agedhorse

That’s the reason for annealing and part of the reason for the dome shape (the other primary reason is to improve the HF radiation pattern). Work hardening and subsequent fatigue fracturing occurs when there is repeated flex in a material that is stiff in nature. As a practical example, JBL successfully perfected this process on their aluminum dust caps and high frequency diaphragms, they can last a lifetime if not abused.

My suggestion is to contact Surrey Amps, this is pro level service and Stan Lawrence is a highly qualified and experienced service engineer who is fully capable of taking care of Subway amps. I can support him at the factory level too. https://surreyamps.co.uk/contact-us

That's a cosmetic treatment, unrelated to the real thing except in looks. Proper aluminum dust caps are formed then annealed so that they resist cracking/fracture. This is no different than how (quality) high frequency diaphragms are manufactured.

It was around 2000-2001, the project was called "Indigo" (a dual concentric + sub music foreground/commercial audio product) and my job was to help them migrate the power amp from BASH technology (which went bankrupt) to ICEPower. This was long before Uli Behringer acquired them.

That's a short lived model that I never ran across. I didn't do anything with the studio side of the market, only the live/install side. It looks like yours is an aftermarket recone because that's not what the factory workmanship would be (based on my experience with the company). This looks to be a factory part:

I wonder if that was some kind of experimental application, doesn't look at all like a factory product. I did a fair amount of work for Tannoy many years ago, and at that time they were all about clean appearance and workmanship.

I can’t say that I have ever seen a Tannoy cone like that. Are you sure that’s a real Tannoy cone/product?

Things get super exciting when neutrons move...

The simplest and most well known example of off axis extension is the Duraluminum dust cap that attaches directly to the bobbin. It radiates the HF component of the signal like a 4” driver. The Duraluminum material is the same material that jBL developed for use in high frequency compression drivers. A wizzer cone is not a separate transducer, it’s just a specialized dust cap that attaches to the top lip of the bobbin just like the JBL aluminum dust cap. How well the HF energy coupled to the secondary radiation device determines in part the effectiveness. Decoupling the dust cap from the bobbin can be used to alter the response, as does the choice of materials.

There’s the basic premise of destructive interference due to the diameter of the driver versus the mid-high frequency wavelength, but this is modified by the size/shape of the dust cap and the geometry of the cone. With the right choices, the mid polar pattern can be altered more than an octave above what the basic predictions suggest. this can be taken to an even higher level by the use of a “wizzer cone” dust cap.

700 watts RMS is 1400 watts peak and 2800W peak-to-peak. You forgot the squared factor in the formula.

Yet there are several mechanisms that are used to improve off axis high frequency extension. This is why some 15” designs are better than others in this regard.

It could be any number of issues. If the OP is in the US then any QUALIFIED, Ampeg authorized service center should be capable of properly diagnosing the fault (the first step in any repair) and then repair the amp. If the OP is in the UK, Surrey Amps is qualified and capable of handling the diagnostics and repair.

As you have probably learned, a little reverb on bass goes a long way, and just a little too much can quickly become a bigger problem than what you are trying to solve.

Cooling an amp in a way that's effective, quiet AND reliable is a lot harder than it appears. The fans that are featured as "super quiet" usually do not have the airflow , especially under any static pressure losses that a cooling system presents. The D-800 has a fairly expensive ball bearing fan that's reasonably quiet (not as quiet as the sleeve version by a dB or two) but is very long life (about 50k hours compared with about 5-10k hours of a typical sleeve bearing fan). It also runs continuously at a slow speed because the resulting air flow allows multiple components to stabilize thermally together as a system. It's extremely unlikely to ever run any faster than the slowest speed, even at 2 ohms. Back about 15 years ago, when I was working at another company, I led a research project (with the cooperation and support of ICEPower R&D) that specifically addressed a whole slew of thermal and dynamics/duty cycle management design approaches, which resulted in a US Patent relating to this application. Some of what we learned, and how they relate to bass guitar ended up being incorporated into the newer ICEPower modules.