Thanks for the clarification.
As for the weight, I can't believe that the Merlin installation weighed more. Think of all the items and equipment that can be yanked if the Allison is gone. All the steel ducting for the turbo, the stainless tub and housing the turbo sits in, the armour plate used to protect the pilot from the turbo, the intercoolers and all that steel intake structure. It'd be interesting to weight the Allison and the Merlin, then start weighing removed components I have around the shop to see what the difference is.
If you consider the power the P-38 had available, you'd have to figure that the props it had just weren't getting the job done. The chin mounted intercoolers just do not add enough drag to absorb all of the HP they allowed the Allisons to produce, after all, the H model and earlier didn't have them, and they were considerably slower that the later models, especially the L model running at full Lockheed/Allison rated boost and RPM (a difference of between 15 and 30 MPH, depending on models compared, and as much as 300HP per engine). Another thing to consider is that a turbocharger loves a load, and the more efficient the prop, the more it loads the engine, since it loses less to slippage.
The performance difference between the K model mule, with ill fitting cowl panels and all, and the J and L models is considerable, not just top speed, but especially climb rate and range (top speed increase wasn't worth a lot at altitude to the P-38, since above 25K you could get it to compress at or before 465MPH), as well as acceleration. The K model didn't have a lot more power (as compared to the J and L models), but evidently those Hamilton Standard Paddle props are a lot more efficient than the Curtiss Electric props are (as well as a lot more reliable according to the pilots). According to Bodie, the K model mule was actually a well worn and abused G model that had the chin intercoolers grafted on and the gear box cowls crudely fitted as well. The gear boxes were bigger and taller, and raised the centerline of the props several inches, so the cowls that covered them had to be quickly cobbled up by hand, they supposedly didn't really fit the prop hubs and spinners or the rest of the plane either, and the same could be said of the intercooler installation. Since there were only one or two (only one K mule was ever photographed, and there are only a couple of pictures of it) were ever built and it/they were built by hand, there's no way to tell how much the poor fit hurt the aerodynamics, but they felt it was enough to make a measurable difference.
I'm pretty sure the Merlin itself weighs a good bit more, the Allison was fairly light for it's size and displacement, Allison was originally a pure racing engine company from what I've read, and the V-1710 was first designed for the Navy to be used in lighter than air craft. I think the weight gain was fairly considerable. I'd have to drag out the book, (I don't know where it is right now, we're remodeling) but I seem to remember the total weight gain for the plane was around 1000 pounds (about 500 pounds per side I guess), despite losing most of the extensive exhaust system, the turbocharger, the intercooler, and the plumbing, which I agree is strange. Best I can remember, they anticipated little gain in speed, and an actual reduction in rate of climb, and possibly range. The desire was actually to reduce cost and complexity.
I don't know that the Merlin can be reassembled to run counter-clockwise the way the Allison can (the Allison was actually designed originally so that it could be shut down and re started running in the opposite direction), that would also cause problems, as one side would have to have a gear box with an extra gear or idler gear to make the props run in opposite directions (or the engine reversed in some other manner). Those helical cut gears in gear boxes don't like running the other way, and they usually don't like having gears added in to reverse them. That's why we use straight spur cut gears in high HP applications, but they bring in their own problems, and I can see where props might not like straight cut gears (harmonics and backlash).
The Merlin, unless tuned so that it gives up power below 22-25K feet, would give up power compared to the turbocharged Allison above 26K feet, further reducing speed and climb above that altitude. The only thing holding the Allison back from performing at altitude without the turbocharger was the crank driven supercharger. Given an equal amount of boost in stock form, the Allison actually makes more power (we tested that fooling with pulling tractors). If you put a crank driven supercharger on an Allison that made the same boost as a Merlin had, the Allison would make more power. Of course, the advantage of the supercharger on the Merlin series is it could be tuned for particular altitudes, the way it was in the Spitfires (there were high and low altitude specific versions of the Spitfire). So you can move the critical altitude to suit your purpose or mission.
But it is really hard to beat a turbocharger for great performance at a wide range of altitudes. In those days, turbocharging still had a ways to go, but even then it held some pretty significant advantages.
A late P-38 with the Hamilton Standard paddle props would have been far easier to produce and offered a lot more than a Merlin conversion. Removing the Curtiss Electric props would have removed a great deal of electrical load (the electrical system was almost overloaded if everything worked properly, and if the props acted up the generators were toast) and solved reliability issues. It would have also increased performance significantly by all measures, top speed, rate of climb, acceleration, and range. It would have required far less work with regards to changes. The gear boxes would fit the plane, props, and engines, all that was really required was a new set of dies to properly form the spinners, shrouds, and cowls. I'd be willing to bet it would have been a lot easier to get more Hamilton Standard paddle props than it would have been to get that many more Merlins. It would likely have been cheaper as well, since Packard was paying a license fee to Rolls Royce to build Merlins.
