HF AC DHT heating

For the three and more legged things
User avatar
Paul Barker
No idea why I do this anymore
Posts: 6889
Joined: Mon May 21, 2007 9:42 pm

Post by Paul Barker » Sat Jun 14, 2014 4:40 pm

you have started with a pretext based on a mis-understanding so all you wrote is meaningless.

I was referring to your lm317 in your AC hf heater adjustment of which changes the heater voltage. You referred to this 317 also in this same thread in the way I did.

The test is to measure the dissipation of the valve. Because with your other 317 that you spouted on about above, you have fixed the valves cathode current the valve can only respond by changing anode voltage.

The point is plain and simple the amount of power in the filament is directly related to the amount of exlectrons leaving the filament. This would normally be expressed in current flow but because you have fixed that it is expressed instead in anode voltage. You cannot change the power of the filament supply without the valve expressing this change.

for the correct RMS power there will be a specific dissipation.

It shouldn't need explanation.
I can do it right or I can do it now, but I can't do it right now.

User avatar
IslandPink
Thermionic Monk Status
Posts: 5398
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

Post by IslandPink » Sat Jun 14, 2014 4:47 pm

I haven't followed all of this discussion, but I still think back to the comments I made years ago when Paul tried this before.
The end-target is that you want the filament to be the same temperature as the DC ( 10V ) case.
If it's the same temperature, the filament will be the same brightness and the same colour .
Now, maybe it would be enough to set up a 'reference' valve with DC heating , next to your HF AC one, and compare the filament colour.
If that's not sensitive enough, then how about putting a white card near to each valve ( same distance, same orientation of valve to card ) and measure the brighness in the brightest bit of the card with a photographic meter ? If you have a camera with a spot-meter , or a separate spot-meter , that would be the best .
For best results you'd want to do this in the dark or at least dim conditions.

Once you've established equivalence on this test you can 'calibrate' your other, easier test ( eg. AC voltage measurement ) to use in the future.
White flags of winter chimneys waving truce against the moon ( Mitchell )

User avatar
Nick
Site Admin
Posts: 10866
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

Post by Nick » Sat Jun 14, 2014 5:33 pm

But again, if I was doing this, I would first try it with two DC heated valves match by colour and see how close I got. Then hope it was not a fluke.
Resistance isn't futile it's V / I.

Alex Kitic
Old Hand
Posts: 369
Joined: Mon Oct 14, 2013 3:20 pm

Post by Alex Kitic » Sat Jun 14, 2014 5:59 pm

I am not using a 317 in the heater circuit (to regulate the DC voltage before the oscillator circuit), which is why I misunderstood. I was suggesting it might be done, but have not yet applied that solution, and probably will not due to heat dissipation concerns when drawing about 440mA. For a 100mA setup (25W and less) it might be the simplest option to get rid of any hum or noise and regulate the heaters voltage).

Of course it makes sense to compare the difference in voltage across the tube (due to the same current draw, the voltage measured at the anode will remain the same, but the cathode voltage will vary, increasing or decreasing the voltage across the tube).

This means measurements should be performed at the cathode, an increase in heater current should lead to an increase in cathode voltage (bias) and vice-versa.

This is a good idea, but it is complicated by the fact that I do not have an AC 50Hz source powerful enough for 50W at 10V. But comparing this cathode voltage change nevertheless seems more precise than the filament brightness comparison.

I will try to make some measurements, but frankly all I want to do is confirm my math in this case...

I have got no light measuring devices (lux-meter) and it is difficult to estimate how white is the white glow... much easier with oxide coated filaments. I recall 9.6V DC was surprisingly pale, and 9.9V not much brighter either. Bright glow begins after exceeding 10.2V, but it may be just an i, pression. The direct comparison method where one is heated by exact DC and the other AC and compared - is not feasible since the two tubes are not identical (one is RCA, the other a Haltron: they might be very similar electrically, but differ in construction, thus perceived brightness is not identical).

The good thing with a two-secondary setup is identical voltage on both heaters. At least I know both tubes are undergoing the same "ordeal" ...

User avatar
Paul Barker
No idea why I do this anymore
Posts: 6889
Joined: Mon May 21, 2007 9:42 pm

Post by Paul Barker » Sat Jun 14, 2014 7:18 pm

Alex Kitic wrote: At least I know both tubes are undergoing the same "ordeal" ...
:D :D :D
I can do it right or I can do it now, but I can't do it right now.

User avatar
Dave the bass
Amstrad Tower of Power
Posts: 10229
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

Post by Dave the bass » Sat Jun 14, 2014 9:50 pm

<token question from the dopey looking kid at the back of the class> ...., surely 'colour-matching' only proves the heater chemistry emits the same light frequency, not that the electrical characteristics are being met equally thus how can you be sure that cathode is emitting as per the manufacturers intent?

Trixy stuff this.

DTB
"'Occasionally phenomenal'"

User avatar
IslandPink
Thermionic Monk Status
Posts: 5398
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

Post by IslandPink » Sat Jun 14, 2014 10:10 pm

Good question there young Dave.
If it was in air, and anything was therefore burning or reacting in some way, then you'd be right that the chemistry would be part of the picture.
However... it's not, and it's essentially an inert piece of metal in a vacuum, for our purposes.
This means that it behaves as what's known as a 'black body' ( paradoxically ) - see Wiki for how this term came about . The photon emission curve ( as a function of wavelength ) and therefore its colour, turns out to be only a function of what its temperature is . As the temperature goes up, the total emission ( light + infra-red ) increases, and the peak of the curve comes down in wavelength . If it's only at 1000 deg K, then the peak is in the IR and you have some emission in the red, so it just glows a bit red. By 2950 K or so , you have a peak in the orange and plenty of emission in the blue and green , and it looks sort of yellow-white and you have a light bulb. Even hotter, at about 3400 K is a tungsten halogen bulb - more like true white . The sun is at about 5500 K and the peak is in the green .

For an 813 , and thoriated tungsten, I think you're about 1800 K or 2000 ( correct me guys if wrong ) so it's kind of yellowy , not quite as white as a light bulb.
White flags of winter chimneys waving truce against the moon ( Mitchell )

Alex Kitic
Old Hand
Posts: 369
Joined: Mon Oct 14, 2013 3:20 pm

Post by Alex Kitic » Sat Jun 14, 2014 10:55 pm

IslandPink wrote: For an 813 , and thoriated tungsten, I think you're about 1800 K or 2000 ( correct me guys if wrong ) so it's kind of yellowy , not quite as white as a light bulb.
I totally agree with you on the yellowish colour, but would estimate that to be more like 2700k. The "surprisingly" pale color is what is actually reflected inside the anode and up towards the glass. With an increase in voltage, the yellowish reflected light gradually turns white as it increases in brightness.

As for the filament itself, it is rather white, but looking directly at it becomes very unpleasant as voltage increases. In the current condition, which I believe to be 9.7V most probably, I can still look directly at the 4 wires that glow. With the other setup, which we believe to be between 10.2 and 10.5V, the filament wires are too bright to look at, but still distinguishable in the haze created by light. The colour of the reflected light is more towards white, but still a long way from really white...

Now this depiction will probably explain better the heaters voltage condition than my DMM. Does anyone's filament and reflected light differ much from my depiction?

Alex Kitic
Old Hand
Posts: 369
Joined: Mon Oct 14, 2013 3:20 pm

THE COLOR OF FILAMENTS...

Post by Alex Kitic » Tue Jun 17, 2014 11:16 am

Image

While not a particularly good photo, you can see several "lights" and their colors: the 813 with probably 9.7V on the filament, the 0A2 with bluish-violet gas color, and the oxide coated filament of the 5AS4 (5U4).

The colors are relatively true to the natural condition...

Now who can estimate the correctness of the filament voltages based on this?

Image

Another picture of the two 813 tubes, again the "chiaro-scuro" looks perfect, doesn't it? Or maybe it does not, and the colors should be different...

User avatar
pre65
Amstrad Tower of Power
Posts: 16748
Joined: Wed Aug 22, 2007 11:13 pm
Location: Ashen Sudbury Suffolk

Post by pre65 » Tue Jun 17, 2014 11:24 am

I had to look up Chiaroscuro. :oops:

http://en.wikipedia.org/wiki/Chiaroscuro
People are like stained-glass windows. They sparkle and shine when the sun is out, but when the darkness sets in, their true beauty is revealed only if there is a light from within
Dr Elisabeth Kubler-Ross

G-Popz THE easy listening connoisseur. (Philip)

Alex Kitic
Old Hand
Posts: 369
Joined: Mon Oct 14, 2013 3:20 pm

Post by Alex Kitic » Tue Jun 17, 2014 12:28 pm

Yes, the play of light and darkness. Just like life: the light at the end of the tunnel.

In my life, though, there is still no light behind some corner. A voyage through darkness. With tubes as candles to keep me from obscurity. But it will end one way or another. Either darkness, or life...

BTW, I speak excellent Italian.

User avatar
Paul Barker
No idea why I do this anymore
Posts: 6889
Joined: Mon May 21, 2007 9:42 pm

Post by Paul Barker » Tue Jun 17, 2014 3:18 pm

I do hope that some sort of work comes your way soon Alex.
I can do it right or I can do it now, but I can't do it right now.

Alex Kitic
Old Hand
Posts: 369
Joined: Mon Oct 14, 2013 3:20 pm

Post by Alex Kitic » Tue Jun 17, 2014 5:13 pm

Paul Barker wrote:I do hope that some sort of work comes your way soon Alex.


Thank you, Paul.
It's depressing... and at moments I get carried away. I am rather tough, actually.

What about the light of the 813s, does it look correct?

User avatar
Paul Barker
No idea why I do this anymore
Posts: 6889
Joined: Mon May 21, 2007 9:42 pm

Post by Paul Barker » Tue Jun 17, 2014 6:07 pm

I couldn't say to be fair. IT is a valid way but I gave you my method earlier in the thread. When the valve does the same work as the same valve does with a measurable voltage you have your answer.
I can do it right or I can do it now, but I can't do it right now.

User avatar
IslandPink
Thermionic Monk Status
Posts: 5398
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

Post by IslandPink » Tue Jun 17, 2014 6:43 pm

Alex Kitic wrote:I totally agree with you on the yellowish colour, but would estimate that to be more like 2700k.
Given that a standard tungsten light bulb is about 2900 K
http://photo.net/learn/optics/edscott/cf000030.htm
( tungsten-halogen is 3200 to 3400 )
..then I'm sure 2700K is too high .
As I guessed above, Harbaugh 's paper here confirms that thoriated-tungsten filament temperature is about 2000 K :
http://electrontubestore.com/thoriated-tungsten.pdf

Thanks for the nice 813 pictures . Maybe as you say it's a bit tricky to estimate the temperature accurately from colour, though .
White flags of winter chimneys waving truce against the moon ( Mitchell )

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest