Thread revival time..... I've been battling with this shuddering (Torque Groan) on my 2.5 hawk sti for over a year now, trying to find a solution. I've changed to a lightweight RCM flywheel and hard race engine and tm mounts with no avail. I've tried all sorts and still in the process but at this point I've resorted to trying to make the turbo spool slower by cracking the wastegate slightly open all the time, crazy but it has helped slightly!

As aside has anyone every linked this shuddering to repeat headgasket failure of cylinder #4 (nearest the flywheel)? Resonance is a powerful and damaging thing and I'm wondering if the vibrations could potential be causing a fretting affect of the bores on the gaskets?

repeat headgasket failure sounds like its being caused by something else, have you had the block/heads checked and made sure there not warped?

For sure, blocks and heads were skimmed flat, I also fitted uprated water pump and radiator (as it was considered it could be a cooling issue) and been running stock fuelling and timing but just slightly more boost since that last rebuild ~2.5k miles ago, with supporting map tweeks etc. I've been running with various brands of HG, RCM, cosworth, but now on stock HGs, and with both RCM and ARP head studs that have definitely been tightened properly every time, but still the same problem.

It's starting bubbling in the expansion tank (and smells of exhaust, like the previous times) and gurgling in the dash. This'll be the 3rd time this year in less than 6.5k miles the hg has gone always on cylinder #4! Though I'm assuming it's that one again but not got it apart yet.

I have some theories currently:

a) ether this shuddering from the drivetrain is causing the failure, as mentioned above,
b) maybe the bores moving and panting under load causes the shuddering and HG failure by fretting the gaskets
c) when the engine has overheated (several times now), this has damaged the head stud threads near that cylinder or somehow softened the blocks. My logic for that thought is that I've needed to helicoil several of the m6 threads on the blocks: all 4 of the water cross over and all of the water pump threads, and that was tightening to the correct 6.4Nm with a new torque wrench! They still rounded out.

Now I'm thinking to get a fully CBD 14mm conversion from AS Performance, which I think would eliminate points b and c, but maybe not point A....

my 2.5 hatch has never had this issue and it has plenty of low down torque , if it keeps blowing head gaskets then it may be worth getting a proper closed deck built and 14-14 head studs

Have the cylinder heads been pressure tested as its not unknown for them to fail around the exhaust spring platform.

no, they haven't been tested, but fail in what way?

Why would you strain the engine and powertrain doing that and not downshift?

For example it's not uncommon that you might be cruising down the motorway in 6th and want to gently overtake something, say going 75-->85 ish, you wouldn't downshift for that. I don't think this scenario is an 'unusual' request when driving.

In any case the shuddering happens in all gears, but is more noticeable in 5th and 6th because of the duration of time spent in the 'affected' RPM range (Always around 3-3.5k rpm). It's also noticeable in 4th for a brief time and 3rd to a much lesser extent.

Amendment to above. Part throttle does not solve this because the turbo is fitted with 12-14psi wastegate and is spooling so quickly in the rpm range that it hits full boost early. As far as I know there is no way to slow down spool more than how the wastegate will naturally operate, and even then it builds boost too quickly. I've tried different wastegate actuators but got rubbish boost control through the RPM range. That's why I've resorted to manually setting the wastegate open a 0.5mm or so to slow down the spool!

75mph in 6th should be ~3.5k rpm, agreed, it shouldn't happen at this but I wouldn't go WOT at once.

coolant gallery failure

I don't know if I fully understand what torque groan means but 10 years ago I had a Saab 900 Turbo with a GT2871R (kind of, jap-spec version) on E85 and every time I got full boost it felt like it vibrated through the entire chassis and gearbox. But they were kind of crappy cars too :)

You can hear it in this old low quality vid, is this what you mean?