Supercharging the MGB-V8
This article appeared in The British V8 Newsletter - Volume X, Issue 2 - May 2002by: Jim Blackwood
Many of us have lusted sincerely since adolescence for a car with a "blower". For me it was a TV commercial where a teenaged boy taunts his dad by playing with the throttle linkage on the bug catcher while the old man is trying to talk. That car was just too cool. In any case for some reason that big belt drive sticking through the hood just struck some primitive nerve.
Having gone the route of the V8 conversion on the MGB, what could be better than supercharging it? For the BOP 215/Rover V8 this is no small order however and as a result this car was first fitted with a factory Olds turbocharger. This was a neat piece of work and gave pretty good performance, but the entire setup was limited in a number of ways and when the shaft bearing began to wobble an excuse was born.
So this is the story of an MGB fitted with a roots type positive displacement blower, along with a modern engine management system including fuel injection and a distributorless ignition. Might as well go the whole route at once I reasoned.
Or maybe reason had nothing whatsoever to do with it. In any case it was
accepted at the beginning of the project that the drive and blower would
stick through the hood. Maybe at some primal level that was the objective.
At a later stage it was seen that it could be possible to do the conversion
and have the hood close on it. Therein lies the potential for a serious sleeper
and this thought will be picked up again later, so if you want to keep the
stock appearance stay tuned. A little about the patient and the direction of
the project is in order. The donor car was a 1971 MGB roadster with a very
checkered past, and not one that would have been selected for restoration
by anyone but a masochist. So despite anyone's preferences in regards to
originality of bodywork, interior, or any other aspect of the car this should
not offend the purists in our midst. They all would have likely consigned
the entire car to the parts bin or the scrap yard long ago, and it is only
by virtue of the radical modifications it has undergone that it has survived.
In short, the car has been lowered a couple inches, has been widened by some six inches, sports about twice the stock tread area, extremely oversized front brakes, and any number of other custom modifications we can touch on later. The engine in question is a genuine original Olds Turbo 215 which was rebuilt at Joe Lemley's Racing Engines in Southpoint, Ohio before I got it. Some of the old drag racers among us may remember Joe as the NHRA class record holder during the late sixties in the six cylinder class, a distinction he held with his grey Corvette for two or three years. Anyway, the turbo engines had unique high compression heads, which in combination with stock Olds 0.030 over low compression pistons, composition head gaskets, and 0.040" milled from the heads gives between 8 and 8.5:1 compression. Olds heads were used rather than Buick because of the even six bolt per cylinder head bolt pattern which gives better clamping force than either the Buick 5 bolt heads or the 300 cu in 4 bolt heads and should do a better job of sealing with forced induction. As boost is expected to be between 5 and 7psi, O-ringing the cylinders is not needed and the factory valve springs were retained since high rpm operation should not be necessary. The factory cam was retained as well for good drivability, and no other significant changes were made to the heads aside from light port matching and cleanup.
Stainless steel was used for all external fasteners, in most cases button head capscrews, which give a larger head bearing surface than socket head capscrews, as well as simply looking better. The choice of stainless was a practical one, to prevent corrosion.
After a good deal of investigation an Eaton blower was chosen. It is somewhat oversized for the 3.5 liter displacement, being designed with a 3.0 to 5.7 liter engine in mind, but that makes it an ideal size for a 4.6 Rover or a stroker motor. It is overdriven at a ratio of 2.3:1, giving a rotor speed of about 12,000 rpm at 5,000 rpm crank speed and if overboost becomes a problem the ratio can be easily reduced with a top pulley change.
The model designation is M-90. It is the 4th generation unit which has an integral bypass valve, a rear inlet, is specifically designed to work with multiport fuel injection, and can be mounted in any orientation. As British car enthusiasts should know, Eaton blowers have been around for a very long time, and this was an element of the decision. The cost was reasonable at around $1100 from Magnuson Products in Ventura, CA (magnusonproducts.com) but at the time of purchase these units were still scarce and I had to twist Jerry's arm a little to get his guys to ship me one. This blower was mounted in the conventional location to a brass plenum bolted to a modified Offenhauser intake manifold. A front support plate was machined from 3/4" aluminum and mounted to the front of the engine using the water pump bolts. This plate incorporates the tensioning arm and pulley for the dedicated 7 rib serpentine belt.
The tensioning arm is the Ford 4.6 unit with the rotation reversed and new stops bolted to the support plate. It is mounted with 1/2" button head capscrews on a boss countersunk into the plate. The alternator also hangs from the support plate, having a long 3/8" bolt attaching to the front of the left head as well. For the lower drive pulley, the hub of the Olds damper was used and modified as follows. The Ford crank timing wheel was fitted to the back side and welded into place. Then the center of the Ford damper was machined to press fit to a machined hub and was tack-welded so that it couldn't move and the assembly was balanced. This was mounted to the damper hub using longer pulley and center bolts.
The outer ring of the Ford damper doubles as the serpentine drive pulley, and the water pump drive pulley is attached in the conventional way. This moves the outer pulley forwards a good distance and was accommodated by using a flat pulley on the Buick 300 cast iron water pump, which is shorter and has the neck angled rearwards away from the serpentine belt. Since the nose or drive of the blower was a custom length which is available in 1" increments the housing for it is made in two pieces. The front cover is cast and is the same as the standard front cover except for having a machined boss for attaching the flange of the extension housing. This extension can be made in various lengths, in my case 8-1/4". That plus the front cover gives a drive housing length of 10" to the back of the pulley. The plenum houses a special experimental intercooler using heat pipe technology, has a sneeze valve on the rear under the blower intake, and raises the blower above the injector fuel rail.
The Offy manifold was chosen because of the single plane runner design and large plenum area after removal of the internal flow dividers which simplified injector placement and made room for the intercooler. The late Rover fuel rail was used with Ford injectors mounted in aluminum bosses. These bosses were heli-arced into recesses cut into the runner top surfaces using a special jig and purpose designed milling cutters, to closely approximate the placement and angle of the Rover injectors. Once the jig is bolted to the intake a1/2" drill is all that's needed to drill the holes and cut the seats for the injector bosses and of course, it will bolt up to any BOP/Rover intake. It would have been possible to use an epoxy based adhesive instead of TIG welding the bosses in, but as I was already having a 3/4" flange welded to the top of the manifold it wasn't much extra to weld them. On the exhaust side of the heads, flow is handled by custom fender well headers having 1-3/8" diameter equal length primary tubes running 35.25" into spiral scavenging 2-3/8" collectors, from where they dump into header mufflers and exit in front of the rear wheels. The exhaust system runs through the bodywork and is accessible by raising the forward tilt hood assembly, and by removing the stainless grillwork from the flared rocker panels.
On the intake side of the blower, fresh air is drawn through a conical K&N filter behind the radiator grille. Wire reinforced silicon hose routes this back to the mass airflow meter near the rear of the blower after which a larger "U" shaped hose connects to the throttle body. A brass coupler matches the throttle body to the blower inlet. An idle speed control valve is plumbed in parallel with the throttle body and no EGR is used. A Ford powertrain control system was used as the electronics package, featuring the EEC-IV PCM and modular distributorless ignition cannibalized from a 1993 Crown Vic police cruiser with the 4.6 liter V8 engine. This system was chosen for its ready availability, high level of sophistication and its proven high performance potential. The system has the ability to modify and store engine tuning parameters based on sensor feedback, and some very good hardware and software packages are available for under $400 which allow virtually unlimited programming changes to be made for custom tuning. In unmodified form the injection system can support power levels over 300 hp. I would recommend using a system taken from a Mustang though as most of the tuning work has been done using those computers. There is a large library of binaries, and GUI software has been well tested, whereas with the Crown Vic I am breaking new ground.
The stock distributor was cut down and capped with a 1" copper tubing cap, O-ring sealed to the housing and reinstalled to provide the drive for the oil pump. This was then reengineered to accommodate a sensor for cylinder identification.
The system will operate without this sensor, but it does provide a useful synchronization function. Much of the Ford wiring harness was reused, including fuse and relay blocks. A power distribution block was added, as well as some 18 fuses and 16 relays. Crown Vic cars utilized a very handy 3 relay module which is quite good for such things as radiator fans and headlights, so I couldn't resist adding a few more relays than the fuel injection system required. The PCM (computer) is mounted in the passenger's footwell using the Ford bulkhead connector.
To do this a small notch roughly an inch square was cut at the front of the outside edge of the existing RHD brake pedal hole for the PCM connector, and a cover plate fashioned to hold the box securely using the recesses molded into the connector body. The result gives a cleaner appearance than the original cover. The fusebox, one relay box, and ignition module are on the unused pedal assembly shelf along with the vapor recovery canister. The other relay box is beside the radiator and in front of the right side coil. The left side coil is just forward of the brake pedal assembly, near the engine compartment ventilation fans which draw air from the engine compartment into the fenderwell area on both sides. A Ford all-electric cruise control unit is mounted on the left frame rail at the front of the engine and the control cable is routed under the intake manifold and up to the throttle-body.
In short, the car has been lowered a couple inches, has been widened by some six inches, sports about twice the stock tread area, extremely oversized front brakes, and any number of other custom modifications we can touch on later. The engine in question is a genuine original Olds Turbo 215 which was rebuilt at Joe Lemley's Racing Engines in Southpoint, Ohio before I got it. Some of the old drag racers among us may remember Joe as the NHRA class record holder during the late sixties in the six cylinder class, a distinction he held with his grey Corvette for two or three years. Anyway, the turbo engines had unique high compression heads, which in combination with stock Olds 0.030 over low compression pistons, composition head gaskets, and 0.040" milled from the heads gives between 8 and 8.5:1 compression. Olds heads were used rather than Buick because of the even six bolt per cylinder head bolt pattern which gives better clamping force than either the Buick 5 bolt heads or the 300 cu in 4 bolt heads and should do a better job of sealing with forced induction. As boost is expected to be between 5 and 7psi, O-ringing the cylinders is not needed and the factory valve springs were retained since high rpm operation should not be necessary. The factory cam was retained as well for good drivability, and no other significant changes were made to the heads aside from light port matching and cleanup.
Stainless steel was used for all external fasteners, in most cases button head capscrews, which give a larger head bearing surface than socket head capscrews, as well as simply looking better. The choice of stainless was a practical one, to prevent corrosion.
After a good deal of investigation an Eaton blower was chosen. It is somewhat oversized for the 3.5 liter displacement, being designed with a 3.0 to 5.7 liter engine in mind, but that makes it an ideal size for a 4.6 Rover or a stroker motor. It is overdriven at a ratio of 2.3:1, giving a rotor speed of about 12,000 rpm at 5,000 rpm crank speed and if overboost becomes a problem the ratio can be easily reduced with a top pulley change.
|
Enjoying this article? Our magazine is funded through the generous support of readers like you! To contribute to our operating budget, please click here and follow the instructions. (Suggested contribution is twenty bucks per year. Feel free to give more!) |
The model designation is M-90. It is the 4th generation unit which has an integral bypass valve, a rear inlet, is specifically designed to work with multiport fuel injection, and can be mounted in any orientation. As British car enthusiasts should know, Eaton blowers have been around for a very long time, and this was an element of the decision. The cost was reasonable at around $1100 from Magnuson Products in Ventura, CA (magnusonproducts.com) but at the time of purchase these units were still scarce and I had to twist Jerry's arm a little to get his guys to ship me one. This blower was mounted in the conventional location to a brass plenum bolted to a modified Offenhauser intake manifold. A front support plate was machined from 3/4" aluminum and mounted to the front of the engine using the water pump bolts. This plate incorporates the tensioning arm and pulley for the dedicated 7 rib serpentine belt.
The tensioning arm is the Ford 4.6 unit with the rotation reversed and new stops bolted to the support plate. It is mounted with 1/2" button head capscrews on a boss countersunk into the plate. The alternator also hangs from the support plate, having a long 3/8" bolt attaching to the front of the left head as well. For the lower drive pulley, the hub of the Olds damper was used and modified as follows. The Ford crank timing wheel was fitted to the back side and welded into place. Then the center of the Ford damper was machined to press fit to a machined hub and was tack-welded so that it couldn't move and the assembly was balanced. This was mounted to the damper hub using longer pulley and center bolts.
The outer ring of the Ford damper doubles as the serpentine drive pulley, and the water pump drive pulley is attached in the conventional way. This moves the outer pulley forwards a good distance and was accommodated by using a flat pulley on the Buick 300 cast iron water pump, which is shorter and has the neck angled rearwards away from the serpentine belt. Since the nose or drive of the blower was a custom length which is available in 1" increments the housing for it is made in two pieces. The front cover is cast and is the same as the standard front cover except for having a machined boss for attaching the flange of the extension housing. This extension can be made in various lengths, in my case 8-1/4". That plus the front cover gives a drive housing length of 10" to the back of the pulley. The plenum houses a special experimental intercooler using heat pipe technology, has a sneeze valve on the rear under the blower intake, and raises the blower above the injector fuel rail.
The Offy manifold was chosen because of the single plane runner design and large plenum area after removal of the internal flow dividers which simplified injector placement and made room for the intercooler. The late Rover fuel rail was used with Ford injectors mounted in aluminum bosses. These bosses were heli-arced into recesses cut into the runner top surfaces using a special jig and purpose designed milling cutters, to closely approximate the placement and angle of the Rover injectors. Once the jig is bolted to the intake a1/2" drill is all that's needed to drill the holes and cut the seats for the injector bosses and of course, it will bolt up to any BOP/Rover intake. It would have been possible to use an epoxy based adhesive instead of TIG welding the bosses in, but as I was already having a 3/4" flange welded to the top of the manifold it wasn't much extra to weld them. On the exhaust side of the heads, flow is handled by custom fender well headers having 1-3/8" diameter equal length primary tubes running 35.25" into spiral scavenging 2-3/8" collectors, from where they dump into header mufflers and exit in front of the rear wheels. The exhaust system runs through the bodywork and is accessible by raising the forward tilt hood assembly, and by removing the stainless grillwork from the flared rocker panels.
On the intake side of the blower, fresh air is drawn through a conical K&N filter behind the radiator grille. Wire reinforced silicon hose routes this back to the mass airflow meter near the rear of the blower after which a larger "U" shaped hose connects to the throttle body. A brass coupler matches the throttle body to the blower inlet. An idle speed control valve is plumbed in parallel with the throttle body and no EGR is used. A Ford powertrain control system was used as the electronics package, featuring the EEC-IV PCM and modular distributorless ignition cannibalized from a 1993 Crown Vic police cruiser with the 4.6 liter V8 engine. This system was chosen for its ready availability, high level of sophistication and its proven high performance potential. The system has the ability to modify and store engine tuning parameters based on sensor feedback, and some very good hardware and software packages are available for under $400 which allow virtually unlimited programming changes to be made for custom tuning. In unmodified form the injection system can support power levels over 300 hp. I would recommend using a system taken from a Mustang though as most of the tuning work has been done using those computers. There is a large library of binaries, and GUI software has been well tested, whereas with the Crown Vic I am breaking new ground.
The stock distributor was cut down and capped with a 1" copper tubing cap, O-ring sealed to the housing and reinstalled to provide the drive for the oil pump. This was then reengineered to accommodate a sensor for cylinder identification.
The system will operate without this sensor, but it does provide a useful synchronization function. Much of the Ford wiring harness was reused, including fuse and relay blocks. A power distribution block was added, as well as some 18 fuses and 16 relays. Crown Vic cars utilized a very handy 3 relay module which is quite good for such things as radiator fans and headlights, so I couldn't resist adding a few more relays than the fuel injection system required. The PCM (computer) is mounted in the passenger's footwell using the Ford bulkhead connector.
To do this a small notch roughly an inch square was cut at the front of the outside edge of the existing RHD brake pedal hole for the PCM connector, and a cover plate fashioned to hold the box securely using the recesses molded into the connector body. The result gives a cleaner appearance than the original cover. The fusebox, one relay box, and ignition module are on the unused pedal assembly shelf along with the vapor recovery canister. The other relay box is beside the radiator and in front of the right side coil. The left side coil is just forward of the brake pedal assembly, near the engine compartment ventilation fans which draw air from the engine compartment into the fenderwell area on both sides. A Ford all-electric cruise control unit is mounted on the left frame rail at the front of the engine and the control cable is routed under the intake manifold and up to the throttle-body.
For the fuel supply, a '68 fuel tank was used in the stock location.
This tank does not have a vent so when the vapor recovery system was
reinstalled a fitting was silver soldered to the fuel filler neck to
accommodate it, as well as an additional fitting for fuel return. Fuel
is drawn from the tank in standard fashion by the stock fuel pump and
sent to a small tank in the spare battery compartment, where a 3/16"
return line vents any vapors back to the main tank and has a restriction
to maintain a small level of fuel circulation. With the ignition turned
on you hear about 3 clicks at a time from the Lucas pump which would
seem to be an acceptable amount of flow. Below the tank in the battery
box is a Ford inline fuel injection pump used on the frame rail of the
dual tank F150 pickups. This pressurizes the line to the fuel rail. The
Ford fuel pressure regulator is referenced to manifold pressure and
returns excess fuel to the small tank. With a capacity of 3 quarts this
will provide ample fuel for short blasts at WOT, and at lower power
levels the Lucas should be able to keep up.
I think that about covers the mechanical end of things, except that I retained the heater box and grafted on a bigger fan. I also added a second brake cylinder and inside adjustable balance bar, in keeping with the car's dirt track racing history. The rear axle is stock late MGB with custom 5 lug hubs and the tranny is a close ratio Warner T-50 which is a combination I've been happy with so far. Since I've got a couple of spare axles lying around I'm not too worried about breaking it. I've seen where one person said it was as strong or stronger than an 8" Ford, so I might as well test it some more.
The electrical system is a whole different can of worms (or spaghetti maybe) that I may open later, so let's get back to that sleeper configuration. The blower housing measures 5-1/2" by 7-1/2" and on my car the pulley is above the hood about 6-1/2". There's about 7-1/2" between the base of the blower and the lifter valley cover. By turning the blower over and mounting it in close proximity to that cover the pulley would be brought below the hoodline, leaving some 4" or so above the blower outlet flange. A custom intake is needed but that isn't overly complicated since there wouldn't need to be a water jacket to warm the floor of the plenum chamber. A simple cross ram IR tube manifold with the runners going up and over the blower to a plenum between the runners and the blower outlet would do the trick nicely, and there might even be room for an air/water intercooler. The injectors could fire nearly vertical into the cylinders and all that's left is a thermostat housing and coolant runners at the front of the heads.
I think that about covers the mechanical end of things, except that I retained the heater box and grafted on a bigger fan. I also added a second brake cylinder and inside adjustable balance bar, in keeping with the car's dirt track racing history. The rear axle is stock late MGB with custom 5 lug hubs and the tranny is a close ratio Warner T-50 which is a combination I've been happy with so far. Since I've got a couple of spare axles lying around I'm not too worried about breaking it. I've seen where one person said it was as strong or stronger than an 8" Ford, so I might as well test it some more.
The electrical system is a whole different can of worms (or spaghetti maybe) that I may open later, so let's get back to that sleeper configuration. The blower housing measures 5-1/2" by 7-1/2" and on my car the pulley is above the hood about 6-1/2". There's about 7-1/2" between the base of the blower and the lifter valley cover. By turning the blower over and mounting it in close proximity to that cover the pulley would be brought below the hoodline, leaving some 4" or so above the blower outlet flange. A custom intake is needed but that isn't overly complicated since there wouldn't need to be a water jacket to warm the floor of the plenum chamber. A simple cross ram IR tube manifold with the runners going up and over the blower to a plenum between the runners and the blower outlet would do the trick nicely, and there might even be room for an air/water intercooler. The injectors could fire nearly vertical into the cylinders and all that's left is a thermostat housing and coolant runners at the front of the heads.
Since the blower inlet is to the rear plenty of options exist for
the inlet tract. Having completed the project except for final
debugging and start-up and with the rest of the bodywork yet to be
done I won't undertake that configuration unless someone else is
particularly interested in it though. I've still got plenty to do
to make the car streetable by warm weather.
So that's where it stood as of a month or so ago. Now it runs, and runs well. Getting from there to here would take another article but to hit the highlights, the first try at start up was a complete failure. From that followed a complete recheck of the wiring, with some fairly minor changes, the ignition timing and injector timing were rechecked about a half dozen different ways, and at last with fire to the plugs and injectors and all but 3 minor error codes eliminated, I noticed that only one injector was clicking. A junkyard trip netted another (older) set of injectors and more importantly, another fuel pump, injector rail with regulator, and yes, another relay box. Cobbling it all together I ended up with a cheap but sophisticated device for back flushing and testing injectors, and was in the end able to salvage every one of the original injectors, which was good because they have a better pattern. So I put those back in, using new O-rings, and by lunchtime cranked her up. Success! (And as I always say, nothing succeeds like success.) And what a great success it was.
The engine sounded great. Smooth and powerful, and more responsive than any of these engines I've seen so far. So smooth in fact that I just had to try balancing a quarter on it's edge on the cowl. Did it too! Winging the throttle brought such immediate power that it was almost scary, and the only disappointment in the entire experience was that when I brought the revs up and held it for awhile it cut out. I'll get back to that one and find out why, perhaps something simple, perhaps not. When I immediately let go of the throttle it dropped to a smooth idle without missing a beat, so at this point I'm not too concerned.
That should be eminently fixable. So now it's back to the bodywork. With a little luck and some hard work I just may have it back on the road in another month or two. At the very least I expect to be at the V8 meet with it in August!
Disclaimer: This page was researched and written by Jim Blackwood. Views expressed are those of the author, and are provided without warrantee or guarantee. Apply at your own risk.
So that's where it stood as of a month or so ago. Now it runs, and runs well. Getting from there to here would take another article but to hit the highlights, the first try at start up was a complete failure. From that followed a complete recheck of the wiring, with some fairly minor changes, the ignition timing and injector timing were rechecked about a half dozen different ways, and at last with fire to the plugs and injectors and all but 3 minor error codes eliminated, I noticed that only one injector was clicking. A junkyard trip netted another (older) set of injectors and more importantly, another fuel pump, injector rail with regulator, and yes, another relay box. Cobbling it all together I ended up with a cheap but sophisticated device for back flushing and testing injectors, and was in the end able to salvage every one of the original injectors, which was good because they have a better pattern. So I put those back in, using new O-rings, and by lunchtime cranked her up. Success! (And as I always say, nothing succeeds like success.) And what a great success it was.
The engine sounded great. Smooth and powerful, and more responsive than any of these engines I've seen so far. So smooth in fact that I just had to try balancing a quarter on it's edge on the cowl. Did it too! Winging the throttle brought such immediate power that it was almost scary, and the only disappointment in the entire experience was that when I brought the revs up and held it for awhile it cut out. I'll get back to that one and find out why, perhaps something simple, perhaps not. When I immediately let go of the throttle it dropped to a smooth idle without missing a beat, so at this point I'm not too concerned.
That should be eminently fixable. So now it's back to the bodywork. With a little luck and some hard work I just may have it back on the road in another month or two. At the very least I expect to be at the V8 meet with it in August!
Disclaimer: This page was researched and written by Jim Blackwood. Views expressed are those of the author, and are provided without warrantee or guarantee. Apply at your own risk.
