Nixie bench supply

236 views
Skip to first unread message

gregebert

unread,
Aug 30, 2016, 1:07:57 AM8/30/16
to neonixie-l
I finally finished my bench supply for testing/experimenting with nixies, and I found that having adjustable current is much more useful than a conventional variable-voltage supply.

Right now I'm gathering I-V data for some b7971 tubes I purchased, and being able to adjust the current with a 10-turn pot is really handy. It's 0-500V at 0-20mA on the high range, so it's ample for any nixie I can afford as well as dekatrons.

Well, now I have egg-on-my-face for the mean words I've said about IN-9 linear tubes.  I put my supply on the lower range (0-180V at 0-100mA), and most of my IN-9's checked out OK. They still aren't linear (5mA half-illuminated, ~20mA or more for almost-full-scale), and at least one of them had irregularities when almost fully lit (a dark band appeared). But they definitely worked much better with a good bench supply than they did with the poorly filtered supply I was using. They also behaved monotonically: more mA = longer glowing region.

Lesson learned - You really do need a good power supply for nixie work.

I have some extra PC boards that I'm willing to give away for cost of shipping, but I'll warn you it's not a cheap project and there is some rework needed. Message me if you're a seasoned designer and interested in building one.
hvps_pic1.jpg
100_9237.JPG

Dekatron42

unread,
Aug 30, 2016, 1:47:45 AM8/30/16
to neonixie-l
Can you publish the schematic for your power supply here?

/Martin

gregebert

unread,
Aug 31, 2016, 12:31:29 AM8/31/16
to neonixie-l
Here are the schematics. Be aware they dont include the front-panel items, such as the variac, pots, switches, meters, jacks, etc.


regulator.jpg
lvds.jpg
isolated_supplies.jpg
dcfilter.jpg
switches.jpg

GeckospotNixie

unread,
Aug 31, 2016, 4:48:30 PM8/31/16
to neonixie-l

Gregebert
Just sent you another private reply about the PCBs.
Don't know if you getting them or not.
Google has no way for me to find out if I even sent a private reply.
GeckospotNixie

Tidak Ada

unread,
Aug 31, 2016, 4:54:55 PM8/31/16
to neoni...@googlegroups.com

Please can I get the PCB artwork too?

 

I think the design is interesting enough to give some explanation about the it to the group.

 

Thanks,

 

eric

--
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 post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/f6abd12a-256f-425e-a6eb-1f7641fb9d02%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

gregebert

unread,
Sep 1, 2016, 1:16:15 AM9/1/16
to neonixie-l
I've received requests for a total of 6 boards, which leaves only 1 remaining. I hope to get time this weekend to post private listings on Ebay. Once I get that setup I'll contact all who have shown interest. The price will be $10 US, and covers my costs for shipping/Ebay/paypal.  I'm also working on the documentation as time permits; I'm horribly busy with my day-job so it may take a few days.

I'm posting the gerber files in case you want to view them, or fab on your own; I use gerbv (free) for viewing.
HVPS1_R1P0.tar

GeckospotNixie

unread,
Sep 1, 2016, 5:48:41 AM9/1/16
to neonixie-l
Gregebert
Thanks for the update!
And thank you for your time and giving back to the community.

GeckospotNixie

Tidak Ada

unread,
Sep 1, 2016, 5:15:19 PM9/1/16
to neoni...@googlegroups.com

Hi ,

 

Does this mean I am in the running for a board set? (I hope so).

Extra costs for shipment outside of the US will be paid!

 

eric

--

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 post to this group, send email to neoni...@googlegroups.com.

Jon Jackson

unread,
Sep 1, 2016, 5:35:36 PM9/1/16
to neoni...@googlegroups.com
I am interested in a set of boards. Count me in if there is still 1 set left.

Jon D.

On Thu, Sep 1, 2016 at 3:15 PM, Tidak Ada <off...@zeelandnet.nl> wrote:

Hi ,

 

Does this mean I am in the running for a board set? (I hope so).

Extra costs for shipment outside of the US will be paid!

 

eric

 

Van: neoni...@googlegroups.com [mailto:neonixie-l@googlegroups.com] Namens gregebert
Verzonden: donderdag 1 september 2016 7:16
Aan: neonixie-l
Onderwerp: Re: [neonixie-l] Re: Nixie bench supply

 

I've received requests for a total of 6 boards, which leaves only 1 remaining. I hope to get time this weekend to post private listings on Ebay. Once I get that setup I'll contact all who have shown interest. The price will be $10 US, and covers my costs for shipping/Ebay/paypal.  I'm also working on the documentation as time permits; I'm horribly busy with my day-job so it may take a few days.

 

I'm posting the gerber files in case you want to view them, or fab on your own; I use gerbv (free) for viewing.

--
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+unsubscribe@googlegroups.com.


To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/986a40d0-9c38-4df6-a590-7e41e52777d9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
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+unsubscribe@googlegroups.com.

To post to this group, send email to neoni...@googlegroups.com.

gregebert

unread,
Sep 5, 2016, 12:55:26 AM9/5/16
to neonixie-l
Thanks for all who have shown interest in this project. All of the available boards have sold via Ebay (Eric- I'm working the details on how I can ship international for a reasonable price).

I've setup another google group called HVPS1 User Forum (https://groups.google.com/forum/#!forum/hvps1-user-forum) and ask that all related discussions be moved there.

Dekatron42

unread,
Sep 5, 2016, 2:14:30 AM9/5/16
to neonixie-l
I think it would be a lot better if you kept the questions and answers here on this forum, so please reconsider.

/Martin

Samuel G. Guss

unread,
Sep 5, 2016, 9:20:27 AM9/5/16
to neoni...@googlegroups.com

I second Martin's suggestion!



On 05-Sep-16 02:14, Dekatron42 wrote:
I think it would be a lot better if you kept the questions and answers here on this forum, so please reconsider.

/Martin

On Monday, 5 September 2016 06:55:26 UTC+2, gregebert wrote:
<snip>

jb-electronics

unread,
Sep 5, 2016, 12:52:41 PM9/5/16
to neoni...@googlegroups.com
I agree, it would be very helpful. Perhaps you can agree on have everybody use [HVPS1] in the subject line or something like that. Jens
--
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 post to this group, send email to neoni...@googlegroups.com.

gregebert

unread,
Sep 5, 2016, 4:22:53 PM9/5/16
to neonixie-l
I'll leave the decision to Darwinism: If nobody uses the HVPS1 group, then it will rightfully die a silent death and we'll just use [HVPS1] as suggested.

Dekatron42

unread,
Sep 6, 2016, 5:14:40 PM9/6/16
to neonixie-l
Can you include larger schematics files, I can't read all of the text in the ones you already put here.

I can't see how the constant current design is working, probably because I can't see the text in the schematic or perhaps because I can't see all components since I couldn't find the wiring diagram for components not on the circuit diagram - or it could simply be because I am not good at this type of power supplies.

Does the power supply work with both constant voltage and constant current, or how does it work?

/Martin

gregebert

unread,
Sep 6, 2016, 6:34:55 PM9/6/16
to neonixie-l
The voltage is not regulated, but it is adjustable via a variac. Nixies really dont need regulated voltage; it's the current that controls the brightness. Many common sources of light (incandescent, LED) operate the same way.

The 500V current-limiter (right-half of regulator.ps) uses a non-precise reference voltage ( a 24 volt DC-DC converter that provides isolation) across a pot. The pot is diagrammed in the schematic at the connector P5. The voltage at the wiper of the pot is applied to the gate of a NMOS device. Once the voltage at the gate is high enough to turn the MOSFET on, there is negative feedback across R12+R13. The voltage across R12+R13 is 1V per mA of load-current. The load is connected to DCOUT500 and ACISO1 (the common gnd for the high-voltage section). As more load current is drawn, the voltage at the NMOS source increases, which tends to shut-off the NMOS. Equilibrium is reached and the current is essentially constant. Note that the Vgs of the NMOS device affects the load-current, but it's essentially constant over the range of currents being used (0-20mA). The voltage produced by the DC-DC converter directly affects the output current, but it too is mostly constant and it has a constant load as well.

Though I could have used a 0-20mA meter to monitor the load-current, there were none available for low-cost but there are plenty of 0-20V meters, so that is why I chose that approach.

The preliminary testing I've done so far shows current holds-steady as indicated on the panel meter; at most it dithers only on the lowest digit. For example, when running at 2mA, the current varies only +/- 10uA which is much steadier than a nixie requires.

Some of the 'extra' components are there for filtering, spike protection, ESD, reverse-polarity applied into the power supply, etc (zener diode, R11, C7, D19). They have no functional purpose.

The 0-170V output, which is adjustable from 0-100mA, operates similarly except it has a real ammeter (0-200mA). The 100 ohm / 2Watt resistor provides current-limiting  in case the NMOS has failed-shorted; in normal operation it will 'penalize' the output as much as 10 volts.

Lastly, there is a voltmeter at the DC output jacks that measures the voltage actually delivered to the load; there are relays to switch the meter accordingly.

Dekatron42

unread,
Sep 7, 2016, 9:36:36 AM9/7/16
to neonixie-l
Thank you for the description, now I understand how it works.

Have you considered using depletion mode mosfets as current regulators? The IXYS (Microchip) IXCY10M45S (450V) or IXCY10M90S (900V) are sometimes used as either a pass element before a LDO regulator or as a CCS in audio amplifier designs.

/Martin

gregebert

unread,
Sep 7, 2016, 11:33:50 AM9/7/16
to neonixie-l
Depletion mode.....the last time I did any work with that kind of device was in college, and even then it was circuit-design/simulation rather than real hardware. Unless you include vacuum tubes; I actually did play around with those in high school back in the '70's.

I've stuck with enhancement-mode mainly because most devices operate that way, and partly because enhancement-mode devices are 'off' by default so there's some implied level of safety.

Thanks for providing the link.
Reply all
Reply to author
Forward
0 new messages