Wednesday, March 27, 2013

Chapter 10 The Art of the Deal

  The race heads were at Precision Engine in Georgia and my crankshaft was ready to be magnafluxed, index ground and clearanced for the Honda rod journals.  The block was ready to have the decks leveled and surfaced.  It was clear I needed a road trip. 

  Another Studebaker Bonneville race car changed owners and moved from the left to the right coast.  David had introduced himself on the Studebaker forum after the purchase and a few of us convinced him it would be more fun to run Studebaker power in it.  He started a buying spree to begin collecting the bones of good cache of potential race parts.  Some of those parts were from legendary Ted Harbit, who hails from central Indiana.  I offered to run up to Ted's and ferry the parts to Georgia.

  Around that same time, my friend Jeff was considering having a few friends over for a weekend of Georgia style Studebaker fun.  A few phone calls later and it became a full fledged event.  Jeff was celebrating what we could only assume was his 100th birthday, because he wouldn't tell us the real number.  I was celebrating the one year survival of my accident.  The owner of the Racing Studebakers web site, Sonny,  Jeff, David, Matthew, Mike and Linda, who are the major partners in a well-known Studebaker shop, all descended on Eastern Georgia.  I came in a couple of days earlier to drop my crankshaft off at Precision Engine and discuss the project.

   On a personal note, backing up a couple of days, we had noticed my left shoulder was swelling and one of the insertion points looked infected.  Knowing I was headed on a week long road trip, I checked with the docs.  Yes, MSRA had infected the surgery sites.  A prescription for antibiotics and a promise to return the day I got home got me out the door and on my way.  But things didn't get better.  The shoulder just got larger and stiffer with accompanying increase in pain levels.  Not wishing to end up in a distant hospital, I sucked it up, ate pain and antibiotics and carried on.  Upon my return home, the doctor checked me into the hospital and went back in to the shoulder.  After scraping the reattachment points for the bicep and flushing 16 liters of rinse to clean everything out came a few more days' stay and IV's.  When they released me, it was with the commitment to complete a 6 week course of daily IV infusions and weekly blood tests.  Very limiting, but better than the alternative.  Back to the important stuff, the car.

  During the celebratory weekend David told us about his plans for the car and the time crunch to finish the first engine built to check out the chassis and safety systems for the April ECTA event.  One of the big time investments was the development of some good cylinders heads.  It didn't take long for me to explain to David about the plans I had for my heads.  Since they would take much less time than starting from scratch, I offered to loan them out.  David quite generously offered to pay the machine shop bill for the titanium valve conversion and the deal was done.  My heads will run in at least the April ECTA event known as the Ohio Mile.  If he needs them longer, I agreed for the loan to last through the race season.  A day or two later I was comfortable enough to offer the loan of my entire engine to David's cause.  We didn't close the details out, but there is a reasonably good chance that Rabid Snail Racing prototype engine number RS-R001 will run this year.  Very exciting and much more immediate gratification than waiting the 2 to 3 years it will take to build the rest of the car.

Tuesday, January 29, 2013

Chapter 9 Just Breathe

  Yet another piece of the puzzle, the cylinder heads, were acquired a few years back when opportunity knocked and kept in mind for a maximum effort Studebaker engine.  Heads with good airflow, or lack thereof, are a major issue for Studebaker racing projects.  With an engine designed in the early 50's for a completely different circumstance, Studebaker's engineering team was looking for a small displacement, high compression V8.  This makes for a good solid bottom end, but they never dreamed the horsepower race would require larger and larger displacement, with ever increasing RPM levels.  That heads designed in the 50's would run out of steam as more performance was demanded, was inevitable.

  The Big Three were able to design new heads with better flow, but Studebaker's finances required creative thinking without much budget.  Throughout the 50's and early 60's Studebaker pretty much used different heads purely for compression ratio changes.  The biggest effort to getting more air into and out of the cylinder centered around compressing the intake air with belt driven superchargers as in the Golden Hawk.  With Detroit getting serious about the horsepower wars in 1962, it was clear that Studebaker had to answer with more potent power plants.

  The announcement of the Avanti and Jet Thrust engines added 4 high performance options.  Known as "R" Series engines, buyers could choose the R1 with hotter cam and reasonably increased compression, The R2 and R3 which included Paxton centrifugal superchargers, and the R4 dual 4 barrel variants.  The R3 and R4 engines also had improved heads.  The intake ports were opened dramatically.  Larger intake and exhaust valves were installed which required moving each pair of valves away from one another and notching the engine block for clearance and increased airflow.  The R3 heads quickly became sought out by engine builders seeking big performance gains.

  The chart above relates a set of stock R2 heads compared to the R3 cloned and ported heads.  Factory R3 heads varied widely in performance but generally fall about half way between the two examples.

  Total numbers of R3 heads cast are hard to come by.  That demand outstripped that figure, whatever it may be, made the price go as high as complete shortblocks.  As a guess, there were less than 200 "B" blocks, the stamping added to blocks destined for R3 or R4 use.  Perhaps, enough to complete those engines and that amount again for replacement parts is close.  Still that is less than 400 sets. The time to buy a set is when you find an owner who will put a price on them.

  With a set of real deal R3 heads out of the budget, the quest began for something just as good or better using more conservativley priced components.  A builder in the Atlanta area was taking common Studebaker heads, welding on more material where needed and matching the R3 intake manifold.  The chambers featured R3 sized valves.  With a lot of grinding and cutting the ports on these heads promised R3 performance plus.

  I picked up a set from the builder and proceeded to install them on my factory R2 engine.  The perfomance gain was immediate and satisfying.  Bottom end torque was vastly improved with careful modulation of the throttle needed to keep the rear tires from spinning at any point throughout 1st and 2nd gear.  I ran these R3 clone heads until I sold the R2 Avanti, when I put the original heads back on for the sale.

  With no running engine to play with myself, I offered to loan the heads to an internet friend who had a flow bench and some Studebaker racing engines.  Tom had been working toward getting the maximum performance from stock Studebaker heads for years and jumped at the chance to experiment with mine.  I dropped them off at a friend's in Georgia who later took them to Tom's in Virginia.

  When Tom got the heads I had been running them very rich and there was considerable soot and carbon.  Tom had to clean them up just to flow test them.  As he cleaned and flowed each cylinder, Tom began tweaking the back cut on the valves and some small changes he felt would help them.  Ultimately, the time came to install them on an engine and see what they did in a real application.

  The test bed was a Lark Tom had.  No cage, no trick suspension, just a base R1 short block, my heads and intake, and a turbocharger.  Running almost every week, testing and tuning along the way Tom got some pretty huge perfomance.  How many stock block, stock chassis Larks can pull the front wheels?  Tom was doing it pretty regularly.  My biggest regret is not getting there to see it in person.  As the season came to an end, the heads were removed from the Lark, picked up by Jeff in Virginia, and currently await their turn at Jeff's for a spa trip to Bill Ford's shop.

  I have been EBay stalking some of the vendors who buy and resell NASCAR parts.  Along the way, I added to my racing nest egg several sets of 7mm stem titanium valves, titanium locks and titanium retainers.  Also a set of Comp Cams' heaviest beehive springs, spring seats and a set of Crower roller lifters.  Bill will cut the new spring seats, add BeCu exhaust seats to improve heat transfer out of the valves, replace the cast ironguides with manganese-bronze valve guides, and cut the intake and exhaust seats for installation of the titanium valves.  Another friend from the Salt2Salt team, Digger, is cutting the titanium valves down to the correct length and diameter.  The goal is a rugged, but very light valvetrain that is capable of swallowing and digesting everything the supercharger will be shoving at it.

Sunday, January 20, 2013

Chapter 8 It's a Grind

  Working on the block has been a parallel project with changes on the crankshaft.  The crankshaft started life as a forged steel Studebaker 289 unit.  It was stroked 5/8" to take the stroke from 3.625" to 4.250".  I had a lot of concerns about getting it to work for a Bonneville application centered around the close quarters and proximity of the camshaft.  More details on the camshaft is coming, but suffice it to say the lobes on the camshaft were very close to the bearing journals.  The cam used by Granatelli was very small in comparison as they had reduced the base circle to 3/4".  The roller cam I wanted to run and designed expressly for Bonneville use with a valve lift of 0.620".  I assumed the cam was reduced because of clearance issues with the stroker crank.

  Since the crankshaft was key to getting the displacement needed, I determined I needed to work on the crank and if it won't clear the giant roller cam I'd design another with reduced base circle.

  In balancing the crankshaft, extra weight was added to the end counterweights.  Otherwise, the crank was just like a standard 289 crank, just made longer in the rod journal placement.  The counterweights were the same shape and rod throws carried the normal contour out to the rod pins.

  Oil control is a big thing for me and I wanted to make this crankshaft enhance the oiling not make it worse.  A phenomenon known as windage occurs when oil from the bearings and cylinders is hit by the crankshaft dissapated into smaller droplets and pulled around the rod throws and counterweights by the rotation of the crankshaft.  Smokey Yunick did some experiments with windows in the oil pan to actually see windage in a running engine.  He reported seeing virtually all the oil in the pan making a larger and larger cloud with the appearance of a taffy pulling machine.  This creates a parasitic drain from the impacts.  Also, as more oil is entrained, less is available for engine lubrication and cooling.  Yes, cooling.  More on that later.  Eventually, with high enough RPM's virtually all the oil is tied up in the windage cloud and, in extreme cases, cause the oil pump intake to be uncovered and system oil pressure and flow goes to zero.  Bad combination.

  So, what can be done?  Various engine builders have done experiments in reshaping the counterweights and rod throws with good effect.  Arguably, the best method shapes the leading edge of the counterweights as a bullnose or rounded profile.  This reduces the impact of the crank slamming into oil draining down to the pan and encourages the oil to run across the surface to an edge where it can be channeled safely back to the oil pump.  The trailing edge is reshaped in a boattail with a sharp edge which sheds the oil in large drops to the main web or block wall, out of the way of the next oncoming rotation.

 The next step is a crankshaft scraper, a thin blade of Teflon which wipes the oil from the crank as it passes 90 degrees on the upward portion of rotation.  This, in conjunction with a well designed oil pan should go a long way to improving oil control.

  It takes a lot of grinding, shaping and smoothing using several tools to reach in and get to the  nooks and crannies.  The pictures show the roughing out of the shapes needed.  There is a lot of surfacing left to do.

  An added bonus is that the crankshaft is lightened during the knifedging greatly reducing bearing wear and crankshaft twist.  The smoothing of forging marks increases reliability as well.  There is another process yet to be done once the entire knifedging proceedure is done and the crankshaft is polished to a scratch free condition. The oil ports that run throught the crank need to be beveled removing the sharp edges and tapering them out to increase the time the bearings are in a place in rotation where oiling can take place.

In all, it's a lot of work.  Some folks pay someone else the $1000 it costs to farm this work out.  And we haven't verified other changes needed for clearance of this crankshaft.

Chapter 7 Concrete Thinking

  With oiling issues partially addressed, I turn my attentions to the block's water jacket.  Studebaker blocks were cast in Studebaker's foundry located at their base of operations in South Bend, Indiana.  Anyone who has peered into the water jackets can attest to the wide variety of foreign objects that occupy that space.  Casting sand, wire used to reinforce and position cores, dirt, rust, a little bit of anything and everything.  This block was no different.

  The hot tanking at Precision Engines II had done a pretty good job of rinsing most of the water jacket clean, but there was still a fair amount of rust and corrosion.  One bank at a time, I filled the water passages with muriatic acid.  If you do this, be prepared for some pretty nasty fumes which will rust any ferrous metals in the vacinity.  After allowing 10 minutes or so of foaming and bubbling, I rinsed the block with fresh water.  Then repeat on side 2.

  The sonic wall testing had shown a few thin spots down low in the cylinders.  This wouldn't be exposed to combustion pressures as the piston rings were well above that point, but any flex would be potential problems for the rocking of the piston at bottom dead center.


  Moroso and several other companies offer a block filler which is close to the expansion properties of cast iron.  I just wanted to reinforce the block at the bottom which is called a partial fill.  I positioned the block level front to back and rotated so that the self leveling filler would be over the main oil galley tapering out to the bottom of the core plugs.  Mixing enough filler for one side, it was simple to pour the mixture through a funnel into several of the block deck water ports.  After tamping and helping the mix level with a long paint mix stick, it was topped off to fill up to the desired point.  The next day was a repeat of the process on side two.

  I need to go back a little and pick up a modification made before the block was filled.  In my trips to Bonneville and reading the adventures of other Studebaker guys I noticed several runs had to be aborted due to core plugs coming out and dumping the block coolant during runs.  Definately not what you want when you're backing up a record run and something like this forces you to shut down.  Explanations abound but center on block distortion loosening the plug in its seat and coolant pressure doing the rest. 

  Pretty simple fix really.  Stealing one of Jeff's ideas he stole from someone else, I'd just thread the block to accept iron pipe plugs.  This takes some cleaning up of the existing casting core opening.  BTW, core plugs, or what some call freeze plugs are not to keep your block from cracking when the coolant inside freezes.  Core plugs exist as an artifact of the casting process.  Foundries have to have a way to get the sand out of the block after the casting and cooling process.  Back to the story.  The major thing to remember when threading for a threaded core plug.  That tap is going to get very close to the water side of the cylinder wall.  Forcing the tap to the point it contacts the cylinder wall and pushes on it may result in cracking the cylinder.

  On to the next project in block prep...

Chapter 6 Divide and Conquer

  Studebaker and many other manufacturers used a partial flow filtration in the late 50's up until midyear 1962.  At that point, with manufacturers making the change to full flow filtration, Studebaker also produced their blocks as full flow.  Race block #0001, as a 1959 product, was a partial flow.  To take advantage of full filtration required some modifications to the block oiling gallies.  Since we were making changes anyway, why not explore other improvements?

  Fairborn Studebaker can modify blocks to function as full flow as seen above.  This is accomplished by interrupting the oil circuit as it comes up through the block to the main galley running parallel to the cam.  The oil is routed out of the block, filtered and returned above a plug installed just above the detour and continues on its merry way just like it would have when stock.

  Rather than use two flex lines, I went with pipe connections.  I moved the return to an enlarged, unused oil galley which elimnates two 90degree turns that the oil must navigate.
  While we were making modifications, why not make other improvements to the Studebaker oil circuits?  In researching the reliability of Studebakers I had noticed the majority of bearing failures were in the front of the block.  Applying some logic, my hypothesis was that oil to those bearings had already travelled through the entire block, picking up heat along the way.  The problem was compounded if the bearings the oil passed by in its journey were worn and allowed more oil to bleed off.  So, a reduced flow of heated lubricants would be more than enough to create a greater possibility of oil related issues.  To address this it seemed only natural to provide a source of clean cooled oil at the front of the block feeding front to back to the point where it would meet the normal oil flow coming back to front.

  Fortunately Studebaker has a fitting on the front of the block connected directly to the main oil galleys.  To regulate oil pressure Studebaker had placed a pressure bypass circuit on the front of the block which had a piston backed by a spring.  When the pressure in the oil galley exceeds the pressure of the spring, the piston compresses the spring and uncovers a port whis bleeds the excess pressure back into the pan.  This in itself was an issue as the oil returned to the pan had also been run through the block picking up heat.  A pressure relief in the oil pump itself would take less effort to drive and the bypassed oil wasn't heated up.

  Removing the contents of the pressure bypass bore and plugging the return port allows use of the bypass to bring fresh oil to the front of the engine.  Boring the bypass bore to 1/2" made it possible to have the oil divided equally to the front and rear of the block.  The oil circuit was now routed from the oil pump, out of the block to a filter and oil cooler, then divided into two equal flows - one to the front of the main galley and one to the rear of the main galley.

  Inside the block all of the oil passages were opened up 1/16" over stock.  So 3/8" passages were taken out to 7/16", 1/2" to 9/16" and so on.  This was a twofold benefit.  The galleys were cleared of all rust and gunk which had accumulated plus it is possible to deliver a larger flow if a bearing clearance begins to open up and still have adequate oil to bearings downstream.

  The emphasis and concentration would shift, but there is much more to do on the oiling front

Friday, January 18, 2013

Chapter 5 Go with the Flow

Now that the worst of the cleanup was done I was less concerned about throwing money into a block only to find out there were problems which would result in having to discard it.  It was time to think in terms of making the engine live and thrive in its high output environment.

The first item I turned my attention to was the lifter valley.  Racing Studebakers all fight one common characteristic; the engine oil as it is pumped through the engine has a tendency to accumulate in the lifter valley and on top of the heads.  So much so that many an engine has been damaged by all of the oil being held on top and none available to the oil pump.  The loss of oil pressure can result in catastrophic damage to bearings, crankshaft, rods and cams.

  With that in mind I ran a little experiment on my block.  I set the block level left to right and angled front to back as it would be when installed in the engine bay.  Then I slowly poured transmission fluid into the lifter valley at the very front and center.  I continued pouring until the fluid filled all the little pockets between the lifter bosses.  It took 10 ounces to fill those pockets before the first drop of fluid reached the opening where it would drain back down to the pan.

This means that fully 1/2 quart of the oil in the engine never gets returned to be recirculated by the oil pump.  Further, when the engine is shut down, that oil is subjected to the heat of the engine and literally cooks into a black thick ooze, sometimes even to hard chunks.  Not the kind of stuff you want coming loose in your engine.

  The answer was simple.  Fill all those pockets in and smooth the oil's way back to the bottom.  A special metal filled high temperature epoxy was used for just that.

  With the next pour of transmission fluid, the liquid flowed quickly and completely to the return port at the rear of the valley.  Of course, that is just one of many steps to getting a handle on the oil control needed.  More on down the line.

  With a block in the condition this one was in, every surface becomes rusted and pitted.  Work in the lifter valley was a constant reminder that the lifter bores were also rusted and pitted just like the cylinder walls.  The lifters must cycle smoothly and be contained in a closely fitted bore.  Clearly, a change had to happen.  The popular engines have a lot of tooling in machine shops all over the US.  Studebakers don't enjoy that widespread availability.  If this were a Chevy engine I could just send it out to a shop with the proper tooling to ensure the lifter bores were centered perfectly over the cam lobes and at a perfect angle to keep the lifter in contact with the cam.  This wasn't possible with this block.

  Jeff called me one day and told me about a company making a tooling set to rebush lifter bores.  I contacted them and ordered the tooling.  It consists of a piloted reamer which centers on the existing lifter bore and cuts the bore larger to accept a bronze bushing.  The second tool installs the bushing in the reamed hole.  The final tool pilots on the inside diameter of the new bushing and cuts it to the final size of the lifter.

Installation proved simple enough. I did add a bushing to center the top of the reamers rather than relying only on the pilot to keep thing straight.  Cleanup prep and a coat of Glyptal added a finishing touch.

Chapter 4 Clean is the Scene

  Even with the choice to run in engine class C a lot of latitude existed.  The displacement of the engine could be anywhere between 306.00 and 372.99 cubic inches.  The block from the woods had started life in 1959 as a 259 cid engine.  The score of the stroker crank certainly provided extra displacement, but now the decision of how close to get to the class maximum had to be made.  If I went all the way to the brink of 372.99 inches and had to bore the engine more at some point I would find myself at the bottom of B class.  We would certainly be the little duck in the big pond then.  So, I backed off a bit and aimed for a size that would allow at least one teardown and rebore without going over the C class max.  Since the R3 engines were bored out 0.093" I arbitrarily chose 0.030" over that.  So the target displacement ended up at 362.8 cid.

  We last saw the block at DeepNHock acres, where Jeff and I had tried to clean up the worst of the mess.  Now it was time to start making the modifications that would bring the little 259 back to life as a bigger, better, faster, stronger genuine race engine.

  The first stop had to be a machine shop where the block could be cleaned up in a hot tank before going any further.  Precision Engine II in Sylvania, GA was just a stone's throw away, so we loaded the block and headed over there.  After exchanging pleasantries, Bill had the block unloaded by a helper who took it back to begin cleanup while we discussed the work we wanted Bill to do.

  First on the list had to be a general inspection for cracks or other problems followed by sonic testing the cylinder walls to be certain we could safely bore the cylinders to 3.686".

  The block was difficult to get readings on because of the rust, but after cutting some light passes Bill was able to declare we were on solid ground.  The block was bored and finish honed to it's final dimensions.

  Running that kind of stroke plus the extra stress from supercharging requires more support for the crankshaft.  Bill installed Ken Evans 4 bolt splayed main caps and align honed the block.  We discussed other modifications I would be making and ordered a set of ARP studs for the mains and heads.  Talk about service !  Bill found out I wanted the work done, did the job and had the block ready for me to bring back to Illinois in 3 days over a weekend.

  The first order of business was a careful documentation of the wall thickness of each cylinder.  The picture above shows the system Jeff and I developed to test and record several areas in each cylinder in a repeatable method.  Hopefully, as we accumulate data on many blocks, we can predict which blocks will be most suited for large bores.  Of course, each block is an individual and will always require checking before investing lots of hard earned cash to build it.

  The system we came up with assigns positions around the cylinder like a clock.  Readings are taken at 2, 4, 6, 8, 10, and 12 o'clock at depths of 1", 3" and 5" from the block deck.  Each reading is recorded for every cylinder.  It takes a little time, but reveals a lot of information about how Studebaker cast its blocks and routed oil galleys and water passages.
  We also developed a chart to graphically visualize how the wall thickness vary and transfer cylinder to cylinder.

  There are always a bunch of little things that add up to a big job on a race engine.  The flash from the casting process had to be ground down.  The water jackets were full of debris.  The accumulated oil sludge now baked dry had to be removed.  The block was certainly better than where it started, but there was a long way to go.

I began with general grinding and smoothing of the exterior casting flash using a small hand grinder, sanding rolls, and carbide burrs.  Then move inside.
The level of miscellaneous parting lines and slag was surprising.  The lifter valley, once it was cleaned up, required a lot of attention.  There is a big difference before and after.

Each threaded hole was cleaned out followed by chasing the threads, more cleaning and inspection for damaged threads and shavings.