LC-531 and SN74141 problem

400 views
Skip to first unread message

mpestkow

unread,
Oct 15, 2012, 12:57:54 PM10/15/12
to neoni...@googlegroups.com
Hi all,

That's my first post, but I hope I get an answer to my huge beginner problem. So, I've recently started to build my own Nixie clock. I have four LC-531 nixie (Polish production) gathered from old Unitra Multimeter. All of them works fine when I'm trying to connetct it to HV source (Voltage converter 9V->150-220V - schematic) I'm using ~180-190V and 10k resistor (is it good?). 

Few days ago I finally get some 74141 and I was trying to test it, so I connected it to ATtiny2313 like this:  














and program MCU with that code. It should count on Nixie lamp from 0 to 9 (2 sec per digit).

Double checked all my connections and as the result I've got Nixie lamp with all digits glowing in the same time. 

I also checked 74141 using LED diods (+ to 5V and - to 74141 outputs) and it works fine (diodes light one after another)

Any suggestions what I'm doing wrong? It seems to in some way all outputs are connected to GND when using HV to my nixie (works fine with LED).

EDIT If I'm missing any electronic part here, please provide me info how to connect it and how it will help, I 'm Software Engineer and unfortunately, I'm not good in whole electronic stuff (yet :) )

John Rehwinkel

unread,
Oct 15, 2012, 1:49:14 PM10/15/12
to neoni...@googlegroups.com
> That's my first post, but I hope I get an answer to my huge beginner problem. So, I've recently started to build my own Nixie clock. I have four LC-531 nixie (Polish production) gathered from old Unitra Multimeter. All of them works fine when I'm trying to connetct it to HV source (Voltage converter 9V->150-220V - schematic) I'm using ~180-190V and 10k resistor (is it good?).

Depends on how much current those nixies want. I'd probably start with 15k or so.

> Few days ago I finally get some 74141 and I was trying to test it, so I connected it to ATtiny2313 like this:
>
> and program MCU with that code. It should count on Nixie lamp from 0 to 9 (2 sec per digit).

There are some quirks when reading from a port on an AVR. Instead of
"PORTB += 1", you might try "PORTB = i". Since it works with LEDs, I
don't think that's your problem here. If you really want to be cute with it,
you can just make your entire loop "for (PORTB = 0; PORTB < 10; ++PORTB) { delay(2000); }".

> Double checked all my connections and as the result I've got Nixie lamp with all digits glowing in the same time.

This can happen for a variety of reasons.

> I also checked 74141 using LED diods (+ to 5V and - to 74141 outputs) and it works fine (diodes light one after another)

That's a good idea. It lets you see that the problem isn't your AVR code, 5V power supply
connections, or data wiring.

> Any suggestions what I'm doing wrong? It seems to in some way all outputs are connected to GND when using HV to my nixie (works fine with LED).

There are a few possibilities here. My first guess is that your voltage is high enough
that the clamp diodes in your 74141 are conducting, causing additional cathodes to
light. Do you see changing cathodes getting brighter? Can you try turning down your
high voltage supply some?

Another possibility is that your HV ground isn't connected to the 74141 ground, which
can cause a variety of issues.

It could also be that your HV supply is causing interference with your digital logic.
If there's a way to put your HV supply a meter or more away from the rest of your
circuit, and power the HV supply with a different source than your logic supply,
it's worth doing. Power supply decoupling capacitors near each of your logic
chips is a really good idea too. 0.1µf between +5 and ground, with short leads,
one each for the ATTINY and 74141. Digital logic cares deeply about clean
power and can do all sorts of bizarre things if it doesn't get it. Those little
capacitors cure a host of ills.

What happens on the nixie if you mash the reset switch? That should freeze the
AVR if it's zipping through digits. Unfortunately, I think it also tristates the I/O
lines, which won't help you. You could try hanging a LED off one or two of the
I/O lines connected to the 74141 - see if those are changing predictably. It'll
probably throw the logic levels off, but at least you can see if the AVR is still
behaving.

Alternatively, you could hardware your 74141 to a specific value (like "2") by tying
its inputs to +5 and ground as needed, verify the expected LED lights, then try it
with the nixie.

> EDIT If I'm missing any electronic part here, please provide me info how to connect it and how it will help, I 'm Software Engineer and unfortunately, I'm not good in whole electronic stuff (yet :) )

You've got the right idea - start small, test with LEDs, and move up to nixies. You already
have your HV supply working, which is good - a lot of people have trouble with those.
You have your AVR code working too. You're really close.

- John

Adam Jacobs

unread,
Oct 15, 2012, 2:06:18 PM10/15/12
to neoni...@googlegroups.com
There is a chance that you have a bad 74141. I would try disconnecting the 74141 inputs from your uC and try grounding them all. Does everything turn off? try pull one of the inputs high and verify that the correct number is lit. I'd do this before proceeding.
If the 74141/Nixie circuit works fine disconnected from the uC, then you've got a uC problem. Figure out where your problem is before you proceed any further.

-Adam
--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To post to this group, send an email to neoni...@googlegroups.com.
To unsubscribe from this group, send email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msg/neonixie-l/-/gXhxb1b78DEJ.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

Krzysztof Więckowicz

unread,
Oct 15, 2012, 5:29:25 PM10/15/12
to neoni...@googlegroups.com

Hi,

just tray to change value of anode resistor. 10k is to high for this tube, so finally all digits glow together.

Tray this:

R=(190V-140V)/0.002A=25k

 

Regards

Krzysztof

mpestkow

unread,
Oct 16, 2012, 6:22:36 AM10/16/12
to neoni...@googlegroups.com
Hi,

First of all jrehwin - thank you for a very detailed answer, it is great to know that there are such a great communities in the Internet. 

I've just resolved my problem thanks to you guys. So what I've done:

 - changed the resistor to 47k
 - set HV source to ~170V
 - add normal diode in 5V line. 

I think that my problem was caused by few things at once, but I think the main was how I set voltage sources. Because of lack of good, stabilized voltage source I have something like this: 
 - From AC 230V i have 12V DC power supply (Made in China, so I realized that this is actually 18,7V )
 - I connect it to my breadboard and than branch it to 7805 and 7809 stabilizer (with capacitors ofc)
 - Output from 7805 goes to logic
 - Output from 7809 goes to mentioned voltage converter and becomes 170V which is my HV line

Thank you very much for all ideas and help. Today I will make an upper board for Nixie Clock (tubes+ TLP627 optocouplers for multiplexing) and start to develop logic board (I'm going to use ATtiny2313+PCF8583+74141). As an alarm sound I will use old-school electric doorbell with metal chime.

Regards,
MP

Adam Jacobs

unread,
Oct 16, 2012, 12:49:34 PM10/16/12
to neoni...@googlegroups.com
Hello,
  You should not place a linear regulator (7809) between the transformer and the boost switching supply. Go ahead and remove it. The only components I would place between the transformer and the input to the high voltage supply are maybe a 1n4007 protection diode and one or two bypass caps.

Also, the reason your transformer is reading 18.7v is because it was not loaded when you measured it. All transformers are like that, you just have to design for it. :) Also, it's somewhat important to use transformers that are rated for something close to the current draw that you intend. A 100w 12v transformer isn't ever going to sink down to 12v if your design only draws 5w.

-Adam
To view this discussion on the web, visit https://groups.google.com/d/msg/neonixie-l/-/OBmodi3PIi4J.
Reply all
Reply to author
Forward
0 new messages