trying to repair panaplex display clock

[paronaram]

Hello All,
I picked up broken Sperry digital clock based on MM5314N and sp-151 panaplex display prototype from 1971
Clock had allot of hours, you can see that from burnout near glow segments, but they still glow, also Anode voltage was allot higher, and I think that contributed to not working in general
Replaced PS caps, all voltages back to normal.
MM5314N - is OK
Display is missing segments, so I found that one of the segment driving transistors are bad they used MPS-H55 (PNP)
Ucb: -80V
Ic: - 01A
F: 185MHz

Can you please help what modern transistor I can use to replace this?
Transistor is connected to multiplexed segment output to MM5314n
(I don't have schematic, and I can't find it online)

Any help, will be greatly appreciated.

 Thank you.

[james_s]

My go-to for that sort of thing is the MPSA92. For NPN applications the complement is the MPSA42.

If the display has darkened you might bump up the current limiting resistors, the lifespan of neon glow displays drops dramatically as current is increased. A 10% reduction in brightness is barely noticeable but will make the displays last several times longer.

[paronaram]

Thank you James!

Have you worked with Panaplex displays?
I have a replacement SP-151 but not sure how to test it, on nixie tubes we have common Anode but on SP-151 is very confusing, I think there is Anode per segment.
This clock has no socket, they used very crappy made single pin connectors, so to swap the indicators, is a big task. Was hopping if I can test somehow before I swap.

Thanks.

[retiredfeline]

A Panaplex is just like a nixie as far as the gas discharge technology goes, the cathode glows. So there would be a cathode per segment and a common anode which is turned on per digit. Work out the schematic and you will see.

[james_s]

Yes I've worked with these and quite a few similar displays. The segments are always cathodes, that's mandatory due to the physics of a glow discharge, the cathode is what glows. Most Panaplex type displays have one common anode per digit. The anode is a conductive coating on the inside of the glass while the cathodes are metal electrodes.

Have you looked at the datasheet? http://www.waynekirkwood.com/images/pdf/Sperry_Gas_Discharge_Clock_Display_Catalog_1973.pdf

Testing is very easy, you need a supply of around 180VDC, you should be able to borrow anode voltage from the clock by using either a clip lead or temporarily solder a wire to a convenient point. Be sure to use a current limiting resistor, the max for these is only about 400 uA (0.4mA) per segment and less for the smaller segments like colons.

[floobydust]

It's been a while since I've replaced Panaplex display tubes, their pins are Kovar and a bit difficult to work with.
What I do is put the soldering iron on the pcb top (display) side, parallel to the pin and apply heat, and then vacuum desolder (solder sucker) from the other side. If it doesn't work, reapply fresh solder and repeat.

The MM5314 datasheet has two Panaplex display schematics, and the Heathkit GC-1093 Digital Car Clock thread it uses a Panaplex display.

I just noticed the MM5314 came out in 1973 at $14.25/100 lot, equivalent to $86 in today's dollars.

[james_s]

ICs were crazy expensive in the early days. Weren't the jellybean TTL logic chips the equivalent of about 10 bucks each?

[paronaram]

Sorry for delay, took me forever to get the MPSA92 and installation.
Good news segment is fixed (see picture)
But now I see "shadow" glowing segments (see picture "11").
Is it possible that electrolytic caps are leaking? They are same age as PS capacitors

Thank you

[james_s]

I suppose it's possible, that looks more like a leaky driver though, or maybe a resistor has gone out of tolerance. Is the display multiplexed?

[floobydust]

I'm wondering about the keep-alive cathode being so bright and the blue glow. The datasheet is 20-50uA for it and 180VDC supply/segments 135VDC lit. It makes me suspect the voltages are too high, either from missing digits drawing less overall current or maybe it's high mains voltage or a bias voltage is off.
I would measure the blue zeners? The big zener diode by pin 14 as well.
With a pic of the clock pcb backside, some of the circuit can be figured out or sketched.

The carbon resistors are pretty old, I think I have some of those IRC parts in my junk box.
I think the 1uF 150V parts are for pulsing the segments on.

[paronaram]

Thank you All for reply!
Clock is fixed, replaced one transistor, replaced all electrolytic caps, inserted SP-151 (original) and all glowing balanced.
All resistors checked good 3 was drifted 1 to 2%
Floodydust -"voltages are too high, either from missing digits" - I think missing SP-151 from "socket" was causing that weird extra glowing.
After power reset I get some strange purple glow (around segment "1" and missing segments) but it's all goes back to normal, after I set the time.

[Renate]

After power reset I get some strange purple glow...

I'd be careful with purple, it usually indicates too much current (density).
Especially on an old thing you have to watch out for life expectation.

Have you ever hooked a spare NE-2 to AC with a MUCH too small resistor?
You get some amazing colors. (And a very hot neon bulb.)

[Gyro]

That keep-alive cathode does seem pretty bright compared to the illuminated segments. It's only there to provide a ready source on ionisation when the display is multiplexed. I'd check that the value of the (carbon composition?*) series resistor hasn't dropped.


Edit: * Ah, yes, they all are. 

[Renate]

Maybe the high voltage goes too high before a bunch of segments go on?