Page 2 of 3
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 13:29
by exxos
Badwolf wrote: 02 Oct 2026 12:35
If that line *only* goes to the CPLD you might be lucky if the chip blew up faster than the voltage could be propagated to the motherboard via the expansion port.
That's basically what happened when the CT60e blew, the Falcon survived, but the accelerator did not.
On the face of it that switch looks too good to be true: same form factor, double function? Genius. But you're putting a lot of faith in its design and contstruction to keep 240 and 5V separated within a 5mm contact box.
Yeah, routing low voltage stuff right through the same switch which is routing mains is a big no no in my book.
I was looking, if the modules themselves had any sort of kickback when they are turned off as that sounds like when the first failure actually happened. But I haven't been able to dig up any sort of information on the power modules themselves saying anything eitherway.
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 13:33
by rubber_jonnie
Hmm, once I read about the switch I was thinking 'Flashover' but you beat me to it.
Why on earth would you even consider dual purposing a switch like that? Separating high and low voltages seems like a fundamental safety mechanism.
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 13:42
by kohli79
That’s exactly the problem I had at the end of last year. A Xilinx chip was destroyed. Fortunately, nothing else was damaged. The problem was definitely caused by the switch on the back. There was an electric arc there, and when you moved it, it crackled and sparked.
However, I have absolutely nothing negative to say about the repair process. I emailed Pavel and then sent everything to him. Of course, you have to allow for some time.
The switch unit was upgraded to the 2024 Rev. and the Xilinx chip was replaced. The switch also seems to be of better quality.
Everything has been working great ever since.

- CT63 - Xilinx.jpg (251.69 KiB) Viewed 85 times
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 14:17
by exxos
On the switch itself, even if the 2024 revision uses a better quality unit, the fundamental problem hasn't gone away. It's still routing a 5 volt signal through the same switch handling 230 volt mains. A better switch might arc less often, but the risk is still built in. It's still a disaster waiting to happen, just with better odds for now.
Do you have any images of the switch? old and new?
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 15:14
by kohli79
exxos wrote: 02 Oct 2026 14:17
On the switch itself, even if the 2024 revision uses a better quality unit, the fundamental problem hasn't gone away. It's still routing a 5 volt signal through the same switch handling 230 volt mains. A better switch might arc less often, but the risk is still built in. It's still a disaster waiting to happen, just with better odds for now.
Do you have any images of the switch? old and new?
Yes, that’s certainly still a risk.
But only with the old switch. Unfortunately, I don’t have a photo of the new switch from when it was still removed.
I think the new switch has a metal casing as well.
It feels much more solid.
The old switch can also be seen on the Centuriontech website.

- Switch new.jpg (285.79 KiB) Viewed 61 times

- Switch old.jpg (321.39 KiB) Viewed 61 times
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 15:36
by alexh
I have Centuriontech CT60PSU and a Willy CT60e. I've not had any problems in the 1-2 years I've had the PSU. (I had a PicoPSU before)
The switch doesn't crackle/arc when thrown.
I have the 2025 model.
I don't remember if I connected the cable that switches between 030/060. I might have already had a switch from when I had the PicoPSU? I never switch from 060 mode anyhow.
I would have bought an Exxos CT60 PSU had he made them.
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 15:48
by rubber_jonnie
Does that switch have a voltage rating stamped on it anywhere as it doesn't look mains rated to me. I'm assuming 230v or similar to the UK.
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 15:53
by exxos
alexh wrote: 02 Oct 2026 15:36
I would have bought an Exxos CT60 PSU had he made them.
Tempting..
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 15:57
by PhilC
If you were to disassemble that switch, you would see that there is very little insulating one set of poles from the other.
I'm an R&D engineer amongst other things and we wouldn't even run low and high voltage through the same cable or even trunking in a panel, let alone the same cheap Chinese switch.
That's a big no no. I would recommend anyone using that switch to remove the low voltage side with a matter of urgency.
Re: Falcon030 dead after Centuriontech CT60/CT60PSU incident - investigation begins
Posted: 02 Oct 2026 17:51
by exxos
PhilC wrote: 02 Oct 2026 15:57
If you were to disassemble that switch, you would see that there is very little insulating one set of poles from the other.
Yeah probably 1mm at best between mains and 5v... *shudders*
I'm an R&D engineer amongst other things and we wouldn't even run low and high voltage through the same cable or even trunking in a panel, let alone the same cheap Chinese switch.
That's a big no no. I would recommend anyone using that switch to remove the low voltage side with a matter of urgency.
100% agree.
Problem is most people don't listen. Been telling people for years to recap their PSU and upgrade the big motherboard cap to avoid killing the videl.. I gave up bothering as most people's attitude was "its worked fine for 20 years.." :roll: