying around. Also tried different lifters, etc. No change...these numbers were consistent!!

 

I searched the web and this forum all week trying to figure out if I screwed up some basic math, wondering if I'm having a momentary lapse of reason, etc. I questioned the cam. Obviously I was leaning to the cam not being machined to the right specs.

 

But this is when I remembered that many people run the stock cam with these rockers. What the heck, let me try a stock cam. Again same centerline and a similar lobe separation. Guess what? When I put in the stock cam with these lifters, I still had to advance that gear to just about the same spot to hit the 107 degree intake centerline! And again, the exhaust valve was opening 30 degrees too early!!!!!!!!!!

 

It finally hit me what the problem was...the roller rockers! They aren't touching the cam in the same position as the stock slappers!!!!!!

 

Next, with the stock cam still in, I swapped out the rollers with the stock slappers and degreed it in. I barely had to move the cam gear from the 12 o'clock position to degree in the stock setup. Although the FSM doesn't give .050" valve degree numbers, I could clearly see the valves were opening when they should. And the intake centerline and lobe separation were spot on.

 

 

Next step was to prove my theory for the different contact position between the stock rocker and the roller rocker. I slapped some of that yellow grease onto the cam lobes that is typically used to check the wear pattern on rear differentials. Then I pulled the lifters from the rockers, slid the rocker (one at a time) onto the shaft, and rocked them onto the lobe (in the base circle region). I did this for the intake and exhaust for both the stock and roller rockers. It is clearly evident that this is the source of my issue!

 

 

 

full

 

full

 

 

full

 

Just comparing the two rockers alone, you can see the difference, but the yellow grease tells the full story! In linear terms, it's an 1/8" off! The stock slappers pretty much contact the center of the cam. The rollers are an 1/8" away...meaning 1/4" total from one another (intake vs exhaust)

 

Conclusion

 

These 3.0L Mitsubishi roller rockers do not contact our camshaft in the correct position!! To hit the intake centerline of any currently developed performance roller cam, as well as the stock cam, you have to advance the cam gear about 8 degrees. In doing this, it also advances the exhaust. There is no way you can ever hit both intake and exhaust centerlines using these rockers!

 

This is some scary stuff. Members have been using these rockers for 10 or so years now. If you degreed in the cam using the intake centerline method, your exhaust lobe is 30 degrees away from where it should be! That means your valve is opening too early in the power stroke, but most importantly it is closing 30 crank degrees too early!!! Where is that last bit of exhaust going on the upstroke once the valve closes? Nowhere, it stays for the next cycle! Very scary!!

 

 

 

Let me know your thoughts. Please try to prove me wrong! I'm back to the drawing board when it comes to a camshaft for my build. Right now I'm looking at just a stock cam and lifters to move on forward.

 

But most importantly, I want to get this info out to the public. Please rethink using these roller rockers on your car....and if you have them on now, you may want to consider pulling them in the near future.

 

kev

Link to comment
Share on other sites

Not really shocking, I would expect the slipper contact to be different than the roller, but I would also expect the grinder of the roller cam to be conscious of that as well and grind accordingly. That is the shocking part here, I'd like to hear Tim C's thoughts. I guess I'm also going to have to pull the Aussie stock roller out of my flatty and check that, you'd think the factory would certainly get this correct.
Link to comment
Share on other sites

I have my daily struggles over the past week or so documented in the mod forum but I haven't heard from Tim yet. But I think I came to a conclusion and wanted to get this out into the public asap.

 

I don't blame Tim for this. Not one bit. He went out of his way to do a great thing for this community. Shame on us for not supporting him by double checking everything on install. It has been almost 12 years since these banshee roller cams were developed. In 12 years, nobody challenged why they had to rotate that cam gear so far to line up the intake center? Shame on all of us! Lots of discussion on the little stainless shims, etc...but nothing on this. I blame myself for letting this thing sit in a box for so long.

 

 

I spent a lot of time awake last night thinking about how the slapper changes contact points along the cam lobe...but still doesn't make sense the base circle and the max lifts should match the card..regardless. Those valve lobes aren't where they need to be for this style rocker arm.

Link to comment
Share on other sites

Its not suprising. Most just stick the cams in without checking anything. Thats why people still run schneider cams.

Its obvious when comparing the contours of the two that they will strike the same lobe in different spots.

Edited by Funky Phil
Link to comment
Share on other sites

I'm not faulting Tim or anyone else but I will repeat, many members have been saying for years not to run roller rockers on the stock cam because the stock cam is not a roller cam, it is a slipper cam. See the wear mark on the slippers in the picture above? It uses a radial rocking motion to follow the lobe, therefore the lobe is ground to a different profile than that used on a roller cam with the limited surface a roller uses and obviously different contact lever (position). So no surprises as far as the stock cam. The surprise MAY be that the cam grinder did not account for the differences in grinding a roller profile onto a stock blank, I say may be cause Kev admits this cam has been sitting around for a decade. That is why it would be nice to hear from Tim and understand the grinding process.

 

As far as Schneider is concerned I think Phil has found they can't follow or grind anything to their own published specs. That's a different (lack of quality) story.

Link to comment
Share on other sites

Of course the foot of a slip rocker will contact earlier and let off later in the lobe. I thought the stats for duration given for slip vs. rollers calculated in those differences?

 

http://www.26liter.us/gallery/albums/userpics/14392/Cam___Rocker_compare.jpg

 

 

For example a Magna roller cam contacts later once the lobe hits the roller, but shoots up much faster to full open and drops off later & quicker to closed than the more gradual slip rockers wide foot imprint which moves throughout the cycle. I'm not a cam/rocker expert so please correct or add to this.

 

Does the recommendation for starting TDC retard include this stock for stock? I can't imagine 20-40 degrees difference from the manual is really what's needed?

Link to comment
Share on other sites

Well this is disappointing news! I have one of Tim's roller cams that I've been meaning to install for a few years now as well as a schneider 284-RH roller cam sitting in the tool box. The slipper schneider 284 currently in my car didn't require but a degree or 2 from the stock cam degreeing to get it to match the card ... at least on the intake side of things. All of this information in recent years makes me want to just put a stock cam back in there!

I hope Tim chimes in on this. Very curious to see what comes of it.

Link to comment
Share on other sites

Yes my cam was from the first batch of monsta banshees. It is definitely possible that the issue may have been found and resolved. Although there is no reference to it that I can find. Thus I am making an assumption that my cam is the same as all monsta banshee cams made and their roller sisters from the same grinder

 

Hopefully this thread will spark someone to verify or counter my results with their cam.

 

One thing is for certain, using these rockers on a stock cam is not a good idea.

 

Keep the ideas coming. Like I said, I WANT to be proven wrong in this!

 

Kev

Link to comment
Share on other sites

Last night I confirmed that the 1.5 ratio roller rocker arms for the Galant, MD106245, contact the cam in the same location as the 3.0L rockers.

 

Starting to do some research on the Magna cams and roller rockers. I'm finding mixed info on the roller rockers...some places say they are the same as the 2.0 Galant, MD106245,...but others say they are not. My guess is that they are the same or nearly identical but the 4G54 engines that came factory with roller rockers had a differrent cam core to account for this.

 

Wish I had more parts to test..like a stock magna cam, a camtech or rpw.au cam, other styles of Tim's cams, etc.

 

I noticed something else yesterday that caught my eye...several places on this site say 'Monsta'Banshee2'. What is the 2? Revision 2? hmmm There may be hope...

 

http://www.starquestclub.com/forum/index.php?showtopic=113208&st=0

 

kev

Link to comment
Share on other sites

Kevin,

 

This is great info. Note how much later the intake valve opens on the slipper cam compared to the roller with the roller rockers....similar to what I found. What I don't understand is why isn't the exhaust valve opening earlier in a similar proportion.

 

I have some homework to do here. I'm going to try to recreate ( in a more crude manner) your curve for the stock cam with the v6 roller rockers

Link to comment
Share on other sites

Kevin,

 

Do you happen to still have the spreadsheet with the data? I printed out the graph as large as I could today. It may be enough for now.

 

I want to clock in my cam gear to the 12 o'clock position and then create points at every .050 lift and max lift/valve center lines all in reference to CAM degrees like your chart has. I'll do it for the stock cam and the monsta banshee with the v6 roller rockers.

 

First off, I want to see if my numbers match yours for the stock cam with v6 rollers. Then see how the banshee cam compares. My data will be much more crude obviously because I'll be doing it all manually.

 

Kev

Link to comment
Share on other sites

TurboRaider did the cam testing. He may still be around on the 4 x 4 wire.

 

Kevin C is still active on the 4x4 wire. http://www.4x4wire.c...ite_id/1#import

 

Warning, The pop up ad's are bad there for me.

Edited by StarquestRescue
Link to comment
Share on other sites

TurboRaider did the cam testing. He may still be around on the 4 x 4 wire.

 

Kevin C is still active on the 4x4 wire. http://www.4x4wire.c...ite_id/1#import

 

Warning, The pop up ad's are bad there for me.

 

Oh wow, sorry about that. My ADHD mind was of course 3 steps ahead and I thought you were TurboRaider.

 

I did have a chance to clock in the Monsta-Banshee straight up and take degree measurements at every .050" lift for intake and exhaust. I plotted the info similar to how Kevin C did. This may not be as bad as I thought. Wish I had his data to better overlay the charts.

 

All in all, the Monsta-Banshee cam with the V6 roller rockers has a 121 degree lobe separation and a 119 degree intake centerline (with no adjustment to the cam gear). The curves fall right on with the measurements Kevin C. took using the stock cam with either the V6 or magna roller rockers. When plotted out, they don't look too bad. With a little bit of advance (not as much as I was going to do), it may make the cam feasible. I still have a bit of research to do though.

 

kev

Link to comment
Share on other sites

Not the best graph, but enough to visualize this cam. The top curve is the Monsta Banshee with no advance/retard of the cam gear. Degrees are in camshaft degrees just like Kevin C's data (on bottom). exhaust valve on the left

 

Note that my data was manually obtained by taking degree readings starting at .006" lift and then every .050" lift. No where near as detailed as the profile Kevin C made in the lower curves.

 

The Monsta Banshee has pretty much the same intake centerline and lobe separation as the stock cam with the 3.0L roller rockers (as expected by the cam cards).

 

full

Link to comment
Share on other sites

 Share

×
×
  • Create New...