Alternatives to 74141's etc

139 views
Skip to first unread message

Richard Scales

unread,
Apr 4, 2019, 11:09:17 AM4/4/19
to neonixie-l
Hello everyone.
Since becoming an addict about two years ago I have built many kits and have now started down the road of building my own clock - albeit a very simple one.
My query centres around the use of 74141's (or the Russian version) as direct drivers for tubes.
From what I have tried so far they seem to work well for all small tubes as well as Z566M, IN-18's and Dalibors tubes.
I know that the issue of the Blue Spot on IN-18's when driven from 74141's has been the subject of much discussion here.
I have a box of IN-18's and I know that at least one of them showed some good looking blue spot(s) when initially tested. I looked in to using an HV5182 in place of the 74141 and that worked well with all tubes though the IN-18 that had the blue spot before, still had the blue spot. I had Vpp on the HV5812 set to about 76V.
I now learn that that HV5812 as an issue sinking much over 3.5mA when connected to the cathode of a digit when the Anode is connected to 170v via a suitable series resistor.
Can anyone advise as to whether I am backing the wrong horse by looking at the HV5812 and whether or not i need to move to another driver such as the HV5622 which I note is used in a lot of designs that drive IN-18's?
For this project I only need to drive two tubes so the HV5812 is ideal in that respect.
I am aware of various solutions involving the use of transistors for switching but would like (if possible) to keep the project as simple as possible. The HV5812 is available in a DIL package which I like, I believe that the HV5622 is only in available in the 44 pin square package (SMD or socket).
All pointers gleefully received!

Kevin A.

unread,
Apr 4, 2019, 11:37:20 AM4/4/19
to neonixie-l
I would say the HV5530 is the best all around driver nixie IC. It comes in a 44 lead quad flat pack or J lead plcc, the latter of which can be placed in a plastic socket that is thru-hole soldered onto the board. It is driven like a regular shift register (like the 74HC595). Shift in 32 bits, each corresponding to a high or low on the 32 high voltage drains. Very simple to implement, and it also includes a blanking pin which can be PWM'd to dim the tubes being driven. 

There are several different versions which functionally perform the same high voltage switching:

HV5522: 44 pins, TQFP OR PLCC, 12 volt logic level, 220 volts max

HV5530: 44 pins, TQFP OR PLCC, 12 volt logic level, 300 volts max

HV5523: 44 pins, QFN, 5 volt logic level, 220 volts max 


--
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+...@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/1e7c6e04-bdb9-4af4-a517-03278ecf8048%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

gregebert

unread,
Apr 4, 2019, 12:01:07 PM4/4/19
to neonixie-l
+1 for the HV5530

I have a 14-tube clock with IN-18's driven by six HV5530's, and have had no issues other than some mild cathode poisoning (my fault) which is easily cured.
I strongly recommend using a level-translator IC when controlling the HV5530 from 3.3/5.0 VDC controllers; others have had success with the translator.


The reason for the cathode-poisoning is simple: several tubes display static values, such as the year. The de-poisoning routine runs nightly from 11PM till midnite, then the clock shuts off until someone presses the wakeup button in the morning. Then it runs all day. Since it's FPGA code, I have to tear it down to reprogram it. Someday it will have a RasPi in it.


David Forbes

unread,
Apr 4, 2019, 12:37:47 PM4/4/19
to NeoNixie
The 74141 has 55 volt transistors. You can use the SN75468 which has 100 volt transistors, for more voltage margin. It can be controlled by the 74HC595 shift register. 

The 44 pin chip does replace a lot of smaller parts, but I prefer to use many smaller parts, because the board layout is more sensible. You have two rows of DIPs behind the sockets.

You can also use the 50 volt TD62083 which is available in a much smaller SSOP package, as I use in my Nixie watches. 



--
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+...@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.

gregebert

unread,
Apr 4, 2019, 1:35:11 PM4/4/19
to neonixie-l
Board-layout with the HV5530 can be an issue, so I connect the tube-pins to the easiest route on the driver IC. That gives a clean board layout (zero feedthrus for tube signals)  but scrambles-up the bits. Unscrambling them in RTL code or software is easy, though. 

Paul Andrews

unread,
Apr 4, 2019, 4:03:06 PM4/4/19
to neonixie-l
But then the HV5530 is a 12V piece. I use the HV5523, which is a 5V part but is also SMD. You can see my reasoning here: https://hackaday.io/project/28220-one-tube-nixie-clock/log/70572-component-selection
Reply all
Reply to author
Forward
0 new messages