I had considered waiting until the project was functional and then posting what I did. Instead, I decided to bring you up to date with where I am and continue posting as I make progress. If I waited, I would probably have glossed over the mistakes I made & corrected, and they wouldn’t be available to others wanting to learn from this endeavor. The project revolves around a 1995 Jeep Grand Cherokee ZJ with the 4.0 I-6, NP242 Transfer Case (I like the AWD feature); the automatic tranny failed & was replaced with an NV3500 5-speed. I’ve owned the Jeep for about 16 years now. It was rust-free when purchased, but the salt spray New Jersey uses on the roads in the winter has created a plethora of rust issues. Still, the interior is is very good shape, and the body doesn’t look all that bad.
![Image]()
mpgmike
High Fuel Economy Stroker Build
I had considered waiting until the project was functional and then posting what I did. Instead, I decided to bring you up to date with where I am and continue posting as I make progress. If I waited, I would probably have glossed over the mistakes I made & corrected, and they wouldn’t be available to others wanting to learn from this endeavor. The project revolves around a 1995 Jeep Grand Cherokee ZJ with the 4.0 I-6, NP242 Transfer Case (I like the AWD feature); the automatic tranny failed & was replaced with an NV3500 5-speed. I’ve owned the Jeep for about 16 years now. It was rust-free when purchased, but the salt spray New Jersey uses on the roads in the winter has created a plethora of rust issues. Still, the interior is is very good shape, and the body doesn’t look all that bad.
![Image]()
Project Outline:
High compression engine, 12-hole injectors, “Rock Crawler” Comp Cams 68-232-4 cam, double roller adjustable timing chain (set straight up), SCAT AMC 258 crank (new), ’96 block (with main girdle reinforcement), 2004 TUPY head for smaller ports, distributor (individual coils considered for a later upgrade, but DEFINITELY NOT the factory DIS coil pack set-up), custom ECU (I’m an electronics guy by trade), factory ’95 manifolds (ultimately I want to change over to the later dual-plane style intake), 1995 Jeep Grand Cherokee, NV3500 5-speed swap (already done), NV242 TC (stock), and basically stock axles, gears, and tire size+1 (235 vs 225). When I did the tranny swap, I had to have the front driveshaft shortened and the rear completely rebuilt to lengthen it. The rear driveshaft was upgraded from 2.5” to 3” diameter.
![Image]()
![Image]()
High compression engine, 12-hole injectors, “Rock Crawler” Comp Cams 68-232-4 cam, double roller adjustable timing chain (set straight up), SCAT AMC 258 crank (new), ’96 block (with main girdle reinforcement), 2004 TUPY head for smaller ports, distributor (individual coils considered for a later upgrade, but DEFINITELY NOT the factory DIS coil pack set-up), custom ECU (I’m an electronics guy by trade), factory ’95 manifolds (ultimately I want to change over to the later dual-plane style intake), 1995 Jeep Grand Cherokee, NV3500 5-speed swap (already done), NV242 TC (stock), and basically stock axles, gears, and tire size+1 (235 vs 225). When I did the tranny swap, I had to have the front driveshaft shortened and the rear completely rebuilt to lengthen it. The rear driveshaft was upgraded from 2.5” to 3” diameter.
Cylinder Head:
- 2004 TUPY primarily for the smaller ports. Smaller ports induce higher velocities at lower throttle angle and RPMs. This is important for torque and fuel economy.
- Custom porting; where a “bias” is created to induce swirl. Looking at the picture, it looks like I only ported half of the port and left the other side stock — which is exactly what I did. Notice the ported side is with tangent to the cylinder side. I want the flow to revolve around the circumference/tangent of the cylinder bore, and inhibit flow from and to the center of the bore.
- Massive combustion chamber reshaping, 59.5 cc net. I radically radiused the squish pad area to capitalize on the Coanda Effect.
- “Somender Singh” Grooves are carved into the squish pad to create “shooters” or “jets” to direct high velocity streams of force that induce swirl on the compression stroke, and when the flame front is creating an expanding pressure front, it directs a jet around the ring lands to flush out liquid fuel scraped by the rings on the compression stroke.
- Powre Lynz are like screw threads carved into the intake ports. They induce an exaggerated boundary layer to force liquid fuel and the air away from the port walls. It is similar in effect to the dimples in a golf ball.
- Powre Ringz are Powre Lynz extended to the back sides of the intake valves.
- 3-Angle seats, I/E (not done yet). I know that 5-angle seats flow better, but are less durable. For ultimate performance, I would apply a 3-angle seat to the intakes and either a 5-angle or radiused seat to the exhaust. However, the 3-angle seat is more durable than the other options, and I don’t want to tear this sucker apart again later.
- Milled just enough to make it flat (haven’t done that yet)
- Bronze guide inserts (not done yet)
![Image]()
- 2004 TUPY primarily for the smaller ports. Smaller ports induce higher velocities at lower throttle angle and RPMs. This is important for torque and fuel economy.
- Custom porting; where a “bias” is created to induce swirl. Looking at the picture, it looks like I only ported half of the port and left the other side stock — which is exactly what I did. Notice the ported side is with tangent to the cylinder side. I want the flow to revolve around the circumference/tangent of the cylinder bore, and inhibit flow from and to the center of the bore.
- Massive combustion chamber reshaping, 59.5 cc net. I radically radiused the squish pad area to capitalize on the Coanda Effect.
- “Somender Singh” Grooves are carved into the squish pad to create “shooters” or “jets” to direct high velocity streams of force that induce swirl on the compression stroke, and when the flame front is creating an expanding pressure front, it directs a jet around the ring lands to flush out liquid fuel scraped by the rings on the compression stroke.
- Powre Lynz are like screw threads carved into the intake ports. They induce an exaggerated boundary layer to force liquid fuel and the air away from the port walls. It is similar in effect to the dimples in a golf ball.
- Powre Ringz are Powre Lynz extended to the back sides of the intake valves.
- 3-Angle seats, I/E (not done yet). I know that 5-angle seats flow better, but are less durable. For ultimate performance, I would apply a 3-angle seat to the intakes and either a 5-angle or radiused seat to the exhaust. However, the 3-angle seat is more durable than the other options, and I don’t want to tear this sucker apart again later.
- Milled just enough to make it flat (haven’t done that yet)
- Bronze guide inserts (not done yet)
Custom ECU:
Here is where things get interesting. I have mapped out each pin on the factory ECU, and captured voltages/scope patterns to verify what it is actually doing. Some of the inputs require a pull-up resistor — like the CPS, CMS, VSS… Whereas the stock ECU processor was based on 5 volts, newer more powerful processors are 3.3 volt. The “brain” for the replacement controller is a Microchip dsPIC33CK256MP506; 64-pins, surface mount, 3.3 volt VREF (as opposed to the original 5 volt), lots of memory, lots of timers, other cool features, and FAST!!
I wanted to use an original style header connector so the factory harness would plug right in. They seem to be obsolete — to the point where I cannot even find a part number let alone order one (or more). I considered sacrificing a stock ECU to pirate the header. Instead, I plan on putting the circuit board into a Cinch LE-60 enclosure with a 60-Pin header/connector arrangement. At least the number of pins matches the factory connector. It looks like a factory ECU (not like our Jeep’s, but still), is weather tight, durable, and functional (the Cinch has 2-30 pin connectors to get the 60 pins). I’ve used many Cinch enclosures for past projects and love them. However, it requires re-pinning each of the wires from the factory connector to the new connectors. Important wires can be spliced & split, where a factory connector is retained, but the signals and outputs are sent/received to either the factory ECU or replacement ECU (only connect one or the other). If I have to troubleshoot issues with my replacement, I can just plug in the factory ECU and I’m mobile.
Having a fancy controller only has value if you can data log and tune on-the-fly. I am creating a “dashboard” to interface the replacement controller via USB for tuning and data logging. I’m looking at ways to use Innovate Motorsports’ LogWorks3 for data logging, though there are still a few hurdles to conquer to get it to work for this project. Plan B is to create data logging software from scratch (not looking forward to that, as it will entail several hundreds of hours). Another feature I want to add is to use the J1962 OBD II diagnostic connector to monitor data. This means a crap load of software so I can use my Actron Scan Tool to monitor what is going on without having to pack a laptop along. It will generate OBD II compatible codes to indicate malfunctions/issues, monitor data stream information, facilitate the “clear codes” function, and perhaps other OBD II features. (This will easily consume over 250 additional hours to implement!!! But, I think it is worth it.)
Added Features:
In addition to what the stock ECU controls, there are a couple open pins not doing anything. I want to add electric fan control — not just on/off, but have at least 4 speed control selections using PWM; low at slightly above threshold, half speed at the next threshold, 75% speed at the next threshold, and full-on at the critical threshold. For my personal project I want EGR. I know most of you cannot possibly appreciate how valuable this can be for low-end torque and fuel economy, but look into what George Arlington Moore did with EGR back in the 1920’s. I believe newer vehicle engineers have taken Moore’s lead with how they use EGR. I want to assign other unused pins for generic digital I/O’s or analog inputs for added capabilities.
Here is why you should care:
I intend to make all of this open source. The schematics and PCB boards are being created with free ExpressPCB software. You can download it at no cost. The PCB board is a 4-layer board you can order yourself from ExpressPCB (or we can do group buys to reduce costs). The software will be regular Microchip XC16 C code using MPLABX IDE that are both free to download from Microchip. All parts (BoM) will be made available so you can duplicate everything yourself. And lastly, I am trying to create this project so that it allows for not only the 4.0 ECU replacement, but also V-8 and even 4-Cylinder ECUs. If you own a Chrysler product 1995 or older, this could be YOUR next controller!
Here’s where I’m at:
The block is basically done (more or less). I left a few things unfinished until I’m ready to bolt the head on (like installing lifters). The head still needs a few finishing touches before I send it off to the machine shop for seats, milling, tanking, and guides. I have been simultaneously building the schematic, controller software, and the dashboard. I have yet to start the PCB board, as that requires everything else to be done first. In other words, I spend about 100 hours on ECU replacement engineering, but probably still have 200-350 hours left before I can even start ordering parts.
I encourage comments. Imagine I’m doing engineering for YOUR Jeep. What would you like to see? What might you do differently? Since I’m making it all open source, your comments now could ultimately become YOUR reality later.
![Image]()
Here is where things get interesting. I have mapped out each pin on the factory ECU, and captured voltages/scope patterns to verify what it is actually doing. Some of the inputs require a pull-up resistor — like the CPS, CMS, VSS… Whereas the stock ECU processor was based on 5 volts, newer more powerful processors are 3.3 volt. The “brain” for the replacement controller is a Microchip dsPIC33CK256MP506; 64-pins, surface mount, 3.3 volt VREF (as opposed to the original 5 volt), lots of memory, lots of timers, other cool features, and FAST!!
I wanted to use an original style header connector so the factory harness would plug right in. They seem to be obsolete — to the point where I cannot even find a part number let alone order one (or more). I considered sacrificing a stock ECU to pirate the header. Instead, I plan on putting the circuit board into a Cinch LE-60 enclosure with a 60-Pin header/connector arrangement. At least the number of pins matches the factory connector. It looks like a factory ECU (not like our Jeep’s, but still), is weather tight, durable, and functional (the Cinch has 2-30 pin connectors to get the 60 pins). I’ve used many Cinch enclosures for past projects and love them. However, it requires re-pinning each of the wires from the factory connector to the new connectors. Important wires can be spliced & split, where a factory connector is retained, but the signals and outputs are sent/received to either the factory ECU or replacement ECU (only connect one or the other). If I have to troubleshoot issues with my replacement, I can just plug in the factory ECU and I’m mobile.
Having a fancy controller only has value if you can data log and tune on-the-fly. I am creating a “dashboard” to interface the replacement controller via USB for tuning and data logging. I’m looking at ways to use Innovate Motorsports’ LogWorks3 for data logging, though there are still a few hurdles to conquer to get it to work for this project. Plan B is to create data logging software from scratch (not looking forward to that, as it will entail several hundreds of hours). Another feature I want to add is to use the J1962 OBD II diagnostic connector to monitor data. This means a crap load of software so I can use my Actron Scan Tool to monitor what is going on without having to pack a laptop along. It will generate OBD II compatible codes to indicate malfunctions/issues, monitor data stream information, facilitate the “clear codes” function, and perhaps other OBD II features. (This will easily consume over 250 additional hours to implement!!! But, I think it is worth it.)
Added Features:
In addition to what the stock ECU controls, there are a couple open pins not doing anything. I want to add electric fan control — not just on/off, but have at least 4 speed control selections using PWM; low at slightly above threshold, half speed at the next threshold, 75% speed at the next threshold, and full-on at the critical threshold. For my personal project I want EGR. I know most of you cannot possibly appreciate how valuable this can be for low-end torque and fuel economy, but look into what George Arlington Moore did with EGR back in the 1920’s. I believe newer vehicle engineers have taken Moore’s lead with how they use EGR. I want to assign other unused pins for generic digital I/O’s or analog inputs for added capabilities.
Here is why you should care:
I intend to make all of this open source. The schematics and PCB boards are being created with free ExpressPCB software. You can download it at no cost. The PCB board is a 4-layer board you can order yourself from ExpressPCB (or we can do group buys to reduce costs). The software will be regular Microchip XC16 C code using MPLABX IDE that are both free to download from Microchip. All parts (BoM) will be made available so you can duplicate everything yourself. And lastly, I am trying to create this project so that it allows for not only the 4.0 ECU replacement, but also V-8 and even 4-Cylinder ECUs. If you own a Chrysler product 1995 or older, this could be YOUR next controller!
Here’s where I’m at:
The block is basically done (more or less). I left a few things unfinished until I’m ready to bolt the head on (like installing lifters). The head still needs a few finishing touches before I send it off to the machine shop for seats, milling, tanking, and guides. I have been simultaneously building the schematic, controller software, and the dashboard. I have yet to start the PCB board, as that requires everything else to be done first. In other words, I spend about 100 hours on ECU replacement engineering, but probably still have 200-350 hours left before I can even start ordering parts.
I encourage comments. Imagine I’m doing engineering for YOUR Jeep. What would you like to see? What might you do differently? Since I’m making it all open source, your comments now could ultimately become YOUR reality later.
CAN Protocol:
For OBD II there are 4 different CAN protocols; 11 or 29 Byte packets, and either 250k or 500k Baud. For this project I want to add the 250k 11 Byte protocol. OBD II Scan Tools try one protocol after another until one works. Therefore all Scan Tools will be able to interface with the controller using this protocol. In addition to OBD II Scan Tool interface, having a CAN network means other controllers can be added where the only inputs/outputs could possibly be fed through the CAN network. You don’t need to run an additional wire to the TPS sensor to get throttle angle, since its value can be accessed via the CAN network. CAN is used extensively by the OEMs because it is durable and robust -- it is extremely tolerant to electrical noise. Because it requires a driver, it can interface with a 5 volt or 3.3 volt MCU. This means my 3.3 volt dsPIC processor can talk to a Holley 5 volt processor with no problems.
For my project, I want to add an HHO system, which requires a separate controller. It needs TPS, RPM, and MAP values to function properly. Instead of splicing into wires, I can simply tap the CAN network; where the main processor (already tracking this data) can tell other processors what the TPS/RPM/MAP values are. Also, I would like to implement an Ionic Feedback controller for Ignition Timing Feedback Control capabilities. There simply aren’t enough pins on the main controller to add the Ionic Feedback interface. Thus, it requires a dedicated controller. Some data can be acquired via CAN (TPS & MAP — but CPS & CMS have to be direct wired for the sake of accurate timing), and the results can be sent to the main controller via CAN. Ionic Feedback determines the intensity of the combustion event, and also the peak pressure in Crank Angle Degrees (CAD). Ideally we want peak pressure at 17-18 degrees ATDC (called Critical Crank Angle, or CCA). It can also detect detonation and misfires.
Hypothetically, a boost controller, water injection controller, and other peripherals can be added to the overall system where critical information is transmitted via CAN to reduce wiring, and to reduce the potential for data corruption when too many controllers try to tap the same signal (yes, this happened to me in the past with previous projects). Holley and other manufacturers are adding CAN networks to their electronic controllers. This feature could dramatically expand its ability to add more stuff.
For OBD II there are 4 different CAN protocols; 11 or 29 Byte packets, and either 250k or 500k Baud. For this project I want to add the 250k 11 Byte protocol. OBD II Scan Tools try one protocol after another until one works. Therefore all Scan Tools will be able to interface with the controller using this protocol. In addition to OBD II Scan Tool interface, having a CAN network means other controllers can be added where the only inputs/outputs could possibly be fed through the CAN network. You don’t need to run an additional wire to the TPS sensor to get throttle angle, since its value can be accessed via the CAN network. CAN is used extensively by the OEMs because it is durable and robust -- it is extremely tolerant to electrical noise. Because it requires a driver, it can interface with a 5 volt or 3.3 volt MCU. This means my 3.3 volt dsPIC processor can talk to a Holley 5 volt processor with no problems.
For my project, I want to add an HHO system, which requires a separate controller. It needs TPS, RPM, and MAP values to function properly. Instead of splicing into wires, I can simply tap the CAN network; where the main processor (already tracking this data) can tell other processors what the TPS/RPM/MAP values are. Also, I would like to implement an Ionic Feedback controller for Ignition Timing Feedback Control capabilities. There simply aren’t enough pins on the main controller to add the Ionic Feedback interface. Thus, it requires a dedicated controller. Some data can be acquired via CAN (TPS & MAP — but CPS & CMS have to be direct wired for the sake of accurate timing), and the results can be sent to the main controller via CAN. Ionic Feedback determines the intensity of the combustion event, and also the peak pressure in Crank Angle Degrees (CAD). Ideally we want peak pressure at 17-18 degrees ATDC (called Critical Crank Angle, or CCA). It can also detect detonation and misfires.
Hypothetically, a boost controller, water injection controller, and other peripherals can be added to the overall system where critical information is transmitted via CAN to reduce wiring, and to reduce the potential for data corruption when too many controllers try to tap the same signal (yes, this happened to me in the past with previous projects). Holley and other manufacturers are adding CAN networks to their electronic controllers. This feature could dramatically expand its ability to add more stuff.
I guess a good place to start is with the NV3500 5-speed conversion.
![Image]()
The original AT took a crap. I was faced with 3 options: rebuild the stock tranny, get a junk yard replacement, or convert to 5-speed. Well, I bought the NV3500 tranny from a local shop that claimed it was removed from a 2004(?) TJ because it had a bearing noise. I tore it down somewhat, but failed to see any issues with it. I replaced the seals and buttoned it back up. The worst issue is the 4th gear synchro (grinds going into gear); but there is some bearing noise as well. I bought a wrong bell housing initially (external slave), sand blasted it, powder coated it, then realized it wouldn't work with the overall set-up I chose. Back to the forums, Craigslist, etc. to find one that used the hydraulic TOB to release the clutch.
![Image]()
![Image]()
![Image]()
The tranny did not come with a shifter. The seller said he acquired the tranny just to scarf the shifter for another project. I ended up with a Dodge Dakota shifter & boot (the boot does not fit the ZJ console!). The pedal assembly was sourced from an XJ (Cherokee). The pedals were too close together for a normal human foot to fit, so I had to heat them with a torch to spread them apart. The assembly didn't bolt up to the ZJ body quite right. This meant welding additional brackets and supports to mount to the existing ZJ body points.
![Image]()
![Image]()
![Image]()
The 5-speed was a bit shorter than the AT it replaced. This meant having the front driveshaft shortened, and the rear lengthened. Speaking strictly physics, it's easier to shorten a stout piece of steel than lengthen it. The shop suggested just rebuilding the rear driveshaft from scratch. I agreed to upgrade from the stock 2.5" diameter to a beefier 3" diameter shaft for the rear; as I intended to tow with it.
![Image]()
![Image]()
The original AT took a crap. I was faced with 3 options: rebuild the stock tranny, get a junk yard replacement, or convert to 5-speed. Well, I bought the NV3500 tranny from a local shop that claimed it was removed from a 2004(?) TJ because it had a bearing noise. I tore it down somewhat, but failed to see any issues with it. I replaced the seals and buttoned it back up. The worst issue is the 4th gear synchro (grinds going into gear); but there is some bearing noise as well. I bought a wrong bell housing initially (external slave), sand blasted it, powder coated it, then realized it wouldn't work with the overall set-up I chose. Back to the forums, Craigslist, etc. to find one that used the hydraulic TOB to release the clutch.
The tranny did not come with a shifter. The seller said he acquired the tranny just to scarf the shifter for another project. I ended up with a Dodge Dakota shifter & boot (the boot does not fit the ZJ console!). The pedal assembly was sourced from an XJ (Cherokee). The pedals were too close together for a normal human foot to fit, so I had to heat them with a torch to spread them apart. The assembly didn't bolt up to the ZJ body quite right. This meant welding additional brackets and supports to mount to the existing ZJ body points.
The 5-speed was a bit shorter than the AT it replaced. This meant having the front driveshaft shortened, and the rear lengthened. Speaking strictly physics, it's easier to shorten a stout piece of steel than lengthen it. The shop suggested just rebuilding the rear driveshaft from scratch. I agreed to upgrade from the stock 2.5" diameter to a beefier 3" diameter shaft for the rear; as I intended to tow with it.
Still working on getting all the Powre Lynz carved into the head so I can send it off to the machine shop. It takes lots of time, and my hands & wrists can only take 2-3 ports before giving out (I'm 56 now).
On another note, I added a tow hitch. When I did, I added a 7-pin and 4-pin plug combo. The Jeep has separate lights for brakes and turn signal, but most trailers do not. Therefore (being the electronics guy), I slapped an XOR circuit together so if either the brakes OR turnsignal bulb lights, the trailer light comes on. However, if both brakes AND turn signal come on, it does not. Thus the XOR Logic Circuit.
![Image]()
I tapped both the brake and turn signal wires from both sides then sent the result to the trailer plug. I understand there might be commercially available solutions, but... it works!
On another note, I added a tow hitch. When I did, I added a 7-pin and 4-pin plug combo. The Jeep has separate lights for brakes and turn signal, but most trailers do not. Therefore (being the electronics guy), I slapped an XOR circuit together so if either the brakes OR turnsignal bulb lights, the trailer light comes on. However, if both brakes AND turn signal come on, it does not. Thus the XOR Logic Circuit.
I tapped both the brake and turn signal wires from both sides then sent the result to the trailer plug. I understand there might be commercially available solutions, but... it works!
The exhaust manifold currently in the ZJ is cracked where the front trio meet the rear. I have another manifold in good condition. I wrapped it with a fiberglass header wrap. The idea is to keep under-hood temps lower by reducing radiated heat from the manifold; keep exhaust velocities high by reducing cooling contraction, promoting better cylinder evacuation; and to keep the catalytic converter hotter for better emissions. Lower engine compartment heat means better performance (lower intake air temps as a result), plus belts and hoses (and probably other parts) last longer. This will be installed when the new stroker goes in. When I swap the engines, I'll also wrap the pipe from the exhaust manifold to the cat, again for more efficient catalyst function and higher velocities..
![Image]()
Referencing the image in Post #4, David Vizard and others tend to bias flow to the center of the port for max VE. I'm biasing flow to the tangent of the port for max swirl and homogenization. The added swirl slows down cylinder filling -- not the best for max WOT power -- but the additional swirl/homogenization works in favor of lower throttle torque and fuel efficiency.
With that said, I have used epoxy to fill intake ports in the past. I used JB Weld 2-part epoxies either the liquid, or the putty types. David Vizard recommended Goodson's PRK-99 2-part epoxy for port filling. I'm seriously considering port filling, and leaning towards the PRK-99. Some of my previous work:
![Image]()
![Image]()
![Image]()
With that said, I have used epoxy to fill intake ports in the past. I used JB Weld 2-part epoxies either the liquid, or the putty types. David Vizard recommended Goodson's PRK-99 2-part epoxy for port filling. I'm seriously considering port filling, and leaning towards the PRK-99. Some of my previous work:
I invested about what you could buy a decent XJ for (running, 1/2 tread on the tires, and a 1/4 tank of gas) into a small genset (about $3g). It sports EFI, ignition coil, crank sensor, cam sensor, MAP sensor, as well HHO, EGR, and a few other options. I figured it would be easier to debug functions intended for the XJ on the Genset, then transfer successful results to the Grand Cherokee.
I got 16 pitch Powre Lynz carved into the roof and floor of the intake ports. With the cold January weather, and no heat in the garage, I finally got a day warm enough to fire up the compressor. I had to put the grinder on a heater in the kitchen periodically to thaw it out. I tried to carve 12 pitch Lynz in the center-of-the-cylinder side of the port(s), but my Tool is dull, and can't get the job done. I'm thinking about what to do about that.
![Image]()
![Image]()
![Image]()
i don't have a clue what you are talking about with most of the computer function stuff, but its really interesting. if this is for gas milage, why no computer controlled shifting on an auto? have you read Smokey Yunicks books on port design and old man Honda's book. also the stuff Oldsmobile did to their engines in the 70's that made them able to pass emissions without a cat.
i forgot.(happens a lot i'm old). ceramic your exhaust inside and out, any powder coat place can do it. brings the radiant temp way down. we found this out go kart racing. little kids come in the pits, exhaust is right at their grab height. my powder coat gut talked me into the inside part, besides the outside. worked great, lot less screaming kids.
9
9
baldfatdad, you may have spent as many hours researching the greats as I have. Inevitably, you studied geniuses I've never heard of, and I spent my time reviewing the works of folks you never heard of. It is probably safe to say we have different backgrounds. That really says that what I'm doing is based on my education, and your interpretation of that is based on your education -- which differs.
I have looked into George Arlington Moore, Smokey Yunick, Vic Edlebrock, Somender Singh, Larry Widmer, and countless others. What really strengthens my background is the fact that I ran a speed shop -- Powre Haus -- for a couple years. I would get an idea, present it to a customer that would bite (at a discounted price), then after doing the work to try it out, I would get feedback. Ultimately I got results that totally debunked stupid ideas, and really incredible results that reinforced others.
What I am presenting here for this Jeep 4.0-to-4.6 Stroker package is based on what I have already done successfully. I toyed with ceramic coatings -- both thermal and friction -- and did find some benefits in specific situations, but usually not worth the cost and effort; with a few exceptions. For what I'm doing, I see no potential benefit from ceramic coatings.
As for the "computer controlled auto", I converted to NV3500 5-speed when the original auto trans crapped out on me. It cost a bit more than a junk yard replacement, way less than a quality rebuild on the auto, but i really wanted a manual trans all along.
It may interest you to peruse through the fuel economy information freely shared on www.Ecoceptor.com/MPGenie.
As for the electronic controls I will be designing & implementing on this project, I do electronics for a living. It is a personal goal of mine to design & build a controller that can replace the factory ECU -- and do EVERYTHING better! I consider it something like a PhD Thesis project. I could probably use a MegaSquirt, MOTEC, or other stand-alone, but it just wouldn't be the same
I have looked into George Arlington Moore, Smokey Yunick, Vic Edlebrock, Somender Singh, Larry Widmer, and countless others. What really strengthens my background is the fact that I ran a speed shop -- Powre Haus -- for a couple years. I would get an idea, present it to a customer that would bite (at a discounted price), then after doing the work to try it out, I would get feedback. Ultimately I got results that totally debunked stupid ideas, and really incredible results that reinforced others.
What I am presenting here for this Jeep 4.0-to-4.6 Stroker package is based on what I have already done successfully. I toyed with ceramic coatings -- both thermal and friction -- and did find some benefits in specific situations, but usually not worth the cost and effort; with a few exceptions. For what I'm doing, I see no potential benefit from ceramic coatings.
As for the "computer controlled auto", I converted to NV3500 5-speed when the original auto trans crapped out on me. It cost a bit more than a junk yard replacement, way less than a quality rebuild on the auto, but i really wanted a manual trans all along.
It may interest you to peruse through the fuel economy information freely shared on www.Ecoceptor.com/MPGenie.
As for the electronic controls I will be designing & implementing on this project, I do electronics for a living. It is a personal goal of mine to design & build a controller that can replace the factory ECU -- and do EVERYTHING better! I consider it something like a PhD Thesis project. I could probably use a MegaSquirt, MOTEC, or other stand-alone, but it just wouldn't be the same
I got 16 pitch Powre Lynz added to the short side radius. I ran them from 5/7 o'Clock position on the cylinder wall side, and 3/9 o'Clock position on the center-of-the-cylinder side. That will push the velocity stream more towards the tangent of the cylinder -- inducing higher swirl -- and catch & break up any liquids hugging the SSR.
![Image]()
![Image]()
Had a rather nice day yesterday, and I managed to cut Powre Ringz into all 12 valves. Intake valves get the Powre Ringz on the port side of the valves, while exhausts get CC side Ringz. Liquids will centrifugally find their way to the back side of the intake valves. It's still hot from the previous combustion event, and adding the texture helps break up liquids and readmit them into suspension. The hottest part of the combustion chamber is the exhaust valve. Adding the textures here helps to pre-ionize the charge for faster combustion. I used the Dremel with a cutting disc and rotated the valves in my lathe. For more info on the concept, look up ST Valves from the Metric Mechanic. Also, here is a link to an Allpar article I wrote that touches on the topic.
![Image]()
FINALLY got the head done! I tried sharpening my 12-pitch Lynz tool, but it still couldn't carve decent Lynz in the cast iron. So, I went with 16-pitch on 3 of the intake port walls, and 20-pitch on the tangent side. While the compressor was fired up, I added 16-pitch Lynz to the sides of the intake manifold ports, only about 1" back in or so, just enough to direct the air around the nozzle of the injector. On the floor of the intake manifold ports, the 12-pitch was able to so a reasonable job with the aluminum.
Called a local NJ machine shop I'd used in the past. Got a price on bronze guides, 3-angle seats, a shave, and a tank cleaning... $700!!!! (plus sales tax!) I called HP Engines in Thompsontown, PA that I've used for over 30 years, and their price was a much more reasonable $300! That also includes a magnaflux, as the Jeep 0331 heads are prone to cracking (even though this is the newer TUPY head, which is more robust, still...). Time to plan 2 road trips; one to drop the head off, and one to pick it up when finished.
![Image]()
![Image]()
![Image]()
Called a local NJ machine shop I'd used in the past. Got a price on bronze guides, 3-angle seats, a shave, and a tank cleaning... $700!!!! (plus sales tax!) I called HP Engines in Thompsontown, PA that I've used for over 30 years, and their price was a much more reasonable $300! That also includes a magnaflux, as the Jeep 0331 heads are prone to cracking (even though this is the newer TUPY head, which is more robust, still...). Time to plan 2 road trips; one to drop the head off, and one to pick it up when finished.
Decided to get the block ready for the head. Step 1 was to assemble the new oil pump and pick-up, then install it. Done.
![Image]()
I still hadn't degreed the cam. I broke out the degree wheel and dial indicators (yes, 2). I wanted to spin the engine around a couple times just to make sure nothing happened since putting it together -- like 4 years ago! What do you know, it spun a couple degrees then locked!! Reversed direction, went several degrees and locked again! Crap!
![Image]()
Pulled the oil pan and started looking for something hitting. Turns out the connecting rods were hitting the girdle. Had I known that in advance, I would have ground away the boss on the CR caps. Next thought was to possibly delete the girdle. I hated to do that since one of the things I like about the 1996+ blocks is that girdle. The next thought was to use washers to space the girdle away from the main caps to gain clearance. Immediately, I embarked on a trip to the local Tractor Supply for some 3/8" washers. Slapped in some grade 8 washers, 2 per stud. Score! Rods now cleared. I grabbed the oil pan and sat it back on the block -- only to discover the girdle was keeping the oil pan from touching the block. Crap Again!!
![Image]()
Now it was time to measure washer thickness and sort through the bag I bought (by the pound from Tractor Supply) and try to get enough clearance without excess. It took some time, but I have sufficient clearance between the girdle and the rods, and I was able to get the oil pan bolted down. (Didn't grab a picture of that, sorry.)
Back to the original goal, degree the cam. The Comp Cams cam card calls for an intake centerline at 110 degrees ATDC. I got a measured 109 degrees. That's 1 degree advanced. I'll live with that. This gives me a little reserve for when the double roller starts to wear out.
![Image]()
The bottom end is together and ready for lifters and a cylinder head.
![Image]()
I still hadn't degreed the cam. I broke out the degree wheel and dial indicators (yes, 2). I wanted to spin the engine around a couple times just to make sure nothing happened since putting it together -- like 4 years ago! What do you know, it spun a couple degrees then locked!! Reversed direction, went several degrees and locked again! Crap!
Pulled the oil pan and started looking for something hitting. Turns out the connecting rods were hitting the girdle. Had I known that in advance, I would have ground away the boss on the CR caps. Next thought was to possibly delete the girdle. I hated to do that since one of the things I like about the 1996+ blocks is that girdle. The next thought was to use washers to space the girdle away from the main caps to gain clearance. Immediately, I embarked on a trip to the local Tractor Supply for some 3/8" washers. Slapped in some grade 8 washers, 2 per stud. Score! Rods now cleared. I grabbed the oil pan and sat it back on the block -- only to discover the girdle was keeping the oil pan from touching the block. Crap Again!!
Now it was time to measure washer thickness and sort through the bag I bought (by the pound from Tractor Supply) and try to get enough clearance without excess. It took some time, but I have sufficient clearance between the girdle and the rods, and I was able to get the oil pan bolted down. (Didn't grab a picture of that, sorry.)
Back to the original goal, degree the cam. The Comp Cams cam card calls for an intake centerline at 110 degrees ATDC. I got a measured 109 degrees. That's 1 degree advanced. I'll live with that. This gives me a little reserve for when the double roller starts to wear out.
The bottom end is together and ready for lifters and a cylinder head.
I've been working on the ECU replacement software, a little here, a bit there, mostly in the evenings. So far I have the tables for Fuel IPW and Ignition Timing (shown below) as far as I can take them for now. I'm currently working on the AFR Table. Visual Studio is extremely memory intensive, and it takes awhile for the computer to process changes and additions.
![Image]()
Perhaps some further explanation is in order. The ECU replacement Controller processor (Microchip dsPIC33EV256GM106 64-pins) works from the Tables to control Ignition Spark Timing, Fuel Injector IPW, a feedback system targeting a programmed AFR, and many other Main & Compensation Tables. There is an external EEPROM (currently planning on using a Microchip AT24C64D, but may move to a bigger part number if Tables and CONFIGurations exceed memory) on the PCB to store non-volatile (Table) values. The screen in the previous post is what I would see on the PC while connected to the ECU replacement Controller (via USB) tuning Ignition Timing.
When you buy a stand-alone MCU like MegaSquirt, Accel DFI, MOTEC, or other; these tables are what you program to get your engine running well. Before you got your new ECU with the software package, someone had to program the ability for you to load the Tables. This is what I'm working on. After I consider it finished, I then must load the software onto a controller, wire it in, and then load the Tables while tuning using a laptop.
Working concurrently with the PC's app (like the Ignition Timing Table in the previous post) the actual ECU replacement has to do something with the values entered into each and every X-Y Table cell to control fuel injector IPW, ignition coil timing, and many other functions. That is performed by the ECU Processor (the dsPIC33EV256GM106). It actually controls ignition coil timing, injectors, EVAP Sol, coolant fan, gauge cluster tachometer, idle air valve, cruise control, air conditioning, and many other functions.
Creating my own version of a stand-alone ECU to replace the factory 1995 Jeep ZJ 4.0 ECM requires a microcontroller (the dsPIC33EV256GM106 processor) with all of the software, a functional PCB with all the capacitors, resistors, MOSFETs, and other hardware packaged in a weather-proof enclosure (I will probably be using a Cinch enclosure). To interface the ECU replacement with a PC requires an app I must custom design with all the Tables and CONFIGurations to get the replacement ECM working to my specs. The previous post merely indicates where I am in this development process (still much to do).
When you buy a stand-alone MCU like MegaSquirt, Accel DFI, MOTEC, or other; these tables are what you program to get your engine running well. Before you got your new ECU with the software package, someone had to program the ability for you to load the Tables. This is what I'm working on. After I consider it finished, I then must load the software onto a controller, wire it in, and then load the Tables while tuning using a laptop.
Working concurrently with the PC's app (like the Ignition Timing Table in the previous post) the actual ECU replacement has to do something with the values entered into each and every X-Y Table cell to control fuel injector IPW, ignition coil timing, and many other functions. That is performed by the ECU Processor (the dsPIC33EV256GM106). It actually controls ignition coil timing, injectors, EVAP Sol, coolant fan, gauge cluster tachometer, idle air valve, cruise control, air conditioning, and many other functions.
Creating my own version of a stand-alone ECU to replace the factory 1995 Jeep ZJ 4.0 ECM requires a microcontroller (the dsPIC33EV256GM106 processor) with all of the software, a functional PCB with all the capacitors, resistors, MOSFETs, and other hardware packaged in a weather-proof enclosure (I will probably be using a Cinch enclosure). To interface the ECU replacement with a PC requires an app I must custom design with all the Tables and CONFIGurations to get the replacement ECM working to my specs. The previous post merely indicates where I am in this development process (still much to do).
Reviewing 2 pages of this thread, I have been tracking my progress post after post. Thus far, there have been 10 likes and 2 other posts (thank you baldfatdad). That feedback would suggest a rather milk-toast level of interest. However, the views are over 2000! This is my build thread -- period. I am not copying on any other forum. I would ask that if you have found something of value, or something that intrigued you, go back to that post and hit "Like". If you have thoughts or questions, make a post asking the question or making your comment. I feel like I'm on a deserted island singing to the palm trees!
I have rented for over 16 years where I live. The matriarch passed away and the property will be placed on the market. I was supposed to be out yesterday, but am still struggling to get everything moved into storage. I got notice the head was done. A bit late to get the new stroker engine installed, but here are some assembly pics.
![Image]()
![Image]()
![Image]()
I will have to wait until I get moved & settled in to continue on. Chat in a month or three.
I will have to wait until I get moved & settled in to continue on. Chat in a month or three.
Since this thread just got bumped, the project is on hold. I am still in transition, moving from New Jersey to Oklahoma. Still waiting on the house to close, so I'm staying in a travel trailer with my family. All my tools (and the engine) are in storage.
On a side note, the front drive shaft has some sort of ball joint in the compound U-joint coupler that dried out and started kicking rusty dust -- and making obnoxious noises. Currently the ZJ is 2WD. Looking for a clean (read that rust free) southern body to swap all the good parts on to. I'm amazed how New Jersey rusted the poor thing worse than anywhere I've ever lived.
On a side note, the front drive shaft has some sort of ball joint in the compound U-joint coupler that dried out and started kicking rusty dust -- and making obnoxious noises. Currently the ZJ is 2WD. Looking for a clean (read that rust free) southern body to swap all the good parts on to. I'm amazed how New Jersey rusted the poor thing worse than anywhere I've ever lived.
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
