Trigger clock revisited

[petehand]

A few years ago - well, maybe seven - I made a trigger clock inspired by the one on Grahame Marsh's web site. I could never get the darn thing to work reliably in the dark - some rings would get stuck at random intervals - so I shelved it. I dug it out recently because I needed to use its HV power supply to un-poison some nixies, and decided to have another go at it. I put rows of UV LEDs down alongside the rings. Now it has the opposite problem! The UV light made the tubes so sensitive that some would trigger just with the touch of a finger. I had to put black card barriers between the columns to stop the light getting to the nixie driver columns, because they were flickering on and off at random, and slide black sleeving over a few counting tubes. It's not that they wouldn't go out, but that they'd trigger on when it was a different tube's turn. I still haven't got all the bugs out - it seems to gain about 12 hours a day, and not in whole hour increments. I THINK this is because the 50 minute trigger tube sometimes fires on the wrong pulse and cuts the hour short. I did once see this - it counted 57, 58, 58, 00 (the glow transferred to 00) and then back to 51, 52 etc, but I haven't been able to catch it doing so again.

Anyway, I just wanted to share that UV (Ultra Viagra?) really does make these old tubes frisky again. Sorry about the terrible picture, I shot it in low light, my camera seems to be rather sensitive to UV, and the clock is behind glass to keep the dust off. It looks in the pic as if the LEDs are really glaring bright, and you can see the UV glow spread out in front of them. In fact to the naked eye, you can barely see any light at all, it's just a hint you see out of the corner of your eye. The LEDs are running at only 1 milliamp.

[Tidak Ada]

Did you already try to diminish the current through the LED's to lower the trigger sensitivity?

An other idea to pack the trigger tubes in sleeves of tubing to separate them from each others. Don't forget also the gas in the tubes has some spectral emission in the UV region that influences the neighboring tubes' trigger level.

 

eric

---

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of petehand
Sent: zaterdag 29 maart 2014 8:59
To: neoni...@googlegroups.com
Subject: [neonixie-l] Trigger clock revisited

A few years ago - well, maybe seven - I made a trigger clock inspired by the one on Grahame Marsh's web site. I could never get the darn thing to work reliably in the dark - some rings would get stuck at random intervals - so I shelved it. I dug it out recently because I needed to use its HV power supply to un-poison some nixies, and decided to have another go at it. I put rows of UV LEDs down alongside the rings. Now it has the opposite problem! The UV light made the tubes so sensitive that some would trigger just with the touch of a finger. I had to put black card barriers between the columns to stop the light getting to the nixie driver columns, because they were flickering on and off at random, and slide black sleeving over a few counting tubes. It's not that they wouldn't go out, but that they'd trigger on when it was a different tube's turn. I still haven't got all the bugs out - it seems to gain about 12 hours a day, and not in whole hour increments. I THINK this is because the 50 minute trigger tube sometimes fires on the wrong pulse and cuts the hour short. I did once see this - it counted 57, 58, 58, 00 (the glow transferred to 00) and then back to 51, 52 etc, but I haven't been able to catch it doing so again.

Anyway, I just wanted to share that UV (Ultra Viagra?) really does make these old tubes frisky again. Sorry about the terrible picture, I shot it in low light, my camera seems to be rather sensitive to UV, and the clock is behind glass to keep the dust off. It looks in the pic as if the LEDs are really glaring bright, and you can see the UV glow spread out in front of them. In fact to the naked eye, you can barely see any light at all, it's just a hint you see out of the corner of your eye. The LEDs are running at only 1 milliamp.

--
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/420542f9-24ad-4258-9473-5cfc6d10be6a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Grahame Marsh]

Well done Pete, I have used a pond water 8W UV lamp to keep the tubes running overnight.  But I just use the clock as a demonstration clock for visitors as it has a lot of "moving" parts.  I run the cicuit at a higher supply voltage than the Dance original and compenstated by increasing the anode resistor value (apart from maintaining the correct current through the tubes I had no reason to choose the anode resistor).  Recollection is that you increased the cathode resistor value instead arguing it was correct for the interstage coupling to do so.  I wonder if the increased cathode pulse is contrinuting to the ring sensitivity?  Your tube density is much higher than mine as well.

I still have a bunch of xc18's left for a second clock so perhaps your success will get me started again.

For new group members here's my clock web page (as always in need of updates)

http://www.sgitheach.org.uk/nixie3.html

Cheers Grahame

[petehand]

I watched it today and saw it come from an hour behind to three hours ahead in the space of less than two actual hours, and I found the culprit - it's the 50 minute tube. It counts 58, 59, 00, 01, 02, and then somewhere in the middle of the ought-minutes the 50 tube strikes again. But, not every time.

Tidak, I will try reducing the LED current further, since it's easy, but it's already down at 1mA. The efficiency of modern LEDs amazes me. I have sleeved the worst offending tubes, but only up to the top of the anode ring - maybe it needs more. People can see the tubes in action and I don't want the sleeves to be noticeable.

Grahame, you are right, I did change the resistor values. For those following after, originally the cathode had a 27k resistor on top of a 56k resistor, with a 56k anode resistor. I changed all the 27k cathode resistors from 27k to 56k, and changed the anode resistor from 56k to 27k. See diagrams above - top is original, second is modified. What this did was increase the amplitude of the carry pulse from 21V, which was right on the margin, to 35V, without altering the tube current. All the stages that previously stuck then worked as intended, except in the dark.

I'm not about to change it back to see if it now works with the original values, as that means changing more than 100 resistors and scrupulously cleaning the PCB afterward, since a little bit of contamination can cause it to stick. But I may increase the anode resistors since there are only a few and they're at the ends of the rows. What this will do is lower the pre-trigger bias applied to the next stage. It's nominally 56V at the moment, which was marginal with the 21V trigger pulse, but with the 35V pulse I can afford to drop it somewhat. At 43k the bias would be 52V. From the XC18 data sheet, the must-trigger voltage is 62 to 74 volts. Some tubes apparently don't make it. But with the 35V pulse, they would still get over 85V on the trigger.

[Grahame Marsh]

I fully appreciate not wanting to change over 100 resistors! I can only say that I have not had the ring failures under lit conditions that you have.  Under normal day/night running the clock would go for 3 or 4 days before failing overnight.  The normal failure would be a ring would have two adjacent trigger tubes fired.  Earlier (faster) rings would still be counting. 

At the moment I have the clock running with the Dance values of 180V supply voltage, 22k common anode resistor and 27k/56k cathode resistors.  The only difference I found was initially no tube fired on switch on (I am not using a XC24 or two XC18s in parallel for the ring starter) and I had to swing the anode voltage upto about 210V when one tube would fire in each ring.  I could then reduce the voltage (I'm using a Heathkit IP-17 PSU BTW) back down to 180V and the rings would operate normally (and fail overnight as normal).  I'm not sure what this all means but I remembered you had far more science in your approach to my more arbitrary suck and see.  I wonder if the lack of radioactivity in the tubes now means their characteristics are too different?  But for now my aim is to replace the simple PSU with a stabilised 180V all-valve PSU to then use the Dance ring resistor values. Something like this maybe


And then use the UV lamp in a more permanent manner and probably on a time or photocell switch.

Cheers Grahame

--
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/9514742b-110a-4eb4-b32f-ee7269d55a9d%40googlegroups.com.

[petehand]

I admire the holy purity of your art, Grahame, to make a tube regulated PSU. I'm afraid the presence in my junk box of a small mains transformer with 200V and 24V secondaries but no suitable heater voltage led me to stray from the path of righteousness and yield to the cheap seduction of semiconductor rectifiers.

I recall my friend from Hivac mentioning that the XC15/CV2486 was radioactive doped, and they had a relationship with Harwell to supply isotopes. They probably used Krypton 85, which has about a ten year half-life. My tubes all have a 78 date code so the activity will only be 12% of when they were new.

Tomorrow I'm going to change the 50 minute tube, since I can't tame it. It now consistently gives me a 10 minute hour. What's very strange is that it always hands over to the 00 minute tube as it should, but then takes the glow back on the next minute pulse at 01. I suspect it's getting electrical noise from the adjacent unit minute track, since the tens ring doesn't get a pulse at that time. Or, it could be board contamination - I've had some trouble with that because of the small clearances.

The clock has an annoying tendency to light several tubes in the same ring when I'm trying to set it - probably switch bounce. It's sometimes recoverable by working the switch, but the totally reliable solution is a push button for each ring, discreetly hidden away on the back out of sight, that directly grounds one cathode. It's not needed very often but it beats having to power down, wait for all the glows to go out, and then start over.

I'm taking it to the local Maker Faire at the weekend. It's totally unrelated to the booth I'll be minding, but I'm going to hang it on the wall to attract some extra visitors. They'll probably ask whether I'm running it with an Arduino or a Raspberry Pi. };>

[Nick]

For the last Lord knows how many years I've been intending to follow in Grahame's footsteps and build an XC-18 clock - I picked up 300 of these many years back when the guy in Southampton only wanted about 10p each for them - hopefully enough for two clocks.

I'm impressed with Pete's results - it's good that the theory is being tested and the results quantified as the "triggering in the dark" issue has been around for a while - Kr85, as has been pointed out, was used for specifically this reason, but is no longer viable. Let's hope that Pete's experiments results are repeatable - do you have the details of the emission characteristics of those LEDs?

Cheers

Nick 

[Grahame Marsh]

I read it that Pete has two problems and I have one of them - poor triggering in the dark which we understand and have fixes for based on UV from LEDs (a semiconductor solution) or a pond UV lamp (i.e a "valve" solution).  But Pete has a second problem that I do not see which is misfiring under lit conditions.  Pete has issues with his construction density perhaps; mine is not very small.  I'm at a loss to suggest anything; Pete seems to have a triggering sensistivity on the rings far higher than mine and the only substantive component change is that Pete has changed one of the cathode resistors and I change the common anode resistor to account for a higher supply voltage.  Our source of XC18s was the same.

I have the XC18 rings running with 180V (Dance) and I have a Z700U ring running as well (this runs in complete darkness as each has a primer electrode) :D

Grahame

[Nick]

On Tuesday, 1 April 2014 15:17:28 UTC+1, Grahame Marsh wrote:

I read it that Pete has two problems and I have one of them - poor triggering in the dark which we understand and have fixes for based on UV from LEDs (a semiconductor solution) or a pond UV lamp (i.e a "valve" solution).  But Pete has a second problem that I do not see which is misfiring under lit conditions.  Pete has issues with his construction density perhaps; mine is not very small.  I'm at a loss to suggest anything; Pete seems to have a triggering sensistivity on the rings far higher than mine and the only substantive component change is that Pete has changed one of the cathode resistors and I change the common anode resistor to account for a higher supply voltage.  Our source of XC18s was the same.

I have the XC18 rings running with 180V (Dance) and I have a Z700U ring running as well (this runs in complete darkness as each has a primer electrode) :D

Indeed - we all seem to be using the same source of XC-18s (why is he the only person with them in bulk?) - I have a box of Z700Us, so will give those a go as well - MTX-90s are also an option - they look a promising alternative too...

Nick 

[petehand]

My tubes also came from the "guy in Southampton", though I've forgotten his name. My UV LEDs, on the other hand, came from Best Buy in Hong Kong, via Ebay, for $28 a hundred, and I have no part number or other documentation. Today a quick search for "UV LED" on Ebay returns this as the first result -
http://www.ebay.com/itm/NEW-100PCS-5mm-Megabright-Ultra-Violet-LED-UV-Lamp-2-500mcd-BESTBUY-/290726116420?pt=LH_DefaultDomain_0&hash=item43b0a07c44
Unbelievably they are now only $6 a hundred, with shipping included, which means after taking off the cost of shipping they're more like 2 cents each. You might suspect for that price they're factory rejects, and you might be right, since about one in 4 of my lot was either dead out of the bag or failed after a minute's use, but who the heck cares when they're cheaper than their own series resistor.

[Tidak Ada]

There is an other source of UV-LED's. Not as cheap as on eBay, but they can serve shorter wavelengths (until 245nm as I remember). [ http://www.roithner-laser.com/ ]  They will ship also small quantities.

 

400nmis actually no UV, but violet. 380 -37 nm should be les visible, because of the peak wavelength is really UV.

 

eric

 

 

---

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of petehand

Sent: dinsdag 1 april 2014 20:22
To: neoni...@googlegroups.com
Cc: graham...@googlemail.com
Subject: Re: [neonixie-l] Re: Trigger clock revisited

--

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/55bd08fc-7e45-43bc-a928-bdc8bc5644de%40googlegroups.com.

[petehand]

Oooh, those are sterilizing wavelength. They could erase an EPROM and do some damage to your eyesight.

[Grahame Marsh]

Perhaps I should use something less vivacious, yes.  At first I just left the strip light over the workshop bench on (40W at 1.5m) and that was fine as well.  I used the pond lamp as I had a spare to hand - it needs to go out to the pond filter box now it is spring.   Perhaps a "black light" lamp for crystal and rock displays? But I think I'm putting my trust in the Z700U for the next trigger tube clock (all valve of course).

Grahame

On 02/04/2014 07:07, petehand wrote:

Oooh, those are sterilizing wavelength. They could erase an EPROM and do some damage to your eyesight.

--
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/6da39607-c809-4c25-8ea1-93ecf057c1d1%40googlegroups.com.

[Tidak Ada]

May be an open mercury vapor rectifier, like a 866 (not an 866A !) does the same. If you can get one with lower heater consumption it is better for your electricity bill.

Don't know what spectrum OSAGE rectifiers radiate. anyhow they have no filament.

Any other small mercury tubes that could do the job?

 

eric

---

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Grahame Marsh
Sent: woensdag 2 april 2014 12:37
To: neoni...@googlegroups.com

Subject: Re: [neonixie-l] Re: Trigger clock revisited

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/533BE842.6060306%40googlemail.com.

[John Rehwinkel]

> May be an open mercury vapor rectifier, like a 866 (not an 866A !) does the same. If you can get one with lower heater consumption it is better for your electricity bill.
> Don't know what spectrum OSAGE rectifiers radiate. anyhow they have no filament.
> Any other small mercury tubes that could do the job?

Argon tubes also emit a useful amount of UV. Maybe an ordinary AR-1 argon bulb, or an 0Z4G rectifier?

- John

[Tidak Ada]

Nice! So a 0Z4G is OK, Thaen is the next sorrow to meet a minimum current of
30mA, needed for maintaining the conduction.
I love that 0Z4G!

eric

-----Original Message-----
From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On
Behalf Of John Rehwinkel
Sent: woensdag 2 april 2014 18:40
To: neoni...@googlegroups.com
Subject: Re: [neonixie-l] Re: Trigger clock revisited

--
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 an email to neoni...@googlegroups.com.

To view this discussion on the web, visit

https://groups.google.com/d/msgid/neonixie-l/735C9E21-802A-4B47-96E8-989BDA3
C1355%40mac.com.

[Grahame Marsh]

Isn't the 0Z4 in a metal can?

:D

[David Forbes]

0Z4 is metal.
0Z4G is glass.
0Z4GT is glass, but tubular, not curvaceous.

[Instrument Resources of America]

There were TWO versions, metal can with octal base, which by far and
away was the majority of them, and also a GLASS version, 0Z4G. Ira.

On 4/2/2014 1:15 PM, Grahame Marsh wrote:
>

[Instrument Resources of America]

P.S. Pic enclosed here of 0Z4G. Ira.

On 4/2/2014 1:15 PM, Grahame Marsh wrote:
>

[Grahame Marsh]

Right, I've only seen the metal can version - that was a few years ago
as well. G

[Instrument Resources of America]

By the way Grahame, FYI, they were used extensively, (and only???) as
the cold cathode full wave rectifiers, in vibrator type power supplies,
for automobile radios. I know of no other use, but might be interesting
to see what others here may say on that. Ira.

[Instrument Resources of America]

Upon further investigation in TCA's data cache, I find the following,
0Z4 first registered by Raytheon on Oct 24th 1935 with a metal shell
which tapered down in size as it got closer to the top of the metal
shell, an 0Z4A first registered by Raytheon on Oct 15th 1956, with the
standard metal shell. The index in the data cache lists an 0Z4G (no GT
ever registered) but when I actually go to the same release number which
is cited, the same as the 0Z4, there is no information there to be had
on the 0Z4G. The only thing I can say is that I have an 0Z4G with
'tubular' glass envelope, not an 'ST' envelope, which is somewhat
curious. Raytheon receiving tube data book bears out the fact that
their 0Z4G was in fact tubular glass and not ST glass. You would think
that the nomenclature would have been 0Z4GT but it was not. Go figure!!
Ira.

[Tidak Ada]

Here 0Z3G at Work

https://groups.google.com/d/msgid/neonixie-l/533C70F4.70302%40dakotacom.net.

[Instrument Resources of America]

P.S. Maybe someone here can elaborate on why Raytheon numbered it as
0Z4G rather than 0Z4GT which would have made more sense, and adhered to
"industry tube standards". As I do not have clue. Ira.

[Charles MacDonald]

On 14-04-02 04:15 PM, Grahame Marsh wrote:
>
> Isn't the 0Z4 in a metal can?

>> Nice! So a 0Z4G is OK, Thaen is the next sorrow to meet a minimum
>> current of
>> 30mA, needed for maintaining the conduction.
>> I love that 0Z4G!

One of the long line of "exception to the rule" the 0Z4G is a cute
little glass enclosed gas rectifier. AND the more common 0Z$ is
actually the only Metal tube that will still work if you open the shell!
inside is a small Glass bulb!

Attachment is a couple of paragraphs from the original data sheet.

The 0Z4 is an oldie but a goodie, having been introduced back in Oct
1935 by Raytheon. the 0Z4A version came out in 1956, so it is a long
running production item.



--
Charles MacDonald Stittsville Ontario
cm...@zeusprune.ca Just Beyond the Fringe
http://Charles.MacDonald.org/tubes
No Microsoft Products were used in sending this e-mail.

[Charles MacDonald]

On 14-04-02 05:19 PM, Instrument Resources of America wrote:
> P.S. Maybe someone here can elaborate on why Raytheon numbered it as
> 0Z4G rather than 0Z4GT which would have made more sense, and adhered to
> "industry tube standards". As I do not have clue. Ira.

I looked and could not find the G version in the JEDEC files.

My guess is that it really did come out as a G version, like a 6V6G,
50l6G etc, but the tiny little bulb is of course much smaller than the
"Bantam" (GT or T9 size) Most G tubes had an ST bulb

[Instrument Resources of America]

Hello Charles, It is there. 0Z4, 0Z4G, along with 0Z4A.This appears to
be an oddity. As far as the JEDEC files in the TCA 'data cache' go the
0Z4G is listed in the INDEX under "Main type", and references the same
exact release number as the plain 0Z4 of #49. Upon opening up release
#49 there is only data on the 0Z4 and nothing for the 0Z4G. I have
looked in the RCA and Raytheon literature for the 0Z4G and find it in
both places with an envelope of T-7, with a small shell octal base. I
went all the way back to RCA 'RC14' copyright of 1940 where I first
found it listed for RCA on pg. 47. If it EVER came out as a "G" version
with "ST" glass I can't find any evidence of it at all, any where. The
0Z4'G' that I have, along with other collectors, is with T-7 glass just
as it is listed everywhere. Does anyone here have or seen an 0Z4'G' with
ST glass?? Ira.

[Tidak Ada]

I have here an addendum from RCA to their 0Z4 datasheet

eric

-----Original Message-----
From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On
Behalf Of Charles MacDonald
Sent: donderdag 3 april 2014 3:56
To: neoni...@googlegroups.com
Subject: Re: [neonixie-l] Re: Trigger clock revisited

--
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 an email to neoni...@googlegroups.com.
To view this discussion on the web, visit

https://groups.google.com/d/msgid/neonixie-l/533CBFAF.4080305%40zeusprune.ca

[Charles MacDonald]

On 14-04-02 10:55 PM, Instrument Resources of America wrote:
> Hello Charles, It is there. 0Z4, 0Z4G, along with 0Z4A.This appears to
> be an oddity. As far as the JEDEC files in the TCA 'data cache' go the
> 0Z4G is listed in the INDEX under "Main type", and references the same
> exact release number as the plain 0Z4 of #49. Upon opening up release
> #49 there is only data on the 0Z4 and nothing for the 0Z4G.

I was using search in the "Mail release file and the 0Z4G did not come
up. I think I only have one of the little guys, and I recall when I
checked it on my Conar Emission Tester it did have a nice glow.

No I suspect that it was only made with the little T7 bulb, which is
about the size of the bulb in the metal version

That is Shown in My old Canadian General Electric Radiotron Manual also.

My guess is that the rest of the metal tubes almost always had a G
version before the T9 GT versions were invented, so the glass version of
the 0Z4 was also called G. The rest of the G version tubes were
originaly made in the ST-14 envelope, and changed to the GT name when
downsized to the T9 bulb. BUT you could not downsize the 0Z4G.

Anyone remember the GT/G designation?

[Instrument Resources of America]

I agree that the "G" version of almost all tubes I can think of that had
that designation, were of the "ST" glass shape, except for at least this
one odd ball 0Z4G, with the T-7 glass. The GT versions applied to 'any'
glass 'tubular' tube regardless of the size, think 6SN7GT versus 5U4GT,
etc. And yes I do remember seeing the GT/G designation, and still not
quite sure what that was all about. I guess that it may have meant that
the GT was a direct replacement for the G. Darned if I know for sure. Ira.

[John Rehwinkel]

"However, in thiis tube, the metal shell serves chiefly as container and electrostatic shield
for the glass bulb, which is required to insulate the contained gas from the grounded shell."

I hadn't thought about that! Just enclosing it in a metal shell wouldn't work with a gas tube!

- John

[Nick]

Mostly, I do valve ("tube" to our transatlantic cousins) amplifiers these days.

Over on some of the UK groups we've been having discussions about the UV risk from Hg rectifiers - it seems though that the glass absorbs pretty much all the dangerous UV

I asked much the same question about a year ago on the rather arcane/eccentric/abstruse UK-based Audio-Talk (lot of clever people there, but we are all slightly deranged) - the thread is at: http://www.audio-talk.co.uk/phpBB2/viewtopic.php?t=4920 

The general opinion was that radiation below 300nm would mostly be absorbed by the envelope and that UV-A would escape (but not far - intensity falls off at a rate of 1/(r^2). 

Main problem seems to be that it will cause a lot of plastics to degenerate... the responses from Mark ("IslandPink"), who is a physicist and commercial (scary) lens designer, and Shane are the ones you need to read...

Nick

[petehand]

Last word on the trigger clock, from me anyway:

Upping the anode resistors to 43k didn't work as expected. I finally settled on the optimum value at 30k. I swapped the 50 minute tube with its associated nixie driver tube. The ring counted just fine but the '5' never went out - the tube was WAY too sensitive and had to be changed out.

Most unexpectedly, it now runs just fine in the light and in the dark without any UV stimulation. It seemed to be getting a bit sensitive, so I removed the LEDs as an experiment. None of the rings has missed a step in over a week. Perhaps after being run continuously for three or four weeks, the tubes aged and settled down.

[threeneurons]

Nice to hear that ! Guess that's why the emphasized the aging process in old neons used for logic. Guess that applies to triggers, too. Kinda makes sense. Great Job ! I love seeing these things !

On Saturday, April 12, 2014 2:44:41 AM UTC-7, petehand wrote:

Last word on the trigger clock, from me anyway:

... I removed the LEDs as an experiment. None of the rings has missed a step in over a week. Perhaps after being run continuously for three or four weeks, the tubes aged and settled down.