If it helps any, you are not alone. There have been quite a number of folks in the last few years that have tried to set up a .223 Rem rifle with the intention of using it with "heavies" in F-TR matches. What many have found is that if you're not dealing with bullet jacket failure, it's poor brass life. If it's not poor brass life, it might be poor precision, or any one of a number of other issues that preclude successful use of the setup for its intended purpose.
Here are a few things I've learned over the years from my own experiments and those of others about how to get a .223 Rem to shoot with heavy (88-95 gr) bullets. First and foremost is dealing with the jacket failure issue. In an F-TR match, even one bullet failing to make it to the target means you will likely not end up on the podium. Bullet jacket failures are generally caused by excessive friction. The heat from friction can actually be sufficient to heat the surface of the lead core underneath the jacket and cause molten or vaporized lead to spew out of any crack or pinhole in the jacket it can find, leading to what are known as "comet tails" on the target (
https://forum.accurateshooter.com/threads/comet-tails.3972620/).
As friction is the primary cause of this phenomena, it is easy to understand how long barrel length (i.e. 28" to 30"+) can exacerbate the issue. Excessive twist rates can also promote this issue (i.e. RPM values over 300K). What was not obvious, to me at least, is that the "tightness" of the bore plays a critical role in this process. To make a long story short, shooters running heavies in the .223 Rem have found that by using a 0.219"/0.224" bore and groove configuration instead of the tighter 0.218"/0.224" bore/groove configuration makes some of the other "undesirable" features in a barrel prone to causing jacket failures go away, such as using excessive twist rates. FWIW - Bartlein makes 0.219"/0.224 barrels. As it turns out, going to the 0.219" bore will allow the use of twist rates as fast as 6.5-twist, without fear of the jacket failure issues associated with 0.218" bore barrels. Also, some types of rifling, such as 5R, may be a little easier on bullet jackets, but the real key to mitigating jacket failures is using a 0.219" bore barrel.
A second major consideration in getting a .223 Rem to work with heavies is high pressure and poor brass life. The kind of velocities we would
like to achieve with 88-95 gr bullets from a 28" to 30" barrel will sometimes result in operating pressures well in excess of the 55K psi SAAMI max. For example, I used H4895 under Berger 90 VLDs for years to achieve velocity in the 2840-2850 fps range, which is a very good precision node for a 30" barrel. Long story short, 3 to 4 firings and Lapua brass is done when run at predicted pressures (QuickLoad) of slightly over 57K psi. On some pieces, the primer pockets will enlarge to the point of being useless after only one or two firings.
There are various methods to deal with this issue. One is simply to accept the poor brass life. However, with the current scarcity of reloading components such as Lapua brass, that is not always a viable plan. Switching to a different powder such as Varget will typically net you about 25 fps less velocity in a tuned load than H4895, but with the added benefit of lower pressure and potentially better brass life. There are certainly other powders that are worth investigation as far as lowering pressure. However, it is not so easy to get around the physics of velocity and pressure. Another approach would be to use a brand of brass that is thicker/stronger in the case-head/webbing region. Brands of .223 Rem brass such as Lake City and Starline seem to be better capable of withstanding high pressure than Lapua. But that advantage doesn't come without a cost. More specifically, if you choose to go with a different brand of brass that may have slightly greater structural integrity, be prepared to spend a great deal more time uniforming the flash-holes, and sorting cases for uniform case/neck wall thickness. Even with the extra effort, other brands of brass have not reliably provided the precision I am accustomed to getting from Lapua brass.
The bottom line is that we're really asking a LOT of the small .223 Rem cartridge when attempting to push heavies at high velocities out of long barrels. Using a 0.219" bore barrel is a good starting point, preferably with 5R rifling and a twist rate that is just
barely sufficient to achieve a gyroscopic stability coefficient (Sg) of very close to 1.5. That usually means no faster than about 6.7-twist, which is sufficient to generate the Sg necessary to achieve the full BC of the bullet in many regions across the country (i.e. altitudes), without going too far. AFAIK, a 6.7-twist is sufficient to fully stabilize the longest (heaviest) .224" lead core bullet currently available, i.e. the 95 SMK, at most elevations/temperatures above 1000 ft/65 degrees or so. Those shooting on the beach in winter may still be giving up a bit of the intrinsic bullet BC, but it should still work just fine. A final note is that some bullets simply have tougher jackets than others. I have never [yet] caused a jacket failure using a Sierra 90 or 95 gr bullet, even in barrels that have routinely caused jacket failures with bullets from other manufacturers.
Finally arriving at the OPs questions, going with a 7.5-twist barrel is really not going to be a good option. You cannot typically push the heavies fast enough to reliably overcome the decrease in BC caused by such a large deficiency in Sg. At that point, you would be better off using a lighter (shorter) bullet with a lower BC, but one that you could fully realize with the slower 7.5-twist barrel. Giving up too much BC with a lower twist rate is not the answer. As to why you had the results you had with the slower twist and faster twist barrels, I'm going to agree with others that have commented above that you may simply be experiencing barrel-to-barrel differences. It could be something as simple as non-uniform bore diameter between the two barrels, or something else, such as how the chambers were cut, etc. Much as we would like to, it is not always possible to know why certain behaviors are observed from certain barrels, at least, without more extensive barrel analysis that the average shooter may not have the equipment to carry out. Rather than trying to understand why those barrels did not work, I would suggest that you may have better luck choosing a specific barrel configuration that has the best chance of generating the results you would like to see. Were I in need of any more .224" barrels at this point in time, they would be Bartlein 5R, 0.219"/0.224", 6.7-twist barrels at a 30" finish length. Yes, those barrels would represent a custom order, and take a long time if you ordered them directly from Bartlein. However, you could probably get one a little faster by ordering through a reputable Bartlein vendor such as Bugholes (aka Southern Precision).
I would then take whatever those barrels gave me in terms of a final tuned load. In other words, I would not be pushing the developed load any harder than necessary in terms of pressure/velocity. Frankly, the gains in terms of reduced wind deflection to be had from increasing bullet velocity by less than about 50-75 fps, or by increasing the twist rate by 0.25 or so are too small to reliably provide a noticeable increase in performance.