Lets see your Nixie Tube equipment

[Vince]

There still seem to be plenty of surplus Russian ones available on the 'bay - I'm not certain of the number off the top of my head, but think it's the IN-12.  They're top view, sort of flat sided ovals similar to what HP used in a lot of gear in the 60s.

You mean the kind of package that Dave Jones used in this video ?

Yes these would do nicely indeed.

]https://www.youtube.com/watch?v=7uogKucrPks

[Cubdriver]

The common Russian ones are IN-12s, I just looked it up.  I got a surplus box of 25 a while back for $50-60, IIRC.







-Pat

[Cubdriver]

LOL - yep - them's the critters.  The only thing I don't like about a lot of the Russian ones is that they use an upside-down '2' for the '5' character.  Smart from an economic standpoint (one less shape to stamp out), but less so from an aesthetic point of view.  Not that they cared about aesthetics, of course, and they are certainly functional.

-Pat

[Vince]

Buying an HP 5340, in my experience, is alot like playing russian roulette. Mine is almost perfect but will only display to 3 digits due to some odd glitch.

Yeah you might be right ! Actually just an hour ago someone I know (the chap I got my TDS 544A from), said he has jsut one of those, what do you know ! 
... got it for free at that, an actual dumpster find !  Says the unit is in a very sorry state and that it had a few bad connections in the wirgin which he failed to locate...

So yes, I am fully prepared for this kind of joy... and that's why I will only buy one if it's cheap enough and in good cosmetic condition, no broken or missing knob to start with. Not going to pay 250 bucks for it, but up to 100, if in good nick, I would go for it...

[neo]

I always liked the upside down 2. 

[Vince]

The only thing I don't like about a lot of the Russian ones is that they use an upside-down '2' for the '5' character.  Smart from an economic standpoint (one less shape to stamp out), but less so from an aesthetic point of view.  Not that they cared about aesthetics, of course, and they are certainly functional.

Yeah it looks weird indeed, don't like it either so would do all I can to find a model with a "proper" 5 digit.

Still, you got a bunch of them for very cheap, so worth using them 

The chap who sold the two nixie counters to me, also told me that a few other Nixie counters he had, would not find any buyer no matter how low the price. So he eventually resorted to pulling all the tubes from them and made a little fortune, sold all the tube for 15 Euros a pop in an eye blink ! So they are definitely sought after these days..

[Vince]

I always liked the upside down 2. 

Well, maybe one day I will look at it differently and grow fond of its "funkiness ", you never know !  Maybe one days they will cost a fortune precisely because of their uniqueness of design... so maybe I will get some anyway, but if I have the choice, for the time being at least, I would prefer a proper 5.
Certainly not a show stopper at any rate ! 

[Vince]

About the HP 5340A, a quick search on Google reveals that the service manual is available ! So at least there is a reasonable chance of being able to fix them, if one is willing to spend many an hour probing the thing...

[neo]

About the HP 5340A, a quick search on Google reveals that the service manual is available ! So at least there is a reasonable chance of being able to fix them, if one is willing to spend many an hour probing the thing...

I have yet to get mine working, at this point i would need much more sophisticated test equipment or a donor machine. A miracle would also work.

[Vince]

Well when you get reaaaaally tired of trying... just sned it my way, I will give it a fresh look, might work  !    No ? Well, I tried... 

Year end is approaching, ie miracle time, just what you asked for !

So... just ask Santa to bring you courage and luck for Xmas  ! 

[Cubdriver]

One of the issues with the 5340A in my experience is a custom IC in the signal path that gets fried.  I have one at the back of my bench that has this problem currently.  (I am also planning to try to help neo out with his blanked digits issue, hopefully soon, when I get back to it.)  I got a replacement board from the bay that (fingers crossed) has a good IC, but haven't gotten to confirming everything else around that circuit is good in order to install it.  I got hung up earlier in the summer by not having a short enough extender card (the board has a coaxial cable running to it, and there wasn't enough slack to use the long extender I had on hand); I ordered a short one and it got lost for a while on its way through Philadelphia, and during that time the mess on the bench grew to envelop the counter.  I'm working my way back to it.  (Patience, neo! )

I'm pretty certain that the deceased part is the nearly unobtanium custom Schmitt trigger IC - the large rectangular part with the SS cover at the top right of the A22 board:


Tonight's project is to shoot a few more pics of the Fluke 8200A and put it back together and get it off the bench and out of the way, and hopefully get a teardown thread of it up in the next few days.

-Pat

[tautech]

There still seem to be plenty of surplus Russian ones available on the 'bay - I'm not certain of the number off the top of my head, but think it's the IN-12.  They're top view, sort of flat sided ovals similar to what HP used in a lot of gear in the 60s.

You mean the kind of package that Dave Jones used in this video ?
YT link

There's a forum thread on those and a series of vids that hasn't been linked in this thread.
https://www.eevblog.com/forum/blog/eevblog-948-nixie-tube-display-project-part-1/

[neo]

One of the issues with the 5340A in my experience is a custom IC in the signal path that gets fried.  I have one at the back of my bench that has this problem currently.  (I am also planning to try to help neo out with his blanked digits issue, hopefully soon, when I get back to it.)  I got a replacement board from the bay that (fingers crossed) has a good IC, but haven't gotten to confirming everything else around that circuit is good in order to install it.  I got hung up earlier in the summer by not having a short enough extender card (the board has a coaxial cable running to it, and there wasn't enough slack to use the long extender I had on hand); I ordered a short one and it got lost for a while on its way through Philadelphia, and during that time the mess on the bench grew to envelop the counter.  I'm working my way back to it.  (Patience, neo! )

I'm pretty certain that the deceased part is the nearly unobtanium custom Schmitt trigger IC - the large rectangular part with the SS cover at the top right of the A22 board:


Tonight's project is to shoot a few more pics of the Fluke 8200A and put it back together and get it off the bench and out of the way, and hopefully get a teardown thread of it up in the next few days.

-Pat

I like to see the situation as me being patient with only the occasional reminder left where you can see it and completely innocuously.   

In all seriousness thanks a lot for your help on such matters, it is how i work best. With someone helping me and someone to bounce ideas off of.

[Vince]

Thanks for the lnik Tautech.

Cubdriver, thanks for the info and pretty picture, all that gold ! Makes me want to have one these machines even more, it's a jewel ! 

I don't have the service manual but looking at your pics, I assume that the coax cable comes from the RF pre-amp, and that Schmitt trigger must be shaping the signal in order to drive properly the RF frequency divider which I assume must be those 4 golden DIP packages sitting on this gold plated ground plane ?
No less than 4 chips... I guess it's much needed if one wants to get the 20GHz input signal down to something the counter can cope with.

So, two thought immediately come to mind :

1) This board/SS can only concerns the RF prescaler board/input, so I guess this means that if you use the low frequency input, the counter works just fine ? So this custom part failing, does not actually render the machine completely useless, better than nothing !


2) Given that the function of this SS can is well identified, and that it is a basic/simple function rather than billion gate ASIC... couldn't there be some hope, short of finding an original replacement part, to substitute a modern part for it ? I mean, 20GHz was no doubt a big deal back then, but is quite trivial these days, RF components can be readily ordered with a mouse click and a few bucks.... maybe there exist Schmitt trigger devices these days too ? Worth investigating...
Obviously it probably wouldn't come in the same SS can foot print, but as long as it functional... and well maybe it would be possible to "stuff" the old can with the modern replacement part, to keep the original look... if one wanted to go extreme about it.

[Cubdriver]

   No worries.  The counter sitting on the bench serves as a constant reminder.  I foolishly volunteered to help my best friend paint a coworker's newly purchased house.  There is some king of confluence of idiocy there that makes us unable to work efficiently - what was supposed to be a week-and-a-half to two week project working in the evenings has somehow entered its third month (with a few breaks for business travel, but still...    )  This time sink has been doing a fine job of hamstringing my other efforts.  Thankfully, the end is finally in sight and we hope to be finished this week.

THEN perhaps we'll continue with the HP 5340A misadventures!

-Pat

[Cubdriver]

Thanks for the lnik Tautech.

Cubdriver, thanks for the info and pretty picture, all that gold ! Makes me want to have one these machines even more, it's a jewel ! 

I don't have the service manual but looking at your pics, I assume that the coax cable comes from the RF pre-amp, and that Schmitt trigger must be shaping the signal in order to drive properly the RF frequency divider which I assume must be those 4 golden DIP packages sitting on this gold plated ground plane ?
No less than 4 chips... I guess it's much needed if one wants to get the 20GHz input signal down to something the counter can cope with.

So, two thought immediately come to mind :

1) This board/SS can only concerns the RF prescaler board/input, so I guess this means that if you use the low frequency input, the counter works just fine ? So this custom part failing, does not actually render the machine completely useless, better than nothing !


2) Given that the function of this SS can is well identified, and that it is a basic/simple function rather than billion gate ASIC... couldn't there be some hope, short of finding an original replacement part, to substitute a modern part for it ? I mean, 20GHz was no doubt a big deal back then, but is quite trivial these days, RF components can be readily ordered with a mouse click and a few bucks.... maybe there exist Schmitt trigger devices these days too ? Worth investigating...
Obviously it probably wouldn't come in the same SS can foot print, but as long as it functional... and well maybe it would be possible to "stuff" the old can with the modern replacement part, to keep the original look... if one wanted to go extreme about it.

Pretty good assumption, Vince - the coax is the input from the direct count amplifier.  Unfortunately, all the frequency signals pass through this Schmitt trigger, low and high.  (Blue below is the high frequency 50 ohm input; pink is the lower frequency high impedance input, and green is the A22 board.)  From there it goes to the count register, then the display register, and on to the glowie numbers those of us in this thread all love so much.

Overall counter block diagram:




And I have considered trying to just substitute a modern IC in the place of the custom jobbie, and may yet wind up doing that.  I first wanted to try to fix things without reinventing the wheel if possible.  It's also remotely possible that there is something else amiss with that board, but I'm pretty sure that with the limited probing I was able to do while it was plugged in that I had signal at U9's gazinta, and a lot of nothing at it's gazoutta.  I need to poke at it while it's on the extender card.

A22 schematic:


A22 block diagram and layout:


-Pat

[Vince]

Thanks for the details Pat.   I wouldn't dream of fixing these complex beasts without the service manual, that's for sure...

Now that you have the adequate extender board you will be able to work much more comfortably but from what you say indeed it looks like it's a done deal sadly 

I am starting to get the impression that this particular model is well known, and not just in the US, so maybe at some point, as people collect them, there will be enough people affected by this problem to warrant a collective effort in designing a replacement. This RF stuff is a specialist job of course but maybe there are enough collective knowledge in this forum or out on the web, to allow one to embark into designing a replacement...

That would be a nice little project for RF design aficionados !  Maybe we could ask Shariar/ The Signal Path, and Alan / w2aew what they think of it from a design perspective, if it's feasible or not, realistically... that could make for an interesting video for them to do !     I think I might contact them about it... both being in the US, I guess it would not cost too much for someone in the US, owing on of these machines, to ship it to them (along with the much required extended card of course, and a copy of the printed service manual if not available in PDF form for them to download). They have the academic and practical knowledge to tackle this, or at least provide us with a well informed insight, as well as the expensive equipment required to play with this thing at will.

Hell... the two of them know each other, why not combine their expertise in a joint video ?! A co-produced video... I guess they live too far apart from each other to actually meet physically, the US being a vast place, but they could still communicate via e-mail to discuss this.

I just downloaded a copy of the service manual... sadly 99% of it is impossible to read, the PDF is massively corrupted somehow. I could only figure out a few snippets here and there, but OMG I see that this instrument came with GPIB ?!!     And the manual is date 1975 !  Sure, HP invented this bus so I understand that they were the very first to implement it on their instruments, before other manufacturers adopted it, but still, 1975 ?!! There was not even an 8 bit micro to run such a thing back then, I think !  Had a quick look on Google to refresh my memory, found an article from 1974 announcing a super duper 4 bit MCU...

The PDF was really too fucked up so I couldn't see how they did it...   my assumption is that they did not use an 8 micro which did not even exist probably, or elese they might have used it to control the front panel and give their instrument a competitive edge in terms of user interface.... so my guess is that they designed their own stand-alone bus controller, a big state machine, which would not necessitate a micro ti run it ?  I am intrigued !

Also saw that this beast is really complex, found a snippet about a PLL ?!

That machine was indeed a marvel of engineering compared to my French nixie counters or the same era !

I want one of those, definitely, working or not !!! 

[neo]

Do you know why it is so complex and nice? Back in the 70s when they sold these units they cost between 12-17k depending on the options. Doesn't sound like much i know but accounting for inflation that is about a 100K in today's money, thats in freedom dollars.

You see the HP office manager who ran the office that designed these only cared about the products efficiency to do its job and its ability to do it accurately, he didn't care the cost because to him no cost was too high. This one model of counter had over a 20 year life that i think stands as proof of his genius and commitment to the design.

[bitseeker]

I'm pretty sure that with the limited probing I was able to do while it was plugged in that I had signal at U9's gazinta, and a lot of nothing at it's gazoutta.

Hey, Pat, watch out with all those highfalutin, technical terms. You must think you're talking to a bunch of engineers or something.

[Vince]

Do you know why it is so complex and nice? Back in the 70s when they sold these units they cost between 12-17k depending on the options.

Ah yes, I am sure 15K or so was a very considerable amount back then !
amazing they can bought for so cheap these days... definitely want one !

This one model of counter had over a 20 year life that i think stands as proof of his genius and commitment to the design.

20 years ?!  Not in Nixie form I guess ? I gather they made a revised version of this model, that used an LCD display ? I guess that helped keep it on the market for so long.
I bet the LCD versions sell for much more, because they look more modern, even though deep down its gutts are the same as the old original Nixie version ?

That's great... this way we can have a super duper counter for cheap, and at the same time enjoy these lovely Nixei tubes....

Definitely worth investigating a way to replace that custom schmidt trigger can, to extend the life of these lovely beasts...

[rsjsouza]

Do you know why it is so complex and nice? Back in the 70s when they sold these units they cost between 12-17k depending on the options.

Ah yes, I am sure 15K or so was a very considerable amount back then !
amazing they can bought for so cheap these days... definitely want one !

This one model of counter had over a 20 year life that i think stands as proof of his genius and commitment to the design.

20 years ?!  Not in Nixie form I guess ? I gather they made a revised version of this model, that used an LCD display ? I guess that helped keep it on the market for so long.
I bet the LCD versions sell for much more, because they look more modern, even though deep down its gutts are the same as the old original Nixie version ?

That's great... this way we can have a super duper counter for cheap, and at the same time enjoy these lovely Nixei tubes....

Definitely worth investigating a way to replace that custom schmidt trigger can, to extend the life of these lovely beasts...

The HP page below indicates the original price was US$5340.00 (about US$100k today) and it seems it used Nixies until the end. Quite remarkable.

http://historycenter.agilent.com/exhibit1

[Specmaster]

Heres my first nixie tube device, I like it.

[Vince]

The HP page below indicates the original price was US$5340.00 (about US$100k today) and it seems it used Nixies until the end. Quite remarkable.
http://historycenter.agilent.com/exhibit1

Hmmm, thanks for that link, very interesting ! 

Introduced in 1973, and retired in 1994 still with Nixies in it, wow !

The article confirms that it was a real beast and intended that way, rather than your average instrument.

With such a long life, they must have built a good number of them, and they must have had time to spread all over the world... so maybe I do stand a reasonable chance of finding one in good nick at a decent price, one day. Brings a big smile to my face already !

[Vince]

Heres my first nixie tube device, I like it.

Hmmm looks real cute indeed, I like it  too !   

[richnormand]

EGG 550-1 radiometer/photometer.

I use it with a fiber optic attachment to map the 360 degree illumination of LEDs.
Got it cheap as one of the Nixie tube was dead. Turned out it was the driver chip (DS8880).
Five minutes with the soldering iron and it has been in use for more than 10 years now.