ScopeClock -- was "Maybe Risk a Guess What That's Gonna be?"

[TurboTom]

What do you think this fun project will be turning into?

Shouldn't be too difficult... 


P.S. It's still hot from the oven...or should I say toaster 

[Alex Eisenhut]

http://buckrogers.wikia.com/wiki/Theopolis

[TurboTom]

http://buckrogers.wikia.com/wiki/Theopolis

Well, I'm not so much of a Buck Rogers fanboy...so...unfortunately...nope.

But I'll provide some additional information -- the complete PCB of the project. Some ideas anyone?

[PA0PBZ]

The lower pcb has the smell of a photomultiplier…

[LukeW]

Some sort of all-in-one integrated scintillation detector for gamma-ray measurements.

Maybe even a full multichannel analyser for basic gamma spectroscopy, all integrated onto the PMT base, with an integrated HV supply.

[chris_leyson]

CRT clock ?

[TurboTom]

CRT clock ?

Hey Chris, you got it! It's a project that I wanted to do since a long time. Actually, I intended to spin my own hardware initially but finally I stayed close to the sparkfun / Durchtronix AVRClock design which also means that I can use the same firmware with only minor modifications.

I bought a Dutchtronix AVRClock kit some time ago and experimented a little with it but in my own opinion, a clock based on this "multiple module design" is far too big and not really attractive. I wanted someting as small as possible. So I designed this two-PCB stack that houses the tube socket and simply plugs onto the base of the tube. The rear PCB accomodates all the digital circuitry and some reference level and gain adjustment stuff while the forward PCB includes all the CRT related stuff. The PCBs are only 63mm diameter and hence the complete package is smaller than the CRT screen. All the required voltages are generated on the CRT PCB, the whole clock operates from a single 15V wall-wart supply and consumes about 3.5 watt. I'm quite proud of the deflection amplifier design. It's not particularly fast (rise time around 1µs) but it's super-efficient at less than 1mA@250V per channel. conventional designs easily draw ten times that much.

One of the more troublesome sections of the design was the high voltage supply. The design is built around an NCP1032 flyback converter. The "oopsi" that occured here is that I didn't realize before that the voltage multipliers at the secondary (to provide -500V cathode supply, 250V anode and deflection supply and finally the second anode acceleration voltage of 1kV) are a complex load that also draws energy from the secondary of the transformer while the primary current is flowing (just like a forward-converter). And the NCP1032 features internally fixed current limiting threshold that, if tripped, causes a restart of the converter. The final tweaks that got the thing going were to inject an external current into the oscillator to reduce maximum duty cycle and a resistor in series with the secondary of the transformer to limit current while the primary is charged. You can see the two bodge resistors around the transformer if you look carefully. Fortunately, this was basically all that needed patching...

The biggest disadvantage of a design like this is that it only fits on a certain CRT or a family with the same socket configuration. I'm not too worried about this since the proportions and design of the electronics aesthetically matches best a CRT with about 70mm screen diameter and I just went for a model that's fairly easily available in my country and around, the B7S2 (-01), made by RFT in the GDR some 30...50 years ago.

The very first specimen of this clock went to a dear friend of mine as a birthday present for his 50th... Currently, I'm working on the second one, which I may possibly keep for myself, it just looks so cool. I simply love designs where history meets the present, especially if they look nice. That's also the reason I call this gadget "Zeitmaschine" (Time Machine) which fits pretty well in my opinion. If someone is interested in more detail (schematics and what not), just post a message.

Cheers & thanks for reading,
Thomas

[chris_leyson]

That was a lucky guess. Nice design thanks Thomas. I can understand the flyback design being a bit tricky especially if it's working as a forward converter as well, nice fix BTW. Nice job on the deflection amplifiers 1mA at 250V, I wouldn't mind seeing the schematic, last time I tried a deflection amp I let the magic smoke out. I like the B7S2-01 nice "modern" crt, cute. "Zeitmaschine" I like the name 

[TurboTom]

Chris -
here's the schematic of the complete device.

The deflection circuits had been optimized with LTSpice and surprisingly, they turned out more or less spot-on. The push-pull design with a relatively high bias resistor and high-gain output transistors made the low power consumption possible. Basically, each deflection driver consists of a complete amplifier with negative feedback, driven by a differential amplifier with a gain less than 0.5. This amplifier permits almost perfect symmetry between the two deflection plates while it permits balance (position) and common mode voltage adjustments.

Btw: regarding the flyback transformer, I was lucky to obtain a complete reel of surplus ones for next to nothing. They turned out to be very versatile due to the multiple windings arrangement. Properly connected, they will perform well in many applications. Albeit, the insulation characteristics won't permit their use for safety-relevant tasks.

Cheers,
Thomas

[chris_leyson]

Hi Thomas, thanks for posting the schematics, I like the deflection amps I never thought of doing it totem pole I was trying cascode amps, I learnt something new today and that is always a good thing. Using a flyback transformer "backwards" I like it a lot and that is something else I wouldn't have thought of, lots of other solutions popped into my head like a reasonant Royer but that would mean having to wind a transformer. An off the shelf small flyback transformer, they quite often have interleaved primary windings but swap it around and you've got two secondary windings.
In hindsight it's obvious but I learnt two things today, thanks Thomas 

[TurboTom]

Yesterday I managed to crack one of my CRTs -- unintentionally  . I must have injured the glass envelope when I fitted on the stainless steel wire stand. Yet it's not really obvious to me how stainless steel welding wire coud possibly cause a scratch in the glass of the tube. The only half-way reasonable explanation may be that from grinding and rounding off the ends of the wire, some microscopic fragments of the abrasives of the grinding disc got embedded in the metal and that caused the damage when I slided it on. There virtually wasn't any pressure at all.

The crack actually took place much later -- without any interaction from me: I put the tube (the stand removed again) aside on a soft rag. While I was doing someting completely different, I heard the sharp noise of cracking glass and immediately looked at the CRT. I was amazed to see a nearly perfect circular crack developing around its neck, close to the socket. I quickly placed a rag on it in case it would shatter but the getters already started to turn white so the vacuum was more or less gone already.

I have never before experienced anything like this and I'm sure it wouldn't have been possible (at least for me) to make such a perfect cut around the tube intentionally. Moreover, the tube anyway wasn't a "perfect" one, some shading at extreme deflection angles was present so I can live with this loss (I anyway have to...  ).

I still wonder what finally caused it to crack since even though it might have been weakened before, it stayed intact for several hours after that. Heavy ion from outer space maybe? 

Anyway, here's the evidence!

[drussell]

Is there any chance that the tube was rebuilt before? 

When they replace the electron gun during a rebuild, they are cut around the neck (usually with small file to make a notch and then a hot-wire) and then the neck is joined back together with the new gun assembly on a CRT glass blowing lathe.  Often the phosphor is also re-done while rebuilding a tube also unless it was a brand new dud tube or something.

[TurboTom]

@drussell -
I doubt that this tube has undergone a rebuilding procedure before.  It's too small, not really rare enough (actually not rare at all) and quite inexpensive. What I could rather imagine is that there were different quality grades of this tube since it had been manufactured in the GDR (German Democratic Republic) that was part of the Warszaw Treaty countries (Eastern Block) and folded in 1989 (for the young players). Usually, in these countries, military stuff was of really decent quality while the consumer grade...well, was something different  . After the crack developed, I could clearly feel a small step with my fingernal at the edge so there must have been several hundredths of a millmeter difference - hence a huge tension in the glass. I just guess it hasn't been properly annealed or cooled down slow enough during manufacture.

Whatsoever, this way I've got a neat show piece and tubes like this are available once in a while on the well-known markeplaces...

It's been a peculiar failure mode anyway, especially due to the "delayed fracture".

Cheers,
Thomas


Edit: Due to a request in forms of a private message, I attached the data sheet for the B7S2 CRT.