CRT "Scope Clock" - Power Supply Question

[Tom Katt]

Hello all - long time lurker, new member here.  I've built several Nixie clocks over the years and am now looking at building a crt based 'Scope Clock'.   I've built a low voltage clock that hooks up to an old crt scope and now I'm looking into building a dedicated clock with a 3RP1A crt tube I have on hand.  But I don't know the best way to go about powering the crt - I understand that I need about 1KVDC, but searching the web it seems like some designs have a negative high voltage for the cathode end and others have a positive high voltage at the anode / deflection end.   I'm trying to understand why one approach is preferrable to the other.

Any thoughts or suggestions would be appreciated!

Thanks!

[gregebert]

It has to do with the voltage difference between the deflection plates and the anode or cathode. From what I recall, the max voltage difference between the deflection plates and the anode is only a few hundred volts (much less than the anode-to-cathode voltage to produce a visible beam), so that leaves 2 options:

1. Make the cathode about 1.5kV negative, so that the deflection plates are driven at voltages closer to 0 volts, and drive the anode at a few hundred volts positive.

2. Keep the cathode at GND, drive the anode around 2kV, and make the deflection plate drivers capable of operating at a much higher voltage. You basically need a level-shifter.

From a circuit perspective, #1 is generally easier to implement.

David Forbes (on this forum) has a very good scope-clock design.

[Mac Doktor]

On Oct 27, 2025, at 3:19 PM, gregebert <greg...@hotmail.com> wrote:

It has to do with the voltage difference between the deflection plates and the anode or cathode. From what I recall, the max voltage difference between the deflection plates and the anode is only a few hundred volts (much less than the anode-to-cathode voltage to produce a visible beam), so that leaves 2 options:

Fortunately, I have too many other things to do and spend money on so a scope clock is out for me. I don't want to go near those kind of voltages on a workbench as I have no practical experience to fall back on. More power to those who can—safely.

Note that over at the GeigerCounters group talking about building negative HV power supplies can get you banned. A couple of people found that out the hard way after having been warned. The -900V supplies in Geiger counters are grandfathered in, of course, and the currents are very low.

Terry Bowman, KA4HJH
"The Mac Doctor"

https://www.astarcloseup.com

"Every kid starts out as a natural-born scientist, and then we beat it out of them. A few trickle through the system with their wonder and enthusiasm for science intact."—Carl Sagan, Psychology Today, 1996

[TomJ]

On Mon, Oct 27, 2025 at 4:31 PM Mac Doktor <themac...@gmail.com> wrote:

Note that over at the GeigerCounters group talking about building negative HV power supplies can get you banned. A couple of people found that out the hard way after having been warned. The -900V supplies in Geiger counters are grandfathered in, of course, and the currents are very low.

Certainly, the voltages involved are no joke and can be lethal.  CRT voltages typically are very low current as well, but I suspect still very dangerous.  Though it seems a bit 

nanny-state' to ban legitimate scientific discussions - you can just as easily discover mortality working on your car or a home appliance.

In any event, I also received a very informative email from the gentleman at Sgitheach...  My understanding is that the crt tube doesn't really care - it comes down how you want to design your circuit.  Having the high potential at the cathode makes the deflection circuit easier to implement because the voltages are lower.  BUT, if you desire to incorporate some kind of intensity modulation or Z axis, you now have to manage that end with high voltages.  The end result from a performance perspective will be the same.  

Thanks for the insight!

[TomJ]

Forgot to mention - if anyone else is considering a crt clock project, brand new Chinese 3RP1A tubes are available from US sellers on eBay (3 for $85 shipped).  But they are listed with the letter "I" instead of the numeral "1" - the eBay search doesn't pick up on that, so they are easily overlooked.  I think these are the same ones that Sphere used to sell.

[TomJ]

Yet one more topic I neglected to address - voltage regulation.  Probably most critical for the deflection circuits.  The majority of examples I find on the web based on simple tube transformer based supplies don't appear to include any kind of regulation (I'd guess switched type supplies are regulated).  But many of the examples of basic oscilloscope type gear implement zeners for the approx +300VDC anode / deflection supply.  

[Instrument Resources of America]

Cautionary note here. MANY older oscilloscopes generated their CRT voltages directly from the A.C. mains connected transformer. ""NOT"" low current. I just finished restoring a Heathkit oscilloscope from not that long ago, where this was implemented.

Ira.

--
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 neonixie-l+...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CAArLthftPeVbaOPxVxHyABmj-YnVNRECZjYp-TCTjGkv6Ua86g%40mail.gmail.com.

[Instrument Resources of America]

In a scope clock it is probably not necessary to regulate the H.V.  In a 'calibrated, laboratory type of oscilloscope', (you can think Tektronix, and others here) it becomes mandatory that the H.V. is regulated, since the vertical and horizontal deflection factors will be affected by wandering H.V. accelerating potentials.  The higher the accelerating potential, the more difficult it becomes to deflect the beam in both the X and Y axes.

Ira.

--

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 neonixie-l+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CAArLthdWyGPOcwfqSukypXgCHOr2e3mBGwqTkK9w4ZXTN4ENtw%40mail.gmail.com.

[TomJ]

On Tue, Oct 28, 2025 at 11:40 AM Instrument Resources of America <IRACO...@hughes.net> wrote:

Cautionary note here. MANY older oscilloscopes generated their CRT voltages directly from the A.C. mains connected transformer. ""NOT"" low current. I just finished restoring a Heathkit oscilloscope from not that long ago, where this was implemented.

Yes, true.  Just like many old tube radios like the All American 5.  ANY tube gear needs to be treated with respect. 

[padler...@gmail.com]

Tom,

Can you post a link to the eBay listing for the 3RP1A tubes?  I was not able to find it searching with the “I” or “1”, looking in both current and completed listings.

Thanks,

Randy

On Oct 28, 2025, at 6:15 AM, TomJ <tomk...@gmail.com> wrote:



--

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 neonixie-l+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CAArLthczeXYLQnthL6gLGVEt%3DsTB1P%2B%2BfOzE-bRh82g2gZEmtA%40mail.gmail.com.

[Tom Katt]

They are hard to find for some reason lol.  I stumbled upon them by accident even though I was searching for 3RP1's.   Mine just arrived today - well packaged and factory sealed boxes.  I haven't hooked them up yet, but they are obviously new and look nice so I do not expect any issues.

LOT OF 3 Oscilloscope Vacuum Tube 3RPI-A | eBay

[Instrument Resources of America]

Pilot Radio Company made a three inch black and white television in the late 1940's early 1950's. It was the first television to break the $100.00 price tag, on the low side. It's CRT IIRC was a 3RP4 with P4 phosphor. These sets are highly prized by collectors, including myself. I have one currently under restoration, with a good CRT. Unfortunately many of these sets do NOT have good CRT's. and the folks who have them have resorted to using the 3RP1's  and this of course is adding to their shortage. Making clocks with them is also adding to their shortage, along with the collector crowd.

Ira.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/8bbf7557-7e62-4da8-a6cc-543318d4c00an%40googlegroups.com.

[J Forbes]

there are an awful lot of P1 CRTs in small sizes for sale on ebay, and laying around in guys workshops. I have at least 3 in the room I'm sitting in. I don't expect there will be a shortage of them any time soon. P4 tubes in small sizes? sure, they're hard to find.

[Tom Katt]

It's the same story with any vintage gear that finds a resurgent popularity...  I built my first Nixie clock before they became 'fashionable' back when old test gear was pennies on the dollar.  I think I got a Systron Donner counter timer for about $30 from a surplus shop with a compliment of 6 Burroughs 8442 nixies.  And I distinctly remember surplus stores selling the big Z568's and NL7037's for a few dollars each - now try to even find a dead one.  It's just the way things go.

That said - I suspect there is still plenty of inventory of small crt's from the Eastern Block and China.  The 3RP1A's I just bought feel like NOS where someone imported a pallet for resale.

[Tom Katt]

On Wednesday, October 29, 2025 at 9:04:16 PM UTC-4 Instrument Resources of America wrote:

Pilot Radio Company made a three inch black and white television in the late 1940's early 1950's. It was the first television to break the $100.00 price tag, on the low side. It's CRT IIRC was a 3RP4 with P4 phosphor. These sets are highly prized by collectors, including myself. I have one currently under restoration, with a good CRT. Unfortunately many of these sets do NOT have good CRT's. and the folks who have them have resorted to using the 3RP1's  and this of course is adding to their shortage. Making clocks with them is also adding to their shortage, along with the collector crowd.

Ira.

Forgot to mention - I have a Waterman S11-A "Pocket Scope" in fantastic working condition (including velvet lined storage carry case!) that dates back to 1948/49...  A compact 5 tube (+ 3MP1 crt) scope that is actually quite versatile for the time period.  Of course it's displaying a clock now or my stereo audio signal depending on the day ;-)  The manual is interesting reading, especially the reference of how to test various signals by observing lissajous curves.  The same model can be seen here: oscilloscope-waterman-s-11-a-s4843 - Oscilloscope Museum

And tech doesn't even have to be that old to become unobtanium - I've got a Teensy 3.6 in a drawer that's been waiting for a project like this...  I went to buy another board and was shocked to discover my $30 purchase now demands nearly $300 on eBay (when they are even listed).  These boards are only about 10 years old - but the controllers they are based on are obsolete.

Back on topic, I'm inclined to try the HV on the cathode end - this makes the deflection drivers a bit easier to manage.  I'm going to borrow heavily from Mr. Forbe's optocoupler Z axis design to handle the high negative voltages on the grid ;-)  I picked up a 500VCT toroid tube power xfrmr from Antek - they are very reasonably priced and made in the US.  I'd like to go with a CCFL type switch mode power design, but my digital skills far outweigh my analog skills - but I can manage a basic rectified power supply lol.

Between the Teensy and an Arduino Due - I have some good experiments planned for these crts.  

And I blame all this on Mr. Forbes lol - I was hooked when I saw that first scope clock way back ;-)  I thank you for sharing your knowledge.   Atari Tempest was my game back in the day, and would you believe that some kid brought in an old Dumont scope for show and tell  - in kindergarten !!!    That day is seared into my brain.  

[J Forbes]

David used to play Galaxia. Tempest was a bit "fast" for us.

[Mac Doktor]

On Oct 30, 2025, at 7:29 AM, Tom Katt <tomk...@gmail.com> wrote:

Atari Tempest was my game back in the day

The best of the bunch. I used to have one but it was too big and underutilized so it had to go. It had the rare early version of the marquee plus the image was square, not pin cushioned.

Paid $65 in working condition, got $500 for it.

Terry Bowman, KA4HJH
"The Mac Doctor"

https://www.astarcloseup.com

"If only you could see what I've seen with your eyes."—Roy Batty, Blade Runner

[Tom Katt]

On Thursday, October 30, 2025 at 1:23:02 PM UTC-4 Mac Doktor wrote:

On Oct 30, 2025, at 7:29 AM, Tom Katt wrote:

Atari Tempest was my game back in the day

The best of the bunch. I used to have one but it was too big and underutilized so it had to go. It had the rare early version of the marquee plus the image was square, not pin cushioned.

Paid $65 in working condition, got $500 for it.

I always yearned for one but alas I never found one I could afford.  I don't play a lot of video games anymore, but I did invest in building a nice arcade quality control panel for a MAME emulator.  Every once in a while I get the urge to spin the knob ;-)  Things are fun until I get to the invisible levels, and then it's chaos lol.

Though I can't really complain - I have stumbled upon more than my share of fantastic deals...  About 20 years ago I saw an advert in the local paper for a _free_ pinball machine - just come get it off their porch.  It was an 80's Bally Centaur - one of the first with synthesized speech!  And can you believe that about $20 in solenoid driver transistors is all it needed?  Between the artwork and the gameplay this pin is still hghly sought after and commands good money.  I'll leave it to a nephew in my will lol. 

And then there's the NYC traffic light (complete with aluminum base, pole and authentic graffiti lol)...  Happened to be driving by construction where they were replacing the old glass signals with leds and asked what happens to the old ones - I found out they fit in the back of an SUV if you position it correctly ;-)

[Dekatron42]

Interesting feedback comment from one buyer of these CRT's:

"Tubes are bright BUT out of focus on the edge typical of 3PR1 A model 3RP1A are not as good as 3RP1.....put them in the oscilloscope then sell it at a ham fest quick"

/Martin

[Nicholas Stock]

I always thought the 'A' designation meant their specs were better....

--

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 neonixie-l+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/9adcf12a-48b8-4121-85e2-d63e639e2349n%40googlegroups.com.

[Adrian Godwin]

A Centaur ! Nice find, especially free :).

--

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 neonixie-l+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/9adcf12a-48b8-4121-85e2-d63e639e2349n%40googlegroups.com.

[Mac Doktor]

On Oct 30, 2025, at 1:40 PM, Tom Katt <tomk...@gmail.com> wrote:

Atari Tempest was my game back in the day

The best of the bunch. I used to have one but it was too big and underutilized so it had to go. It had the rare early version of the marquee plus the image was square, not pin cushioned.

 I have stumbled upon more than my share of fantastic deals...  About 20 years ago I saw an advert in the local paper for a _free_ pinball machine - just come get it off their porch.  It was an 80's Bally Centaur - one of the first with synthesized speech! 

My buddy had Gorgar for a while. The speech board cost extra.

And can you believe that about $20 in solenoid driver transistors is all it needed? 

I had a Bally Lost World, the first one wtih non-traditional artwork. There were spare transistors/SCRs on the logic board so I used them to replace the three that had blown.

Between the artwork and the gameplay this pin is still hghly sought after and commands good money.  I'll leave it to a nephew in my will lol. 

You can always ship it to me. Not that I have room for one. You don't realize how big and loud one is until you have it in your basement.

nd then there's the NYC traffic light (complete with aluminum base, pole and authentic graffiti lol)...  Happened to be driving by construction where they were replacing the old glass signals with leds and asked what happens to the old ones - I found out they fit in the back of an SUV if you position it correctly ;-)

We scored a pile of railroad stuff when Norfolk-Southern ripped out sidings. Signals, cross-buck lights, some really cool relays and lots of selenium rectifiers.

That dwarf ground signal was a lot of fun to carry up a hill.

Are we off-topic yet?

Terry Bowman, KA4HJH
"The Mac Doctor”

https://www.astarcloseup.com

(The notorious thread hijacker)

[Tom Katt]

On Thursday, October 30, 2025 at 3:38:32 PM UTC-4 Mac Doktor wrote:

Are we off-topic yet?

Probably - but a good statistician can tie everything together ;-)

But to get back on topic - does anyone know the original designer or source for the Chinese scope clock kit control boards?  Like most low cost China electronic modules, I'm guessing it's a copy of a controller designed elsewhere.  They seem to have gone through a few iterations, but the current ones look like this (Alibaba example).  They are marked Xiao ni / oscilloscope_driver 4.6 and are a single board power/controller/driver powered by 12VDC with onboard CCFL type xfrmr and a DIP style microcontroller daughterboard.  I know someone who has one but haven't had the opportunity to inspect it closely or identify what mcu its using.  There are a few jumper pads to select various voltages - one set is marked -550V / -850V / -1100V and the other is marked +200 / +300.  No doubt to select cathode and anode/deflection voltages for driving different tubes.  I haven't been able to find any information about them on the web and I don't recall seeing similar boards elsewhere that must be the original designer.  They seem to be running the Dutchtronix firmware and the menu has options for things like GPS and Serial terminal, but there are no connectors for those features on the board and I didn't notice any unpopulated pins on the board for them.  Perhaps they just omitted the hardware but left the firmware in true cheap clone style?

If they're going to steal the firmware, I doubt they're going to invest in designing original hardware lol.  This is probably what happens when someone contracts a Chinese pcb factory to make their boards - and the factory sees an opportunity to make a few extra bucks.  Of course, once one Chinese company starts making them the other companies down the street start making them too.  It's like the wild west of intellectual property over there... 

But the kit seems to work - I've seen the 8S31J crt version and it looks decent.  The crt seems like a 3RP1A copy with a thicker neck and is another component that Google cannot find - no datasheets anywhere.

On Thursday, October 30, 2025 at 3:32:11 PM UTC-4 Nicholas Stock wrote:

I always thought the 'A' designation meant their specs were better....

I always understood the 'A' to indicate the flat face version as opposed to the standard radiused face of the 3RP1. 

[Nicholas Stock]

3BP1A tubes still have a rounded face, so not sure that's correct...

Sent from my iPhone

On Oct 30, 2025, at 13:56, Tom Katt <tomk...@gmail.com> wrote:



--
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 neonixie-l+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/4a6bc79f-ca4b-408f-a87e-7b5ea588c1a5n%40googlegroups.com.

[Tom Katt]

On Thursday, October 30, 2025 at 5:08:04 PM UTC-4 Nicholas Stock wrote:

3BP1A tubes still have a rounded face, so not sure that's correct...

LOL - it appears my 'understanding' is not correct.  Thankfully, I can rely on my browser AI to reveal the facts - the 'A' versions are manufacturing revisions incorporating minor engineering or tolerance changes.  So, something changed and the 'A' version replaced the original.  It's likely there were performance improvements made, though it's also possible the changes were made for manufacturing reasons.   Perhaps the flat face of the 3RP1A was created out of demand for a flat face tube that was interchangeable with designs using the 3RP1.

Regarding the earlier comment about the 3RP1A, I suspect the flat face might result in some optical focusing issues at the edge of the face due to the geometry of the glass at that point.  I've seen several scope clocks with the 3RP1A tubes and never noticed any issues - though the display on a scope clock rarely goes edge to edge, unlike test gear such as an oscilloscope where the trace may extend beyond the edge of the face.

[Tom Katt]

On Thursday, October 30, 2025 at 4:56:46 PM UTC-4 Tom Katt wrote:

If they're going to steal the firmware, I doubt they're going to invest in designing original hardware lol.  This is probably what happens when someone contracts a Chinese pcb factory to make their boards - and the factory sees an opportunity to make a few extra bucks.  Of course, once one Chinese company starts making them the other companies down the street start making them too.  It's like the wild west of intellectual property over there... 

Did more Googling and wanted to correct a likely errant hypothesis - the Dutchtronix firmware is open source, so no issues with anyone copying it or otherwise using it as the foundation for their own project (though I wonder if that applies to commercial sales for profit?).  And while the hardware shares many similarities with other projects documented on the web, I haven't found it to be a straight copy of anyone claiming authorship - and again, it's likely that whatever project was the basis of the Chinese boards the designs are frequently open source as well.  It's possible that some Chinese engineers saw the opportunity to sell kits and created their own boards - too bad it's impossible to find documentation for the current boards.  I'd very much like to examine the deflection design.

I did find a discussion of the (much) older kit versions in this group back in 2020 (CRT clock kit instructions), where member Paolo generously linked the kit documentation here.  This version was a two board design with the controller and power/drivers on separate boards.  This version has a much more elaborate differential amp design that uses multiple transistors in each leg along with what appears to be passives for  frequency compensation.  But it also uses the optcoupler style Z axis as seen in the Cathode Corner design.  I will have to check out my friend's kit because I believe I only counted 4 transistors in the deflection circuit while the older design uses 6.  I'm not sure what dac they are using or what the resolution is - they do not seem to be using the sin/cos circle method but rather the conventional point to point.  

I will try to take some photos of the clock kit and board the next time I get an opportunity.  They seem well laid out and use all surface mount components.

[David]

The 3RP1-A CRT is a flat face tube. It cannot maintain focus across the face as well as a curved face 3RP1 tube because of that. And it’s true that the Chinese tubes tend to defocus at the right side of the screen, possibly due to the electron gun structure not being entirely symmetrical. The H deflection plates are closer to the screen, so they require more voltage per mm of deflection. 

I have adjusted the displays of literally hundreds of these CRTs. They are remarkably consistent in their quirks. 

Also, I don’t have any information about when they were manufactured. 

David Forbes, Tucson AZ

On Oct 30, 2025, at 2:32 PM, Nicholas Stock <nick...@gmail.com> wrote:



To view this discussion, visit https://groups.google.com/d/msgid/neonixie-l/CAOX%2BRHJRKVCvNLsZZ6NXLbqmz06xFQoektrBat4kO27R-2yocw%40mail.gmail.com.

[Tom Katt]

On Friday, October 31, 2025 at 2:05:39 PM UTC-4 David wrote:

The 3RP1-A CRT is a flat face tube. It cannot maintain focus across the face as well as a curved face 3RP1 tube because of that. And it’s true that the Chinese tubes tend to defocus at the right side of the screen, possibly due to the electron gun structure not being entirely symmetrical.

Thanks for the insight!  It may be coincidental, but my friend with the Chinese scope clock told me that he needed to reverse the deflection wiring because the 8SJ31 crt seemed to have something loose or out of alignment, requiring him to rotate the tube 180'.  Apparently, the display was off-center as built and if the tube was rotated the display would move further off center, as though the gun was shifting - and occasionally a "shadow" would develop on the left side that he attributed to elements in the gun 'drooping' depending which side was up,  He found that if he rotated the tube, the display greatly improved.  If you slowly rotated the tube while held horizontally as it would be normally mounted, you could hear slight metallic 'pings' from the gun assembly moving inside the tube.  In fairness, he also said the package showed signs of rough handling during shipping, so it's possible that his issues result from shipping damage.  Still, I found your comment slightly coincidental.

Thankfully, I don't 'hear' anything loose in the 3RP1A tubes I just purchased - my B12 sockets should arrive this week and I'll be able to test their performance.