Re: [neonixie-l] Push-pull driving cathodes

[David Pye]

Yep.  A lot of designs just use individual HV transistors for each cathode.

I recommend mpsa42 personally as they are ideal for cathode driving and not
expensive.

They come in several different packages including sot23 SMD if you need
small..

David

On Sat, 9 Mar 2019, 15:31 Jasper C.,  wrote:

> I was planning to not multiplex my nixies, probably leaving all cathodes
>> connected directly to the HV line, and placing current liming resistors on
>> the cathodes.  Was interested to see if I'd need to tweak any of the
>> individual resistor values uniform current/brightness.
>>
>
> I only just reread what I wrote there, and realise I made a typo.  I
> should have said I was going to connect all the anodes to the HV line...
>
>
> On Thursday, 7 March 2019 00:41:00 UTC+8, nixiebunny wrote:
>>
>> Another important fact about Nixie tube cathodes: turning on one cathode
>> by pulling it to zero volts causes it to steer all the available current
>> away from the other cathodes, causing them to be dark. This is why you only
>> need a 50V switch on each cathode.
>> The caveat is that if no cathodes are pulled to zero volts, then there
>> will be leakage current flowing through the tube that will destroy a 50V
>> transistor. This is why I used the TD62083 with its set of commutation
>> diodes, and connected the diode common anode pin to a 50V source in my
>> Nixie watch circuit.
>> It's also why I made the blanking mode that drops the anode voltage to
>> 100V, so that the cathodes all remain dark when blanked.
>>
>>
> Ah, well that comes back to the page I referenced in the opening:
> http://www.decodesystems.com/re-how-nixies-work.html.  Maybe that was
> written in reference to having leaky cathode transistors? (With no voltage
> clamp, as in Fig. 4.)  If I had say 300V+ transistors with sufficiently low
> leakage when off (< 10 uA) and a 170V power supply, would I be able to
> blank the tube safely by turning all transistors off?
>
> I'll have another look at power supplies. :)  I was a little concerned
> about how to calculate the value of the loop compensation components.  For
> my first attempt I figured I'd avoid the problem altogether by using one of
> Yan's power supplies, and focus on the segment driving circuitry.
>
>
> --
> 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 email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/7026df09-ef22-408c-9e92-7daa3f5f2db6%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+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/CAOQ6x0He3aJSDhbhb%3DyfW3ZzNeJYouoWX5syECYM2WD1XEOHGw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[David Forbes]

Jasper,

I didn't use any fancy switching power supply knowledge - in fact, I
copied the switching power supply schematic straight out of the LT1308
data sheet!

The clever part was figuring out that I could do things with the
feedback resistor network to achieve current regulation and blanking
voltage control. The power supply adjusts the output voltage as needed
to make 1.2V appear at the feedback node. The rest is just simple
resistor networks, as taught in any intro to electronics class.

On 3/6/19, Jasper C.  wrote:
>> So you can get good results with very little circuitry, if you apply a bit
>> of cleverness to the problem.
>>
>> http://www.cathodecorner.com/nixiewatch/firmware/nwrf-schem.gif
>>
>> The feedback regulated power supply is a very elegant solution, and I'd
> love to implement something like that, but my knowledge of switching power
> supplies isn't good enough to pull it off.  I envy you.
>

-- 
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/CAPbqtvf%3DdwrJuzNmTWMQU3E1mXK3Aw%2Bp9Q-XTTL6kYumDsMR4g%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[David Forbes]

Another important fact about Nixie tube cathodes: turning on one cathode by
pulling it to zero volts causes it to steer all the available current away
from the other cathodes, causing them to be dark. This is why you only need
a 50V switch on each cathode.
The caveat is that if no cathodes are pulled to zero volts, then there will
be leakage current flowing through the tube that will destroy a 50V
transistor. This is why I used the TD62083 with its set of commutation
diodes, and connected the diode common anode pin to a 50V source in my
Nixie watch circuit.
It's also why I made the blanking mode that drops the anode voltage to
100V, so that the cathodes all remain dark when blanked.


On Wed, Mar 6, 2019, 7:57 AM gregebert  wrote:

> The required voltage rating of a cathode driver has been debated in this
> forum a few times. There are 2 opinions that I know of
>
>- The driver needs to be rated at the full anode-supply voltage
>- The driver only needs to be rated at (Anode_supply_voltage -
>Voltage_drop_across_nixie_tube)
>
>
> Based on my knowledge of semiconductor physics, it's fine to use a bipolar
> (NPN) cathode driver that is rated at a lower voltage than the anode supply
> because the breakdown mechanism is non-destructive as long as the current
> is limited. When an NPN is off, presumably when it's base terminal is
> grounded (NOT open...), there will not be any current gain from collector
> leakage current, so it will stay off. I suspect an OPEN base could result
> in some visible glow, depending upon the collector leakage current Ico.
> This is why legacy drivers like the K155 or 7441 can drive a nixie that
> requires about 160V even though the IC itself can handle only 50 V.
>
> If you are using a MOSFET as your segment driver, which is typical in
> IC's, then you need to use a driver rated at the full anode supply voltage
> because a MOSFET's construction is sensitive to excess voltage (oxide
> breakdown) as well as current. Long-term overstress of the oxide will cause
> reliability problems and lead to failure.
>
>
> --
> 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 email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/c0c015b5-15e9-4387-b7f7-911193e174a4%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+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/CAPbqtve77G0UfxxoBaoi1JdXgcBR7fW%2BTYZV7K0OTmSK%2B_oQrA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[gregebert]

The required voltage rating of a cathode driver has been debated in this 
forum a few times. There are 2 opinions that I know of

   - The driver needs to be rated at the full anode-supply voltage
   - The driver only needs to be rated at (Anode_supply_voltage - 
   Voltage_drop_across_nixie_tube)


Based on my knowledge of semiconductor physics, it's fine to use a bipolar 
(NPN) cathode driver that is rated at a lower voltage than the anode supply 
because the breakdown mechanism is non-destructive as long as the current 
is limited. When an NPN is off, presumably when it's base terminal is 
grounded (NOT open...), there will not be any current gain from collector 
leakage current, so it will stay off. I suspect an OPEN base could result 
in some visible glow, depending upon the collector leakage current Ico. 
This is why legacy drivers like the K155 or 7441 can drive a nixie that 
requires about 160V even though the IC itself can handle only 50 V.

If you are using a MOSFET as your segment driver, which is typical in IC's, 
then you need to use a driver rated at the full anode supply voltage 
because a MOSFET's construction is sensitive to excess voltage (oxide 
breakdown) as well as current. Long-term overstress of the oxide will cause 
reliability problems and lead to failure.


-- 
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/c0c015b5-15e9-4387-b7f7-911193e174a4%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[Jasper C.]

On Monday, 4 March 2019 04:12:55 UTC+8, nixiebunny wrote:
>
> The voltage applied to the elements of a Nixie tube is not as 
> straightforward as you may think. Keep in mind that there is a voltage drop 
> from anode to an 'on' cathode of ~130-140 volts, due to the ionized gas. 
> Also, the capacitance of the tube elements and wiring has an effect on the 
> switching behavior.
>
> I have successfully made several varieties of multiplexed Nixie clocks and 
> watches using 60V or lower cathode switches. I used printed circuit boards 
> to reduce capacitance, and adjusted the timing to allow the recently turned 
> off cathodes to drift up to their natural 'off' voltage before turning on 
> the next digit's anode.
>
> My Nixie watch design uses 50V cathode switch arrays. It also has no anode 
> resistor. This is done by enabling only one cathode at a time, leaving the 
> other tube with no cathodes enabled. 
>
> The power supply has a resistor current sense and feedback mechanism to 
> regulate the total current as well as the open-circuit voltage. Blanking is 
> achieved by reducing the anode voltage to 100V, with a digitally controlled 
> shift to the regulator feedback. 
>
> So you can get good results with very little circuitry, if you apply a bit 
> of cleverness to the problem.
>
> http://www.cathodecorner.com/nixiewatch/firmware/nwrf-schem.gif
>
> The feedback regulated power supply is a very elegant solution, and I'd 
love to implement something like that, but my knowledge of switching power 
supplies isn't good enough to pull it off.  I envy you.

-- 
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/892ef4d6-d71e-45a8-ab96-30aa6beb5cc8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[Jasper C.]

So I was originally looking at the Microchip HV513 
.  It's a 250V totem pole 
driver with 5V input logic and a latch.  Was going to do a quick sketch of 
what I had in mind for the output, but haven't had the time to get round to 
it.

I was planning to not multiplex my nixies, probably leaving all cathodes 
connected directly to the HV line, and placing current liming resistors on 
the cathodes.  Was interested to see if I'd need to tweak any of the 
individual resistor values uniform current/brightness.

If I implemented only a low side switch, I was looking at something like 
the MMBTA42, available from a few manufacturers and is good for 300V, but I 
haven't yet found any information about leakage currents when the device is 
off.  Alternatively I think I found a mosfet or two with sufficient V(DS) 
and logic level switching, but I need to recheck the D-S leakage.

-- 
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/80ca1625-c43a-47fc-a115-eeff6130f4e2%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[Jasper C.]

On Sunday, 3 March 2019 20:53:04 UTC+8, David Pye wrote:
 

> Yes, that's it.  If the other segments are high, they just work as more 
> anodes.
>
> The best thing imho is just to leave them floating, but they will then 
> float up to the anode voltage. So provided you have drive circuitry that 
> can switch the full anode voltage, that is ok.
>

I'm a little confused there, about it being OK to switch the full anode 
voltage.  Do you mean that the low side switch can handle the full HV, or 
that I switch my cathodes to HV?  The latter would contradict the "working 
as more anodes" issue.


-- 
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/76834d66-5acc-40b5-920a-2023067f14bf%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[John Smout]

I am currently working on a set of multi-digit panaplex displays. One is a 16 
digit, 16 segment alphanumeric display and the other is 6 digits and 7 
segments. Both were originally intended to be multiplexed across all the 
digits. There is a 'keep alive’ cathode and an 80v mid rail to keep the display 
just below firing level. A consequence is that both displays like some off-time 
before a new selection is made, otherwise I am seeing a very ‘hot’ display with 
arcing over to supposedly unlit segments. It’s a bit of a fine balance to get 
right, but software can cope with this.

John S


> On 3 Mar 2019, at 10:03, Jasper C.  wrote:
> 
> I'm still new when it comes to nixie tubes, and was reading this page 
> , in particular the 
> "partial glow" that occurs if the prebias voltage of an "off" cathode is too 
> low.  Wouldn't that imply that the ideal solution would be to pull the 
> cathodes to the supply voltage of the anode when their respective segment is 
> off?
> 
> Only in this EEVblog video, starting at 14:25 
> , he does just that and it doesn't go 
> well.  I get that the anode voltage here is lower than the HV supply because 
> of the 22k resistor, though I don't understand why that then has the effect 
> that it does.  Is the segment shorted to HV acting as an anode? (Which might 
> explain the higher current he's seeing.)
> 
> Would a totem pole circuit like that work if the anode resistor was removed 
> and each cathode connected to the totem pole output by it's own resistor?

-- 
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/6D62EC78-FC6D-45BA-9A76-0448433FC928%40jsdesign.co.uk.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[David Forbes]

The voltage applied to the elements of a Nixie tube is not as
straightforward as you may think. Keep in mind that there is a voltage drop
from anode to an 'on' cathode of ~130-140 volts, due to the ionized gas.
Also, the capacitance of the tube elements and wiring has an effect on the
switching behavior.

I have successfully made several varieties of multiplexed Nixie clocks and
watches using 60V or lower cathode switches. I used printed circuit boards
to reduce capacitance, and adjusted the timing to allow the recently turned
off cathodes to drift up to their natural 'off' voltage before turning on
the next digit's anode.

My Nixie watch design uses 50V cathode switch arrays. It also has no anode
resistor. This is done by enabling only one cathode at a time, leaving the
other tube with no cathodes enabled.

The power supply has a resistor current sense and feedback mechanism to
regulate the total current as well as the open-circuit voltage. Blanking is
achieved by reducing the anode voltage to 100V, with a digitally controlled
shift to the regulator feedback.

So you can get good results with very little circuitry, if you apply a bit
of cleverness to the problem.

http://www.cathodecorner.com/nixiewatch/firmware/nwrf-schem.gif

On Sun, Mar 3, 2019, 8:54 AM gregebert  wrote:

> For direct-drive, you can just let the cathodes float, and they will be
> completely dark unless your driver has leakage. I've done 5 full designs
> with direct-drive that floats the cathode, and none has any issues. All
> told, there are 52 nixie devices in operation with no hint of failure or
> leakage. 24 of those tubes have been in service 24/7 for the past 8 years.
>
> There are some variations in the design, mainly for cathode and/or
> anode-current limiting but they all float the 'off' cathodes.
>
> Admittedly, I've never done a multiplexed design with nixies, only with
> LEDs, and unless the nixie tube is specifically designed for multiplex
> operation I would stay with direct-drive. The reason I follow is that
> multiplexed digits need a higher pulsed current to maintain visible
> brightness, and higher current leads to tube wearout.
>
>
>
>
> --
> 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 email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/e168a850-f3d7-4e88-adc8-1b3bd32b1902%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+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/CAPbqtversF-P0-jzSGjs0iXvSz-skJagg8JV23DCxENt8Kae_w%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[gregebert]

For direct-drive, you can just let the cathodes float, and they will be 
completely dark unless your driver has leakage. I've done 5 full designs 
with direct-drive that floats the cathode, and none has any issues. All 
told, there are 52 nixie devices in operation with no hint of failure or 
leakage. 24 of those tubes have been in service 24/7 for the past 8 years.

There are some variations in the design, mainly for cathode and/or 
anode-current limiting but they all float the 'off' cathodes.

Admittedly, I've never done a multiplexed design with nixies, only with 
LEDs, and unless the nixie tube is specifically designed for multiplex 
operation I would stay with direct-drive. The reason I follow is that 
multiplexed digits need a higher pulsed current to maintain visible 
brightness, and higher current leads to tube wearout.




-- 
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/e168a850-f3d7-4e88-adc8-1b3bd32b1902%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Push-pull driving cathodes

[David Pye]

Yes, that's it.  If the other segments are high, they just work as more
anodes.

The best thing imho is just to leave them floating, but they will then
float up to the anode voltage. So provided you have drive circuitry that
can switch the full anode voltage, that is ok.

If your drive circuitry cannot handle the full anode voltage, you need to
either keep at least one cathode on at all times (ie not use cathode
blanking) or have a pull down resistor that holds them at say, 70v so they
don't glow, and use the drive circuitry to put them from that holding
voltage to ground.

David

On Sun, 3 Mar 2019, 10:03 Jasper C.,  wrote:

> I'm still new when it comes to nixie tubes, and was reading this page
> , in particular the
> "partial glow" that occurs if the prebias voltage of an "off" cathode is
> too low.  Wouldn't that imply that the ideal solution would be to pull the
> cathodes to the supply voltage of the anode when their respective segment
> is off?
>
> Only in this EEVblog video, starting at 14:25
> , he does just that and it doesn't go
> well.  I get that the anode voltage here is lower than the HV supply
> because of the 22k resistor, though I don't understand why that then has
> the effect that it does.  Is the segment shorted to HV acting as an anode?
> (Which might explain the higher current he's seeing.)
>
> Would a totem pole circuit like that work if the anode resistor was
> removed and each cathode connected to the totem pole output by it's own
> resistor?
>
> --
> 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 email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/144bce53-142d-44be-9a9b-fae08fafab7b%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+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/CAOQ6x0G9yc34nETMzQENzAZDOQv1_KoEeJ5j%2Bx3ZQrDeG0odoA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.