BMW M5 Forum and M6 Forums banner

Building a 4.7L low compression, supercharged S65 stroker motor

65K views 70 replies 23 participants last post by  herrubermensch 
#1 · (Edited)
Table of Contents:

Overview:
It's been over a year since I dropped my car off at Auto Talent to build the 4.7L low compression, supercharged S65 stroker motor. The custom pistons had been ordered in April 2010, and the plan was to wait for the pistons to arrive and then drop off the car as soon as they arrived.

RD Sport had built the original 4.6L high compression stroker motor. A custom ESS VT2 supercharger was then added as an afterthought, and we came up with the idea to hit 200 MPH in the Texas Mile. I knew it was a long shot already because I knew I would need about 725whp in order to hit 200. We sat with 615whp -- knowing that 200mph wasn't going to happen. But we tried anyways and hit 186.1mph. We learned whenever we turned up the boost past 6.25 PSI, two things happened: 1) the fuel system couldn't supply enough fuel, and 2) the high EGT's of the 4.6L supercharged motor would burn O2 sensors. The only solution was to redesign parts of the fuel system and build a low compression motor. So I decided to order the pistons in April 2010 and patiently waited for them to arrive.

Once the piston landed in customs, I drove my car down to Auto Talent. Once the motor was out, we inspected everything and began to notice some changes we wanted to make. (Pictures shown here[/url].) After pulling the motor, we received our first piece of bad news. This was going to be an omen of many delays that were to come in the future. Basically, Mahle made the wrong pistons. Mahle made 94mm pistons for high (12.0:1) compression, not the 10:1 low compression that we ordered. It was going to take Mahle another eight weeks to manufacture and ship the replacements. So, the waiting game started.

During the eight weeks, I didn't sit idle and uninvolved. I had a big decision to make: who was going to build the motor? I considered my top choices Auto Talent, RD Sport, or possibly my former racing engine collegues at Van Dyne Engineering in Huntington Beach, CA (formerly from Drake Engineering -- from the famed "Drake Offy" racing motor). Using Van Dyne had a certain appeal because I know their standards are very high and they do things quite a bit differently than most street engine builders. The downside was Van Dyne had never built a BMW S65 motor. I untimately worked out a deal where Van Dyne would build the motor to their standards, and Zolti from Auto Talent would be directly involved as a participant and consultant during the motor building process.

In hindsight as I would later discover, chosing Van Dyne might have been the smartest decision I made. Today, the bottom end has been completed: it is now built. But the road leading to this point was filled with delays, setbacks, and challenges. Over the course of the next few days, I will write one article focussing on each of the challenges and decisions we faced. As usual, I will provide plenty of pictures of the build and the painstaking perfectionist process that we have been going through. For the most part, I will share everything we discovered -- with one exception. Certain technical details of the measurements we made and the tolerances we measured and adjusted will remain confidential because some of these measurements and adjustments will make their way into future products.







 
See less See more
4
#2 · (Edited)
Pistons:
The pistons eventually arrived in early July, 2010, and the motor had already been out of the car for a few months. By this time, I had already decided to use Van Dyne Engineering to build the bottom end, and I was still undecided about using Van Dyne for the top end as well. I drove down to meet Federico (RD Sport), Stewart (Van Dyne), and look at the new pistons. It was a great opportunity to see both sets of pistons side-by-side (high compression and low compression).

I snapped a bunch of pictures of both sets of pistons, and talked to both of them about my project. Below shows pictures of the pistons side-by-side. The high compression stroker motor pistons are those with the higher dome, while my low compression pistons are the ones with the dish. At some point in the discussion, Federico pulled me aside and told me Carrillo wanted to remake my connecting rods. It was my option, but it would mean another eight weeks of delays. (That story, and pictures, will follow in the next article.)

The entire set of piston pictures can be seen here.
New Pistons





 
#3 ·
Very interesting. Keep it coming!
 
#6 ·
Connecting Rods
For the most part, the connecting rod bearings looked fine when we disassembled the motor. Other than one or two uneaven wear marks on one or two of the bearings, they all looked like they were in good shape. As I would find out a year later, the uneven bearing wear would be addressed when we assembled the bottom end. (That will be the subject of another update.) I thought everything was in decent shape, so I was taken a bit by surprise when Federico (RD Sport) pulled me aside to tell me that Carrillo wanted to redesign the connecting rods. The complete set of disassembly pictures and pictures of all bearings from multiple angles can be seen here.
BMW M3 Motor Tear Down - Robert & Mary Ann Collins















The problem with the connecting rods
When Federico pulled me aside, he told me that he had sent one of my connecting rods back to Carrillo for analysis. After inspection, he said Carrillo hadn't designed the connecting rods for a boosted motor, and wouldn't warranty the connecting rods if I continued to use them. If was my choice, and I could continue to use my existing set of connecting rods and might not have any problems, or take Carrillo's advice and redesign them, spend another few thousand $$$, and wait another eight weeks. To me it was an easy decision: we must redesign the connecting rods to be safe.

Federico and Carrillo had seen something in the connecting rods bearings so small that I never saw it. On three or four of the connecting rods, precisely where the connecting rod cap bolts together, three or four of the bearings were worn to the copper. At such high power and high rotational velocity (high RPMs), the connecting rod cap was deforming, becoming oval shape, and pinching the connecting rod bearings. Pinching the bearings was causing them wear down to the copper in this location. When I inspected the bearings, I didn't even notice it because it's a very small area -- about 1mm width. The pictures below show the wear near the bearing caps.









The new connecting rods
Once the new connecting rods arrived, I was very eager to see what had changed. I opened the box, removed a connecting rod, then stared at it for a few minutes because I didn't see any difference. Only when I placed the rods side-by-side did I begin to see the difference. Precisely at the spot where the connecting rod cap bolts together, there is a little extra metal. That's about all I noticed. However at the 2011 SEMA show in Las Vegas, I happened to be talking to a guy in the Carrillo booth. I described my project to him and how Carrillo redesigned the connecting rod. Just by coincidence, I was talking to the guy who actually designed my connecting rods and he remembered the project very well. Not only did he add a little more metal at the junction, but he thickened the H-Beam a little, changed the H-Beam radius, and slightly modified by bore size for the connecting rod bearing.

All of these latter changes aren't really noticeable to the naked eye, but you can definitely see the extra metal in the new rods when you compare them to the older ones. A complete picture set of all of the new connecting rods can be seen here:
New Connecting Rods - Robert & Mary Ann Collins













Calculating Rod and Piston Velocity
While I was waiting on the connecting rods to arrive, it seems that I had a little extra time on my hands. So I decided to see how the stock connecting rod compares to the RD Sport RS-46 connecting rod (they are not the same length). So I wrote a spreadsheet and compared rod length, piston velocity, piston acceleration, and rod angle. Since I had specs for the factory S65B40 (OEM Motor), S65B44 (GTS Motor), and RS-46, I decided to compare all three.

I apologize that I don't remember the units of measure when I did this (it was a long time ago). But from these graphs, it's pretty easy to see how each one compares against each other.





 
#7 ·
Impressive work Robert,how many miles did it take to cause the damage on the rods, perhaps i should check mine.

one other thing, have you checked if the 4 wheel drive tranny from 335xi would fit on a M3, i am starting to get tired of wheel spinn ..
 
#10 ·
Impressive work Robert,how many miles did it take to cause the damage on the rods, perhaps i should check mine.
20000 miles on the stroker + supercharged motor. I lost track of how many of those were pure stroker, and how many were stroker +supercharger. If I'm not mistaken, I think it was about 11000 stroker and 9000 stroker+supercharger. Included in those miles were two track days, the Mojave Mile, and Texas Mile.

[/quote]one other thing, have you checked if the 4 wheel drive tranny from 335xi would fit on a M3, i am starting to get tired of wheel spinn ..[/quote]

I haven't checked, but I doubt the M3 chassis would accomodate the AWD setup very easily. I also think there would be mandatory ECU programming and vehicle coding as well.
 
#11 · (Edited)
Theres a lot of wear on the bearings, were the tolerances right? I would check the radius on the crank journals and side to side float clearance of the rod big end to crank.

20000 miles, high revs and a supercharger should not have done this; the oil film separating the bearings and crank has failed so I would be concerned if I was you and would want answers and remedial work done before buttoning up the engine.

The rod pinching at the break line is very strange, were the caps fitted with dowels and was there signs of movement/distortion?

What were the bores like when it was pulled apart? From what I can see of the pistons all looks well there. With such short piston skirts and supercharging especially on a stroked engine puts a lot of load on the thrust side of the piston and bore.

615BHP is pushing things that far so I am surprised at the visual clues I am seeing.

Cheers

Jay
 
#12 ·
Theres a lot of wear on the bearings, were the tolerances right?
BMW factory tolerances are very tight. Factory tolerances aren't published by BMW in any of their repair manuals. We had to measure stock cranks and connecting rods to figure them out. The stroker was built using factory tolerances. This was reconfirmed after we pulled the motor apart and measured it up after the fact. In other words, whether right or wrong, the motor you are seeing was built with factory tolerances.

I would check the radius on the crank journals and side to side float clearance of the rod big end to crank.
All of these issues have been addressed on the rebuild.

20000 miles, high revs and a supercharger should not have done this; the oil film separating the bearings and crank has failed so I would be concerned if I was you and would want answers and remedial work done before buttoning up the engine.
You're getting ahead of the story. :M5thumbs: But yes, all of these issues were addressed with the rebuild.

The rod pinching at the break line is very strange, were the caps fitted with dowels and was there signs of movement/distortion?
No dowels. I'm not sure how to check for signs of distortion and movement. The connecting rod bores all measured up nice and round when we pulled them apart...if that's what your asking. I have all of the rods here at home. If you tell me what to look for, I will check.

What were the bores like when it was pulled apart? From what I can see of the pistons all looks well there. With such short piston skirts and supercharging especially on a stroked engine puts a lot of load on the thrust side of the piston and bore.
A few hundred pictures are posted here. It sounds like you already checked them out. I photographed just about every hole from multiple angles. There was some side wear on a few holes, but not all of them.
 
#13 ·
Hi PencilGeek,

I'm not trying to poke holes in your project or thread, sorry to jump ahead, I'll hold tight and read your updates with great interest.

As for signs of movement in the rod big ends there might be areas polished or gualing from movement, if there had been dowels damage might be present. I'm surprised there's no dowels, what locks the 1/2's alignment? Again if these questions are answered later in your project story I can eagerly wait.

So many questions... oh the wait...

Cheers
Jay
 
#15 ·
Hi PencilGeek,

I'm not trying to poke holes in your project or thread, sorry to jump ahead, I'll hold tight and read your updates with great interest.

As for signs of movement in the rod big ends there might be areas polished or gualing from movement, if there had been dowels damage might be present. I'm surprised there's no dowels, what locks the 1/2's alignment? Again if these questions are answered later in your project story I can eagerly wait.

So many questions... oh the wait...

Cheers
Jay
It's all good. Nothing wrong with asking questions. Not sure I know the answers. But I will look at the link you gave me, and see if I have a disassembled Carrillo around here to look at...maybe take some pictures to help answer your question.
 
#14 ·
See here:

con-rods « RET-Monitor

Location of con rod caps, part 2
Thursday, December 16th, 2010
In the previous article on the subject of maintaining the accurate locations of the two parts of a split con rod design, we looked at dowel pins and ring dowels, and the relative merits of the two methods.

The subject of joint shear stiffness was raised, and it was noted that the ring dowel, having a greater cross-sectional area, provides more stiffness to the joint. A stiffer joint is more stable and less likely to suffer from joint face fretting wear. While the con rod bolt may not Read more…

Tags: con-rods
Posted in con-rods | 2 Comments »

Location of connecting rod caps, part 1
Wednesday, November 10th, 2010
In an article published in 2009, I looked at some of the design features of the joint face of a split con rod. One important requirement is that the two parts of the con rod - the ‘blade’ of the rod and its cap - must be positively and reliably located with respect to each other.

It is important that these location features are machined into the rod before the big-end bore is finished to size. This guarantees that when the rod is assembled and the Read more…

Tags: con-rods
Posted in con-rods | No Comments »
 
#17 ·
Crankshaft journal and bearing adjustments
For the most part, all of the bearings looked great when we disassembled the supercharged stroker motor. There were some discolorations in the bearings, but nothing looked out of place. That's why it was a bit of a surprise when Federico (RD Sport) pulled me aside and told me Carrillo wanted to redesign the connecting rods.

After I gave RD Sport permission to redesign the rods, I left Van Dyne Engineering and went directly back to Auto Talent. I wanted to take a much closer look at the bearings and take as many pictures as possible of every connecting rod and every bearing from every possible angle. I was able to immediately see the excessive bearing wear at the joints, just as Federico had told me. And the more I looked at the bearings, the rattier they began to look to me. But I'm not the engine builder, and I have no idea what Van Dyne planned to do about it (if anything).

As you can see below, the good bearings look pretty good, and the others vary from decent to rather ratty looking. Here's a sample of the bearing pictures. The complete set of all bearing pictures can be found here, starting at the bottom of page-2. If you click on the photos, they will expand to a full screen to get a closer look. The photos actually look much worse than the bearings look in real life. In real life, there's really nothing more than a few discolored spots on the bearings. I think the lighting has amplified the discolorations and makes it look much worse than it actually is. But the lighting also makes it real easy to see what I'm talking about.

Good bearing photos:





















Ratty bearing photos:















Measuring the journals
Since BMW doesn't publish their bearing journal clearances, we measured a few crankshafts and connecting rods to figure out what those factory clearances are. Van Dyne doesn't use platigauge to measure the clearances, they measure it much more accurately and in a more sophisticated manner. Van Dyne pre-assembles and correctly torques the main and connecting rod caps. Using an expanding micrometer that can measure inside diameters of a circle, Van Dyne measures the assembled bearing size. Then using a micrometer on the crankshaft main and rod journals, those measurements are subtracted from the bearing size to calculate the bearing clearance. If Van Dyne doesn't like the clearance they measure, they will recommend to change it.

Many people might have read that and thought BMW knows best and you shouldn't make any such adjustments. But BMW doesn't publish these specs and doesn't offer any different rod bearing sizes to allow you to make adjustments. This means BMW literally has a one-size-fits-all mentality to connecting rod bearings -- even though main bearings come in three different sizes. Van Dyne isn't the type of shop that simply slaps the motor together with the parts available. Van Dyne has a very long and prestigous racing pedigree (see side-bar below). As I mentioned earlier, if they don't like the measurements, they will want to make adjustments.

For weeks we had been getting ready to assemble the bottom end. All of the parts were at Van Dyne and ready to assemble. The only remaining job was to meaure the bearings and journals. Now that everybody's schedule was in sync, Van Dyne made the measurements and wasn't satisfied. 48 hours before we were supposed to assemble the bottom end, everything came to a screeching halt. Van Dyne wasn't satisfied with the clearances and believed they were too tight; they wanted to make some adjustments. Van Dyne recommended to resize the main and rod journals to obtain the exact clearance they wanted. Using an exact journal-by-journal measurement, Van Dyne sent the crankshaft down to the machine shop for resizing, re-treating, and re-balancing. Resizing the journals cost us another two week delay. Hopefully it would be our last!

Unfortunately I don't have any pictures of the measurement process. But I do have pictures assembling the rest of the bottom end after the adjustments were made. The following pictures are just a few of the many pictures I took. To see all of the photos, please view this photo album.

















Who is Van Dyne Engineering?
Stewart Van Dyne was the principle engine builder for Drake Engineering (formerly Meyer/Drake Engineering). Stewart and I worked together at Drake Engineering in the early 1980's. Stewart built Mark Donahue's Indy-500 winning motor, along with motors that have won endurance races in Lemans and Sebring. Stewart's winning race motors are far too numerous to list. Drake Engineering was no lightweight racing shop. The Drake "Offy" won the Indy-500 twenty-six times of which seventeen of those wins were consecutive. During my tenure at Drake, I witnessed our engine designer, in conjunction with Stewart design and build prototype motors motors for Chrysler, Buick, Mc Laren, and Carol Shelby. When I left Drake Engineering, they were in the process of designing a five-valve OHC cylinder head for a small block Chevy. When John Drake retired, Stewart Van Dyne bought all of his engine molds and patents, thus ensuring the ability to remanufacture any of the historic Drake racing engines and other Drake racing engine parts.
 
#19 · (Edited)
Checking the connecting rod side play

After assembling the crank plate and installing the connecting rods, it's time to check the connecting rod side play. The connecting rod side play is the amount of side-to-side play after the two connecting rods have been installed on the same journal. If the clearance is too small, then the connecting rods will rub against each other, create extra heat, and not allow enough oil to escape. It's a vicious cycle of generating too much heat. The side play is measured by spreading the connecting rods as far apart as possible and then measuring with a simple feeler gauge. If the side play is too tight, then the motor must be disassembled and adjustments must be made.



Factory Measurements:

This is another one of those areas where BMW doesn't publish any specifications, so without measuing a few crankshafts and connecting rods, it's impossible to know what BMW intended. A typical measurement is 0.010" (ten-thousanths of an inch). For our application, we wanted a slightly different side-play tolerance. So we first had to measure the factory crank and connecting rods, then measure the stroker crank and connecting rods to see how they compare. The complete set of photos measuring the side play may be found here.





Stroker Measurements:







Adjusting the side play:
Even though Carrillo connecting rods are made to the highest standards, we did find a few of them were 1/2 of a thousanth of an inch different than the others (0.0005"). Since Van Dyne already planned to use his own tolerance specifications, this was our opportunity to make the necessary corrections. To ensure that Van Dyne machines the correct face of the connecting rod, each rod is marked with dye. The dyed face is the side that gets machined.







After all of the connecting rods are properly marked, each one is machined on a high precision grinder to obtain the exact dimensions Van Dyne desired. The entire process took more than two hours because the grinder only removed about 0.00025" on every pass. Each connecting rod is measured and re-measured. If it's not perfect, then it goes back on the grinder for one more pass. The complete set of photos documenting the connecting rod resizing process may be found here.











Once all of the connecting rods are properly resized, then it's time to clean up the parts. Each connecting rod is slightly sanded to remove any burrs, then dropped in the solvent tank to clean it all up. The pistons are reattached and the motor is put back together again.







Once the bottom end is finally assembled, then final measurements are taken. Now that the bottom end is fully assembled and no more adjustments are necessary, the block is sealed by injecting a special block sealer provided by BMW. Once the two halves of the block are assembled, the crank seals are installed. Two one-way check valves are pressed into the block. A sealant is injected through the check valve. The sealant fills a channel that runs between the block and the crank cradle. The injection process continues until the sealant squirts out a tiny opening at each crankshaft seal. You don't want to waste this sealant because it's $100 per tube, and there's just barely enough to complete the job.











 
#20 ·
Looks like they are taking the extra time and effort to check and recheck everything. Nice to see the correct tools used too! I like the slide wedge, haven't seen one of them for a while!!

Thanks for taking the time to document the process and post here, I for one really appreciate it as I have a passion for well built engines.

Good luck with the next stage.

Jay
 
#29 ·
Yes.

This is a very long build thread, do not know I could live without my car for this long.
Still waiting on head studs. Engine builder is at Indy-500 this weekend as part of celebration of past drivers and crews. I guess he's bringing some book or something to give to Roger Penske from back in the day when he worked on Mark Donahue's motor.

Anyways we're getting real close on head studs. Think we found the 625 material we were looking for without having to buy 75,000 pounds of it. Then head studs can be made.
 
#28 ·
This is a very long build thread, do not know I could live without my car for this long.
 
#32 ·
625 material fell through. We ended up going another route with 8740. Studs are now in progress. We have enough material for at least two sets of head studs...maybe three. Final specs will be about 220k PSI. 625 material would have been 260k PSI.

If you need a set, PM me.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top