Agree on not blindly trusting the 12v adaptors.

 

If you have clocks that use the NCH6300HV or NCH8200HV supplies, the issue is 
that they WILL die catastrophically if they see inputs over 15v.  And they will 
potentially take out other elements in the neighborhood when they go.  Check 
the 6300 datasheets 
<https://nixieclock.org/wp-content/uploads/2023/01/NCH6300HV-Datasheet-EN-v1.0-final.pdf>
  and 8200 datasheets 
<https://omnixie.io/assets/products/nch8200hv/nch8200hv_datasheet_en_v2.0.0.pdf>
  for the limits.  

There are a lot of clocks that use them.  Often, the clocks that use them don’t 
have much (or any) filtering or input protection on the board and just assume 
that the 12v power supply that is used will have sufficient 
filtering/protection.  That’s the flaw.  Not all wall warts or bricks are 
created equal and they may seem to be working fine, right up until they aren’t. 
 You may get lucky and notice flickering or one day you may see nothing 
working.  The additional components that were suggested are insurance for when 
the PS fails.  And they do fail.  Even though a crowbar circuit serves no 
functional purpose for the operation of the power supply, it is there to 
protect everything else.  This is just a simple response when the clock has no 
protection on the input and is easy to add.  They can be added right at the 
input jack.  Choose a different cap size if this is “too big”.



 

The 6300HV and 8200HV are great little converters, small, efficient and they 
power lots of cool nixie and panaplex clocks.  Don’t skimp on input protection 
and don’t count on the fact that the 12v PS is always going to be perfect.  If 
the clock can be designed to run on 5v and uses the 6300HV or 8200HV 
converters, you are better off using 5v because there is more margin for safety 
than using 12v.  If you need to have level conversions and need 12v, then 
protect yourself with some filtering and a TVS.

 

Just my thoughts…

 

Jeff

 

 

 

From: [email protected] <[email protected]> On Behalf Of 
Paul Andrews
Sent: Saturday, February 21, 2026 10:03 AM
To: neonixie-l <[email protected]>
Subject: Re: [neonixie-l] HV Power Supply failues

 

I use my own designed converters and it’s real nice knowing how the supply 
works so you can diagnose issues like this. I’ve used the design in a lot of 
clocks and never had a problem except in one clock which, after a while, 
started showing signs of distress - tubes not being full brightness, 
flickering, noise etc. my converter can be driven with anything from 3V to 24V, 
but this one was using an apple 5V USB adapter rated at 2.1A. Turned out the 
adapter was flaking out on me, which was a relief. I didn’t want to discover an 
issue with my design! Anyway, the moral is: Don’t trust the adapters, but also 
start the diagnosis with the stuff that is easy to change before you open up 
the case and get in to the electronics.

On Friday, February 20, 2026 at 7:03:36 AM UTC-8 JBro63 wrote:

What is the brand of the power supply you are using for this clock? 

Was the same wall wart used during both failures? 

This was from an old BT router I'd hoarded I think, 12v 1.2A, branded I.T.E. It 
was the same PSU in both failures. Output is 13.8v so lesson learned not to 
make assumptions about these things and to check in future before use.

Is it the same model of wall wart as your other clocks are using?

No, others use mix of 12v 1A for the IN-12s (4 & 6 tube versions) and 12v 2A 
for an IN-18 (6 tube version) - bought new recently unbranded from Amazon. I've 
measured these and they're ok for 12v. Not checked for spikes yet. The IN-18 
uses the bigger Omnixie 6300.

Do you run another clock which has exactly the same PCB and design as the 
device of concern?

Yes, I've had a build frenzy recently and have 5 identical IN-12 / K155ID1 / 
8200 models. It's 2 of these that I've had the failure on - both were powered 
by the exBT PSU.



Do you own an oscilloscope to look at the 12 V rail?

Yes, a very old one. I have recently bought new probes so will be a good excuse 
to fire it up / learn how to use it. Should this be measured under load?


I've added the capacitors as suggested above. Still awaiting the TVS diodes. 
Luckily there is enough room in the cases for them to be added unseen. One 
other benefit of the capacitors has been reduction in high pitched whine the HV 
boards emit.

Thanks for the tip the PSUs. Farnell stock them

Thanks again all.

On Thursday, 19 February 2026 at 08:12:55 UTC Tomasz Kowalczyk wrote:

What is the brand of the power supply you are using for this clock? 

Was the same wall wart used during both failures? 

Is it the same model of wall wart as your other clocks are using?

Do you run another clock which has exactly the same PCB and design as the 
device of concern?

Do you own an oscilloscope to look at the 12 V rail?

 

I suspect the issue is just failing/poor quality wall wart. I've seen cheap 12 
V LED power supplies create spikes exceeding 16 V. Some time later the 16 V 
rated capacitor inside the PSU exploded. In your case, the NCH8200HV gave in 
first, for one reason and another. The absurdally large capacitor the designer 
recommended is meant to filter any garbage poor quality wall warts generate, 
but I think it's smarter to just invest in reputable wall warts and skip all of 
this nonsense altogether.  My recommendation are Mean Well power supplies, they 
are not only reliable, but also electrically quiet. I used mostly GSM series. 

poniedziałek, 16 lutego 2026 o 15:43:11 UTC+1 Richard Scales napisał(a):

I've just tried one of those and for sure it works though I don't think that 
USB-C with PD capability is as ubiquitous as some suggest. 

For myself I prefer to distribute 12v from  beefy 10A supplies to individual 
clocks but that's only really because of the number I like to display. 

For regular folk I can see that  USB-C-PD could work well. 

 

- Richard

 

On Mon, 16 Feb 2026, 14:23 David Pye, <[email protected] 
<mailto:[email protected]> > wrote:

I'm moving to try to power mine from usb-c as there are a range of 
autonegotiation boards that can request supply voltages from 5-24volt from a 
suitable PD PSU.

 

David

 

On Mon, 16 Feb 2026, 14:08 newxito, <[email protected]> wrote:

By the way, my attempt to make a similarly flat nixie power supply failed 
miserably. I tried using the CJ5143-ALC flyback transformer which is 4.6mm 
high. It worked somehow but I was never happy with my design. 

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected] <mailto:[email protected]> .
To view this discussion, visit 
https://groups.google.com/d/msgid/neonixie-l/2976622b-1c50-4cf0-804a-6a3d37c95346n%40googlegroups.com
 
<https://groups.google.com/d/msgid/neonixie-l/2976622b-1c50-4cf0-804a-6a3d37c95346n%40googlegroups.com?utm_medium=email&utm_source=footer>
 .

-- 

You received this message because you are subscribed to a topic in the Google 
Groups "neonixie-l" group.
To unsubscribe from this topic, visit 
https://groups.google.com/d/topic/neonixie-l/myKhY_ZZl4Y/unsubscribe.
To unsubscribe from this group and all its topics, send an email to 
[email protected] <mailto:[email protected]> .
To view this discussion, visit 
https://groups.google.com/d/msgid/neonixie-l/CAOQ6x0F8Gm6prhsG6CYEdEh8NOi-dS94quk6-D%2BXh%2BH_jCsXkg%40mail.gmail.com
 
<https://groups.google.com/d/msgid/neonixie-l/CAOQ6x0F8Gm6prhsG6CYEdEh8NOi-dS94quk6-D%2BXh%2BH_jCsXkg%40mail.gmail.com?utm_medium=email&utm_source=footer>
 .

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected] 
<mailto:[email protected]> .
To view this discussion, visit 
https://groups.google.com/d/msgid/neonixie-l/9c3c0632-c473-4b38-8dee-fa84d8166e71n%40googlegroups.com
 
<https://groups.google.com/d/msgid/neonixie-l/9c3c0632-c473-4b38-8dee-fa84d8166e71n%40googlegroups.com?utm_medium=email&utm_source=footer>
 .

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion, visit 
https://groups.google.com/d/msgid/neonixie-l/010701dca380%247fa2f680%247ee8e380%24%40gmail.com.

Reply via email to