The secondary side actually has two fuses, plus thepolyswitch. One is on the PCB.
Yes. hence my description of the black wire ended one as the 'extra ordinary fuse'. Knowing that there's a fuse on the PCB, its sheer presence hanging off the secondary terminal indicates a design process SNAFU.
Did anyone note the ratings of all the fuses?
Both are 4A. Why you'd have two 4A fuses in series is beyond me.
The secondary fuse + PTC combo... I'll give a perspective and hope the thread S/N ratio stays up there.
From an engineering, physics point of view, if the fuse clears first- the PTC can do nothing. If the PTC trips first, the fuse can do nothing. It doesn't make sense.
With only a secondary fuse, a problem is passing the transformer overload heating tests and short-circuit test.
Generally, the transformer is long term tortured at the fuse trip point OR the protective element (PTC) can trip and hold.
Without the PTC, 61558-1; 15.3.2 test is constant current 210% times the fuse rating for one hour, then checking temperatures, insulation etc. which is 4A*2.1 or 8.4A very high hence the need for the thermal fuse to pass this.
UL 1585; 27.1.2b "... {with} positive temperature coefficient device (PTC), the current is to be measured after 5 seconds of operation."
Here, the PTC gives a lower test current, passing is easy but skirts the transformer's lack of protection.
UL 1585; 27.2 "... Protective devices are to be shorted out during this test."
It doesn't say all or both devices and may be why a second fuse is present. Even the transformer short-circuit test, where the secondary is shorted... must have been done
after the terminal pins, after the windings.
I'll check if I have access to UL 5085 - Low Voltage Transformers safety standard which would apply for the US. It's an old standard with some issues:
"During the regular review and maintenance cycle of the harmonized standards, it was noted that there were several areas in the standards where the requirements were either redundant and/or not clear or that additional clarification is needed. Revisions to the standard to correct the above stated deficiencies also included corrections to typographical errors."
CSA Urgent Bulletin re: C22.2 No. 66 and UL 5085Proof these old standards have cracks and loopholes and engineering common sense should be used.