Ryan, if you would like to understand what happens to a nixie tube
when presented with an unrestricted (by the anode resistor) power
supply, I suggest you pick an old unloved nixie tube, arrange a good
deal of open space, and try it! I also suggest you also have one of
those fire extinguishers designed for electric fires handy, too :).
You'll need it.

Once lit, nixie tubes are basically constant voltage devices. Think of
Zener diodes. An individual nixie tube might run at 135V with a
current of 2mA. Push that current to 4mA and the voltage across the
tube will rise a bit to say 137V. A different sample of the same type
might do this but at 139-141V, instead. So, this makes it very hard to
set a Vht to get the right current for each of six tubes. And if your
Vht is typically 180V, then the attached nixie tubes will draw
WHATEVER current they can get from the supply (tens of mA or more)
until something gives out in an explosive manner.

So basically, that's what the anode resistor is there for. In the
above examples, a value of 22K will ensure that the nixie tube ends up
running at about 2mA, regardless of tube-tube variations. Yes, you
lose a bit of power in the resistor, but that's better than losing
your precious nixie tubes...

You can dim nixie tubes to some degree by droping the Vht, but you
will run into other problems of part-lit segments. Even in my direct-
drive designs, I dim the displays by PWM'ing them - all on for 20% of
the time, then all off for 80%.

The 555 is a good general purpose timer, but it was never DESIGNED to
be an SMPS device - or more particularly, it was never designed to
switch power MOSFETS quickly and cleanly, in the way that most
specialised SMPS ICs are. And I dare say that it is as easy to design
an unstable 555 SMPS as it is easy to design a good MAX1771 one. Just
keep everything short, especially high current and feedback wires/
tracks.

To summarise the Three Neurons' advice, the ideal MOSFET would have
infinite V(ds), zero R(on) and zero Q(g). For reasons that escapes me,
the components industry doesn't seem to have come up with one of these
yet (!), and any current real device is a comprimise. Buy a couple of
each type, and experiment.  I personally favour a range of mosfets in
plastic cases from Toshiba.

Also, remember inductor losses, too. A good quality inductor that can
handle the PEAK (not average!) current at the top of the ramp will be
required.

Laurence




On Aug 7, 7:44 am, Ryan McDonald <[email protected]> wrote:
> Well, I was going to build the 555 using the plans 
> atwww.ledsales.com.au/kits/nixie_supply.pdfwhich calls for a IRF740.  I
> noticed that the MAX1771 plans always call for a IRF830 or a
> IRF644PBF.   If I build the 555 would I be better off swapping the
> IRF740 for a IFR830 or IRF644PBF?  Or are the 555 plans tuned to the
> IRF740.   I'm absolutely not against building with the 1771, but I
> guess Im scared of reports of its instablity if not done *exactly*
> right.  I do understand the fundamentals of a switchmode supply but I
> do not understand them well enough to debug and tune them with a
> scope.  The 555 seems more "forgiving".  Does anyone know what the
> output specs are of the 555 design are anyway?  I've never seen them
> stated.
>
> Is it correct to say that the only way to purposely dim the tubes is
> to use PWM on the tubes themselves (and not use PWM in the HVPS to
> reduce V or C)?   I totally get uC's and digital logic but V/C
> considerations are not my strength.  Am I correct in saying that a
> tube will attempt to draw too much current if directly connected to
> the HVPS (even if the HVPS is running at the correct V) and the anode
> resistor prevent this (while also requiring a V adjustment to get the
> V *and* C correct simultaneously?)
>
> I completely understand the IN-14 datasheet when it says the current
> is 2ma for the tubes in direct mode, but can someone check my head
> about what the datasheet means when it says "Average current for
> digits 0.7 - 1.5mA" and "Pulse current for digits 7 – 13mA".  I take
> this to mean literally what it says (during the "on time" of a mux,
> current is 7-13 and the average current is .7 to 1.5).  In mux mode,
> how would I go about measuring the current using a scope (I have a
> fairly decent 20mhz analog scope I use for logic analysis) and/or
> multimeter.   Why is it ok to drive a mux at 7 to 13 ma "pulse
> current", when the tube is rated at no more than 3.5ma?  I do
> understand the duty cycle aspect here but it still seems that during
> the "on time" of the mux, we are nailing the tube VERY hard (well
> above it's direct drive specs).  I'm concerned that my ignorance about
> V/C is going to cause me to run my tubes too hard and cause sputter/
> poisoning.
>
> I would really appreciate some of you veterans helping about a
> newb.  :)  BTW, I will post my BASCOM-WWVB code soon.  It turned out
> really nice!
>
> -Ryan

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To post to this group, send an email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/neonixie-l?hl=en-GB.

Reply via email to