Re: [neonixie-l] Re: Dekatron Experiments

[John Rehwinkel]

> Oh on those. If you get a regular nixie supply; 170V. Make a divider out off 
> 3 22K resistors, and the two voltages are 57V and 113V, which are just 
> rounded off to 55V and 110V. That's assuming it was dead nuts on 170V to 
> start with. Throw that 170V on a tripler and you get 510V. Again rounded off 
> to 500V. All these voltages can vary quite a bit, and not effect operation.

But I want to effect operation!

- John

-- 
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/7CF6E914-DF16-4034-8DD6-438F56D00473%40mac.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Original evaluation PCB for the Burroughs BG12205-2 bar graph displays?

[John Rehwinkel]

> In the early 80's I was doing 8 mil lines & spaces, getting two traces 
> between pads on .1" spacing. It was pushing the limit of the local board 
> houses. It was also just the ticket for high speed memory boards for a couple 
> decades, until the DDR stuff came along. Now 6 mil is considered large, 4 mil 
> is the bottom size without paying for low yields. I still use 8 mil on boards 
> that can afford it, from a density standpoint.

I do 10/10 if it will fit, a lot of 8/8 and a few 6/6, but most of the stuff I 
design isn’t terribly tight.

> Gold plating has been making a big comeback in recent years -- ENIG -- or 
> electroless nickel immersion gold. It's a good finish for lead free solder.  
> Almost everything  I've done for 7 or 8 years now gets ENIG. At many board 
> houses it's not even a premium cost adder. It's a soft finish -- not suitable 
> for edge connectors.

ENIG is much better for high lead count surface mount packages than HASL.  It’s 
also pretty.  I did one run of lead free HASL for a group that wanted lead free 
but didn’t want to spring for ENIG.  It worked, but wasn’t as good looking.

> PCB design has come a long way. I've been doing it 30 years now. There are 
> things done on these early PCBs that we avoid like the plague now.

Care to share?  I’m always up to learn something new.  I’ve been amazed at the 
things that are common knowledge one place and unheard-of at another.

- John

-- 
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/98848533-FCFE-40BE-82C6-76FF72E5EA69%40mac.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: One slot on my clock keeps burning out tubes, while the others have been good for years. Thoughts?

[A.J. Franzman]

Either bad luck and coincidence getting three bad tubes, or THIS:

On Wednesday, August 17, 2016 at 7:56:44 AM UTC-7, Nick wrote:
>
> Either it's a mechanical issue with the socket stressing the pins to the 
> extent the tube eventually becomes leaky
>
> ...or it's to do with the anode driver (the cathodes share the same drive 
> across all tubes) - either the anode resistor has failed in some way (or 
> isn't the correct value) or the opto-coupler is leaky/faulty (but I'm not 
> sure how that would cause premature tube 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/c6b9264f-ec47-400e-97f3-c20efefe9a81%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: Dekatron Experiments

[Keith Moore]

OMG! That 0A5 arc discharge dekatron spinner is amazing, too! 

I am ignorant as to the ways of you EE masters. But I want to make one now. 
  Damn you!


On Tuesday, August 16, 2016 at 5:56:58 PM UTC-4, threeneurons wrote:
>
> Well, those who know me, know that I have attraction to dekatrons. Here's 
> some of my latest tinkering. A long while back I designed a 4000 CMOS base 
> pendulum circuit. It used a 4017 counter, and a 4013 flip-flop. I revisited 
> that circuit, and came up with one that just uses on 4518 dual counter:
>
> 
>
> In addition to the one chip, and the HV interfacing transistors, there's a 
> clever little piece of what Don Lancaster referred to as "Mickey Mouse 
> logic". The one transistor and its associated components, in the pink 
> field, form an exclusive-OR gate. Half the chip along with XOR ckt, make 
> the guide timing circuit. The other half of the chip is used as the 
> direction flip-flop. Circuits quite nicely. HV supplies, and clock, not 
> shown.
>
>
> In my test circuit, I decided to run two dekatrons. One master, and one 
> slave. To align the slave tube, I added a reset circuit to K0 (normally 
> used as the NDX). This forces the glow to K0, instantly. Here's that 
> portion of the circuit:
>
> 
>
> I hooked it such that when QD (pin 14), of the IC went high, the reset was 
> issued. That meant a reset every 10 'flips'.
>
>
> It worked ... mostly. I noticed something strange. If I used a A101 as a 
> slave, the glow reset to K0, as expected. But using several 6802's and a 
> GC10B, the glow appeared to reset to K9. After some poking around, I 
> decided to check the reset, with the guides (G1 & G2) off. Then, 
> indifferent to tube type, the reset forced to glow to K0. So this new 
> pendulum circuit "backsteps" some tubes from K0 to K9, almost instantly.
>
>
> Here's a photo of the test circuit, and video of it running:
>
>
> 
>
>
>
>
> Video  of above circuit.
>
>
> Enjoy
>
>
>
>
>

-- 
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/ebc1bffe-734a-47ef-8736-93d2738bf2b0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: Maybe a stupid question...

[gregebert]

After studying the circuit, I'm fairly certain the way it works is that 
when a particular bulb fires, the other bulbs that are capacitively coupled 
to it will extinguish. Neon bulbs have built-in hysteresis (firing voltage 
is higher than sustaining voltage), which means that circuits can be 
constructed that self-oscillate.

So, back to debugging, if the capacitors are leaky I think it could prevent 
oscillation. Also, I'm not sure what resistors R2, R5, R8, R11, R14, and 
R17 do, so you may want to snip them off temporarily for debug. I dont 
think the transistor circuit will cause any interference.

-- 
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/fc60c218-f116-4a88-8f27-e2d62fc0391b%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: Dekatron Experiments

['threeneurons' via neonixie-l]

> Yeah. I saw the odd voltage levels too!
>

Oh on those. If you get a regular nixie supply; 170V. Make a divider out 
off 3 22K resistors, and the two voltages are 57V and 113V, which are just 
rounded off to 55V and 110V. That's assuming it was dead nuts on 170V to 
start with. Throw that 170V on a tripler and you get 510V. Again rounded 
off to 500V. All these voltages can vary quite a bit, and not effect 
operation.

-- 
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/d2f7a576-ba23-46b3-a1d3-c39e13633b3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: Dekatron Experiments

['threeneurons' via neonixie-l]

Oh, yes, its a timing issue. It works fine with the guides detached. Its 
just more of a curiosity. Flipping the guide pins does make it switch to 
K1. I'll have to analyze my circuit timing more closely to discover the 
actual mechanism details. The reset is issued at the same time as the flip. 
Ideally this means its at a main cathode, and the guides should be off. But 
this is a simplistic circuit, where the prior "main" state, now becomes the 
new "G1" state. To fix it, here, would add complication. If someone wanted 
to use the pendulum circuit, with the reset on the second dekatron, its 
just simpler to rotate the tube such that K9 is on top. Its quite 
consistent.

In a microcontroller circuit, its a non-issue. To have the glow rest at a 
"main" cathode, the guides need to be off. So, you just turn both bits off, 
then issue your reset pulse. Also if its used in a microcontroller circuit, 
Cr is omitted. Recommended minimum pulse duration is 100uS. The reset can 
be used instead of monitoring the glow position, and waiting for the glow 
to be at K0 (NDX). 

On Tuesday, August 16, 2016 at 3:05:29 PM UTC-7, Dekatron42 wrote:
>
> Nice!
>
> Could it be that you have a timing issue with the guides not being totally 
> "off" when you do the reset?
>
> Or that the reset does not load the cathode enough so that the glow steps 
> to a lower potential, being K9? You can also check what happens if you 
> exchange the guides, then it should step to 1 instead of 9 - unfortunately 
> not telling you what the problem is but it should at least be another way 
> of measuring what's happening. One more thing to try is to add a resistor 
> in the RTN +55V power line as that then changes the balance between K0 and 
> the other guides.
>
> /Martin
>

-- 
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/363edb00-3b07-4248-8d6b-37443f46a2ef%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Original evaluation PCB for the Burroughs BG12205-2 bar graph displays?

['Terry S' via neonixie-l]

I am sure you are right -- notice some of the odd angles and mixed 45/90s. 
Also notice the very poor registration of the holes to pads.

In the early 80's I was doing 8 mil lines & spaces, getting two traces 
between pads on .1" spacing. It was pushing the limit of the local board 
houses. It was also just the ticket for high speed memory boards for a 
couple decades, until the DDR stuff came along. Now 6 mil is considered 
large, 4 mil is the bottom size without paying for low yields. I still use 
8 mil on boards that can afford it, from a density standpoint.

Gold plating has been making a big comeback in recent years -- ENIG -- or 
electroless nickel immersion gold. It's a good finish for lead free solder. 
 Almost everything  I've done for 7 or 8 years now gets ENIG. At many board 
houses it's not even a premium cost adder. It's a soft finish -- not 
suitable for edge connectors.

Terry

On Wednesday, August 17, 2016 at 12:46:11 PM UTC-5, gregebert wrote:
>
> I'm certain it's a "tape-and-donuts" PCB layout. The first CAD-developed 
> PCB's I saw in the 1980's had telltale 45-degree bends everywhere, and 
> everything else was perfectly orthogonal.
>
> Notice there are no traces going between 0.1"-spaced pads, due to the 
> primitive manufacturing capabilities at that time.
> Remember when they used to gold-plate everything ? Some were so thickly 
> plated they appeared to be solid gold.
>
>

-- 
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/6781acc4-fdb6-4139-b111-20e0303406ac%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: Dekatron Experiments

['threeneurons' via neonixie-l]

On Wednesday, August 17, 2016 at 8:16:44 AM UTC-7, Nick wrote:
>
> Very nice!
>
> I like the discrete XOR gate but its a lot of complexity when you can get 
> a single "tiny logic" XOR gate from Fairchild, Texas or whoever for $0.10 - 
> one SOT-23 (as opposed to 4 diodes, 4 resistors and a transistor) and 
> you're there... :)
>
> Yeah. I saw the odd voltage levels too!
>
> Nick
>

That's actually the case with everything, when you compare old with new 
electronics. I can go the whole 9-yards (US term) and just use a 
microcontroller. Oh ... I do, its that damn DoHickie kit ! Modern 
technology has turned the economics upside down. So, this project is mostly 
a mental exercise. At least its more stable, than those log amp multiplier 
circuits they use to make us suffer thru, in lab experiments, back in 
school !
 

-- 
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/c15d0715-f7f1-4df9-a96c-36d13db839e9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Original evaluation PCB for the Burroughs BG12205-2 bar graph displays?

[gregebert]

I'm certain it's a "tape-and-donuts" PCB layout. The first CAD-developed 
PCB's I saw in the 1980's had telltale 45-degree bends everywhere, and 
everything else was perfectly orthogonal.

Notice there are no traces going between 0.1"-spaced pads, due to the 
primitive manufacturing capabilities at that time.
Remember when they used to gold-plate everything ? Some were so thickly 
plated they appeared to be solid gold.

-- 
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/04c72e1f-c3a4-4a46-9ba7-eb9e01203f30%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: One slot on my clock keeps burning out tubes, while the others have been good for years. Thoughts?

['Terry S' via neonixie-l]

I think that it's a mechanical issue. Any electrical issue that over-drove 
the cathodes would be visually apparent during operation -- especially at 
levels that could destroy cathodes in a matter of days or weeks.

Terry



On Wednesday, August 17, 2016 at 9:56:44 AM UTC-5, Nick wrote:
>
> I'm with the others - there is something about this slot which 
> differentiates it from the others.
>
> Either it's a mechanical issue with the socket stressing the pins to the 
> extent the tube eventually becomes leaky
>
> ...or it's to do with the anode driver (the cathodes share the same drive 
> across all tubes) - either the anode resistor has failed in some way (or 
> isn't the correct value) or the opto-coupler is leaky/faulty (but I'm not 
> sure how that would cause premature tube failure), Vceo for a PC817 is 80V, 
> which should be OK, but this one may have failed - they are really cheap, 
> so maybe change it anyway :)
>
> Check the values of R11-R16.
>
> Have you tried the "faulty" tube in a known-good position in the clock? 
> Are you certain the tube has failed, not the surrounding circuitry cooking 
> slowly over time, then failing, then you turn the clock off to replace the 
> tube so it all cools down and the cycle repeats?
>
> Cheers
>
> Nick
>
> .
>

-- 
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/3084282c-d4c3-483f-97f5-eb3d3e961f31%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: Dekatron Experiments

[Nick]

Very nice!

I like the discrete XOR gate but its a lot of complexity when you can get a 
single "tiny logic" XOR gate from Fairchild, Texas or whoever for $0.10 - 
one SOT-23 (as opposed to 4 diodes, 4 resistors and a transistor) and 
you're there... :)

Yeah. I saw the odd voltage levels too!

Nick

-- 
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/76f329d1-4ab2-481e-96e5-1f4c83871542%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[neonixie-l] Re: One slot on my clock keeps burning out tubes, while the others have been good for years. Thoughts?

[Nick]

I'm with the others - there is something about this slot which 
differentiates it from the others.

Either it's a mechanical issue with the socket stressing the pins to the 
extent the tube eventually becomes leaky

...or it's to do with the anode driver (the cathodes share the same drive 
across all tubes) - either the anode resistor has failed in some way (or 
isn't the correct value) or the opto-coupler is leaky/faulty (but I'm not 
sure how that would cause premature tube failure), Vceo for a PC817 is 80V, 
which should be OK, but this one may have failed - they are really cheap, 
so maybe change it anyway :)

Check the values of R11-R16.

Have you tried the "faulty" tube in a known-good position in the clock? Are 
you certain the tube has failed, not the surrounding circuitry cooking 
slowly over time, then failing, then you turn the clock off to replace the 
tube so it all cools down and the cycle repeats?

Cheers

Nick

.

-- 
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/bb0c17cf-e5e2-4109-9515-0d16ace15ce2%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] One slot on my clock keeps burning out tubes, whilethe others have been good for years. Thoughts?

[chuck richards]

Sounds like a multiplexing issue to me.
I am guessing that this particular tube in question
is getting cooked for quite a bit longer than the others.

Chuck
>
>
> Original Message 
>From: jwalton...@gmail.com
>To: neonixie-l@googlegroups.com
>Subject: Re: [neonixie-l] One slot on my clock keeps burning out
>tubes, whilethe others have been good for years. Thoughts?
>Date: Tue, 16 Aug 2016 21:28:09 -0700
>
>>
>>
>>When all of the cathodes in a single tube lose large patches of
>coverage in a short period of time, it still seems that it would be a
>failure of the gas mixture.  Over driving the cathodes for a period
>of time would cause an eventual loss of brightness but it should
>cause variations among the cathodes and not a sudden loss of coverage
>on all cathodes.  
>>GregEbert suggested to look at the socket to make sure that pins
>aren't stressed, which seems to match your description of how the
>tube failed...  Three failures in the same socket does seem to be
>more than coincidence and it would be a good question to determine
>whether mechanical or electrical.   You should follow up with the
>group regarding future failures and/or findings as it is
>interesting. 
>>
>>
>> Original message 
>>From: Nicholas Stock  
>>Date: 8/16/2016  8:26 PM  (GMT-07:00) 
>>To: neonixie-l@googlegroups.com 
>>Subject: Re: [neonixie-l] One slot on my clock keeps burning out
>tubes, while the others have been good for years. Thoughts? 
>>
>>Could the photocoupler at HC2 be the culprit at all? When you change
>the tubes, is the brightness of the IN14 at that position the same as
>the rest?
>>On Tue, Aug 16, 2016 at 8:10 PM, gregebert 
>wrote:
>>Since it's multiplexed, it might be worth comparing the 'on' time of
>each tube (I'm guessing it's H2 on the schematic ?). You'll need a
>scope. Maybe there is a quirk in the controller firmware that leaves
>this particular tube on longer ?? There's also a slight difference in
>the circuit around tubes H2, H3, and H4 though nothing pops out as a
>possible culprit.
>>Any chance there's a source of heat next to tube H2 ?
>>
>>Only other thing I can think of is if the particular socket puts
>more stress on this tube.
>>
>>
>>
>>-- 
>>
>>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/14912c65-1aa1-46a3-b2dd-
>f2320a6d966c%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 email to neonixie-l@googlegroups.com.
>>
>>To view this discussion on the web, visit
>https://groups.google.com/d/msgid/neonixie-l/CAOX%2BRHJ4H9qTzQ0OrEh2L
>YmrN%2Bn7nVhCjfc9coGRKwFy9yyQwA%40mail.gmail.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/5hnpmx7sytvq69mp5yfocrcp
>.1471406207133%40email.android.com.
>>For more options, visit https://groups.google.com/d/optout.
>>



$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 
http://www.All2Easy.net for more details!

-- 
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/380-220168317142733655%40all2easy.net.
For more options, visit https://groups.google.com/d/optout.