The most likely culprit that I can think of would be the voltage regulator MOSFET. Do you have a P54 or P55C Intel that you can swap in to see of that changes the noise, or makes it go away?

FYI, the capacitors do not need to be bloated to be dead or dying. Capacitors near hot areas, or even old capacitors, also just tend to dry out, showing no visible signs of damage.

Have the electrolytic capacitors on your board ever been changed?
If not, failing capacitors might be part of your problem.

JF_Sebastian wrote on 2024-03-09, 14:43:

But since I have been running the K6-III on the motherboard, there are strange whining- and crackling noises from the VRM/CPU area and it depends on the CPU load.

A faint whining or buzzing noise from the core voltage regulator (the regulator that provides the 2.2V) that depends on CPU load is not necessarily an indication for a fault. One of the core components of the core voltage regulator is an inductor with a ferrite core. (Do I get bonus points for putting the word "core" three times into that sentence without making it sound weird?) This inductor consists if a coil wound around the core, which is made from magnetic material. Do you know what other component also has a coil in vincinity of magnetic material as primary operating mechanism? No, I don't think about the transformer, but about a loudspeaker.

Designers of the voltage regulators try to minimize the noise that gets generated by the regulator, but this minimization has its limit. The K6-III you are using now uses more power on the 2.2V supply than the K6-II did, because the integrated L2 cache is quite power hungry. Using more power means more magnetic interaction in that coil, and possibly cause more intense mechanical movement of the coil wire around the core, starting to make more noise. The noise is most likely emanating from one of the two coils below the processor socket. As long as they don't get excessively warm (more than like 55°C to 60°C), you don't have to worry about the coil being overloaded. If you can dampen the noise by touching one of the coils, you can consider applying a drop of super glue to that coil to prevent the vibration that causes the noise.

But while the noise might be just normal under sufficitently high load, it might also (as already suspected in this thread) be an indication of the regulator operating erratically due to worn capacitors. If you have a capacitor ESR measurement device, it would make sense to check the ESR of all capacitors around the processor socket.

I've never run a K6-III in a P55T2P4, but I have used a number of them with a K6-2+, and I never encountered a noise issue with those. They are significantly less power hungry than a K6-III, and if the K6-2 was silent in this board, a K6-2+ or K6-3+ will most likely be silent as well. I lost my P55T2P4 in a fire, but I have a very similar Asus TXP4 and it works well with a K6-2+ 533 that I modified into a K6-3+.

Hi folks!

Firstly, thanks for all the replies! I have two P55T2P4 and on both boards the CPU is the cause of the whining and crackling. So I'm guessing that the CPU is pushing the VRM to its limits. But yes, it could also be that the capacitors on both mainboards are too old and dried out. However, it is interesting that the K6-x+ on the same motherboard does not produce this noise. It seems to be much less power hungry 😁.

My next step will be to test the capacitors on both boards. Maybe this is the right solution 😉

Results will follow.

Have a nice weekend!

JF_Sebastian wrote on 2024-03-09, 19:22:

However, it is interesting that the K6-x+ on the same motherboard does not produce this noise. It seems to be much less power hungry 😁.

As the K6-2+ / K6-3+ were designed as notebook processors, it's not very surprising that they use a lower power design. Furthermore, they have a newer core that likely is generally less power hungry due to process improvements.

I can 'cause' coil whine by using the 'throttle.exe' utility on my MS6156 (440BX chipset). It uses ACPI CPU throttling feature in slowdown steps of 8.

To stop the whine I wedged a rubber (any insulator) slightly against the inductor with a ferrite core.

Does this mobo have a piezo speaker onboard? I've had boards that I thought had coil whine but it turned out the be the piezo picking up bad ripple from the +5V line due to bad caps.

Hello folks,

I need your help. I am trying to find suitable capacitors to replace the old ones. For a low ESR type from ELNA next to the VRM, I was able to find an equivalent type from Nichicon. But around the socket there are two pairs of three capacitors each from I.Q. with 6.3V/1000uF and I can't find any information about their type. Normally these capacitors are just for buffering, right? So I conclude that they could be replaced by general purpose capacitors for high temperatures. Are there any arguments against my assumption? Also, there are still footprints from older revisions around the socket. Would it also useful to fit these with capacitors from same type for more stabile buffering?

Any help would be greatly appreciated. Maybe I can solve my "noise" problem.

Greetings!

PS: I have nospeaker onboard 😉

The "I.Q." electrolytics were just chepo low ESR 105° ones.
It´s best practice to replace them by some current equivalent type of a renowned brand like Rubycon, Panasonic or Nichicon

https://www.reichelt.de/elko-radial-1-000-f-1 … tct=pol_0&nbc=1