Modern-day Miura

Randy V

Staff member
Lifetime Supporter
The adapter plate approach hasn't been ruled out yet, it may still prove to be the better way to go. I've learned over time that until something complicated like this is mocked up, you just don't have enough detail and perspective to make an informed decision. R&D work can be tedious, slow and duplicative but sometimes required to get to the more optimal answer.

I would think a steel wedge shaped adapter would work better than trying to get tubes in there that tight. I say this from past experiences in trying to keep the header tubes from cracking due to excessive heat right at the mounting flange. Make those short radius bends in the steel of the adapter. You can work the radius of the bend easier with a burr. Also, if necessary, you could open up the top of the cylinder head’s port a few mm.
I would think a steel wedge shaped adapter would work better than trying to get tubes in there that tight. I say this from past experiences in trying to keep the header tubes from cracking due to excessive heat right at the mounting flange. Make those short radius bends in the steel of the adapter. You can work the radius of the bend easier with a burr. Also, if necessary, you could open up the top of the cylinder head’s port a few mm.

Interesting, can you give more details on the cracking issue? For example, was the header material mild steel or SS? How tight of a tube bend radius did you have? Did the cracks occur in the joint between flange and primary or were they just in the primary tube? I ask these questions as I'm using 1 3/4" with 2" radius bend as the first primary tubes in my mockup. I'm using mild steel for the mockup but plan to use 304SS for the actual headers. I'd like to make sure I'm doing "valid" mockup that don't have an inherent design flaw in them.

Hmmm...when you said, "steel wedge shaped adapter", this triggered an idea in my head. How about if a pair of 3/8" thick header flanges were welded together to make an adapter plate of sorts? That would give a 3/4" plate which should be thick enough for a decent amount of short radius bend. One of the things that has given me pause about the adapter plate alternative is all the machine work it would require if the starting point was a billet of aluminum. Another aspect has been that if the adapter were to be bolted in place prior to the transaxle, then it would require having the header flange fasteners in new locations and thus some sort of custom header flanges and gaskets.

Using off the shelf CNC cut header flanges as the starting point would avoid the vast majority of machine work and enable the use of the existing studs and standard Coyote exhaust gaskets. One negative is that it isn't a separate adapter and thus couldn't be bolted on prior the the transaxle. It would need to have provision for installation that includes sliding downward over the transaxle shift mechanism. It looks like this type installation should be possible from my current header flange mockup by having slots (instead of holes) for the bottom studs and screw in the top studs after the header is in place. So this alternative is really a hybrid of the other two alternatives.

One other thing Randy touched on is porting the head to raise the port. I can see where this would really help to smooth the flow but I'm really, really, really trying not to open up the engine here. It's a very low mileage engine and being DOHC means head removal is at least twice as complicated as a push rod engine. It seems like there must be a way to securely block the exhaust ports such that aluminum particles are kept from getting in the cylinders.

Question: Has anyone successfully ground away exhaust port material with the heads on the engine and without getting grinding shavings in the cylinders? If so, what means did you use to block the port? Being a 4 valve head, the exhaust ports are fairly short in that I can touch the exhaust valves with a finger. So the blocking of the port would be crucial to keeping all shavings out.

Amazing how the light bulbs can go off in your head with a little bit of good idea stimuli. Please keep it coming!
Glue a "plug" of polycarb in the port with RTV, thread a hole in the poly and install a screw. After the porting, remove the screw, insert a longer screw and pull it out. Any remaining RTV can be scraped off and vacuumed out.
Glue a "plug" of polycarb in the port with RTV, thread a hole in the poly and install a screw. After the porting, remove the screw, insert a longer screw and pull it out. Any remaining RTV can be scraped off and vacuumed out.

Excellent approach!! I have plenty of aluminum sheet scraps around the garage that could serve as the "removable plug". A seal of RTV should fully take a compressed air blast from shaving cleanup. I'm guessing any RTV flakes that might make their way into cylinder after plug removal should be harmless to internal engine components as they burn up.
Header Mockup

I spent a day starting the mockup of an engine right side header where there is the transaxle clearance challenge. The header flange is the old rusty one cut from the factory header and I used 1 ¾ mandrel tube bends for the mockup. Primary tubes on ports #1, 2 and 3 (right to left in first picture) are 2” CLR bends which is the tightest available for this size tube. The aluminum tube laid horizontal from the shift shaft is simulating the extension I plan to fabricate to keep shift cables away from the hot header zone.




This view provides a good perspective on how the primary tubes can be routed to clear the shift mechanism and still get under the rear chassis cross member.



There is 1/8” clearance between the #3 header tube and shift housing and ¼” between #2 header tube and shift shaft. So there is clearance, but not much, so radiant heat transfer will be great. I estimate that the top of shift mechanism/housing can be lowered by 3/16” by machining it from a single piece of billet instead of the current combination of cast aluminum and adapter plate. That would give 5/16” clearance which is better but I’d like to get more like ½” air space for clearance.

I guess it’s time to mockup the double thick header flange alternative to see if it results in more clearance. Fortunately, the right and left Coyote header flanges are mirror images of one another so I just need to cut up the left side factory header to get a second flange for the mockup.
Header Design Progress

This is a progress update on the Miura engine headers.

I’ve been working with Vince Roman at Burns Stainless on header design options and specifications. Burns offers free header design services if you buy the various header pieces from them. The primary objective on the Miura header design is to make the Coyote V8 sound as close to a V12 as possible and feasible. A 180 degree header design (primary tubes mixed across engine banks) would do that but it just isn’t feasible given the transverse engine orientation and configuration. So the next best design option to achieve the v12 sound is 4 into 1 headers followed by Y pipe with all primary tubes equal length along with equal length secondary tubes into Y pipe.

The spec Vince worked up on their header design software is 28.5” tapered primary tubes starting at 1 5/8” going to 1 ¾” and then up to 1 7/8” prior to a merged collector. If the left side header is routed under the oil pan, it looks like there’s barely enough room to fit headers of this specification and keep them equal length. It will require a “flat collector” instead of the standard stacked collector design but that should be all right.

As to the right side header and maximizing the transaxle shift mechanism clearance, Vince confirmed that starting a short radius bend in the flange is a viable design and it’s used in some racing engines that also have tight clearance issues. He confirmed they can provide a ¾” thick header flange with 1 ¼” pilot holes for the ports. This way there will be sufficient metal in the flange to cut in the bends and have the primary tubes exit at an angle to the flange instead of perpendicular. It will likely be time consuming to grind out the flange for a good flow match but this appears to be the best way to maximize the clearance over the shift mechanism without compromising exhaust flow. In addition, I’m having the shift mechanism made up from a single piece of aluminum billet which should shorten it by 3/16”. Between these two measures, it looks like a clearance of something like ½” will be possible where prior it was 1/8”.
AC Compressor & Engine Mount Progress

When it comes to mounting the engine/transaxle package in my Miura, it’s coming down to fractions of inches, literally fractions of inches. The critical spot is between the upper suspension joints, specifically the heim joints at the front of upper A arm. To make room for the AC compressor, the engine/transaxle needed to be moved 3/8” to the drivers side from where I originally intended to place it. Fortunately I hadn’t yet fabricated the engine mounts so it’s become a matter of making room for the AC compressor and then building out engine mounts for that placement.


My first choice was the Ford AC compressor that came with the Coyote engine as it mounts directly to the engine block with 3 long M8 bolts; nice, simple and clean. I couldn’t make it work though as it has some sort of hub that sticks out about ¼” as part of the electro-mechanical clutch and this hits the suspension mount prior to all 3 mounting bolts lining up. I had already moved the engine/transaxle as far to the drivers side as possible.


There’s 1/8” clearance between the transaxle case and the heim joint and that’s as close as I’m comfortable running them. Like I said, fractions of inches. I’ll shave a bit more metal off the chassis so there will be the same 1/8” clearance there as well. So plan B is to use the Sanden 508 AC compressor that came as part of the Old Air aftermarket AC system. While the Sanden unit is a nicer piece from an eye candy perspective (it’s polished to a nice shine), nobody makes the hardware to mount it to a Coyote engine. Oh well, in for a penny, in for a pound. I’ll have to spend a couple of days fabricating mounting brackets for it.




After building part of the mounting bracket and lining up the serpentine belt with the crank pulley, there’s 3/16” clearance, just enough to sneak a belt through and hopefully with good engine mounts enough to keep the AC clutch from hitting the chassis. I knew clearances would be close but I didn’t envision they’d be this close.

Ok, onto engine/transaxle mounts. The primary requirement for the mounts is to control the potential lateral movement for the engine/transaxle with a secondary requirement to dampen engine vibrations. So I decided to use “biscuit style” mounts that are secured with a 7/16” bolt through the middle. Ideally, the engine/transaxle package could be suspended at 3 mounting points, but I decided 4 points would be better given the tight clearances. I started building mounts at the back side of the final drive.



I started with a scrap piece of 6” by 4” by ½” rectangular tubing and with some machine work, it slips snuggly over the final drive case and through bolts can be used to fasten in place. I’m thinking 3 bolts are better than just 2 so I’ll add some ears to the bottom to pick up a 3rd bolt hole. As often happens when starting from metal scrap pieces, it has a couple of extra holes in it, maybe they will come in handy for some sort of mounting bracket. I’ve ordered up some more metal and special bolts to complete the engine mounts.

More to come…
Engine Mount Progress

I’ve made more progress on fabricating engine mounts with 2 of 4 now complete.

On the mount attached to the final drive case, the metal cut away from the main part is used to extend the mount downward for the 3rd bolt attachment. The metal was cut to shape, a ¼” bevel made along the weld zone, and a big fat bead laid down. I used a ½” thick steel backer plate during welding to minimize the warpage from shrink in the weld bead.


After metal finishing the weld, some extra metal was cut away in a curved cut to match curves in the final drive case. The surface was cleaned up with a belt sander and 100 grit sanding belt leaving a satiny “grained” finish.



The next mount to be fabricated was the drivers side front. I started with a scrap of 3/8” thick aluminum angle. I was able to cut it to pickup 3 bolts on the bellhousing and this time all the extraneous notches and holes in the scrap went away with the cut. After welding on a foot and a gusset, it was drilled with lightening holes to remove some weight and belt sanded to get the same finish as the other mount. The short piece of channel the biscuit mount sits on still needs to be welded to the chassis but I’ll wait to do that until all the engine mounts are completed.



2 mounts down, 2 to go. For some reason I always think fabricating stuff like this should go quickly but they really do take quite a bit of time to make.
Engine Mount Completion

The engine mount fabrication is now complete; all that remains is to weld the mount pedestals to the frame rails. I’ll need to pull the engine out to get access for the welding so I’m going to try to complete more things that need the engine in place before pulling it. The front passenger side mount was the hardest to make as it is wedged between the oil filter and oil pan. I probably made more than 25 trips under the car on a creeper to get it sorted out.



I would have preferred to have only 3 engine/transaxle mounts instead of 4 but I was worried that the transaxle would contact the chassis and the 4th mount is to ensure that doesn’t happen. The transfer case part of the transaxle has about 3/16” clearance above the lower A arm pivot. The cedar door shim in this picture was used to set the clearance while the engine mounts were being fabricated.



It’s back to the transaxle shift mechanism and header mock-ups now.
Progress on Transaxle Shift to Header Clearance

While I was busy fabricating engine mounts, I put out a request to Pete Aardema to make a new billet, one piece shift mechanism with the objective to increase header clearance as compared to the two piece unit I had cobbled together. As a reminder, I had paired up the cast aluminum shift housing from a Tremec 6060 transmission with an adapter plate so it could mount on a Tremec TKO type housing. It resulted in about 1/8” clearance with the header mockup. Here's a picture of the two piece unit.


Pete’s machinist, Kevin Braun, machined out this very nice, one piece shift mechanism that uses the same basic internal components scavenged from the old unit.


The clearance is now about 3/8 to ½” with the same header mockup. Kevin carved it from a piece of 1 ½” aluminum billet on a CNC mill. While we were at it, the unit was upgraded so the shift shaft is now carried on ball bearings instead of bushings. This should help it to keep from tightening up when the inevitable heat soaks into it.

In addition, I’m adjusting the header design to also increase the clearance. More to come on the header re-design as I redo the header mockup for it.
Chassis Member Reinforcement

Now that I have cut out the chassis next to AC compressor and belt to get sufficient clearance, I decided it would be wise to add some metal back for reinforcement. I don’t know that this reinforcement was strictly needed but I thought it better safe than sorry. Here’s a picture of the vertical member that spans between the lower and upper A arm pivots after it was fully trimmed for the AC compressor clearance.



The only real option is to add the reinforcement to the outside of the frame rails. I had already installed a similar reinforcement on the drivers side so I had a good sense for how to go about it. Fortunately, I had plenty of ¼” aluminum plate still on hand to make this odd shaped piece.


Given the current heat wave here in N. CA, I decide to glue this reinforcement in place during the early morning cool to give me enough time to fully spread the glue over both surfaces and get it fully clamped in place prior to expiration of the 15 minute glue work time. One more loose end checked off the Miura project list.
Header Design/Mockup

Header fabrication is now underway, well at least the first stage which is full mockup fabrication to determine header tube routing optimized to get to an equal length design. The specification calls for 28 ½”primary tubes that taper from 1 5/8” to 1 7/8” over their length to 4 into 1 collectors and then 3” secondary tubes to a “Y” so all 8 exhaust pulses are mixed into to a single 4” pipe prior to routing to a muffler. The primary objective for the exhaust system design is to mimic the sound of a V12 as closely as possible even though it’s produced from a Coyote V8.

The transverse engine placement is forcing me to be very creative with primary tube routing because I can’t just duplicate tried and true designs from similar situations. There just aren’t enough mid-engine transverse V8 cars out there to have much to draw ideas from. I did get the ¾” thick header flange for use on the aft bank (aka passenger side) where there’s only minimal clearance with the transaxle shift mechanism.


I transcribed the ports and mount holes onto a ¾” thick piece of wood so I could work out how best to maximize transaxle clearance while still getting good flow into the headers. I decided to have primary tubes attach to the flange at a 50 degree angle so the tubes are basically horizontal as they pass over the transaxle. The main challenge I ran into is lack of clearance to get bolts onto the upper mounting studs. So the porting in the flanges needs to be angled away from the upper studs in addition to angling upward. I had this worked out by the 3rd from left port in my wooden flange.


I had Burns Stainless fabricate the merged collectors for these headers. The aft header uses a traditional stacked collector design and the fore header requires a flat collector design. The fore header is to be routed under the engine and the flat collector is positioned under the short part of the oil pan. I chose this under engine routing as it’s the only chance at making the secondary pipes anywhere near equal length. These collectors almost look like jewelry with perfect, tiny TIG welds on them.


I chose to do mockup headers in carbon steel using cheap Chinese made mandrel bends. Once I had the port spacing and primary tube to flange angle worked out, it was a bunch of cutting, tack welding, measuring, etc. etc. I found magnets removed from old computer hard drives to be very helpful in holding tubes temporarily in place for fitting and tack welding. These magnets are very strong for their small size.


There is a very small envelope of space for this header. It goes over the top of the transaxle, dives down under the rear chassis cross member, proceeds over the top of the final drive case and curving out a small hole at chassis rear. All this while bringing in each primary tube at 28 ½” length, well close anyway as I got them all within ¼” of each other and can see how to easily get them exact when I build the actual header.



And here it is with the chassis cross member back in place.



I do think some notches might be needed in the cross member for primary tube clearance. I’m not sure yet because I chose to use 1 ¾” primary tubes throughout for the mockups where the actual headers will use 1 5/8” tube there going up to 1 ¾” about 4 inches further down the pipes.

Well it’s onto mockup on the fore side header. That one will have a whole different set of challenges with very little in common with the aft header except 28 ½” primary tubes.
Header Design/Mockup (cont.)

Header fabrication continues with the mockup of the fore side (drivers side) header. I start the mockup process by starting with the exhaust port that is furthest from the collector, in this case that’s the rearmost port as the collector will be routed under the oil pan near the engine front. I try to run this tube in as direct route as feasible so I have the best chance at making the other tubes equal length without having to redo the whole works.

The main considerations for this header is to minimize heat transfer into the car’s interior by providing an air gap to the firewall while not heat soaking the starter motor by routing to close to it. Trying to strike the best compromise, the first two primaries pretty much wrap around the starter solenoid with ½” clearance at a minimum.



I had to get creative with routing the front two primaries in order to get the full 28 ½” lengths into them.



The flat collector is positioned at an angle under the non-sump side of the oil pan. The challenge here is the collector outlet needed to be angled away from the final drive case and back under the transmission so the secondary pipe would end up near the secondary pipe from the aft side header.


With these mockup headers now complete, I have the information needed in order to order up all the stainless steel mandrel bends and make the headers for real.
Chassis Wiring

While waiting for the header components to arrive, I decided to start figuring out some of the chassis wiring. I had prior purchased a wiring kit from Painless Performance that includes all the commonly used wiring circuits pre-terminated at a fuse block. So basically a fuse block with about a dozen bundles of wires of various lengths snaking out from it. The first step to install the wiring is then to find a suitable, central location where the fuse block can be mounted. I like to have the fuse block hidden from sight but also easily accessible for troubleshooting and fuse replacement so I factor this in when evaluating potential locations.

The instruction book that came with the kit recommended laying the various wire bundles out on the chassis to position the fuse block such that the provided wires would be long enough to reach their end points. I did this and under the dash on the passenger side looked like the best location for the fuse block with adequate space and accessibility.


If the fuse block was mounted high under the dash, it would be out of sight but wouldn’t be easy to get to. So I decided to mount it on a hinge of sorts such that it could be stored up high for normal car use and lowered down for maintenance. I used a scrap of ¼” aluminum (lightened with holes) as a mount for the fuse block and a 5/16” bolt through the mounting plate with a nylon spacer behind to provide a pivot for the hinge.



The other big electrical pieces that needed a home are the controls for the Holley Terminator X engine/EFI system. Specifically there’s the ECU, Variable Valve Timing (VVT) controller, and two ignition/spark controllers (one for each bank of cylinders) and the maze of harness wiring connecting these boxes. While the EFI instructions said these controllers could be mounted in the engine compartment, my preference is to locate them somewhere they would stay dry and away from heat sources. The most logical space looks to be a cubby hole located behind the drivers seat and next to the engine compartment. I had prior cut some of the chassis metal away here to make room for the transaxle so it’s easy to pass the wiring harness through to test out various mounting options.


After trying a couple of alternatives, this vertical mounting scheme seems good. It enables good routing for the wiring harnesses and good access to the control boxes. There are four wiring harness bundles between the EFI and chassis wiring that needs chassis bulkhead pass through in this area. So I want to make more progress on figuring out what additional wiring is needed before I weld in the metal to re-seal the chassis here.
Alternator Belt Routing

In working through wire harness routings, it became clear that I needed to figure out the placement and mounting for a central component of the electronic system, the alternator. The original Miura had a somewhat quirky, non-standard element to the alternator in that it used a toothed, Gilmer style belt to drive it. For my Miura, I have no choice but to go with something non-standard as there just isn’t any space to mount even the smallest of alternators in a standard arrangement. Given the EFI fuel pump, electric cooling fans, electric water pump, ECU/Ignition/VVT controllers and A/C, this configuration requires a high output alternator and thus the alternator itself will be decent size and need some mounting space.

I decided to start with the alternator that came with the Coyote engine from the Mustang GT. It’s a Nippondenso unit, put out 140 amps when tested and there’s a 200 amp version in the same case if the one I have doesn’t have quite enough output. Given the lack of space, the only choice is to mount and run the alternator backward so the first modification was to swap out the clutch drive pulley it had with a solid pulley so the alternator could be spun counter clockwise.

The Coyote engine is typically configured with 2 serpentine belts having the alternator located on left side which is the forward bank of cylinders. To fit the engine in the chassis, I had to remove the front belt drive off the crank damper pulley, so the engine is now only single belt capable. So to determine potential belt routings, I had to source a belt tensioner that fits on the right side of engine where the idler pulley needs to be right next to the engine front. I also had to add an extra idler pulley so the belt would clear the newly added belt tensioner.



It’s all very compact and tight with only one belt rubbing issue but I think it can be made to work (pulley at top is simulating the alternator pulley). Out of the two possible belt routings, this one seems to be best. The other potential routing has the belt going around the coolant intake neck. While that routing would enable placing the alternator higher (providing more clearance between it and chassis tube), I don’t like the notion of needing to dump coolant and removing a hose to do a belt change.

Here’s the rubbing issue.


The timing chain case drops down and touches the belt just upstream of the idler pulley. I removed just a bit of metal from the timing chain case where it rubs and ended up with a hole. The casting is maybe 1/8” thick where I thought given the shape it would be thicker. I put a cotton swab through the hole and poked it around. I felt only solid walls and it came out dry without any sign of oil on it. Best case, it’s only a hollow cavity and no repair is needed. Most probably, I’ll need to pull the timing chain cover off and weld some metal in behind it.

So question out to anyone who’s torn down a Coyote engine; what’s in that part of the timing chain cover? Is this a space that is somehow connected to the windage part of the engine?
Probably already looked into, but could you get a smaller diameter Idler Pulley?

Thanks for that suggestion. Oh my, don't I wish for the good old days when there were knowledgeable counter people at the auto parts store. I used to walk in with a close enough description for what was needed, they'd leaf through some paper catalogs behind the counter and they could tell you what part options/sizes were available without asking for a year, make, model. These days, the counter people seem to just be cashiers and are lost if you don't have a year, make and model which you don't have when your car is a compilation of OEM and homemade parts.

I was just able to find a Dayco dimensional reference catalog online and it does look like they make this idler pulley in 59mm OD where what I have is 76mm OD. It's plastic instead of metal but that's not a show stopper. So I think the smaller pulley will give more clearance and resolve the rubbing part of the issue.

My challenge is that I still have a hole that I already made in the timing cover by removing some metal. Best case, I could poke some JB Weld into the hole (it's about 1/8" wide by 1/2" long in size) and cross my fingers that oil doesn't start leaking out down the road. Like I said, a cotton swab comes out clean and doesn't give any sign there's oil behind the hole. My worry is that its very rare to find empty, sealed off voids in engine casings. So I'd feel a lot better if someone with first hand knowledge can either confirm that or tell me what can of worms I've opened up.
I did more investigation today and was able to answer my question. Yes, the hole I opened in timing chain cover is open to crankcase windage. I poked around through the hole some more with a cotton swab and this time came up with a trace of oil. I'm guessing that since this is a 15k mileage engine and it uses 5W20 oil that very little oil sticks to the engine casing walls.

So I've now added one more task to the project list: weld up the self inflicted hole in timing chain cover :(