Transformer for Nixie clock

137 views
Skip to first unread message

Jose Casas

unread,
Jul 13, 2015, 12:18:35 PM7/13/15
to neoni...@googlegroups.com
Hello,

I would like to use a transformer for generating the high voltage needed for the nixie tubes instead of a boost converter approach.

I was thinking to use a high freq. transforcer (because is smaller) and basically rectify the output. But... honestly I was looking a bit around and I can not find anything easy.

I heared that could be possible with a simple audio transformer, but after cheking that I see that most of them are 1:1...

Any idea?

Thanks!

Jose

John Rehwinkel

unread,
Jul 13, 2015, 12:52:43 PM7/13/15
to neoni...@googlegroups.com
> I would like to use a transformer for generating the high voltage needed for the nixie tubes instead of a boost converter approach.

There are a couple of custom 50/60Hz units out there for that purpose that provide a low voltage winding for the logic and a high voltage one for the nixies.

> I was thinking to use a high freq. transforcer (because is smaller) and basically rectify the output. But... honestly I was looking a bit around and I can not find anything easy.
>
> I heared that could be possible with a simple audio transformer, but after cheking that I see that most of them are 1:1...

Look for a 1000Ω:8Ω output transformer. They got used in the old Radio Shack kits with a 1-transistor blocking oscillator to produce high voltage.

Würth Electronics offers some 1:1:1:1:1:1 transformers that can be configured in various ways. Possibly paralleling two windings as the primary,
with the remaining four windings in series as the secondary, would give you some voltage boost with an off-the-shelf part.

For more step-up from commonly available parts, ordinary CCFL transformers offer ratios from 1:50 to 1:100.

One last idea is to get an ordinary inductor with plenty of turns on it to serve as a secondary, and wind a few turns on it to serve as the primary.

Let me know if you have any questions.

- John

Jose Casas

unread,
Jul 13, 2015, 1:05:59 PM7/13/15
to neoni...@googlegroups.com

Hello,

Thanks for your answer. I was thinking more about a high frequency transformer used in the led drivers but in aliexpress or so they don't put any spec. I would like to buy 100 or so therefore parts of old radios or so is not possible. The idea would be go from 5v to 170v@12 mA

Best regards,

Jose

--
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/NchB4jYwR2o/unsubscribe.
To unsubscribe from this group and all its topics, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send an email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/7E8D14B9-9ABD-4099-9809-6E69D60F89BF%40mac.com.
For more options, visit https://groups.google.com/d/optout.

gregebert

unread,
Jul 13, 2015, 2:33:37 PM7/13/15
to neoni...@googlegroups.com
What voltage are you starting from, and what voltage do you want to achieve ? Those are critical parameters for selecting a transformer and designing the converter circuit.

My first nixie clocks have *no* transformer, and use a voltage-doubler. Simple & easy to design. Just be careful about live voltages.
I've also done a tiny DC-DC converter for a wristwatch, and that was a lot of design and debug work.
A good middle-ground if you are doing a custom design is to use a power-transformer with dual primary windings, then use one of the primaries for your HV supply.

In the US, the 120V line-voltage will produce +170V when rectified & filtered. This is a bit low for reliable nixie firing over the lifetime of the tube, especially if you use a dropping resistor instead of a constant-current driver. Using a secondary winding you can 'boost' the HV a bit more, or you can use a voltage-doubler though the latter will waste more energy.


jf...@my-deja.com

unread,
Jul 13, 2015, 5:36:46 PM7/13/15
to neoni...@googlegroups.com
On Monday, July 13, 2015 at 11:33:37 AM UTC-7, gregebert wrote:
A good middle-ground if you are doing a custom design is to use a power-transformer with dual primary windings, then use one of the primaries for your HV supply.
 
In the US, the 120V line-voltage will produce +170V when rectified & filtered. This is a bit low for reliable nixie firing over the lifetime of the tube, especially if you use a dropping resistor instead of a constant-current driver. Using a secondary winding you can 'boost' the HV a bit more, or you can use a voltage-doubler though the latter will waste more energy.
All Electronics (www.allelectronics.com) has a transformer with dual 120V primaries and dual 12V/830ma secondaries (PCTX-1304) for $9.20.  This is a fairly standard combination of windings for companies like Signal Transformer, and they offer other VA ratings (but much more expensive that from surplus suppliers like All Electronics).
 
I agree that the HV from a single primary winding is a little low, but with this transfoermer, you can stack one of the 12V windings to boost the HV, and use the other 12V winding for the rest of the stuff.
 
Disclaimer:  I am an occasional customer of All Electronics and Signal Transformer.

gregebert

unread,
Jul 13, 2015, 8:12:26 PM7/13/15
to neoni...@googlegroups.com
One other thing: When I use one of  the primary windings of a dual-primary transformer for generating HV, I select a transformer with twice the VA-rating I actually need. It might be overkill; my concern was cranking too much current thru the primary-winding and potentially causing overheating.

My reasoning is that for a given VA-rating, a dual-primary transformer should 'share' VA equally across both primary windings, therefore each primary winding normally delivers half of the load VA. When only 1 primary winding is used for pushing energy into the transformer, you can only supply 1/2 the rated VA for the same primary current.

While this isn't strictly correct, because the VA rating is largely determined by the peak flux in the magnetic core, there is another consideration. Most power supplies use a bridge-rectifier --> filter-cap topology for generating DC. With a low-ripple design, the bridge rectifier is 'topping-off' the filter capacitor for a short time during the AC 'peak' and is otherwise off; the load current in the transformer is not continuous like the DC load current (it's a spike). Short-duration spikes have larger RMS current values than the DC load-current. And, it is RMS current that causes I^2R losses in circuits (aka heat), notably the transformer. So, you may want to over-rate your transformer to account for this.

Running a spice simulation is a good way to determine the RMS current; LTspice has a nice built-in utility that calculates the RMS value of an arbitrary waveform. Also, it's a good idea to select a filter cap that has a rated ripple-current well above your RMS current.


Charles MacDonald

unread,
Jul 13, 2015, 8:36:17 PM7/13/15
to neoni...@googlegroups.com


I wonder if the Hammond 261/262 series might be simpler to use..

http://www.hammondmfg.com/261.htm

Several of the 262 units have a 120v secondary, as well as a 6.3 or 12.6
secondary. They were designed for small tube type test equipment and
similar projects..

It can be ordered from most of the distributors, but of course it will
cost more than a surplus unit, but earlier in the tread the requirement
to have many available was mentioned if I am not getting my threads mixed...

--
Charles MacDonald Stittsville Ontario
cm...@zeusprune.ca Just Beyond the Fringe
No Microsoft Products were used in sending this e-mail.

jf...@my-deja.com

unread,
Jul 13, 2015, 9:33:37 PM7/13/15
to neoni...@googlegroups.com
I can think of two other minor advantages for the Hammond transformers, neither of which I would consder deal breakers.  First, the HV secondary is rated at 125V, while using half the primary probably produces about 105-115V, so you might not need to use one of the LV secondaries to boost the HV.  Second, the primary and secondariy windings look like they are on sepeerate bobbins, which increases the isolation.  Also, if you really nee lots of them, you can pick one with just enough VA rating, saving weight and board space.

David Forbes

unread,
Jul 14, 2015, 12:53:19 AM7/14/15
to neoni...@googlegroups.com
Jose,

Designing a high frequency switching power supply for a Nixie clock is
not an easy job, unless you have a lot of experience doing electronics
design.

If your goal is to have high voltage DC power for a Nixie clock, then I
recommend that you buy an assembled power supply.

One maker that is highly regarded by members of this group is John Taylor.

http://www.tayloredge.com/storefront/SmartNixie/PSU/comparison.html
--
David Forbes, Tucson AZ

Reply all
Reply to author
Forward
0 new messages