Sufficient electrical knowledge obviously does not exist. First indication is multiple replies that ignore all spec numbers. Only wild speculation ignores specification numbers. Start with impedance. Is that connection 'less than 3 meters"? No. Obviously not. You have no idea what impedance is. Your circuit does not have required low impedance connections.
How does a fuse disconnect a surge when it does not blow until after 300 surges? It doesn't. Now learn what blows that fuse - ie follow through current. A current that does not exist if well proven 'whole house' protection had been installed. You did not address the problem. And did not learn of well proven solutions. Blown fuses are only reporting that effective protection still does not exist.
A surge enters your circuit. Gas discharge connects it from one incoming wire to all other wires. Now that surge is on *more* wires incoming to attached appliances. It makes damage easier. The obvious was ignored. Surge is incoming because a best connection to earth is via electronics. Your circuit gives that a surge even more potentially destructive paths through electronics. Please learn how surges do damage (see reference to metallic current).
More facts must be learned. Best protection at the appliance is always inside that appliance. Electronics will routinely convert most surges into rock stable, low DC voltages to safely power its semiconductors. Best protection converts and consumes those near zero surges into well regulated and safe electricity. Afterwards, your fuse blows. A blowing fuse doesn't say that circuit did protection. That fuse blew long after a surge was incoming to or outgoing from attached appliances. Facts and numbers were ignored to only assume protection.
Follow through current also explains why fuses would blow long after a surge existed. Another path is incoming on AC mains and outgoing from an ADSL port. Either way, that surge was still incoming to electronics. You ignored spec numbers and fundamental concepts such as low impedance, follow through currents, and the question that defines all protection: where do hundreds of thousands of joules harmlessly dissipate?
You only assumed ADSL damage meant phone line was the incoming path. Usually it is not. Another example of what must be known before a solution can exist. Best protection is already inside electronics. If your circuit was effective, then it is already inside all electronics. It is not for one simple reason. Not effective.
Start by learning what a surge current really is. Surge is not a metallic mode current as you have only assumed. Just another reason why gas discharge tubes may give a surge more incoming and potentially destructive paths.
Protection is always about no surge inside a building. Always as in the only solution implemented in every building that cannot have damage. Your circuit might be supplemental protection if AFTER effective solutions are implemented. If that 'whole house' solution does not already exist, then read above paragraphs. Your circuit gives a surge MORE potentially destructive paths through attached electronics.
Impedance also says why protection increases when separation increases between appliance and protector. Learn what impedance is. Your circuit obviously has no low impedance connection to earth. Again, numbers: less than 3 meters.
Where do hundreds of thousands of joules dissipate? You cannot say. So you pretend it does not matter. Effective protection always says where hundreds of thousands of joules dissipate. Well proven protection connects low impedance (ie hardwire has no sharp bends) But that means learning spec numbers, a type of current that is called a surge, the most common incoming path that causes ADSL damage, and what low impedance is. Everything defined by numbers,
10 Kamp gas discharge tube says it can connect up to 10 Kamp into attached appliances via more (other) wires. What that tube might do, that is useful, is already inside electronics.
Wow, you are really confused. I understand impedance and high frequency earthing far better than you do. I happen to be formally educated in this field.
Some of your misconceptions:
1. I did not claim the circuit protects from a direct hit. As you noted, I stated "nothing can".
Furthermore, this means I don't have to deal with your "1000s of joules", because they aren't there. If they were there, I said the case is hopeless? Reading comprehension is an important part of understanding what people are saying. I am saying that the circuit has a good response time and a good capability to dissipate moderate transients and surges.
2. "Start by learning what a surge current really is. Surge is not a metallic mode current as you have only assumed. Just another reason why gas discharge tubes may give a surge more incoming and potentially destructive paths."
I know exactly what a surge current is, having dealt with a system designed to create very short pulses (read high frequency content, if you're as knowledgeable as you claim to be). It was capable of up to 2MV, I believe. Pretty scary machine. You're upset because I'm using fuses - the fuses are not to disconnect the device during a surge, they're to disconnect the device in the event of component failure caused by a surge or repeated surge). Is that really so hard to understand? They protect the component if you get anything more than a transient.
3. You actually ignore the practical instances where this system has worked, and this design has been used in commercial application. This implies that there exist surges that this is a practical and useful defense against.
4. No, I don't assume ADSL damage is from the phone lines always, but unlike you I don't randomly claim it is not. If the device was not effective in the way I state it is, then it follows I would be replacing modems a lot more often than I do. (I have two working modems that we replaced in a box somewhere, because the Asus DSL-N11 kept overheating, and the other did not support the port-forwarding I needed).
I'm sorry, but this is as far as I go with you. You're talking nonsense, that that is quite plain to see. You know only enough to be dangerous.
Are you really trying to sell something? I doubt you can provide adequate low impedance earthing at high frequency (Plain language for your level of understanding; the waveform of the lightning strike has sharp edges, which means a frequency analysis includes lots of high frequencies) that is sufficient for a direct hit on a building.
In short, you're not really as knowledgeable as you pretend, and you're quite confused as to the purpose of the circuit I supplied, seeming to think it is designed to stop a direct hit. This is where I leave it with you. Cheers.