Choke Heating Up to 140°C – Zyxel Power Supply Issue

Kosmakow

Hello everyone,

I’m facing an issue with a Zyxel GS1008HP, specifically with one of the power supply components — the choke. It hasn’t burned out, but it gets extremely hot during operation — the surface temperature reaches around 120–140°C, which seems abnormal and potentially dangerous in the long term. Input Voltage 52V,

For context:

• I’ve already replaced the PWM controller and electrolytic capacitors, but that didn’t solve the issue.
• The device still works, but I’m concerned about the high temperature of the choke and possible degradation or failure over time.
• There is no visible damage to the choke (no burns, no cracks), but I will attach a photo below for reference.
• The input voltage is 52V — if I supply 32V, there’s no heating at all. PoE is being tested, but that’s not a viable solution.

My questions:

1. Has anyone else encountered this exact behavior?
2. Could this be a design flaw or a defective component?
3. Would it make sense to contact Zyxel support?
4. If I replace the choke myself, what parameters should I be looking for? (inductance, current rating, core type, etc.)
5. Could there be another root cause that I’ve missed?

Any advice, shared experiences, or technical suggestions are appreciated. Thanks in advance!

Zyxel_Melen

Hi @Kosmakow

GS1008HP is a fanless design switch. It uses its case to dissipate the heat.

1. How many PDs did you connect to the GS1008HP? What PoE mode are these PDs?
2. How did you measure the surface temperature?
3. Does the adaptor come from the brown box?

Kosmakow

https://community.zyxel.com/en/discussion/comment/76575#Comment_76575

I understand that the case is used for passive cooling, but the temperature of the inductor is extremely high — borderline critical. I came across a similar case on the forum (Link).

As far as I can tell, the L2 inductor is responsible for stepping down the voltage from 57V to 12V. It doesn’t seem to be directly related to PoE delivery, but likely powers the internal logic or something similar. What's interesting is that the temperature of the inductor changes depending on the input voltage — when I supply 12–24V, the inductor stays cool. But as I increase the voltage, it gets significantly hotter. Of course, running the switch at voltages below the standard PoE range is not correct operation, but this behavior is noteworthy.

Also, none of the other components on the board show any signs of heating — only the inductor. It gives the impression that the inductor might be saturating and heating itself, but this is just speculation on my part and could be incorrect.

I tested it with zero connected devices and then with two PoE cameras — the heating occurs in both scenarios.

I measured the surface temperature using a thermal imaging camera. While the exact readings may not be perfectly accurate, the choke radiates noticeable heat — you can feel it easily by holding your hand nearby — so I believe the thermal image is reasonably close to the actual temperature.

Regarding the power supply: I did not use the original adapter. I tried a 52V third-party adapter, a MeanWell 48V power supply, and, after disassembling the switch, also tested it with a laboratory power supply in the 12–57V range.

Thanks.

Zyxel_Melen

Hi @Kosmakow,

Could you share a screenshot or a video so we can confirm the temperature?