[NEONIXIE-L] Anti-ghosting technicques for multiplexed Nixie clock with 74141
Gelu
Dear all,
I'm trying to build a quite simple Nixie clock (with Russian IN-12 tubes) based on specialized IC clock (quite similar with MM5309: multiplexed BCD outputs, fixed 1024 Hz multiplex frequency) and I used, like as tubes' driver, the IC 74141. Unfortunately I'm facing with ghosting phenomenom and kindly I'm asking to you to share me from your experience to prevent it (if this it is possible in the presented configuration).
Thank you in advance.
Gelu
threeneurons
> "Gelu" <sahist_amator@...> wrote:
>
> I'm trying to build a quite simple Nixie clock ... based on
> specialized IC clock ... BCD outputs, fixed 1024 Hz multiplex
> frequency) and I used, like as tubes' driver, the IC 74141.
> Unfortunately I'm facing with ghosting phenomenom ...
If you could post your schematic somewhere, and provide the link, that would be nice. But here are a few things you SHOULD (or SHOULD NOT) be doing:
1) You should have inter-digit blanking intervals. ~200uS ballpark.
2) You should NOT use the cathode driver for the blanking. ie sending a 'f' to the 74141, so that all cathodes are OFF. One cathode should always be ON.
3) Use the anode drive for your blanking. During blanking, all anodes are OFF.
4) Make sure your anode drive switches fast enough. Just because a transistor is rated to switch in the MHz, doesn't mean it will, if your surrounding circuit slows it down. Make sure there is a resistor across from base to emitter on each transistor, if its driven from a single ended source. You're using 1024 Hz, which is pretty fast. I assume that's roughly 1mS per digit. For a 6-digit clock, that would translate to 170.7Hz, per complete frame. Note: nixie takes around 10 to 50uS to turn ON, once voltage is applied.
My 1st experience with multiplexing involved 7-seg LEDs with discrete transistor anode drive. I omitted the base-emitter resistor (not to be confused with the base resistor), and it ghosted badly. On the scope, I could see that the prior transistor was still delivering collector current, during the next transistor's turn. Putting 10K resistors across base to emitter, cured the problem, and the transistors turned OFF promptly, and cleanly. Nice square waves.
This should keep you busy for a while.
Gelu
[edited by A.J. - please trim quoted material]
--- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@...> wrote:
<snip>
> If you could post your schematic somewhere, and provide the link,
> that would be nice.
<snip>
I'm attaching the electric diaghram (sorry for the poor quality of the scanned file); one more thing: it is a four-digit clock.
http://i46.tinypic.com/2saghhe.jpg
threeneurons
> "Gelu" <sahist_amator@...> wrote:
>
> I'm attaching the electric diaghram (sorry for the quality of the
> scan)... four-digit clock.
> http://i46.tinypic.com/2saghhe.jpg
>
Quality is just fine. Nothing to be sorry about. You haven't seen my scribbles.
The anode drive circuit looks fine. There are still other areas that could be causing the problem:
1. Cathode side blanking. Sending an out-of-range code to the 74141 (A-F aka 11-15). This is a no-no.
2. No blanking interval, or too short of a blanking interval.
3. Output polarity inverted, on either the cathode or anode signals, from the clock chip. If its on the anode side, it can be remedied by hooking up the emitters of those MPSA42s to their respective chip outputs (instead of the base). You still need current limiting resistors from the chip outputs, but they now hook up to the emitters. The bases of the MPSA42s should be tied to some midpoint bias, ~2.5V, which can be made with a simple resistor divider (10K+10K). Collector stays connected as is. If the chip cathode outputs are inverted, then add an inverter chip, like a 4069, between the clock chip and the 74141.
BTW, do you have a datasheet on that clock chip ?
Common emitter transistor hookup, inverts the output. Emitter tied to some fixed point (ie gnd, or Vcc).
Common base transistor hookup, stays in the same phase. Base tied to some fixed point (gnd, Vcc, or some fixed point in between).
Both have voltage gain.
David Forbes
On 6/10/10 6:47 PM, threeneurons wrote:
>> "Gelu"<sahist_amator@...> wrote:
>>
>> I'm attaching the electric diaghram (sorry for the quality of the
>> scan)... four-digit clock.
>> http://i46.tinypic.com/2saghhe.jpg
>>
>
> Quality is just fine. Nothing to be sorry about. You haven't seen my scribbles.
>
> The anode drive circuit looks fine. There are still other areas that could be causing the problem:
>
I would consider two other points:
1. The base to emitter resistor on the MPSA92 could be much smaller,
perhaps 10K.
2. The controller must turn off the anodes and wait ~200 microseconds
BEFORE changing the 74141 BCD value. Otherwise the new cathode value
will display on the previous tube for a brief time, since there is a
capacitive time delay on the anode turn-off.
--
David Forbes, Tucson AZ
http://www.cathodecorner.com/
Ron
Why is cathode-side blanking a no-no? I am currently running a clock I built that does this and I have had no problems. I once tried changing the software to do anode-side blanking but I found I could not do it on this clock. I'm using the digit outputs to drive the setting switches. When I changed to anode-side blanking, switch presses were not being detected.
--- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@...> wrote:
>
> > "Gelu" <sahist_amator@> wrote:
> >
> > I'm attaching the electric diaghram (sorry for the quality of the
> > scan)... four-digit clock.
> > http://i46.tinypic.com/2saghhe.jpg
> >
>
> Quality is just fine. Nothing to be sorry about. You haven't seen my scribbles.
>
> The anode drive circuit looks fine. There are still other areas that could be causing the problem:
>
> 1. Cathode side blanking. Sending an out-of-range code to the 74141 (A-F aka 11-15). This is a no-no.
>
> 2. No blanking interval, or too short of a blanking interval.
>
> 3. Output polarity inverted, on either the cathode or anode signals, from the clock chip. If its on the anode side, it can be remedied by hooking up the emitters of those MPSA42s to their respective chip outputs (instead of the base). You still need current limiting resistors from the chip outputs, but they now hook up to the emitters. The bases of the MPSA42s should be tied to some midpoint bias, ~2.5V, which can be made with a simple resistor divider (10K+10K). Collector stays connected as is. If the chip cathode outputs are inverted, then add an inverter chip, like a 4069, between the clock chip and the 74141.
>
> BTW, do you have a datasheet on that clock chip ?
>
> Common emitter transistor hookup, inverts the output. Emitter tied to some fixed point (ie gnd, or Vcc).
>
> Common base transistor hookup, stays in the same phase. Base tied to some fixed point (gnd, Vcc, or some fixed point in between).
>
> Both have voltage gain.
>
Dan Harboe Burer
Quixotic Nixotic
On 11 Jun 2010, at 20:48, Dan Harboe Burer wrote:
> ..are back..
>
> http://www.surplusshed.com/pages/item/m2302.html
Do those things light up or something? If so, what technology? What
do they look like in real life? Orange? Pink? Blue? I am guessing
red. I think I see a red filter. Have they neon inside?
John S
Nick
--- In NEONI...@yahoogroups.com, Quixotic Nixotic <nixotic1@...> wrote:
>
>
> On 11 Jun 2010, at 20:48, Dan Harboe Burer wrote:
>
> > ..are back..
> >
> > http://www.surplusshed.com/pages/item/m2302.html
>
> Do those things light up or something? If so, what technology? What
> do they look like in real life? Orange? Pink? Blue? I am guessing
> red. I think I see a red filter. Have they neon inside?
John - when you drop round at the weekend we can have a play with some - I've a few sitting in the workshop...
Cheers
Nick
Theodore Johnson
From: NEONI...@yahoogroups.com [mailto:NEONI...@yahoogroups.com] On
Behalf Of Quixotic Nixotic
Sent: Friday, June 11, 2010 4:03 PM
To: NEONI...@yahoogroups.com
Subject: Re: [NEONIXIE-L] Surplusshed display units
On 11 Jun 2010, at 20:48, Dan Harboe Burer wrote:
> ..are back..
>
> http://www.surplusshed.com/pages/item/m2302.html
Do those things light up or something? If so, what technology? What
do they look like in real life? Orange? Pink? Blue? I am guessing
red. I think I see a red filter. Have they neon inside?
John S
[Non-text portions of this message have been removed]
Gelu
--- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@...> wrote:
>
> > "Gelu" <sahist_amator@> wrote:
> >
> > I'm attaching the electric diaghram (sorry for the quality of the
> > scan)... four-digit clock.
> > http://i46.tinypic.com/2saghhe.jpg
> >
>
> Quality is just fine. Nothing to be sorry about. You haven't seen my scribbles.
>
> The anode drive circuit looks fine. There are still other areas that could be causing the problem:
>
> 1. Cathode side blanking. Sending an out-of-range code to the 74141 (A-F aka 11-15). This is a no-no.
>
> 2. No blanking interval, or too short of a blanking interval.
>
> 3. Output polarity inverted, on either the cathode or anode signals, from the clock chip. If its on the anode side, it can be remedied by hooking up the emitters of those MPSA42s to their respective chip outputs (instead of the base). You still need current limiting resistors from the chip outputs, but they now hook up to the emitters. The bases of the MPSA42s should be tied to some midpoint bias, ~2.5V, which can be made with a simple resistor divider (10K+10K). Collector stays connected as is. If the chip cathode outputs are inverted, then add an inverter chip, like a 4069, between the clock chip and the 74141.
>
> BTW, do you have a datasheet on that clock chip ?
>
> Common emitter transistor hookup, inverts the output. Emitter tied to some fixed point (ie gnd, or Vcc).
>
> Common base transistor hookup, stays in the same phase. Base tied to some fixed point (gnd, Vcc, or some fixed point in between).
>
> Both have voltage gain.
>
Thank you (and to David, too) for the tips and, further these, please find my comments:
1) I don't know how can I put in evidence the fact that there are codes out-of-range on 74141. Any ideas in this way?
2) Unfortunately I'm dealing with a specialized IC clock (more technical details about it at: http://www.datasheetcatalog.com/datasheets_pdf/M/M/C/3/MMC351.shtml) not with a uC, so I cannot define/edit a blanking interval
3) Indeed, I used 4049 to invert the clock outputs polarity.
4) I'm working on to optimise the working area of MPSA42 - by adding at its input a resistor divider, and I observed an improving of its switch-off scope trace (and, further this, the ghosting phenomenom decreases).
Dmitri Vorobiev
> 1. Cathode side blanking. Sending an out-of-range code to the 74141 (A-F
> aka 11-15). This is a no-no.
I'm currently designing a counter where I'd like to send an out-of-range
code to K155ID1 for zero blanking. Is there anything bad in such an
approach?
Thanks.
Dmitri
[Non-text portions of this message have been removed]
threeneurons
> Dmitri Vorobiev <dmitri.vorobiev@...> wrote:
>
> > 1. Cathode side blanking. Sending an out-of-range code to the
> > 74141 (A-F aka 11-15). This is a no-no.
>
> I'm currently designing a counter where I'd like to send an
> out-of-range code to K155ID1 for zero blanking. Is there
> anything bad in such an approach?
>
> Thanks.
>
> Dmitri
>
The outputs of the 74141 & its Russian equivalent (K155ID1) don't go all the up to 180V (nixie supply) when they are OFF. The 74141 clamps to a max of 60V or so. The Russian part somewhat higher (~100V). In other words, they are kinda 'leaky', and still MAY not fully turn the nixie OFF. This may be expressed as the nixie tube presenting a hazy orange glow, when its suppose to be OFF. If yours don't, consider yourself lucky. But keep this in mind, if in few years, the display gets a little hazy.
In a direct drive circuit this never happens, since never happens because always one cathode is ON (at, or near, gnd potential), and draws all the current.
If transistors like the MPSA42 (300V NPN) are used, then cathode blanking would be perfectly okay. These transistors won't leak until well above any nixie supply voltage used. This also applies to some HV serial drivers that are also rated at 200V or more.
threeneurons
> > > scan)... four-digit clock.
> > > http://i46.tinypic.com/2saghhe.jpg
> Thank you (and to David, too) for the tips and,
> further these, please find my comments:
>
> 1) I don't know how can I put in evidence the fact that
> there are codes out-of-range on 74141. Any ideas in this way?
> 2) details about it at: http://www.datasheetcatalog.com/datasheets_pdf/M/M/C/3/MMC351.shtml
> 3) Indeed, I used 4049 to invert the clock outputs polarity.
> 4) I'm working on to optimise the working area of MPSA42 - by
> adding at its input a resistor divider, and I observed an
> improving of its switch-off scope trace (and, further this, the
> ghosting phenomenom decreases).
>
I looked at the datasheet, and there are somethings I could pull from it.
The sample circuit they used, uses a 4511, which is a common CMOS decoder chip. The 4511 is used with common cathode 7-seg LED displays. What this means to you, is that all the signals are NOT inverted, so do NOT add that 4049 inverter. It means your circuit as originally shown, is at least of the right polarity. The ghosting is caused by something else.
The datasheet mentions a duty cycle of 22%. For four digits, 25% would be the max, without any blanking interval. They are using 3% intervals (25%-22%). 1024Hz, is 1/32 of the 32KHz xtal thats to be used. So I suspect each digit is ON for 31 clocks, blanked for 1 clock, then on to the next digit. I hope.
You can see if the anodes are used for blanking using just one scope channel. Make sure your cicuit is electrically isolated from 'earth gnd', because your scope gnd is probably tied to it. Connect the probe to one of the select outputs, and the gnd clip, of that probe, to the next select output, in the sequence. You should see a positive pulse of 22% duty-cycle, followed or preceeded by a negative pulse, of the same width. If the blanking is done here, then there should be a small gap between the two pulses (~3%).
Also a 3% blanking interval, is only 30uS, for this circuit, if I interpret the datasheet correctly. You can also measure, and record, the pulse width (above), and let us know, what it is.
Looking for the blanking on the BCD outputs may be a bit trickier with only a single trace scope. With a dual trace, it may be possible, but you may have to hunt around to find a pair of signals that are different, for at least two adjacent digits. Then you can look for gaps at logic-1 levels, of around 3% (30uS ?). This would indicate blanking here. This could be done, because the 4511 blanks the display with codes A thru F (10-15). But it may not, if the designers intended to use it with other unknown BCD decoders.
It would be more universal if the blanking is done on the 'digit select side'. For you, that would be good, since that means anode side blanking.
Finally, David is right, you should make those base-emitter bleeder resistors a little smaller (to 10K-22K, from 100K). The base current is now around 380uA. A 10K resistor will only divert ~70uA, leaving plenty to drive the MPSA92, and it will turn it OFF faster. A 30uS blanking interval is marginal, but acceptable for most nixies.
guus.a...@wolmail.nl
On a clock with 7 segment (flat) nixie tubes I've had ghosting as well.
This was fixed by slowing the multiplex rate down.
Any other method was not working.
BR/
Guus
>From: "Gelu" <sahist...@yahoo.com>
>Date: Thu, 10 Jun 2010 09:59:31 -0000
threeneurons
> guus.assmann@... wrote:
>
> On a clock with 7 segment (flat) nixie tubes I've had
> ghosting as well.
> This was fixed by slowing the multiplex rate down.
> Any other method was not working.
>
> BR/
> Guus
http://i46.tinypic.com/2saghhe.jpg
http://www.datasheetcatalog.com/datasheets_pdf/M/M/C/3/MMC351.shtml
And in this case, with a small 30uS blanking interval, that may also be the true solution. Unfortunately, if the scan rate is slowed down, and the blank period increased, the time will also be affected, since they all share the same 32KHz timebase.
30uS is just about the turn-ON time of nixies of that type. Speed can be increased by either raising the supply voltage (180V -> 200V), by 'priming' it, by shining purple LEDs at it, or all of the above.
If those IN-12s are the 'B' type (IN-12B), then they'll have a decimal point, which may act as a primer. In that case, you may want to add 'mid-pull' resistors to all the anodes and cathodes. Say 470K resistors to 60V to 80V. You'll also need a negative bias voltage so that the primer is 180V of from the mid-pull potential. -120V for 60V, to -100V for 80V. The primer will be connected to that bias (-120V to -100V) thru a larger series resistor (~1M).
The negative bias can be made from your existing switcher, by tapping off the coil-FET node, with a small capacitor (~10nf) and resistor (~10K) in series. From the other end, you can connect that signal (node 2) to a pair of ultra-fast rectifiers. One with the anode to node 2, and its cathode to gnd, and the other with its cathode to node 2, and its anode to the primer bias (node 3). A 1uf 250V cap is tied to node 3 (primer bias); neg to node 3, and pos to gnd. The 4 decimal points (one per digit) are tie to this primer voltage thru 1M resistors.
When all is done, this should drop the turn-ON time to the 10-15uS territory. I've gotten ZM1000 nixies to switch at 7uS, with a 190V supply in combination with a -100V primer, but then, ZM1000s come with primers.
James
--- In NEONI...@yahoogroups.com, "Dan Harboe Burer" <dhb@...> wrote:
>
> ..are back..
>
> http://www.surplusshed.com/pages/item/m2302.html
>
>
> Regards
> Dan
>
Shoot, I saw those a couple weeks ago and almost bought some, I figured they were probably just LED though.
Dan Harboe Burer
I bought 6 units the first time they were up for sale.. That's why I know
there aren't LED's. I have regularly kept snooping around on surplussheds
site ever since. Lots of fun stuff (But I haven't bought anything else
there..yet )
Dan
----- Original Message -----
From: "James" <james...@gmail.com>
To: <NEONI...@yahoogroups.com>
Sent: Monday, June 14, 2010 7:20 PM
Subject: [NEONIXIE-L] Re: Surplusshed display units
>
>
> --- In NEONI...@yahoogroups.com, "Dan Harboe Burer" <dhb@...> wrote:
>>
>> ..are back..
>>
>> http://www.surplusshed.com/pages/item/m2302.html
>>
>>
>> Regards
>> Dan
>>
>
>
> Shoot, I saw those a couple weeks ago and almost bought some, I figured
> they were probably just LED though.
>
>
James
--- In NEONI...@yahoogroups.com, "Dan Harboe Burer" <dhb@...> wrote:
>
> I bought 6 units the first time they were up for sale.. That's why I know
> there aren't LED's. I have regularly kept snooping around on surplussheds
> site ever since. Lots of fun stuff (But I haven't bought anything else
> there..yet )
>
> Dan
>
>
They have a bunch of cool stuff. I bought a little 3 phase induction motor for my drill press, it's the smallest one I've ever seen, and 1/3HP is not bad.
Gelu
[edited by A.J. - please trim quoted material]
--- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@...> wrote:
<snip>
> The sample circuit they used, uses a 4511, which is a common CMOS
> decoder chip. The 4511 is used with common cathode 7-seg LED
> displays. What this means to you, is that all the signals are NOT
> inverted, so do NOT add that 4049 inverter. It means your circuit
> as originally shown, is at least of the right polarity. The
> ghosting is caused by something else.
<snip>
> Finally, David is right, you should make those base-emitter bleeder
> resistors a little smaller (to 10K-22K, from 100K). The base
> current is now around 380uA. A 10K resistor will only divert ~70uA,
> leaving plenty to drive the MPSA92, and it will turn it OFF faster.
> A 30uS blanking interval is marginal, but acceptable for most
> nixies.
Further last trials I can say the followings:
- even in the the sample circuit 4511 is used, if I don't use 4049 inverter, the Nixies are not working properly (they are displaying just characters 0, 1, 2 and 6) - I know that it is strange thing but that it is, I'm suspecting the 74141 driver (maybe it recquires inverted input signals ...). In the first instance I tought to interface in a right way a CMOS circuit output (the clock - 351) with a TTL circuit input (the driver - 74141) and I used for this the buffers included in 4050 IC, but the wrong displayed mode persisted and the remmedia was to use the inverter gates 4049.
- I didn't succeed to improve the switch off time of MPSA92
(I changed the 100k base-emiter resistor with the suggested 10K value but the turn off time remains the same: around 30 usec)
Now, with a proper resistors polarisation network for MPSA42 the circuit works in a satisfactory way - there are no ghosting shadows on the Nixies.
A.J.
--- In NEONI...@yahoogroups.com, "Gelu" <sahist_amator@...> wrote:
> --- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@> wrote:
> <snip>
> > The sample circuit they used, uses a 4511, which is a common CMOS
> > decoder chip. The 4511 is used with common cathode 7-seg LED
> > displays. What this means to you, is that all the signals are NOT
> > inverted, so do NOT add that 4049 inverter. It means your circuit
> > as originally shown, is at least of the right polarity.
> <snip>
>
> Further last trials I can say the followings:
> - even in the the sample circuit 4511 is used, if I don't use 4049
> inverter, the Nixies are not working properly (they are displaying
> just characters 0, 1, 2 and 6)
<snip>
I don't understand where you are using the 4049 inverting buffer. Mike is right, you shouldn't need one. Can you post a new schematic diagram, with the circuit you are using now?
A.J.
Gelu
--- In NEONI...@yahoogroups.com, "A.J." <a.j.franzman@...> wrote:
> --- In NEONI...@yahoogroups.com, "Gelu" <sahist_amator@> wrote:
> > Further last trials I can say the followings:
> > - even in the the sample circuit 4511 is used, if I don't use 4049
> > inverter, the Nixies are not working properly (they are displaying
> > just characters 0, 1, 2 and 6)
> <snip>
>
> I don't understand where you are using the 4049 inverting buffer.
> Mike is right, you shouldn't need one. Can you post a new schematic
> diagram, with the circuit you are using now?
This is the complete diagram that I'm using now; it is working ok, without ghosting issues.
http://i48.tinypic.com/2mfykgw.jpg
A.J.
--- In NEONI...@yahoogroups.com, "Gelu" <sahist_amator@...> wrote:
> --- In NEONI...@yahoogroups.com, "A.J." <a.j.franzman@> wrote:
> > --- In NEONI...@yahoogroups.com, "Gelu" <sahist_amator@> wrote:
> > > Further last trials I can say the followings:
> > > - even in the the sample circuit 4511 is used, if I don't use 4049
> > > inverter, the Nixies are not working properly (they are displaying
> > > just characters 0, 1, 2 and 6)
> > <snip>
> >
> > I don't understand where you are using the 4049 inverting buffer.
> > Mike is right, you shouldn't need one. Can you post a new schematic
> > diagram, with the circuit you are using now?
>
>
> This is the complete diagram that I'm using now; it is working ok, without ghosting issues.
> http://i48.tinypic.com/2mfykgw.jpg
So apparently, either the data is wrong for the MMC351, or you actually have a different part which has inverted BCD outputs???
A.J.
threeneurons
> > This is the complete diagram that I'm using now; it is
> > working ok, without ghosting issues:
> > http://i48.tinypic.com/2mfykgw.jpg
>
> So apparently, either the data is wrong for the MMC351, or you
> actually have a different part which has inverted BCD outputs???
>
> A.J.
>
http://www.datasheetcatalog.com/datasheets_pdf/M/M/C/3/MMC351.shtml
http://www.datasheetcatalog.org/datasheet/microelectronica/mmc4511.pdf
Or the datasheet is wrong (! ?)
I figured maybe their MMC4511 isn't the same as a generic CMOS 4511, but the datasheet seems to indicate that its just an ordinary 4511, both on the inputs and outputs. All 'active-high'.
Maybe the MMC351 datasheet is incorrect ?
Is it really a 4049 ?
You can't argue with a working circuit. Sometimes you can. If its some analog detail, such as a level threshold, or slight timing difference. But in this case, its the polarity of 4 DIGITAL signals. I suspect one or both of those MMC--- datasheets aren't what they're suppose to be.
If these chips are still readily available, from some online hobby store or such, then we may want to stash some of this data here in this group's FILE section. I know a lot of group members like chips like this (Westdave), and they don't want to rediscover those details all over again.
Gelu
--- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@...> wrote:
> > > This is the complete diagram that I'm using now; it is
> > > working ok, without ghosting issues:
> > > http://i48.tinypic.com/2mfykgw.jpg
> >
> > So apparently, either the data is wrong for the MMC351, or you
> > actually have a different part which has inverted BCD outputs???
>
> http://www.datasheetcatalog.com/datasheets_pdf/M/M/C/3/MMC351.shtml
> http://www.datasheetcatalog.org/datasheet/microelectronica/mmc4511.pdf
>
> Or the datasheet is wrong (! ?)
>
> I figured maybe their MMC4511 isn't the same as a generic CMOS 4511, but the datasheet seems to indicate that its just an ordinary 4511, both on the inputs and outputs. All 'active-high'.
>
> Maybe the MMC351 datasheet is incorrect ?
>
> Is it really a 4049 ?
>
> You can't argue with a working circuit. Sometimes you can. If its some analog detail, such as a level threshold, or slight timing difference. But in this case, its the polarity of 4 DIGITAL signals. I suspect one or both of those MMC--- datasheets aren't what they're suppose to be.
>
> If these chips are still readily available, from some online hobby store or such, then we may want to stash some of this data here in this group's FILE section. I know a lot of group members like chips like this (Westdave), and they don't want to rediscover those details all over again.
>
I still have some doubts about 74141 IC inputs ...
Related to MMC351, this clock circuit is obsoleted from almost 20 years ago, but I made the Nixie clock around it based on the stock that I have in my basement (a couple dozen ...)
Also I have a couple pieces of MM5309 and I'm thinking to give to it the same destination.
A.J.
--- In NEONI...@yahoogroups.com, "threeneurons" <threeneurons@...> wrote:
> > > This is the complete diagram that I'm using now; it is
> > > working ok, without ghosting issues:
> > > http://i48.tinypic.com/2mfykgw.jpg
> >
> > So apparently, either the data is wrong for the MMC351, or you
> > actually have a different part which has inverted BCD outputs???
>
> http://www.datasheetcatalog.com/datasheets_pdf/M/M/C/3/MMC351.shtml
> http://www.datasheetcatalog.org/datasheet/microelectronica/mmc4511.pdf
>
> Or the datasheet is wrong (! ?)
>
> I figured maybe their MMC4511 isn't the same as a generic CMOS 4511, but the datasheet seems to indicate that its just an ordinary 4511, both on the inputs and outputs. All 'active-high'.
>
> Maybe the MMC351 datasheet is incorrect ?
I think I said that... I did notice that it isn't a datasheet, but pages from a book, and there's a fairly obvious error on the second page (did you spot it?), so the possibility of additional errors elsewhere seems likely. It could also be preliminary data, and the production part was deliberately changed.
A.J.
