Mitsubishi HS-M55 VCR VFD display tube

Remember what Sherlock Holmes said?

If the filament itself seems to be okay, but there's no voltage across the filament went it's connected, what is the only possible explanation?

as my recent time-waster electronix project I'm troubleshooting a Mitsubishi HS-M55 VCR which actually works fine expect for a seemingly dead vacuum flourescent display.

I checked all the voltages on the VFD, and they are within spec: the cathode filament is driven by 4VAC straight from the PSU and biased to -30VDC, while the anode is grounded. There's activity on the segment pins between -30V and GND, so the drive voltages look ok.

For reasons beyond me, there's no voltage across the filament; when I unplug the display, I measure the 4VAC on the connector coming from the PSU, but when I hook it up, the 4VAC is nowhere to be seen on the filament -- only the -30VDC bias. Not suprisingly, the filament doesn't glow.

But when I apply about 6-8VDC across the filament from an external PSU, the VFD lights up!

Since there appears to be nothing wrong with the drive voltages, I'm guessing I'll simply have to tweak this thing. Is there any simple way to apply the extra 6-8V from the PSU without blowing it? (Simply tapping the

5VDC supply, for instance, would surely short the PSU, since the cathode is at -30VDC).

Thanks for any hints,

--Roland

**************************************************************** (WINDOZE not quoting properly...)

Seems easy. Bad connection, bad resistor, something like that.

Mark Z.

If you don't have a print you can get one at :

Do not just up and click on things there. Below where it shows a preview is some text, "This manual is ..." bla bla bla, and "Processing. After a shor t wait the text changes to "Get manual. THAT is where you click.Other hting s want to give you new readers and toolbars or who the hell knows what else .

Page 70 of the PDF shjows what is supposed to be feeding that, and it ain't good news. If you don't find a bad connection, you need a power trandforme r. that is probably pure unobtainium.

BTW, 6-8 volts is WAY to much vor that. They call for 4 VAC, you go to 8 DC and the thing is superheated. Hopefully it will be alright. That hting is even more unobtainable than the transformer.

If it tunrs out to ne the trandformer, stop back for some - umm - engineeri ng tips. I'm sure Digikey has a dandy off the shelf solution but it can pro bably be done cheaper.

't good news. If you don't find a bad connection, you need a power trandfor mer. that is probably pure unobtainium.

Hi Jurb,

thanks for the info. I have the service manual, but even with that at hand I'm stumped (alas, no EE degree). As Mark pointed out, it *seems* easy enou gh.

There is continuity from the filament pins all the way to the PSU, so it's definitely not a bad connection. There's an 82ohm resistor between each 4VA C supply and the -30VDC centre tap which check out fine. There's also an el ectrolytic cap and a 4.7V zener diode in parallel in the -30VDC supply; the y also check out fine. There's nothing else in that circuit on the display board, so I can't think of anything else that could fail there.

The circuit is basically the same as fig. 17 in Noritake's Guide to Fundame ntal VFD Operation

s), only the whole thing is biased to -30VDC.

DC and the thing is superheated. Hopefully it will be alright. That hting i s even more unobtainable than the transformer.

The filaments don't begin to glow until about 5VDC. at 8VDC I'd say the dis play is bright enough to be readable through the plastic cover.

I assume these things do alter their resistance with age, right?

ring tips. I'm sure Digikey has a dandy off the shelf solution but it can p robably be done cheaper.

Well, the xformer does churn out the 4VAC unloaded BUT I can't measure any continuity in the secondary, which is unsettling. Am I getting some spuriou s induction from the other windings?

Um, what was that about engineering tips again? :^)

Thanks & best regards,

--Roland

Assuming its a failed winding or internal join or fuse, can you add some overwind turns to get enough for the CCD

If it works with a higher voltage and does not with its spec voltage the VF D could be tired. It may somewhat recover if run for some time with the hig her voltage (just like rejuvenating a CRT) but be careful you could blow th e filaments by applying excess voltage for too long.

OK, then, there's your problem. Apparently, the transformer has an open winding. There may be enough capacitance that the meter, in Volts mode, picks up about the right voltage (with a 10 Meg Ohm load) but that the transformer can't deliver any current. You can check the terminals to make sure the windings are soldered well to the pins (assuming this is a typical PC board mount transformer, that is usually pretty easy to do.) If you are really lucky, it may be bad soldering at the terminals, where the windings are soldered on. If not, then you can try to figure out how to get the required voltage from a tiny add-on transformer. The filaments of typical VFDs do not even need to light up at all, visibly, to produce electrons.

Jon

The worse news is when a connection to a transformer goeslike this, otherss can as well. the not so bad news is that if the transformer mounts to a PC board, the solder maty hve simply oxidized over the years. If so, resolder them ALL, or you are likely to be doing it again.

I do not know if it does solder to a PC board, you got the thing right ther e so just look.

Be CAREFUL taking it out, I suggest a solder sucker followed by wicking and forcibly bending each pin with the iron, and then cold to make sure the co nnection is broken. You are delaing with thirty year old plasic here, and w hile breaking that is not a big concern, you could pull on one of the wires going into the actual windings. I had this happen on a flyback once, and i t had to be replaced.

But the main thing is not to screw up ay more of them, and even if you cano t fix it, resolder them all anyway, put it back in and then we look for tho se "injunearin" solutions.

I think you'll get lucky maybe. If it is PCB mounterd and you got good eyes ight, you will probably be able to do this.

I know this reply is not in the exact hierarchy, but too bad.

Anyway, you should be able to do this.

wanna hear a wierd thing about this ? when I worked ion TVs a long time ago , the wires started breaking on a transformer in a Zenith chassis. Where I worked it came in with no CRT filaments. (solid state, those werew the onl; ly filaments). It was the chassis that stood up with the modules and would flip down to work on the stuff behind. A really good chassis actually, vert reliable and produced a good picture. Some even had really good sound.

So I work on this thing at work, and they had a filament fuse. It was not b lown, the voltage was not coming out the transformer. Most people would obt ain a 6.3 volt transformer and sub the filament supply, but I did it cheape r, and this is why I got the big bucks. Well not that big, but OK. I simply put a winding of HV wire around the flyback. In those days they used tripl ers, so the voltage around the flyback was not all that high.

Years, and I mean yrears later I was buying tradeins and actually selling t hem out of the house. Damn, one Tradin' Times ad would sell five tVs, and b ack then I am gettiong like $125 for these things.

So, eventually my source comes up with the same model. Not only the same mo del, but the SAME TV ! I recognized my handywork around the fly, and the wi res going to the CRT socket.

The problem with the set is that the picture was shifted down vertically. I t also was eating vertical boards. the centering control ha a very limitesd range and would not center the picture. If you got it to fill the top of t he screen, the overscan was too excessive. And I mean too excessive. A doub le positive.

This was a DC coupled circuit, one of the earlier ones that used a ramp gen erator and a comparator. ALL DISCRETE COMPONENT as well.

Anyway, it was fed by a sl=plit supply from the power transformr and the center tap for that winding opened up. The circuit apparently pulled a bit more current on one side and that caused the offset DC. It had to be the de sign and not just certain cumulative tolerances and shit like that because several vertical board were installed and exhibited the same symptom.

I did not attempt to reapir that transforme because we could get them back then. Piece of cake, just take one out of a chassis in the boneyard, no pro blem.

BTW, this is not my thing so I am not looking, but really there are alot of people selling parts on eBay. I would at least give it a shot.

Quick update before heading to the office...

Finally decided to check out the PSU (after managing to extract the other e nd of its #*@& captive connector from somewhere deep in the control/video b oard cluster). Lo and behold, one of the green 100uF caps had gunked up and seriously corroded the -30V and one of the neighbouring 4VAC traces.

How the PSU still managed to output the correct voltages (at least unloaded ) is beyond me. Kudos to those Mitsu engineers, I guess. The caps incident ally check out fine with an ESR meter.

I guess the lesson learned here is not to blindly trust your scope and othe r gear! :^\

Will replace the caps and bridge the traces, then see what happens.

See ya,

--Roland

That did the trick. As Jon pointed out, the xformer winding appeared open-ended to the scope due to the severed trace on the PSU, yet still picked up 4VAC but no current.

Btw, the AC across the lit filaments now actually measures 8VAC...

Thanks for the help, guys!

--Roland