Help in understanding the fuse rating of this amp

[BrianS]

Caveat:
-As an authorized service tech that has signed an NDA for this particular brand of guitar amplifier, I cannot disclose the make, model or much in the way of circuit design that is not "generic".
The amp has a 6V6, cathode biased p/p output section; solid state rectifier.  It came in with literally burned up screen grid resistors (rated for 2 watts), a shorted power tube and ultimately, a bad power transformer.
-The transformer looks to have a fusible link in the primary that had opened up, rendering it inoperable.
-Repaired amp is fully functional, but I'm hoping to understand...at least at a fundamental level...their choice of mains fuse.
The amp is rated at 100 watts of power consumption.  Ohm's law tells me that at 120VAC of input voltage, the amp will draw a maximum of .83 amps of current.  From what I've read, a basic calculation for mains fuse selection would be max current draw as calculated above, times 1.5.  In amplifiers with high in-rush current, a slo-blo fuse is often selected.  So, it would stand to reason that a 1.5A slo-blo fuse would be sufficient here.
Looking at the current meter of my variac shows that at power up, the amp's inrush current barely exceeds 1.5A.  There are some other components on the primary side of the PT that may limit inrush.  Again, I cannot describe the circuit in detail.
For a mains fuse, the amp has a 3A fast blo fuse.  This seems like a high value, and I'm wondering if this seemingly high value contributed to failure of the power transformer.  I'm not an amp designer, just a repair tech, so it's very likely I'm missing some key design element that makes this fuse selection sensible. 
Insights from the more engineer minded/educated folks here?  Thanks.

[astronomicum]

The transformer looks to have a fusible link in the primary that had opened up, rendering it inoperable.

What is/was the link rated?

[BrianS]

The fuse is temperature rated for 130C...266 Fahrenheit.

[tubeswell]

Modern silicon-grain-oriented laminations can cause severe transient surges at startup, and a fast-blo fuse would be more susceptible to this. So I agree that it’s puzzling (if it was a 3A fast-blo from the factory)

[acheld]

I started using in-rush limiters a few years ago -- seemed like a good idea in my amps with SS rectifiers.  They've worked flawlessly.  Now, how much they really have decreased stress in my builds is, of course, debatable.   But, I'm a belt and suspenders guy.

Check out Amphenol NTC current limiter thermistors  https://amphenol-sensors.com/hubfs/AS-920-325F-Thermotrics-NTC%20Inrush-0525-web.pdf.  I often use a CL-70, but that of course will vary by the build.

[BrianS]

Yes, 3A fast from the factory, and as indicated on the amp panel.
Got more info from the customer:  he's used the amp regularly for gigs and rehearsals for 9 years with no issues.  Never replaced the tubes or needed any servicing of any kind.  In reading up on modern transformers with fusible links, some people believe that these links can fail simply due to use over time.  Seems unlikely, but...
I have sent an inquiry to the manufacturer regarding the fuse.  I don't hold too much hope in getting a real explanation, as they don't owe it to me.

[stratomaster]

I can just about guarantee your equipment didn't catch the inrush current peak, but that's irrelevant at the moment.

The 3A is a wild choice for this amp.  For comparison an original Deluxe Reverb (also 2x 6V6) uses a 1A slo blo.

[Merlin]

I can just about guarantee your equipment didn't catch the inrush current peak, but that's irrelevant at the moment.

The 3A is a wild choice for this amp.  For comparison an original Deluxe Reverb (also 2x 6V6) uses a 1A slo blo.

3A is not that unreasonable for an amp that draws close to 1A normally. Most people would probably use 2A. But fast blow? That's mental.

[JPK]

I can just about guarantee your equipment didn't catch the inrush current peak, but that's irrelevant at the moment.

The 3A is a wild choice for this amp.  For comparison an original Deluxe Reverb (also 2x 6V6) uses a 1A slo blo.

3A is not that unreasonable for an amp that draws close to 1A normally. Most people would probably use 2A. But fast blow? That's mental.

Thank you. I was also wondering why the fast blow. Wrong fuse. I built a 50 watt Ceriatone head recently (120v). They offer it in a 100 as well. The kit he sent me had a 4 amp slow blow and it was for the 100. When I emailed him he said it was 3 for the 50 and 4 for the 100. I ordered a 3.

[BrianS]

Just for the sake of discussion, WHY is the fast blo type fuse a "wild choice" for this amp?  As I stated in the OP, the amp does have some kind of current limiting circuitry on the primary side of the PT... a thermistor at a minimum...so wouldn't that negate the need for a slo-blo fuse? 
Again, this is all academic for me.  I've repaired the amp with all original spec parts and it's fully functional.  I'm not one to second guess a "successful" amp design, it's just that the mains fuse choice in this case seemed to do nothing to protect any part of the amp.  I'm guessing that the fusible link in the transformer kept the thing from actually catching on fire.

[Merlin]

the amp does have some kind of current limiting circuitry on the primary side of the PT... a thermistor at a minimum...so wouldn't that negate the need for a slo-blo fuse? 

Not really. Thermistors are not tightly-controlled, hard limiting devices, they're more of a 'better than nothing' device. For example, you could flip the amp off and on again quickly while the thermistor is still hot, so it would not prevent a second inrush event. Fast fuses are rarely suitable for power-supply fusing.