开云体育

Ok, so here is my predicament: after searching the internet for what seems like days I'm still unable to find a reliable explanation of how a Russian K155ID1 (74141) BCD-Decimal decoder or a 74HC595 8-bit shift register works. I know that a BCD-Decimal decoder does just what it says; it converts binary coded decimal into decimal binary. I also know that a shift register does some kind of shifting or switching operation, logging information and dispersing it through clock pulses. What I don't understand is: a) How?, b) What is the purpose?, and c) What job does it do in a nixie clock?

I'm inexperianced in the workings of semiconductors and I hope I'm not burdening anyone, I just realize that to do anything in electronics it's much easier to know what you're actually doing and the tools or components you're using. I don't know anyone but you intelligent fellows who can help me with these problems, so I thank anyone who contributes to me losing my unwillful electronics ignorance.

The shift register:


franklinmknight wrote:

franklinmknight wrote:

Ok, so here is my predicament: after searching the internet for what
seems like days I'm still unable to find a reliable explanation of
how a Russian K155ID1 (74141) BCD-Decimal decoder or a 74HC595 8-bit
shift register works. I know that a BCD-Decimal decoder does just
what it says; it converts binary coded decimal into decimal binary.
I also know that a shift register does some kind of shifting or
switching operation, logging information and dispersing it through
clock pulses. What I don't understand is: a) How?, b) What is the
purpose?, and c) What job does it do in a nixie clock?

Your question reminds me of the PhD student who went to the exam and was asked "why is the sky Blue" - it can be answered on so many different levels.

In any of these TTL IC's there are a bunch of circuits which form locic gates. Give an AND gate Two "true" inputs and it will give you a "TRUE" output, anything else and you get FALSE. An OR gate will give you TRUE if either input is true. An XOR gate will give you TRUE if ONLY ONE of the inputs is true.

Now if you do a truth Table of the BCD decoder, you can come up with an arrangement of gates that will give a true on one given output for each one of the posible inputs.

NOW The gates themselves are made up of a bunch of bipolar transistors/Diodes and such. and the designer has a large number of valid ways to design one. so that is another "layer of the onion" if you like.

IF you look at the TI datasheet for their 74141 at

You can see a logic diagram and a truth table of the TI version. The Russian version is slightly more complicated as it will not light up a segment if you give it an illegal input.

--
Charles MacDonald Stittsville Ontario
cmacd@... Just Beyond the Fringe

No Microsoft Products were used in sending this e-mail.

[edited by A.J. - please trim quoted material]

On Sat, Jan 2, 2010 at 3:50 AM, chuck richards <chuckrr@...> wrote:

<snip>

There are two books written by Don Lancaster that explain these
things, and many more in great detail. Have a look on
Don's web site and look for:

For all of Don's contributions over the years, and engineering know-how, his
site is a disaster area. There's some good info, but it's nearly impossible
to find. Reminds me of a Geocities refuge.

--jc

Eagles may soar, but weasels don't get sucked into jet engines

Yes, maybe tinaja.com is a mess, but finding the TTL and CMOS
Cookbooks is about as easy as stumbling over a big rock.

Those 2 books are the best ever written about the
subject that I know of. They explain how the different types
of logic ICs work, and how to use them by showing example
circuits.

The concepts are presented clearly, with theory, and a schematic
is shown, along with a truth table of the expected results.

Don leaves it up to the reader to follow along and build the
example circuits so as to be able to see them working
first hand.

Those books are like a huge toolbox full of reliable, working
ideas that can be used as building blocks to make whatever
one wants to make.

Navigating a somewhat clunky old website to find them is
well worth the effort.

In this fast-paced time of speedy processors, and all sorts
of things only dreamed of when those books were written,
one can still get right down to each binary bit, and where it
goes, and exactly what it does. Those books even show the
basic internal transistor circuits used to make the gates.
That helps explain the "why" of how they work.

Chuck




$4.95/mo. National Dialup, Anti-Spam, Anti-Virus, 5mb personal web space. 5x faster dialup for only $9.95/mo. No contracts, No fees, No Kidding! See for more details!

On Fri, 2010-01-01 at 22:12 -0500, Charles MacDonald wrote:
[snip!]

IF you look at the TI datasheet for their 74141 at

You can see a logic diagram and a truth table of the TI version. The
Russian version is slightly more complicated as it will not light up a
segment if you give it an illegal input.

There's a version from TI, the (SN*)74(LS**)145, which acts
(FWICT) /exactly/ like the russian one. And, Mouser has them in fair
supply, unlike a 74141 which are more and more hard to come by;

Truth table as follows:
N DCBA 0123456789
0 0000 1000000000
1 0001 0100000000
2 0010 0010000000
3 0011 0001000000 IN THIS CASE, 0 = HIGH, 1=LOW
4 0100 0000100000 In datasheet, "High level = Off,
5 0101 0000010000 low level = On"
6 0110 0000001000
7 0111 0000000100
8 1000 0000000010
9 1001 0000000001
I **** 0000000000 <~ Invalid values

So basically, if you read the datasheet for this series, it tells you
the logic diagram. you then can figure out from there how it works.

Go read a book called "Bebop to the Boolean Boogie" -- great book that
really tries to make you rethink how electronic design works. Fun book,
to boot.

*SN appears to be TI's way of marking things as theirs; I dont know why
the prefix other than "Its a prefix"
**the LS specifier is for a D, N or FK package

--
Morgan Gangwere <0.fractalus@...>

SN stands for 'semiconductor network', a TI name for the IC that seems to
have replaced 'solid circuits' and a prefix that they still use today.
From . I don't know
how *authoritative* that is, but it sounds familiar from the olden days.

--jc

On Sun, Jan 3, 2010 at 2:23 PM, Morgan Gangwere <0.fractalus@...>wrote:

*<**The vorpal blade** went **snicker**-**snack to make AJ happy>*

*SN appears to be TI's way of marking things as theirs; I dont know why
the prefix other than "Its a prefix"
**the LS specifier is for a D, N or FK package

--
Morgan Gangwere <0.fractalus@... <0.fractalus%40gmail.com>>

Eagles may soar, but weasels don't get sucked into jet engines

--- In NEONIXIE-L@..., Morgan Gangwere <0.fractalus@...> wrote:

**the LS specifier is for a D, N or FK package

The "LS" stands for LOW POWER SCHOTTKY, whereas an "HC" (or "HCT") would stand for High-Speed CMOS.

snip

segment if you give it an illegal input.

There's a version from TI, the (SN*)74(LS**)145, which acts
(FWICT) /exactly/ like the russian one. And, Mouser has them in fair
supply, unlike a 74141 which are more and more hard to come by;

snip

Ok, but wait a minute now...
The TI data book says that 74LS145 has 15 volt output transistors
that will sink up to 80 mA. So, that IC is more of a lamp,
or relay, or MOS driver. Not a nixie tube driver.

Chuck


$4.95/mo. National Dialup, Anti-Spam, Anti-Virus, 5mb personal web space. 5x faster dialup for only $9.95/mo. No contracts, No fees, No Kidding! See for more details!

* * Big F-n Snip * *

I see that everybody is concentrating on the digital side. Which is all fine and dandy. But, what I notice a lot about newbies, are analog side issues. Mostly, how does a chip that's hooked up between +5V and Gnd, light a tube that needs 170V !?

Well, most logic chips (74xx, 4xxx, ...) have push-pull (source-sink, either totem-pole, or complimentary) outputs. These are mostly intended to connect to the inputs of chips of the same 'ilk'. Those outputs can only swing from 0V (gnd) to (+5V, or whatever Vcc your using).

A few chips, have what are called "open-collector", or "open-drain" outputs. The 74141 (both Western & Russian) is that kind of chip. "Open-collector" (OC) outputs, provide a current path to GND, which can switched ON or OFF. If you look at the 74141 truth table, you will see that for any BCD code, only the desired output shows a low, all the others are high. The LOW output is the ON output, which is only one passing current. All the other outputs (HIGH) are OFF. The current path is from the nixie +HV supply (+170V min) thru a current limiting anode resistor to the nixie's anode, across the internal neon atmosphere of the nixie, to the selected cathode, out that cathode lead, to the chip, and down to gnd.

An important thing to know about voltage, is that its electrical pressure. Current (amperes, milliamps ...) is the actual electrical flow. To get current flow, there has to be a pressure difference. For a nixie to ignite, it needs to be at least +170V, from anode to cathode. That means, if a +180V supply is used, the cathode voltage needs to drop to +10V or less to ignite the tube. Once ignited, current starts to flow. The current will rise until the voltage across the nixie drops to its "maintaining voltage". A pressure difference usually around 140V (130-150V). The (required) anode resistor takes up the slack, by having the remaining 40V (180V supply - 140V nixie) drop across it. The current is limited to a value determined by that 40V divided by the resistor value (40V / 15K = 2.67mA).

There is also an "extinguishing voltage, which usually is only a tad lower than the maintaining voltage. The "extinguishing voltage" also means that the OFF outputs don't need to rise all the way to +HV (+180) to stay OFF. All they have to due is rise to a value so that the voltage difference is less that the extinguishing voltage. In this case that's only to +40V (180V - 140V). The Western 74141's have protection zener diodes on their outputs, that limit their rise to 60V. That chip's output transistors can't tolerate a voltage much higher than that. If the voltage try's to rise higher than that, the zener will start conducting, and current will flow. This will protect the chip, but is bad for the circuit, since an undesired output, will be partially ON, and maybe visible.

Some nixies have extinguishing voltages as low as 100V. You should "characterize" your tubes before designing your circuit. See what voltage is required to turn them ON, and how low the voltage needs to drop before they go OFF. And write those numbers down. If the difference is bigger than 60V, then you should not use the western version of the 74141. The Russian versions seem to clamp at a higher voltage. I got a batch of 50 or so, a few years ago, and they all seemed to get as high as 90V (or higher), before there was any "leakage".

If all else fails, you can still use a 4028 (BCD-decade decoder) and a set of (10) MPSA42s. You can skimp on the base resistors, since only one transistor will be ON at any one time. Just tie all the emitters together, and hook them up to a single emitter resistor (1K), with its other leg tied to Gnd. This one resistor limits both base current, and collector current, to a maximum of 4.3mA (Ie=Ic+Ib, using a 1K R & 5V supply).

At 10:27 PM +0000 1/3/10, threeneurons wrote:

If all else fails, you can still use a 4028 (BCD-decade decoder) and a set of (10) MPSA42s. You can skimp on the base resistors, since only one transistor will be ON at any one time. Just tie all the emitters together, and hook them up to a single emitter resistor (1K), with its other leg tied to Gnd. This one resistor limits both base current, and collector current, to a maximum of 4.3mA (Ie=Ic+Ib, using a 1K R & 5V supply).

Instead of a bunch of separate transistors, you can use a TD62084 or an SN75468 which have 8 or 7 transistors in one DIP package, which saves a lot of wiring. They are good to 50V (TD62084) or 100V (SN75468) and both work well. They have a clamp diode on each transistor to protect it; tie that to a resistive divider form 180V to the proper clamp voltage.


--

--David Forbes, Tucson, AZ

[snip]

On the topic of the SN74LS4145,
Using these as a Nixie driver is perfectly legitimate, from what I can
tell (I'm not an expert, but I live with a few). Current is pressure,
yes; at that point, the fact the nixies on average that I'm using for
my project use 2mA (average, at most 3mA according to Burroughs
catalog 616) makes these /overkill/; Sinking Current in this case is
using the 74....* as a means of simply tying the cathode of a nixie to
the common ground like a transistor. If I wanted to be /really/
paranoid, I'd put them on transistors (something like a MPSA42RLRAG
from ON Semiconductor, load capacity 300V w/625W dissipation) but the
according to 2 analog guys, that'd be paranoia unneeded as the 2mA
that my nixies pull, the 74.... can handle sinking that to ground.



* I'll just write the .... because i'm lazy tonight.
--
Morgan gangwere

“The future’s already arrived. It’s just not evenly distributed yet.”
William Gibson

On Sun, Jan 3, 2010 at 8:54 PM, Morgan Gangwere <0.fractalus@...> wrote:

[snip]

Also, if you look at the datasheet for the 74145:
page 2 of TI's datasheet from 1974:

Typical of all outputs:
,------ OUTPUT
----------------------|'
`>---- GND
[and some more but this is the pertinent part of it]

Because of this, all we care about is the sinking capacity because of
our usage in this case; the transistor should be able to handle the
load fine, voltage is irrelevant.


--
Morgan gangwere

“The future’s already arrived. It’s just not evenly distributed yet.”
William Gibson

At 9:04 PM -0700 1/3/10, Morgan Gangwere wrote:

Because of this, all we care about is the sinking capacity because of
our usage in this case; the transistor should be able to handle the
load fine, voltage is irrelevant.

--
Morgan gangwere

Morgan,

You are using the 74145 well over its rated output voltage of 15V. The high voltage on the 'off' outputs generated by the nixie tube cathodes may destroy the part eventually.

It may work for years; it may never fail. It could fail next week or in the next hour. That's why people who design circuits for a living would never use a part in this manner. It's asking for trouble.

--

--David Forbes, Tucson, AZ

David Forbes wrote:
[snip]

Would you then instead suggest using the MPSA42 transistor in with the 74145?

Upon some thinking and research I've figured out what the difference is <sarcastic grin> the difference is that the 74141 is an open-collector while the 74145 is /not/. I've taken one of them, shoved voltage through and watched the thing go <pop!>. Whoopsy~.

(I guess this is what I get for writing rants late at night: my mind goes "Durrrr" and forgets to read things right.)

--
Morgan Gangwere
"testing, testing and testing: Three things you should never do without." (John Carmack)

Ok, so I've found a schematic I like, it doesn't, however, use either of the ICs we have reviewed (kinda rude I know, didn't intend on that happening, circumstances just made it that way). I do appreciate the help though, I've learned a deal here. Now I'm moving to new questions though; starting with these:
1. What's the purpose of the anode resistor?
2. How do I find the range of resistance I need for a specific tube(like a formula)?
3. Anode and cathode? Or just anode?

Any other information you might be able to share that you think is even remotely related. By the way, this is what I settled on . I'm sure one of you've heard of this version.

Thanks to you all,

Frank

[edited by A.J. - please trim quoted material]
franklinmknight wrote:
<snip>

Now I'm moving to new questions though; starting with these:
1. What's the purpose of the anode resistor?
2. How do I find the range of resistance I need for a specific tube(like a formula)?
3. Anode and cathode? Or just anode?

Any other information you might be able to share that you think is even remotely related. By the way, this is what I settled on <> . I'm sure one of you've heard of this version.

I'll chime in with my 2 cents.

1) the purpose of the Anode Resistor is to limit the current through the Nixie to the specified current value in your particular nixie's datasheet. It is important to try to get fairly close to the nominal current defined in the datasheet. If you completely omitted the limiting resistor, your nixie would have a VERY short and eventful life.

2) The range of resistance will depend on the supply voltage and the characteristics of your particular nixies. Use Ohm's law to get a good idea of the resistor value you will need, and then fine tune it from there. You will find that with higher current, the maintaining voltage goes down.. So it makes this not exactly a straight-forward proposition. Also, if you are going to use a non-regulated high-voltage supply (like the one in Mike's clock) the power supply voltage will sag as current increases, making this an even more not straight-forward proposition.

3) Just anode.

4) Mike's clock is certainly the easiest to understand, and it's what I used as my first clock. But be careful! It wasn't too long before I switched to isolated high-voltage supply clocks, because I was sick of electrocuting myself with 250vdc of wall-current.. Also, it's a bugger to debug since you'll need an isolating transformer before you can use a grounded scope on it. Oh, and also it's a lot of work to build that circuit! There are 28 transistor drivers on that thing (without the 6 digit option) and all of the decade counters need to be cascaded. After building that once, you'll decide that a microcontroller design is less work and more flexible. This is my bread/butter microcontroller nixie clock design:




Good luck!

-Adam

--- In NEONIXIE-L@..., "franklinmknight" <franklinmknight@...> wrote:
<snip>

Now I'm moving to new questions though; starting with these:
1. What's the purpose of the anode resistor?
2. How do I find the range of resistance I need for a specific tube
(like a formula)?
3. Anode and cathode? Or just anode?

Any other information you might be able to share that you think is
even remotely related.

Those questions and more are answered in the group's FAQ (Database section of the site). It was mentioned in the email you should have received upon joining, if you provided a working email address.
<>

Side note:
If the email address you provided doesn't work, you might want to change it, as anyone who "bounces" group emails for a month or more may be deleted from the roster, without notice of course. (How can I notify you, if your email address doesn't work?) Please note that this applies to ALL Yahoo Groups, as even those who elect not to receive any email from this group will appear in our Bouncing list if their emails from any Yahoo Group are being returned as undeliverable.

A.J.

---Snip---

Ok, so I've found a schematic I like, it doesn't, however, use

Any other information you might be able to share that you think is even remotely related. By the way, this is what I settled on . I'm sure one of you've heard of this version.

Thanks to you all,

Frank

Frank,

Before you start on Mike's clock circuit you may want take a look at Peter Wendt's design, . I have one about 85% complete, just waiting on a few odds and end to arrive. I didn't use a crystal for timing, basically because I couldn't get it to work right. I opted for a DS32KHz module which worked out great.
I like Peter's better because it's a little more straight forward and he explaines every section very well.

Good luck!!

John