Before we look at preparing the Arduino/Freeduino microcontroller for connection to the nixie tube driver modules described in Part I and Part II, you can build this power supply to provide the high firing voltage required by the nixie tubes. This switch mode power supply easily outputs 50 mA, which is more than most, and offers a variable output from 150 to 220 VDC, when driven by a 9 to 16 VDC source.

Step 1: About the Circuit

A 12 volt source at one amp will easily drive this nixie tube supply. There is sufficient power produced by this switch-mode supply to drive at least eight of the nixie tube driver modules (I've had 12 of the nixie tube driver modules running off of one of these boards, that's 24 IN-12A nixie tubes!).

A typical nixie tube power supply offers 170 to 250 VDC at 10 to 50 mA. A switch-mode power supply is desirable because it's small and very efficient. You can fit it inside your clock and it won't heat up. The schematic for the project is taken directly from the MAX1771 datasheet, however, because of the large voltage jump from input to output, board layout and low ESR type components are critical.

Step 2: Parts List

The Following are Digi-Key Part numbers for all components:


CAP CER .1UF 25V Y5V 0805
C2, C3



CAP .1UF 250V PEN FILM 2420 5%

J1, J2, J3


RES 10.0K OHM 1/8W 1% 0805 SMD

RES 1.5M OHM 1W 5% 2512 SMD

RESISTOR .050 OHM 1W 1% 2512





Step 3: Preparing Parts for the Printed Circuit Board

These parts I Ieave to solder conventionally after I've got all the smaller surface mounts parts on the board.

Step 4: Oven Soldering

Here are the smaller parts that we'll apply to the printed circuit board with solder paste, and then toast in our oven.

Step 5: Solder Paste

Get with the gooey stuff. Pull the solder paste out of your fridge and give it a chance to warm up. Then it's not so stiff when you try to force it out of the tube. The best part is that if your board has a good solder mask, you don't have to be quite so precise. Once the paste hits the oven it will flow to just where you want it (most of the time - see step 9).

Step 6: Solder Paste Application

Settle in and hold the caffeine because you need steady hands for this work. Put your thumb over the plunger and gently squeeze the paste onto the pads. Don't worry so much if you're not always on the mark. Excess paste will clog fine pitch parts, so go easy.

Step 7: Pre-Heat Oven

Once you know where the components go, it's quick to apply this amount of paste to a small board. This is about the right amount of paste for successful toasting. Get out your pick-up tool and lay on the SMDs.

Step 8: Seat Components Into the Paste - and Toast

The solder paste used here is lead-free, and though it looks dull and murky now, just wait until it pops up in the oven. The standard issue toaster oven I'm using I got for $20. It's got 3/8" wide quartz heaters above and below the oven rack. I can toast six of these boards at a time. Here's the temperature curve you'd like to adhere to:

Preheat your oven to 200 deg F

1. insert the board into the oven and hold it at 200 deg F for 4 minutes
2. Bring temperature up to 325 deg F for 2 minutes
3. Hold at 450 deg F for about 30 seconds until solder pops up, then wait 30 more seconds
4. Tap the side of the oven, and drop temperature to 300 deg F for 1 minute
5. Allow to cool, but not too fast. You don't want to thermally shock the components.

Step 9: Post-Toast Inspection

After the board is cool, examine it for shifted parts and solder bridges. You can see some beads of solder in places where they might get into trouble. Gently tap them away and off the board.

Uh oh. It looks like we've got two solder bridges on the right side of the 8-pin IC.

Step 10: Solder Wick Is Your Friend

Here's where the truly deft work occurs. Fan open the end of the braided solder wick mesh so that it will pick up molten solder. Place it over the solder-bridged location, and press down with a hot iron. Apply heat for no more than 5 to 7 seconds. This is usually all you need to do to remove the solder bridge. If it doesn't work for you the first time, maybe try approaching the board from a different angle.

Step 11: Solder Remaining Components to the Printed Circuit Board

Ok, pull up to your solder station, and locate the components set aside in Step 3. The MOSFET is static sensitive so don't run across the carpet with this one. We're almost done.

The two solder bridges on the step-up converter have been removed with the solder wick, and the board is now complete.

Step 12: Connecting HV Power to Nixie Tube Driver Modules

If you are connecting this high voltage nixie tube power supply to a nixie tube driver module, here is a simple test set-up. Refer to the markings beside the green terminals on the printed circuit board.

For main PWR input voltages supplied to the nixie tube power supply that are lower than 15 volts DC, you can connect the PWR and Vcc terminals together.

For main PWR input voltages supplied to the nixie tube power supply that are higher than 15 volts DC, you will need to insert a regulator (7812) to provide 12 volts DC to the Vcc terminal.

If using a 12 volt AC adapter, for example, the PWR terminal and Vcc terminal should be connected with a short jumper wire. For normal operation, also connect the Shdn terminal to GND with a jumper wire. This will enable the nixie tube power supply to produce an output when input power is supplied.

Step 13: Power Input Pins

The HV+ and HV- labels on the nixie tube power supply correspond with HV and gnd on the nixie tube driver module.

The HV- lead connects to pin 1 of SV1 (gnd), and the HV lead connects to pin 4 of SV1. For SV1 and SV4, pins 1, 2, 5, and 6 are all connected to gnd.

Only pins 3 and 4 of SV1 and SV2 carry the high voltage required by the nixie tubes.

Step 14: High Voltage Threading Throughout the Modules

Now that you have power supplied to the nixie tube driver modules, you should see all the elements in both nixie tube digits illuminated.

Use caution not to touch the high voltage output to the nixie tube driver modules. There is potentially enough energy here to cause a severe shock.

When nixie tube driver modules are connected edge-to-edge, left-to-right, both high voltage power and serial data from the external microcontroller are threaded through to all the boards.

A microcontroller is required to take full advantage of the nixie tube driver module shift register chain. The nixie tube driver module permits a microcontroller ( Arduino, etc.) to address two nixie tube digits, and via this shift register chain, multiple pairs of nixie tube digits.

For an example of how the nixie tube driver modules may be supported by an external microcontroller, see the sample Arduino digits driver code. Multiple nixie tube driver modules are seen operating together in the nixie tube driver module movie.

Depending how brightly you wish your nixie tubes to be illuminated, you can adjust VR1 to generate output between 170 and 250 volts DC. Increasing the output power will also allow you to drive more nixie tubes simultaneously.

Stay tuned for Part IV, where we'll hook up an Arduino Diecimila, and make some very long numbers. Extra special thanks to Nick de Smith.

See also this nice bit of work by Marc Pelletreau.

<p>hi, can you upload the board files?</p>
radiation markings. were.<br> <br> http://www.sphere.bc.ca/test/nixies.html Krypton85 was used.
A toaster oven? Hmmm? I remember using a very dainty weller iron, and now I have the hot air rig. I may start using SMDs a lot more. Bush took away all the WMDs. ROTFL Seriously the through hole is OK, and I like it for my guitar stomp boxes and amps for ruggedness, but SMDs are a whole lot faster. A neighbor works at a board etching facility. I usually forgo any solder mask and just have him etch stuff. Thanks for the tip(s)! I have some Boroughs Nixies and do want to make a clock or thermometer, or both. I used a tranny from an old Fisher tube tuner that ran 155V CT. How do you do the time keeping and display parts? These Buroughs were to be used at a Nuke power plant, and have the radiation symbol on them. By the time the plant was finished 7 segment LEDs had replaced the NIXIE tubes. I can probably still get them new for $4 each, maybe $5 or $6 now.
Hi Satchmoeddie,<br> <br> <a href="http://www.ogilumen.com/arduino-c-58.html" rel="nofollow">Microcontrollers</a> make everybody happy. You can try some <a href="http://www.ogilumen.com/arduino-c-58.html" rel="nofollow">Arduino</a><a href="http://www.ogilumen.com/nixie-2xd2x-p-92.html" rel="nofollow"> flavour solutions</a>, and drop in some <a href="http://www.ogilumen.com/pub/fileroom/001/drivermux.zip" rel="nofollow">code</a>.
Thanks! I saw some people referencing power supply options. I got a Digitech guitar synth stomp box, midi in/out/thru, and while I was at the thrift store tried buying the missing power supply. Anymore the low volt DC stuff is rectified on the mains/line side and then puts out whatever the mfgr wants using a regulator on the secondary side so the mains can run from 90 VAC to 260 VAC, After that an automotive inverter or neon/florescent transformer can be used. I just dropped the supply voltage to I believe 4.5 VDC and got a nice supply for NIXIE tubes. It replaced the dead one in an old Freq counter/clock my 2nd cousin once used. He was the engineer in charge of keeping the Atomichron at a stable temperature back in the 50s-60s.I got a lot of neat stuff from that estate that everyone else deemed as obsolete &quot;junk&quot;, including lots of engraved base 300B tubes, 350Bs, and other WECo stuff. Since the primary coils that ran just the display on the old freq counter were fried and it was designed to work anywhere I opted for the DC Wall wart and an step up from a low DCV. Jim used to work on the clock in France too. All his stuff was universal mains power.
What if, anything would have to change to allow for voltage in the range of 80 - 220 VDC. Also very well done.
Hello Sealman,<br>Much has gone in to designing this SMPS circuit so that it operates over its range of 150-220VDC. Due to the nature of the way SMPS supplies produce their power, it would be quite a challenge for it to both boost power and operate over that broad of an output range. You might get a solid 80-150 VDC, but it would involve completely redesigning the circuit.
Thank You for your quick reply. I had already decided that it was going to take a lot of change. Thank you! Great work
Any chance we can get the eagle cad layout of this board like with the other 2? or can we still order one from you guys??? Any help would be apreciated.
If you look closely, it's included in <a href="http://www.desmith.com/NMdS/Electronics/NixiePSU/MAX%201771%20V4%20Eagle%20files.zip">the Instructable</a>.<br> <br>
Awesome! found it! Thanks :)
Is the PCB available for purchase somewhere or does anyone know of a through-hole equivalent for the&nbsp; Maxim 1771? Thanks.<br />
The DIP version of the Maxim 1771 is available <a href="http://avnetexpress.avnet.com/store/em/EMController/DC-to-DC-Controller/Maxim-Integrated-Products/MAX1771CPA/_/R-7405766/A-7405766/An-0?action=part&amp;catalogId=500201&amp;langId=-1&amp;storeId=500201&amp;listIndex=-1" rel="nofollow">here</a>, or <a href="http://www.newark.com/jsp/search/productdetail.jsp?SKU=99K3053&amp;CMP=AFC-OP&amp;CMP=AFC-OP" rel="nofollow">here</a>.&nbsp;<br />
Wow!<br /> <br /> Helluva quick response.<br /> <br /> Thank you so much.<br />
The original author of this circuit (my dad) designed it in 2003/4.<br/><br/>See: <a rel="nofollow" href="http://www.desmith.net/NMdS/Electronics/NixiePSU.html">http://www.desmith.net/NMdS/Electronics/NixiePSU.html</a><br/>For full details.<br/>
Credit to your father's circuit is in the post. &nbsp;I did ask him if it was okay to offer his kit for sale, and he did confirm that this was okay.
Too true Robbie. I&nbsp;know your dad and he's a really cool and generous person. Ogilumen also copied your father's nixie power supply design.<br /> I do think people should give credit where credit is due.<br />
sweet tell him i said thankx for the circut
Would this design be suitable for driving the Nixie IV-18 VFD tubes? I haven't worked with any of the Nixie Tubes yet, so forgive my ignorance. Here are the specs I got for the IV-18 VFD tubes: • Cathode voltage: 5V • Working Cathode current: 85 mA • Segment voltage: 35-50V • Working segment current: 8 mA • Grid current for single position: 11 mA • Grid pulse voltage: 35-50V • Grid pulse duty cycle: 10 • Max reverse grid bias: -7V • Nominal reverse grid bias: -5V
There's a really great resource on these tubes <a rel="nofollow" href="http://web.jfet.org/inGrid/">here</a>. Note that the maximum anode voltage is 50 volts only. The HV power supply in this Instructable operates between 150 and 220 VDC.<br/><br/>
Thanks. I thought that might be the case. Thanks for the great link. It's super helpful. Looks like playing with vacuum tubes is a big step up from playing with LEDs.... :)
Such a good device I would make it (if i had a Nixie tube) Awesome 5 stars
Awesome, awesome, awesome. Thank you!
yay high voltage!

About This Instructable




Add instructable to: