@Philip_Travis it is hard to determine whether a power supply or tube is bad in a K40 that is acting up like this.
In cases like this its is useful to be systematic in thinking & troubleshooting and “trying stuff” can reveal hints as to what is really wrong.
A long rambling of “thinking out loud” below.
In this case both LPS and tube seem to be capable of producing the correct power under certain conditions.
As you have observed there seems to be a threshold where additional electrical power is not being converted to light but is being applied to the system properly. At this point the tube discharge is bright and stable except for the occasional jump at the anode (which may or may not indicate a problem).
At the threshold the LPS is under stress (the whistling noise). The meter reads increasing current as it is adjusted to higher than normal levels. That amount of current is at the LPS limit. The key question is what happens to the light conversion at this threshold since:
…The power supply puts out more current
…The discharge glow is brighter
1a). Conductive water: Although your water exchange suggests the water is ok, I would get a meter and measure the water. Having a conductivity meter is a good thing anyway for laser cutter water management. They can be gotten for reasonable $ on eBay and amazon.
…Its unlikely this is the problem but a measurement would check it off and insure we did not miss something …
1b). HV breakdown: Perhaps somewhere there is a HV short circuit that occurs at higher power settings. You may be able to see this with the lights out.
What discounts this possibility is that usually breakdowns can be seen as pulses in the meter and power drops. This machine is not behaving this way.
In my experience these supplies are stable unless the HV transformer is damaged and in that case the output power and ionization would be lowered which it is not!
the power supply is open-ended in which case it will arc violently and there will be lots of corona from the HVT and the tube will not discharge !
Neither seem to be your case…
I bet that the LPS is ok since it puts out current proportional to the meter setting. The noise from the supply is because it is stressed at more than normal current when the knob is adjusted all the way up…
To me the video shows the tubes ionization operating pretty normally.
… the tube operates normally up to the threshold and
… the discharge gets brighter when the LPS is turned up
… laser output light gets dimmer
It seems the problem is inside the tube!
The tube is drawing more current as the LPS output is increased and the discharge suggests it should put out more light!
At the threshold something is happening with the internal optics or gas conversion?
Is the water at the proper temperature?
What does the ionization at the cathode end look like at the point of threshold?
Test: adjust the current to the point of threshold. [Don’t go much above as all that will do is stress the supply.] At threshold take a video showing a closeup of the anode and more importantly the cathode end and mirror(s). Maybe that video will give us a hint?