First does the "other" dispenser has the same functions (mainly the filtration, need for sufficient inlet pressure vs ability to operate when the water pressure gets low...)?

And seconds I do not see that much strange when different manufacturers use their own different solutions.
Plus there are very big differences among how these pumps behave (mainly the noise,...) among various manufacturers and also depending on the way how exactly they are installed in the machine (how much of the vibration gets transmitted towards the machine body - even in coffee machines, the noise these pumps create varies from loud rattle to completely inaudible, while operating the same pressure). So I see no surprise some cheaper model creates more noise than some more expensive, better refined one.

It is true, the aluminum is more than double the conductivity of copper per conductor weight (not cross section, really weight), also way cheaper, but these aspects are not that significant on thinner wires. There the surrounding accessory (the insulation, restriction to the cable assembly materials to prevent corrosion, more care so more expense needed for termination,...) either cost the same or even is more expensive with aluminum, so if the material/weight savings are not substantial, it does not make much sense.

Regardless of the exact composition, aluminum are all more prone to oxidation and creep problems. The improved composition made them less severe, but compare to copper, they are still there. And some of the ways how to address these problems means more expensive terminal designs.
Now these problems affect mainly thinner conductors, the thicker the thing is, more resistant it becomes. So for relatively thin home installations they are way too much problem vs how much cost and weight savings it may offer vs copper, or the fixes for the aluminum problem become relatively expensive (big part of the extra termination cost to fix the aluminum problems does not depend on the size/rating of the connection, so for low current ones becomes pretty expensive). But for higher current, thicker conductors have inherently less problems to start with, plus the amount of material in question becomes so much it starts to make sense to invest into the more complex connection solutions.

And this is reflected in the standards - aluminum is allowed only 16mm^2 and above (AWG6 and thicker)

For normal home 16A circuit installation the 2.5mm^2 of copper is used (4mm^2 was used with Al),
that in the equivalent 4mm^2 Al, was and in older installations still is causing permanent headaches with connection resistance faults.

Only the main feed and distribution lines, generally significantly thicker than the 16mm^2, are still installed and used with aluminum without any problems. But the important differentiator is the conductor size.

If a house costs like €300.000 to 800.000 to build and buy the land for, i'm not gonna care about €500 in extra cost of cable. I'd rather have copper in my walls and just be able to terminate anything to everything using bog standard connectors.

Copper wire is expensive but not that expensive. I pay €1 per meter (give or take a few cents - less for larger rolls) for single core 2,5mm² copper residential installation wire, rated for 16a 230v in the Netherlands and 20a 230v in Belgium. You write that stuff off in maybe 50 years or so if you install it right. The yearly cost is small compared to all the other things you gotta manage on a house.

But in other water dispensers like TAMI 4, I don't hearing this noise.

Do have a way to measure the line current, such as with a Kill-a-Watt?  If the line current with the 34(s) is higher than the specified current for 40s, then it is not safe.  If it's lower then it's probably OK at least for a little while.  I'm basing that on the fact using F34s on an NPF ballast for F40s will result in them using more power and burning up the ballast in no time.  By contrast using F34s on an HPF ballast not rated for might make run slightly hotter and stress the capacitor, but usually won't result in an immediate failure provided the ballast was in good condition to begin with.  My impression is that HPF ballasts not rated for F34s will likely have a higher failure rate in the long term, but you can generally expect most ballasts to operate them for years without trouble.

Most of what I am saying has to do with the US, I have no idea what happens in other countries. From what I can tell, this is the background information that is true about aluminum use in the US.

- In the 1960s they started using 1350 alloy aluminum wire, had problems with creep/cold flow, oxidation, strength, and thermal expansion
- 1350 would oxidize to make high resistance connections, thermally expand and creep to make low surface area connections, and was a weak material to work with
- Revised CO/ALR devices with indium plated terminals can usually be used with 1350 because the indium metal seals tightly to prevent corrosion, but doesn't solve the other problems (expansion, creep, strength).
- After that fiasco, 8000-series aluminum alloys were used for aluminum wire. Stronger, less creep, less oxidation. With anti-oxidant compounds this works A-OK, problem pretty much solved.
- Aluminum terminal blocks and bus bars use 6101 or 6061 (most of the time, for the ones that weren't recalled), more structurally optimized, not as oxidation resistant but often tin plated.

---

So we solved aluminum wire, but it never really returned to low-amerage branch circuit wiring in houses. Smallest aluminum you typically see is 4AWG for 50A circuits, but even there 6AWG copper is still used a lot. These installations of aluminum cable must use an anti-oxidant compound on all terminations, and they all much be rated for use with aluminum.

Now if you are gonna have to do that for every termination, using aluminum for branch circuit wiring would suck. But that isn't always true, as copper clad aluminum wire exists. You can get CCA wire in sizes as small as 12AWG for 15A circuits, and of course it is cheaper and lighter than the 14AWG copper counterpart.

This wire uses 8000-series aluminum alloy, and is coated in copper to prevent oxidation. No anti-oxidant needed. Almost all terminations rated for Cu are rated for Cu-clad Al. But the thermal expansion issue still persists, and some say the cold flow issue was not entirely eliminated with 8000-series alloys.

What do you think? Will aluminum catch on for residential again? Or is this another disaster?

Ah, I see. Probably not then.

I have found a 7.5 volt transformer in my stock of extra parts. This light works perfectly upon installation, without any overheating or odd behavior. I will go head and install it permanently after some extended testing, but I think it's pretty much good to go!

Thank you so much for the help!! Another vintage emergency light saved from the trash!

Because it just likely uses the same style pump.
It would make sense, if it has to pass the water through some denser filters, which pose considerable backpressure.

Yes. For lesser lamps this is a messy fog on the internal surface of the bulb. Definitely a product of filaments coating deterioration.