Helium-Neon PSU repair (old laserdisc player)

[mad_lux_it]

Hello, and thanks for trying to help me!

I'm searching for an help replacing/rebuilding the high-frequency transformer component of the PSU of a HeNe laser tube inside a laserdisc player (a glorious but broken Philips LV-720 from the 1978).

After quite a research and study (and painful translation from german I'm able to first describe you a good part of the PSU operation, with the real schematics! (see attached)
The oscillator at 20KHz is driving a sort of darlington configuration to drive a step-up ferrite-core transformer from a mere 12.3V regulated supply (labeled 12.3A in the schematic). Reading the specs of the laser tube, the PSU should be able to feed 1150V at 5mA.
The voltage tripler labeled with 1002 is here only to create the spike necessary to light the laser tube on (something like 8Kv). After that the tube starts sucking up enough current (5mA) to completely disable the tripler effect.
The circuit on the right should be able to pickup the feedback tension to 'regulate' the PSU at 5mA, on the resistor 3067(+3066): 5V should be there. The transistor 6131 and the 6.2V zener 6022 closes the loop (limiting the amplitude of the square wave instead of his duty cycle actually?).

Now, after some test and a serious attempt to fry my multimeter (I don't have any tool for measuring HT) I can see that the PSU is only able to produce a mere 0.2mA. The 22V square wave at the collector of the driving transistor 6129 is there. Around 5V AC are across the primary windings.
I've tried to replace the electrolytic cap 2021, hoping for a dried one, with no luck.

Luckily, after a test with the room completely darkened (I was trying to spot flashes inside the laser tube), I've discovered a nice blue shiny arc coming from the inside the 5003 transformer! Found the culprit! 

Ok, calm down. What now? How hack I'm supposed to replace that transformer? I will bet that the He-Ne technology is now only an memory from old nans. No way to buy a new one I think. You can find attached some photo of it (it doesn't seems a 40 years old transformer to me

And here the idea to 'rebuild' it from scratch. TME has the U-U ferrite of exactly this size, so most probably I could try to rebuild a new one with the plastic core as well. But... with what turn ratio? And with AWG wire gauge?

The primary is only 15+1 turns, so easy to count it! (I cannot explain the role of transistor 6128 driving this single coil turn BTW).
But the secondary (those that causes arcing) I cannot tell. The windings are not opened nor shorted: I can measure something like hundreds of ohm in the secondary. I don't have any inductance meter, nor capacitance meter...

Obviously I had tried to take the ferrite apart to reach the secondary coil and... simply count it.
But the damn glue seems invincible. Before breaking the ferrite I had tried to 'boil' the transformer in water, trying to soften the glue, but with no effect.

Ok, before crying on a pile of ferrite dust, I had tried to measure some voltage with... a tone generator app and an audio amplifier.
Here the readings using a 220K 1W load resistor (that actually become quite hot and fuming after few seconds), and using the 4-5 pins as primary (so 15 turns only):

5Khz sine wave: 4.5V~ -> 530V~
7KHz sine wave: 3.6V~ -> 510V~

So? 1:140 or 1:110? 

Mmm my multimeter can be not so good at AC reading at these frequencies. In addition, is it possible that the transformer change its response at different frequencies? (it is designed to run at 20kHz but my ghetto tone generator made with a Lumia smartphone starts to cut at 10Khz 'by design'
No visible arcing during test, but who can say it?. The resistor ensured that I had sourced at least 1mA during the test.

Perhaps my only chance is to destroy this little thing only to count the exact turns, and then hoping to be able to rebuild a working replica.
I have harvested producers of SMPS transformers, but it seems that I'll be luck to find brand new parts up to 300/400V only.

Searching for complete PSU for laser tubes gives me only 'out-of-budget' results.

Now, what chances I have to fix this vintage (and heavy) brick in your opinion? 
It will be funny to discover that the laser tube is kaput as well. And this time the 'rebuild' option will be unavailable this time

Well, thank you all, and be thankful for laser diodes!

Luciano

[floobydust]

I would take it apart, it's dead anyway and do surgery.
The ferrite core is like glass, be very careful. But it should pull apart when hot. Put in your oven at about 250F, not enough to melt the bobbin but hot enough to melt the core epoxy glue (including center pole). I use pliers and a very gentle tug will pull it apart, or heat it more. Don't chip the ferrite if you grab hard.

The transformer bobbin should have been varish-impregnated for the high voltages. I'd use a drill when rewinding the secondary.

It's very similar to a CCFL backlight transformer, you could look at using one. Just add the small osc winding on the primary.
But hard to do with different core sizes, on this saturated-core oscillator design. Then the turns ratio prob. need adjusting but experts here can discuss.

Coilcraft CCFL transformers, some with Royer/feedback winding.

[f5r5e5d]

It's not a safe bet that a HeNe from 1978 has the right amount of He to work anymore - it diffuses through just about anything non-metalliac

epoxy seal HeNe life was single digit years

[floobydust]

I have HeNe tubes from 1986 that work OK, but all glass seals

[CJay]

That's a very Phillips looking transformer.

While it might still be a difficult task, it looks a lot like the type used in many Phillips CRT televisions of the era so if you do manage to break it, you *may* find replacement cores out there in a junk bin or vintage radio forum.

I also have HeNe tubes from the early 80s which still work very nicely so while there's a possibility the tube may be faulty, there's a good chance it will work.

[mad_lux_it]

Thanks for the many hints.

I cannot peer inside the tube: i's a completely sealed unit, with a black metal shield and no immediate way to remove it. So I don't know which type of construction it can be, nor try to guess the joint material.

Mmm.. So instead of wasting time right now to rebuild the trafo, perhaps I can try to temporary patch the PSU with a transformer scavenged from a CFL lamp, just to tell if the laser blinks...
Then if I'm lucky, start with the real repair.
Ill keep you posted. Thanks!

[NiHaoMike]

Replace the tube laser with a diode module?

[mad_lux_it]

== UPDATE

I would take it apart, it's dead anyway and do surgery.

Indeed! A lucky strike with a flat screwdriver, and.. surprise! Not a pile of ferrite dust but a clean separation!
See photo in the attachment section.

But what was yet more unexpected is the construction of the trafo: the two bobbins are coiled on two separated plastic supports, that are mounted one inside the other, with even the soldered pins separated! A photo is better of hundred words.

The inner winding seems so perfect that I would be crazy to unroll this thousand of coils of a hair-thin wire, nicely stacked layers upon layers separated by a sort of insulating tape.

Well, once dismounted like this, arcing becomes less and less probable.
So I've cleaned everything up with isopropilic alcohol, removed some glue stain in the recess of the inner bobbin support, and then I've put everything back in place (no glue for the ferrite, only a tight rubber band.

Let's power up the unit... some buzz (less than before)... nothing.
Still nothing. No laser.

 

Ok. After some night thinking about it, I'm close now to a couple of ideas.

Idea #1:
Since I'm missing a minimal workbench for HeNe lasers (well, I was even not aware of that monstruosity a couple of month ago), and since eBay is disappointingly empty in that category (in contrast to Nixie tubes that are coming back to life with violence), the plan is to build a quite simple unregulated PSU using the above transformer and the voltage tripler.
The idea is to implement a basic 555-based flyback converter with a bit of regulation, more or less like this one: http://www.repairfaq.org/sam/ichi3sch.pdf

Plus I need to build sort of ghetto-high-voltage-probe to double check that output voltage is correct under a dummy load and without load for the startup.
Well, I'll start to fill my BOM some day.

Once done, I'll proper test my tube outside the quite complex Philips schematic.

Idea #2:
(If Idea#1 fails, or if it actually works but it comes that the laser tube is really kaput).

Replace the tube laser with a diode module?

Well, this will be fine for the repair. And this will even reduce the power consumption of the player

However I think that a modification like that will require unaccessible tools and skills. The repair manual says that a simple laser replacement requires a complete optic realignment (each mirror has 3 screws!).
Changing the color of laser, the light focus and the beam dimension will be easily something that would require me to be retired and with a looot of time to spend on it.
But, well, a try can be done (even with a battery-powered laser pointer) to let this unit back to life.

For now the tube is a nice ornament on my desk

Thank you all!

[chris_leyson]

Hey nice job on the transformer it looks like you've fixed the arcing problem

Replace the tube laser with a diode module?

Don't think that would work, you would need a very well colimated laser at the same wavelength 633nm because of the quarter wave optics. Old bar code scanners are good source for HeNe lasers, I bought some on ebay about 20 years ago and they are all still working,
they're quite large 150mm long and 25mm diameter Melles Griot 05-LHR-004-146.