Need help understanding issue about filament voltages (using a Sencore CR70)

[spacediver]

Hi everyone,

I have a few Sony GDM-FW900 crts that I hope to keep healthy for as long as possible. I have a Sencore CR70 that works for the most part but has a few issues.

I'm trying to figure out how best to set the correct filament voltage, but there are a few areas of confusion.

In the manual of my CR70 (my unit didn't come with the operation manual, so I had to download it), it says that the filament voltage reading on unit will be an AC voltage if the unit is a run 16 or lower. Now I have no idea if I have a run 16 or lower, but my multimeter readings correspond to the unit readings much more closely when I have my multimeter set to DC as opposed to AC.

The manual also says the CR70 uses a regulated, DC power supply to provide the proper RMS filament voltage to the CRT. I'm guessing that this means that the CR70 emulates an AC waveform.

So, now for the big questions.

Because the setup manual is older than my CRT, the filament voltage for my tube isn't listed. I do have the service manual for the CRT, but I'm unable to glean from it what the correct filament voltage is. I am assuming it is 6.3V, but it would be nice to know for sure.

The second issue is that I don't really trust the needle on my CR70 - it doesn't seem to be very reliable, so I'm in a pickle even if I do know what the correct filament voltage is. Would it be a good idea to use my multimeter to adjust the filament voltage (so I connect the leads of the adaptor to my multimeter, and turn filament knob on CR70 until it reads the target voltage - e.g. 6.3V), and then connect the leads back to the pins on the CRT? Would I then be applying the correct filament voltage? Could I be confident that my multimeter is reading the correct voltage under the DC setting, even though the CR70 is applying an emulated AC waveform (assuming I'm correct about this emulation issue)?

I'm also including some images from the service manual that may help here. In the first image, H1 and H2 refer to the two filament pins. I believe the numbers in the red circles correspond to the waveforms shown in the second image. I have no idea what the red numbers next to the solid triangles mean (e.g. 4.7 next to H1).

I can upload the whole service manual if someone thinks they could use that to get more information. I'd really appreciate any insights that could help me figure this out. Thanks!

[buttersoft]

I can't help with most of this, though I know someone I can ask for more details about rejuvenators using AC waveforms.

However.... the tube won't represent the same load as the meter. If the voltage isn't controlled somehow, you'll deliver far more current than you should, or possibly far less voltage? The meter is best connected in parallel, so if you can work out the inductance and impedance across the filaments, you could replicate the correct load and connect that to the Sencore...? Either that, or set the Sencore really low and connect it to the tube, and use your meter across the same terminals - i.e. in parallel?

Next, I think those pink numbers without circles might actually be DC voltages. They look about right, or certainly possible. You could check them for R, G & B while the set is running - I'd use the connector block CN407 or CN408 as a test point, don't try to mess with the neck itself. Try a white screen and a black screen and see what gives. Same goes for the heater voltage (filament voltage). And be careful of the G2 voltage - it really will be at ~500VDC! Your meter may not handle it.

(EDIT3: Looking at things again, I may be completely off there. There's a spark gap on the heater line, which might mean it's running several hundred volts over ground. Still, you can use the meter over the connector block CN408 and measure voltages. If the results don't make sense then you'll want another source of info. Most digital meters will simply shut down if you try to measure voltage above their limits, but it is possible to damage them.)

I really don't know much about a heater running AC, but I know the other lines won't be AC. The G2 grid cannot be, it wouldn't work.

Lastly, are you sure you want to do this? Rejuvenators severely reduce the lifespan of a tube - it's like ten years use on the filaments and the cathode coatings in one hit. It can help a dim, blurry old tube, and it can help shorts and such, but if it's not already too dim or too bad to use, it represents a real risk. I hear, and I'm no expert, that you *really* want to practice on a throwaway tube first, like some junk consumer CRT - the meters for the guns will often improve and move into the "green" or "ok" zones, and then get a little stronger, and a little stronger, and then just fall off a cliff with no warning and the tube is dead. The higher the rejuvenation setting, the greater the risk at any given moment. Use the lowest settings first, and then if one or more of the guns still isn't testing ok, move up one notch. Never be tempted to overdo it. If the tube tests ok, stop then and there - unless the other option is the bin anyway because it's utterly shot.

I realise this is all pretty general, and I hope it's not a waste of time. I'd love to hear how you get on, especially if you do decide to rejuvenate.

EDIT: oh, and there are a few tube resources on the web, lists of voltages and stuff. I can't remember the addresses though, one of them is actually called Tubular, but google isn't helping me. I'll see if I can find out more. I'm fairly certain a set like the FW-900 won't be on the lists though, the one's i've seen were arcade and consumer tubes.

EDIT2: It also occurs to me to mention, as it should have at first, that rejuvenators like your Sencore may have been designed with consumer tubes in mind. Arcade tubes are very similar, but other tubes can be different. Trinitrons and PC tubes are difficult to use with an arcade chassis, for example. What you have is both of those, being a Trinitron PC tube. And it's as high-end as Sony ever made. That rejuvenator might work, or it might kill the tube, or it might not have the ability to remove grid shorts, or I really don't know. The pins are numbered differently to tubes I'm more used to, and they might be the same but the ability to use them in the right way to rejuv the tube may not.

By the time the need to restore tubes of that ilk could arise, LCD was on the up and up, so while the facilities might exist industrially, kit-rejuvenators probably don't. Again, this is speculation, but I'd recommend making sure of as much as you can before going ahead. Once you kill a tube, there's no going back.

[spacediver]

Thanks so much for the awesome reply. I have a friend who might be able to help me sort out the voltage reading issue. I'll also need to upgrade my understanding of circuits to have a better grasp of what's going on.

I think you might be correct about the numbers indicating voltage. I did find out the parameters for my tube (and yes, I think it was through Tubular), and it turns out that the filament voltage listed was 5V, so maybe the 4.7V is correct according to the schematic. Also, the bias is -116V, so my Sencore can't even do proper cutoff and emission testing, being limited to -68V. I just ordered a B&K 467 for a decent deal, as that one goes to -100V. Alternatively, I might be able to somehow modify my Sencore (although it is having other issues, as shown here: )

I do aim to do some more experiments on a less valuable tube (it's a 21 inch trinitron) to learn more. Thanks for the heads up about the dangers of rejuvenation. I plan to use restoration/rejuvenation as a last resort, and start very gently if I do. I think one of my FW900's does have a short, so I'll report back when I attempt to remove it.

When you refer to kit-rejuvenators, do you mean rejuvenators that were specifically built for a particular tube, or class of tube?

Again, really appreciate the thoughtful response...I had already kinda given up on the thread, and was a nice surprise to see a response.

[buttersoft]

Not to derail things, but I don't think I know enough terminology, or what's connected to what, and I'd really like to. As far as I was aware, the heater/filament voltage heated the cathodes. So what does the R,G & B voltage do, what does it energise, specifically? It's at -116V relative to common ground, correct? And the signal is laid on top of that? If the other end of the RGB voltages is common ground, what do those pink numbers mean? They're not all -116V, so do the different guns require different bias/gain voltages? (Which last would make sense, if the different phosphors aren't perfect, even emitters; and there's no reason they should be).

(Incidentally, I have a PC monitor that's having problems with bias voltage (Should be lower for a bright screen, I think. I watch it rise to -55V as the screen fades within about a 1 second). What should I check? Is there a name for that line?)

Happy to help, of course. Most of that post was warnings and guesswork though! And by kit I simply meant the portable suitcase stuff local TV repairmen would use. Heathkits used to come as just that, I believe, you built them yourself from the parts. As opposed to the tailored facilities available at the factory.

Any chance you're in Australia and want to sell me one of those FW-900's?

[spacediver]

Not to derail things, but I don't think I know enough terminology, or what's connected to what, and I'd really like to. As far as I was aware, the heater/filament voltage heated the cathodes. So what does the R,G & B voltage do, what does it energise, specifically? It's at -116V relative to common ground, correct? And the signal is laid on top of that? If the other end of the RGB voltages is common ground, what do those pink numbers mean? They're not all -116V, so do the different guns require different bias/gain voltages? (Which last would make sense, if the different phosphors aren't perfect, even emitters; and there's no reason they should be).

Here's my (ever evolving) understanding, and forgive me if much of what I say is obvious to you.

The -116V indicates the voltage between the cathode and G1 that is associated with a video signal level of 0. I'm still not conceptually fluent enough with circuits and electricity theory to grok how the concept of ground plays into this, but the way I understand it is that G1 has a negative voltage with respect to the cathode (does this mean that the cathode has a positive voltage with respect to G1? Or do we need ground here to make sense of it all?). There are three cathodes, one for each of the three channels. I believe, but am not 100% certain, that there is only one G1 common to all three cathodes (and one G2).

Now when G1 is negative with respect to, say, the green cathode, this means that electrons originating from the cathode, which would otherwise be accelerated by the positive voltage at G2, are repelled. When, say a video level of 100% for this channel is sent to the tube (i.e. RGB [0,255,0]), this causes the magnitude of the voltage difference between the green cathode and G1 to decrease. And now, because there's less repulsion between G1 and the green cathode, full beam current is achieved (relative to the brightness setting, which controls the bias at G2 (relative to G1?) ).

Now I believe that those pink numbers indicate the G1-cathode voltages required for appropriately balanced peak luminances (probably for a D65 chromaticity - see the intro to my WinDAS white point balance guide for some basics in color theory)

https://hardforum.com/threads/windas-white-point-balance-guide-for-sony-trinitron-crts.1830788/

Assuming this is correct, the reason that the voltages aren't identical is due to the following reasons:

The relative balance of the luminance of the red, green, and blue phosphors (SMTPE-C phosphors, which correspond to the primaries associated with the BT.601, or SD, standard, and which are very close to the primaries for Rec.709 HD standard/sRGB primaries) needed to achieve a white point of D65 is not 1:1:1. It's more like  22% red, 71% green, 7% blue.

Secondly, the spectral energy of each of these phosphors does not contribute equally to luminance. You need to apply the luminosity function, which is a spectrally weighted sum, to calculate luminance. For example, as the spectral power distribution of the green phosphor overlaps the luminosity function more than that of the blue phosphor, this means that you need more radiant flux from the blue phosphor than from green to achieve the same luminance (you can think of luminance as perceptually weighted radiance).

Finally, the radiant efficiency of each of the three phosphors (how much radiant flux is produced per unit beam current) is not necessarily equal.

When you combine these three factors, you presumably end up with the voltages listed in the schematic, and this also presumably does not take into account cathode aging.

Interestingly, in the service manual for the Sony CPD-G520, which is a 21 inch trinitron, the pink annotations are different.

(Incidentally, I have a PC monitor that's having problems with bias voltage (Should be lower for a bright screen, I think. I watch it rise to -55V as the screen fades within about a 1 second). What should I check? Is there a name for that line?)

What do you mean by "that line"? Based on my understanding (see above in this post), the magnitude of the bias voltage should rise as the video signal goes to 0. Are you suggesting that the bias voltage should rise to something like -50V with a video signal of 0 (instead of -55V)? What symptoms are your monitor displaying? If it's too bright at video signal of 0, then you probably need to recalibrate the cutoff voltage at G2 (i.e. reduce it slightly).

Happy to help, of course. Most of that post was warnings and guesswork though! And by kit I simply meant the portable suitcase stuff local TV repairmen would use. Heathkits used to come as just that, I believe, you built them yourself from the parts. As opposed to the tailored facilities available at the factory.

Interesting! Just looked it up, looked like a good way to learn and save money

Any chance you're in Australia and want to sell me one of those FW-900's?

Ah, sorry. Toronto, Canada. Plus I may only have 2 working FW900s now, depending on whether or not I destroyed one of my units when I first tried using the Sencore on it 

Set up a google alert though, you never know what might appear on your local craigslist or ebay.

Also, there's this thread, which has been going strong for over 11 years:

https://hardforum.com/threads/24-widescreen-crt-fw900-from-ebay-arrived-comments.952788/


p.s. this verification step where I have to input the letters in order to post is annoying and frustrating. Will it stop once I've become a more regular poster?

[buttersoft]

I think it goes away, yes. I don't even remember it, and I 've only been active on here for like a year or so.

Cool, thanks for that description. The difference in the voltage between the guns, and the different emission efficiencies and so forth, I did understand. The grid structure I wasn't too sure on. So what is the heater connected to? Nothing but itself? Just there to warm the cathodes up? And the individual colours, the RGB lines, normally 50-60V below ground, are connected to each gun, but the G1 Grid is further below ground? Thus the operation as you described when the R/G/B voltage is increased.

And regarding the CRT I have: from memory, when suddenly fed a steady white screen from nothing, the filaments hit about -65V, and then over the course of about half a second rise to -55V, during which period the screen dims to about half its original brightness. Then it stays like that. Given the filament voltage is dropping, I'd say the G2 voltage was steady. And I can see that B+ is steady as well. (I think this is all correct, it's been a few months.) So I should probably check the power going into the last stage of amplifiers next.

[ed12]

spacedriver
i see u have read up on theroy (good )
but i do have a question for u
in what part of the equation are u referering to ?
are u looking only for :luma: or :the effects of chorma: or the effects of chorma on luma: ?
or in general heater volt's ?
which btw if u look at the full circuit u will find it is un-filtered dc
thus making it :new word: quasi-ac :ie chopped dc at refresh rate:

ed

[buttersoft]

I was aware that the Flyback/LOPT worked that way, timed to the refresh, but not the heaters as well.

[spacediver]

The grid structure I wasn't too sure on. So what is the heater connected to? Nothing but itself? Just there to warm the cathodes up? And the individual colours, the RGB lines, normally 50-60V below ground, are connected to each gun, but the G1 Grid is further below ground? Thus the operation as you described when the R/G/B voltage is increased.

From what I understand, in some cases, the heater is basically a separate circuit. Current runs through it and it generates heat (it's a filament). And yes, its purpose is to heat up the cathode(s). In other cases, I believe the filament voltage is generated by the flyback (which I think would explain the "quasi AC" point that ed12 made). But I believe the filament is electrically separate from the triode (cathode, g1, g2).

As for the RGB cathodes, if G1 is negative with respect to the cathodes, and G1 is neutral with respect to ground, then perhaps this means that the cathodes are positive with respect to ground, and when a video sends in an image above black level, this voltage is reduced. But like I said, I don't really understand ground and circuits very well. All I know is that G1 is negative with respect to the cathode, and the more negative it is, the less is the beam current (and the darker the resulting picture).

btw, when you say "lines", are you referring to the drawn lines in the schematic? Or something else.

And regarding the CRT I have: from memory, when suddenly fed a steady white screen from nothing, the filaments hit about -65V, and then over the course of about half a second rise to -55V, during which period the screen dims to about half its original brightness. Then it stays like that. Given the filament voltage is dropping, I'd say the G2 voltage was steady. And I can see that B+ is steady as well. (I think this is all correct, it's been a few months.) So I should probably check the power going into the last stage of amplifiers next.

Are you sure it's the filament voltage that you're measuring here? As far as I understand, the filament voltage should not change depending upon what the video signal is.

[spacediver]

spacedriver
i see u have read up on theroy (good )
but i do have a question for u
in what part of the equation are u referering to ?
are u looking only for :luma: or :the effects of chorma: or the effects of chorma on luma: ?
or in general heater volt's ?
which btw if u look at the full circuit u will find it is un-filtered dc
thus making it :new word: quasi-ac :ie chopped dc at refresh rate:

ed

Thanks for the post. Yes, this quasi AC thing makes sense. Even the more recent Sencore tube testers claim to generate an RMS filament voltage from a DC power source. Given that RMS is a measurement that only makes sense in the context of a waveform, this seems to imply that the Sencore modulates the DC voltage to emulate an AC waveform. Perhaps the CRT itself does a similar thing to generate the filament voltage.

Not sure how luma and chroma are related to my question - my original question was around how to properly ensure that I generate the correct filament voltage when using a CRT tester. I now know what the correct voltage is (4.7 or 5V), but as I don't trust the needle on my sencore metre (see the last 2 min of my video: ). I take it the solution is to connect my multimeter directly to the CRT, so I can measure the filament voltage off the pins directly, or off the connecter block as buttersoft suggested, while I'm using the CRT tester.

[ed12]

well
if u dig into the theroy
u are getting grid/r/g/b points rather well
the heaters are a inductive load >light bulb effect<
there voltage is what is refered to as (sweeped) ie runs at scan rate.
tube testers will give u retalive volts (hence the fine tune knob)
so u set general range 4.7-7volts (course) and fine tube the heater as it warms up.
u see a heater dose not get to full current at first.
now in use (general veiwing) u will see it swing with the brightness/contrast ..load on tube..
thus it loads the sweep..the futher u take a tube to cut-off the more current it needs,hence the more load

ed

[spacediver]

interesting, perhaps this explains buttersoft's observations. Buttersoft, are you measuring the heater voltage with your multimeter set to AC or DC?

[ed12]

btw for futher refernce a heater is best seen using a scope >set to p2p value<

ed

[spacediver]

awesome thanks. My friend has a scope so he might be able to help me out here.

[buttersoft]

btw, when you say "lines", are you referring to the drawn lines in the schematic? Or something else.
Are you sure it's the filament voltage that you're measuring here? As far as I understand, the filament voltage should not change depending upon what the video signal is.

I had things back to front, I believe. By line i meant that I thought I was measuring some sort of common voltage for one of the grids. This was clearly incorrect. It's the voltage on each pin, R, G & B that I'm measuring at around -55V. I believe it should be lower to get a brighter picture, but it's been a little while. Might be the caps or the power to one of the late-stage amps. I'll take a look sometime in the future and see what I can see.

Thanks for the replies.