Re: [neonixie-l] Not-Nixie-Related: Unusual Discrete Component LED Clock.

[Arne Rossius]

Hi,

5-ht wrote:
> See: http://techno-logic-art.com/clock.htm

That's some very nice work, and incredibly clean construction. I can't
even begin to imagine how long it must have taken to build this.

I'm quite surprised by the total absence of capacitors though. I assume
it uses a power supply that isn't shown, which should at least have a
filter cap, but this amount of digital logic without any decoupling
capacitors still doesn't seem right to me.

There is another discrete component clock here:
http://www.transistorclock.com/
It's a more straightforward design on a PCB, and it uses plenty of
capacitors, mostly for coupling the counting stages. Full schematics
with explanations are available in the assembly manual.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/549D5BD2.5060708%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Nick Desmith smps board question

[Arne Rossius]

Hi,

RHC wrote:
> http://desmith.net/NMdS/Electronics/NixiePSU.html
> with the exception that its a through hole board rather than smd.

Since your friend apparently didn't use the exact same PCB layout, I
think some photos of your board might be helpful as well to diagnose the
problem. With the MAX1771, it is very important that the ground
connection between the IC and the current sense resistor is very short,
and that little current flows through it. It's a good idea to feed the
input ground to the sense resistor first, and then add another trace
directly from the resistor to the IC. The same goes for the CS pin
connection, it should be connected as close to the resistor as possible.
Are you sure that your circuit actually has a 50 Milliohm resistor? I
haven't often seen values this small as through-hole components.

>  Im using a 9v 3amp wall wart supply. 

Have you tried using a higher input voltage, e.g. 12V or 15V?


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/549C3677.7010501%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: Nixie Clock Cool Project- a few questions

[Arne Rossius]

Hi,

Dman777 wrote:
> Using my camera and zooming in, I see some strange markings on PCB board 
> surface and some brown/burn spots also. Could this have anything to do with 
> the heat concentration? What are these from? I attached a couple of 
> pictures.

The brown gooey-looking stuff is probably flux residue from soldering
the component (which is on the other side). The component warms up the
flux residue and it turns from transparent or yellow-ish to brown. This
is nothing to be worried about, but if you want to get rid of it you can
clean it off with some isopropanol.

By the way, is that the footprint of the MOSFET? If unsure, take a
picture of the same spot from the other side of the board. This would
confirm my suspicion that it is the main heat source.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/54388D79.5030402%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Nixie Clock Cool Project- a few questions

[Arne Rossius]

Hi,

Dman777 wrote:
> 4) I don't understand why, the heat concentration is in the dead center of 
> the clock, but the source of what I think is the source of the heat...the 
> 7805 voltage regulator... is of to the side of the source of the heat 
> concentration area. 

The 7805 only powers the electronics, which I expect draws very little
current, so the regulator will remain cool.

I think the heat source might be the MOSFET in the switching power
supply for the nixie voltage. Unfortunately it's the rather common, but
not very good MC34063-based design, you can see the schematic in the kit
instructions here:
. They use
an IRFD220 MOSFET which has way too high R_DS,on for this application IMHO.

If you don't want to replace the MOSFET (which wouldn't look too good as
I don't think you will find a better replacement with the same
footprint), having a fan blow air onto it might indeed be a good idea if
you can stand the noise.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/54354E29.4030209%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] How to digitally adjust output voltage of Taylor Edge HVPS 1363

[Arne Rossius]

Hi,

John Rehwinkel wrote:
> I suspect that wouldn't work the way you want.  I think that resistor 
> parallels
> part of the voltage divider in the feedback loop.  By introducing a separate
> voltage, you would reduce the sensitivity of the feedback loop and thereby
> produce poorer regulation.

Possibly, yes. But it looks like this is an "official" way to change the
output voltage, especially for voltages below 150V, so I expect the
regulation won't get too poor. I don't own one of the modules in
question though, so OP might want to try if it works or not by supplying
an external voltage from a lab supply or similar.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/540CF801.5010508%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] How to digitally adjust output voltage of Taylor Edge HVPS 1363

[Arne Rossius]

Hi,

jb-electronics wrote:
> I want to adjust the output voltage of the Taylor Edge HVPS 1363 [1] 
> digitally via a microcontroller. I drive the HVPS with 12V to have 
> maximum output current.

If I understand it correctly, the output voltage is a function of the
resistor R_Adjust and the voltage it is connected to, V_Radjust. It
seems that chosing the resistor for maximum output voltage (0 Ohms) and
then increasing the voltage decreases the output voltage, so it might be
possible to use a well-filtered PWM output to generate a voltage for
adjustment. For an adjustment range between 150V and 200V, it seems that
the necessary voltage is between 0V (for 200V output) and 1.27V (for
150V output), so well within the limits of what can be generated from a
PWM output. Depending on the input current of the adjust pin, an opamp
buffer might be necessary as well.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/540CE261.6070006%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Nixie Ionisation and Maintaining Voltage

[Arne Rossius]

Hi,

'Alaba Baju' via neonixie-l wrote:
> I understand the IN-1 needs about 170v to light up and about 133v to
> stay on. One thing I've noticed though is that most people only seem
> to build switching supplies for the higher voltage. I don't 
> understand!

The higher voltage (ionization voltage) is the *minimum* voltage your
power supply needs to provide. If the output voltage of your power
supply is lower, the tubes may not light up at all. The lower voltage is
the (approximate) voltage at the tube anode when the tube is on, similar
to an LED's forward voltage. You use a resistor between the voltage
supply and the tube's anode to get these two voltages automatically.
When the tube is off, (almost) no current is flowing so the resistor
doesn't drop any voltage, and the full power supply voltage is available
at the anode. Once the gas ionizes and the tube lights up, it starts
drawing current, thus dropping voltage across the resistor. In effect,
the difference between the two voltages is dropped across the resistor,
and you don't have to do any switching.

A simple way to approximate the resistor value is this:

R = (V_supply - V_sustain) / I

where V_supply is your power supply voltage (may be higher than 170V),
V_sustain is the tube's sustaining voltage (133V) and I is the desired
current (a typical or maximum value can usually be found in the tube's
datasheet).

The formula isn't quite exact because the sustaining voltage will be
dependent on I, but it is good enough to get an idea for the correct
anode resistor value. If you are unsure, start with a high resistance
and measure the current.

If you use a lower current than the datasheet recommends, the tube life
will be prolonged but the digits will be less bright (obviously) or not
light up completely. Some datasheets also list a minimum current for
this reason.

In my nixie clock with four NL-840 tubes, I used a 180V power supply
with 18k anode resistors (one for each tube). The datasheet recommends
10k at 170V and 47k at 250V. From that, we get I = 2.2 mA and
V_sustain = 148V, so with my 18k resistors the current should be about
1.8 mA, just above the minimum value of 1.5 mA and well below the
maximum value of 3.5 mA.


HTH,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53E9306E.3010303%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Accutrace inc aka pcb4u.com

[Arne Rossius]

Hi,

Quixotic Nixotic wrote:
> Has anybody used this outfit for PCBs? Accutrace inc, trading as
> pcb4u.com.
[...]
> I am asking, because they are doing a US $30 + $10 FedEx shipping
> deal (is this US only?) on ten double side boards up to 20 square
> inches, 129 square centimetres. This seems a very attractive price.

I haven't used them, but if you don't mind ordering from China and
waiting a while for your boards, you can get them even cheaper. There
are a few suppliers with similar prices, I have used ITEAD
 several
times. For ten double-sided 10cm x 10cm boards (about 4" x 4") with
green silkscreen, they ask 20 USD + shipping. I have found that shipping
is very cheap (less than 10 USD for 10 boards) if you choose Hongkong
Post, but takes 2-3 weeks (to Germany) in addition to about one week for
manufacturing, and you get (very basic) tracking. As far as I know,
shipping times to the US are similar.

The quality was usually very good, although in one case there was a
board with a clearly visible break in one of the trace, despite ordering
with "100% E-Test". I got 11 boards instead of 10 though.
Also the silkscreen isn't very high resolution and they put an order
code somewhere on the PCB, so if you need a board that *looks* perfect,
this one isn't right for you.

The other cheap chinese manufacturers I know of are




some of which are even cheaper than ITEAD in some cases, but I haven't
used any of those so far.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53E76780.4000505%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: Strange Strobe Behavior on my Nixie Clock

[Arne Rossius]

Hi,

Darin Hensley wrote:
> Here is the diagram I tried to capture with my phone. With this helps any
> I'm trying to see what is causing it?

The PC817 optocouplers are only specified for a collector-emitter
voltage of 35V, so it is very likely one of them is breaking down when
its tube is turned off. It might be a good idea to replace them with
optocouplers that can withstand a higher voltage.

Biasing to 85V won't work with these optocouplers, because
(175V - 85V) = 90V > 35V.

If the pin numbers in the schematic are correct, the optocoupler
transistors are also connected in reverse (pin 3 = emitter to +HV), in
which case the specified voltage is only 6V. However the position of the
pins makes me think that very likely, the pin numbers are incorrect and
the transistor is connected correctly (usually, pin numbers are counted
in counter-clockwise direction around the package, however here it is
1-2-4-3).


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53D3A8A5.5010002%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: O.T.but...

[Arne Rossius]

On 22.07.2014 18:00, Nick wrote:
> Two bits of ply glued at 90 degree to each other.
> At one end, glue a strip of wood at 45 degrees to the other two such
> that one side of the strip intersects with the vertex of the two other
> pieces.

And here's a DIY version of the tool you described (the page is in
German but the pictures should speak for themselves):
http://www.zabex.de/site/helferlein.html#mianvoru

Of course, it is best to use a pen that draws as close to the guiding
edge as possible.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53CE8E3E.10107%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] A use for all those Itron displays I've got in the garage

[Arne Rossius]

Hi,

petehand wrote:
> I can't program the QFP processor chip until 
> it's soldered into the board.

If you really want to do that for a small number of kits, I found that
it can be done by just putting the chip on top of the footprint on the
PCB (otherwise empty, except for the programmer connection) and pushing
it down with your thumb during programming. A gold-plated PCB seems to
work best. I have programmed about 100 controllers this way over the
last few years and didn't have any problems.

Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53AB4B4B.6040405%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: ITS1B Clock

[Arne Rossius]

Hi,

Joseph Bento wrote:
> Your clock is the first practical use I've seen for these tubes. 

There is one other guy who built a clock, but as far as I know there are
no details or schematics available. Here is his video:
http://www.youtube.com/watch?v=ULKdAg0mLhQ

> I would be most curious to know what their intended purpose was - why
> would they design a display tube that had such odd power requirements. 

That has be baffled too. Mine even have a datecode from the year 1990,
so regular 7-segment LED displays were already available for quite a while.

> Perhaps on the plus side, is it's a high-voltage tube that can be
> controlled directly with TTL logic (if I understand other references
> I've read).

Yes, that's correct (although I'm not sure about actual TTL levels, i.e.
2V = high). It certainly works great with 0V/5V CMOS levels.
There is also the latching feature of the thyratrons, so you get the
advantage of few pins needed for a large display without the
disadvantages of actually multiplexing the tubes (flicker, reduced
brightness, increased CPU load).


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53A5A6F0.1060308%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] ITS1B Clock

[Arne Rossius]

Hi,

I saw some talk about the ITS1B thyratron indicator in the archives, so
I thought I'd share my clock design I did a few months back. I got the
tubes from ebay for about 15 Eur each (it looks like they have already
doubled in price by now).

There is a russian datasheet available for these tubes at
http://www.tube-tester.com/sites/nixie/dat_arch/russian-book-0001.djvu
(starting at the bottom of PDF page 211, original page 210). After
having some fun with Google Translate's on-screen keyboard (I don't
speak russian), I figured out that these voltages are required:

1st anode:   40V ±  4V
2nd anode:  100V ± 10V
under-cathode: -250V ± 15V

I also found that the current on the first anode is negative, so
clamping it to ground with a 39V zener diode is sufficient.

To reset the fired segments, the 100V power supply has to be
disconnected for at least 500 us (I used 1 ms). The shortest pulse to
fire a segment (on the 5V CMOS compliant grid inputs) has to be 100 us
(I used 1 ms as well).

As expected, the power supply was the most difficult part of the design.
I used the UC2843 fixed-frequency (PWM) controller IC for the 12V to
100V step-up and connected a voltage multiplier circuit to to switch pin
to generate -300V (since the main output is 100V, it is impossible to
generate the required -250V directly through a multiplier) and used a
resistor + zener diode circuit to drop the voltage down to about 250V.
The current draw on the under-cathode is pretty constant, so just a
resistor with a carefully chosen value probably would have been good
enough, but I wanted to be on the safe side.

Due to the load on the switch pin, the power supply was very unstable,
especially when only drawing current on the -300V output. I finally
managed to get it stable in discontinuous mode, when in continuous mode
there would always be subharmonic oscillation (a few short pulses
followed by a long pulse or vice versa on the gate drive output, causing
audible noise). I fiddled with the operating frequency until I managed
to get the full current without the power supply entering continuous
mode at about 30 kHz. The timing capacitor value is rather critical, if
you use my design make sure to use a low tolerance cap here (definitely
not ceramic) and measure the frequency at the gate drive output. Also
use a beefy enough inductor (my first try with an axial resistor-style
1.5mH, 6.5R inductor failed miserably).

I uploaded my files here:
http://elektronik-kompendium.de/public/arnerossius/temp/its1b/
It's all there: schematic, PCB layout (both as PNG and EAGLE 4.x files),
a scan of the partially translated datasheet, the code for the AVR
microcontroller, some photos and oscilloscope screenshots (in the "oszi"
directory, ch1 = switch pin, ch2 = -300V output).

The AVR program is for use with the German DCF77 radio time transmitter,
but it should be easy to modify the program to be used as a
quartz-controlled clock (using PD5 and PD6 for the set buttons) or even
write a new one from scratch.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53A59EB6.6000402%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Panaplex display on eBay

[Arne Rossius]

Hi,

On 18.06.2014 12:14, Paul Parry wrote:
> Just spotted eBay  item 141318162100 It's a very old Multimeter ending
> today (18th June ) Might be worth a bid as I think it has some old
> Panaplex type display if anyone is interested.

Looks more like incadescent to me ("Minitron" displays), 1x
plus/minus/one and 3x 7-segment, but I couldn't find any info on it
(probably homemade). Those minitrons are hard to find and usually quite
expensive too, so it's still a good find.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/53A1689A.5060209%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Camera Neons

[Arne Rossius]

Hi,

H. Carl Ott wrote:
>  That 200v strike was probably something special order just for use in
> flash circuits to let  the user know the cap was charged. 

They were also used to prevent the cap from overcharging. I wonder how
this is done in the cameras that use an LED indicator.


Best Regards,
Arne

-- 
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+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/5389386C.5050208%40blinkenarea.org.
For more options, visit https://groups.google.com/d/optout.