Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Apr Mon 22, 2019 12:08 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 23 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Fri 11, 2008 12:43 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 7503
Location: Minneapolis, MN USA
I bought a bunch of pulls. 35L6 and 50L6.
Some show HK leakage on an Eico 666.

I know what the manual says. Don't wanna test in radio.
I'm curious what your opinion is.

1-5 megs leakage OK?
10 megs leakage OK?


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 1:05 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 34103
Location: Maryland 20709, USA
Hi Dale,

To my mind, NO measurable leakage is acceptable.

My tube tester is designed to measure leakage to 10,000 Megohms. Once I get it up and running I'll be able to characterize tubes with considerable accuracy. Then we can determine what a reasonable number might be.

_________________
73 de W3NLB
http://www.AtwaterKent.info
Click "Grebe Stuff" for Synchrophase info


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 1:14 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4444
Location: Connecticut. USA
Back in the 1970's I remember the 6GH8 tube showed leakage most of the time, even some of the new ones.

_________________
Bill Benson


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 1:42 am 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 10823
Location: Latham NY
The Eico has a very sensitive H-K leakage test which is a good feature. However as a practical mater for output tubes on a table radio that amount of leakage is not going to mean anything. You can do some experiments for yourself using a pot on a working radio and see just how much leakage is required to produce noticeable hum on a table radio. In RDH4 the max for new power amp tubes is 1-2 megs (pg123). Even less for an old table radio, don't throw away good tubes if they work good in your radio. Preamp tubes in a hi-fi amp is a whole different story.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 3:10 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 7503
Location: Minneapolis, MN USA
Lou deGonzague wrote:
You can do some experiments for yourself using a pot on a working radio


I Like this idea. Should be an interesting exercise in theory VS reality. Can always you a fun diversion on the bench.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 4:57 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1070
Location: 98802
I don't know how that particular tester checks for leakage. I wonder if it really is leakage the tester is measuring.

While troubleshooting some unwanted modulation in an old HP audio oscillator, I discovered an interesting effect. My tube tester did not flag any of the tubes as leaky, yet still I suspected hum was getting in somewhere. I wondered if a tube could have some h/k leakage the tube tester missed.

I set up to test leakage the same way I do for capacitors. I used a bench power supply in series with a VTVM. Using this method, if you know the internal resistance of the VTVM, using Ohms Law you can calculate the leakage resistance at the voltage you test at (and the leakage current if you're interested). Most tubes have a maximum heater to cathode voltage, and sometimes it is different depending on polarity. Stay below the maximum and you should be fine. I don't remember how much voltage I used, but suspect I was not very close to the maximum.

With the tubes cold, the leakage resistance was *extremely* high, in other words they all tested good. The tubes in question were 2 6j7's, a 6f6, and a 6v6. Since they checked good cold, I wondered if the insulation would break down when they got hot. I powered up the heater on the 6v6......

It leaked a fair amount. I thought I had found the problem. I started digging through all my 6v6's looking for a replacement. They all did it. So did the 6f6 from the generator, as well as all of my spares. The 6j7's did too, though a very small amount by comparison. In all cases, when I turned the heater voltage off the leakage went away.

At some point I switched the grabbed the wrong knob on the VTVM and switched it to AC. I found some. Where did *that* come from? :shock:

I then put the VTVM back on DC and switched the polarity. It leaked quite a bit less in the other direction. At that point I broke out the oscilloscope. The cathode assembly was behaving as a diode, a lousy, leaky thermionic diode, with the tube heater as cathode, and the tube cathode as plate. The oscilloscope showed half-wave rectification of the heater voltage.

Further experiments showed every old prewar octal power pentode or beam power tube I could lay my hands on that night did it, Some more modern ones, 6bq5, 6aq5, 12by7, barely did it, if at all.

The Radiotron Designer's Handbook mentions this effect, but doesn't say much. I found only one other book that mentioned it at all. One book suggested running the heaters with a bias voltage on them to reduce hum. I suspect they meant to bias this "diode" off, even though they did not exactly say so. I never tried to bias the diode off, so I'm not sure where that would take you.

Since the 35l6 and 50l6 have a lot higher heater voltage, I would expect the rectified DC pulse on the cathode would be of a higher voltage too, though you would have to try it.

I would think a tube tester with a sensitive leakage test would have a tough time differentiating between this effect and leakage. In a set, some hum would make it into the output, but the signal level is high in the output stage. It might not be noticed.

John
.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 6:23 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 34103
Location: Maryland 20709, USA
blue_lateral wrote:
The cathode assembly was behaving as a diode

That's very interesting information, John.

I'll set up some specific tests to look for that effect.

Thanks :D

_________________
73 de W3NLB
http://www.AtwaterKent.info
Click "Grebe Stuff" for Synchrophase info


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 6:38 am 
Member

Joined: Jul Sun 02, 2006 9:15 pm
Posts: 884
Location: Norwalk, CT
That is interesting, and as you have mentioned there is often talk of biasing heater supplies to cut hum.

Denis


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 3:23 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1974
Location: Hinsdale, IL, USA
Very good information. My interest is peaked as well.

I have some leaky tubes I will play with also.

How about this: put a pot across the filament, then apply the DC to the wirper of the pot and balance the hum. From your description it sounds like if you use a DC source and reverse bias the heater to cathode i.e. negative on the cathode, then what ever current flows should be due to leakage. What do you think?

Chuck D. KB9UMF


Top
 Profile  
 
 Post subject:
PostPosted: Jan Fri 11, 2008 6:28 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 30698
Location: Pocasset, Cape Cod, MA
My audio amp, which I built from plans in (I think) a 1959 Electronics World, has its heaters biased 40V positive with a voltage divider from B+ to ground. I always understood that was to eliminate heater-to-cathode emission.

I thought that tube testers were designed to be insensitive to that. Whether by polarity, or applied AC voltage, I don't recall.


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Mon 30, 2012 6:22 am 
New Member

Joined: Jan Mon 30, 2012 3:57 am
Posts: 3
Let's focus for a moment on one of the MOST valuable vintage tubes - the "coke bottle" 6X5G. These were put into lots of ~1940 radios, especially old Zenith radios. However, smaller 6X5GT tubes came out soon thereafter, but they don't look nearly as good in the old radios as the original 6X5Gs. As 6X5Gs died, they were replaced with 6X5GTs. There are still some 6X5Gs floating around, mostly salvaged from old radios that were scrapped out. However, ALL 6X5Gx tubes test as shorted!!! You can easily confirm that it is a heater emission problem and not a short by dropping the filament voltage on the tube tester and seeing that the "short" disappears, and on doing that, I have yet to find one that was actually shorted. I suspect that one of the reasons that 6XS5Gs are SO rare is that most of them have been thrown away as being "shorted", when there is absolutely nothing wrong with them!!!

BTW, you can also "make you own" 6X5G tubes by slapping an octal socket onto a #84 tube.

These tubes are indirectly-heated cathode dual power diodes, commonly known as "power tubes". If you look at typical power supply circuits, the heater is at ground potential, while the cathode is at B+ potential, so there will be PLENTY of unintended (but apparently not unexpected) emission from heater to cathode. I don't think that this would "impair" operation, but it could conceivably get carried away and produce a sort of dynamic short, and conceivably burn out the power transformer.

Does anyone have any experience or real data concerning this? I would think that some ancient tube manual would have an upper limit for heater emission. I don't seem to have one that is old enough, so I would REALLY appreciate it if someone who has a really OLD tube manual take a look at it and see if there is any spec for maximum heater emission.

I suspect that the trick would be to setup an emission tube tester to test the tube like it is a simple diode, but with the heater being setup as the cathode, and the present "cathode" being setup as the plate, and only if it tests BAD is the tube acceptable. I can figure out how to set this up on Heathkit testers, but I will first need the maximum heater emission specification.

Any thoughts?

Thanks - especially to anyone who can dig up some old heater emission specifications for 6X5G tubes.

Steve.Richfield@gmail.com


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Mon 30, 2012 10:07 pm 
Member
User avatar

Joined: Oct Fri 16, 2009 11:10 pm
Posts: 455
Location: Bergen The Netherlands
From R.B.Tomer´s book :¨Getting the most of Vacuum Tubes¨ I learned three things.

a. All tubes have some heater-to-cathode leakage.
b. The leakage is in high-powered tubes greater than in small-powered tubes.
c. There is the possibility of short tests causing tube damage

I think, that many good tubes have ended their lives in the bin
just because a ¨false¨reading of a tubetester.

Jard N.

For those interested in the story: page 121 +122
(Total Book 164 pages)
http://www.nutshellhifi.com/library/MostVacuumTubes.pdf

_________________
If You Ain't Dutch, You Ain't Much


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Tue 31, 2012 4:45 am 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 8327
Location: Long Island
If you want some background into heater-cathode insulation phenomena, see "Materials and Techniques for Vacuum Tubes" by Walter H. Kohl, p. 300-302.

The insulation that separates the heater from the cathode is a form of aluminum oxide. Guess what else uses an aluminum oxide dielectric. If you said, "electrolytic capacitors," you were right.

Just as electrolytic caps are polarized, and break down rapidly if connected in the wrong direction, the heater-cathode insulation in tubes is also polarized. In most small receiving tubes, if the heater is positive with respect to the cathode by 100 volts or more, breakdown of the insulation by "electrolysis" is far more likely than if the heater was negative with respect to the cathode.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Tue 31, 2012 4:58 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 37248
Location: Livermore, CA
After a lot of use many higher power filament tubes will have heater to cathode leakage. Besides 6X5 mentioned, 50L6, 25L6, 43 etc may have leakage between cathode and filament.

Testers like Eico 666 show amount of leakage on a meter. A person needs to set their own limits for this leakage when replacing tubes. Higher filament voltage increases leakage. New tubes usually won't register leakage on the meter.

_________________
Norm


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Tue 31, 2012 5:55 am 
New Member

Joined: Jan Mon 30, 2012 3:57 am
Posts: 3
I think I found a reasonable qualitative way of gauging heater emission aka heater cathode leakage on any ordinary tube tester. I set my Heathkit emission tester up, using the same TYPE and PLATE settings as for normal testing of the same tube, but with all pins grounded (levers in center position) except for one side of the filament that remained connected to the filament supply (lever down), and the cathode in the "up" position that connects it to the tester's B+.

In this test, GOOD is bad, and BAD is good, but only if it is REALLY "BAD", like <5%.

I tested a 6X5G using this setup, that had previously lit the "short" light very brightly, and the gauge barely budged, maybe 2% of the way up, and VIBRATED so much that the thin end of the needle seemingly disappeared!!!

I interpret this that the heater cathode issue will suck off ~2% of the output of the power supply, so the tube will be OK in service. I suspect that a rational good/bad cutoff would be at the first line on the scale, at 5% of full scale.

An obvious caution - feel the power transformers in those old radios from time to time, to see they aren't getting extremely HOT. A heater-cathode near-short in a power tube could easily cremate a power transformer. Better yet, put a fuse into the circuit.

Any thoughts?

Steve


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Tue 31, 2012 2:05 pm 
Member

Joined: May Sat 22, 2010 4:42 pm
Posts: 2987
Those tubes with a high voltage heater do show more H-K leakage, as there is sooo much more heater wire to leak from.

It's not very significant, as those tubes are usually run with a small cathode resistor, like 220 ohms, and that will limit the induced voltage to a very few millivolts, even less if there is a cathode bypass capacitor.

Where you really don't want H-K leakage is in a tube that's running at low current, with a large cathode bias resistor, unbypassed. For instance a 12AX7 preamp first stage. To minimize the leakage, those heaters are usually put near the grounded end of the heater string. Some Bogen amplifiers run the 12AX7's off the DC voltage at the cathode of the power output tubes, a clever way to get DC on the heaters.


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Jan Tue 31, 2012 6:46 pm 
Member

Joined: Oct Sat 20, 2007 3:36 am
Posts: 14393
Location: Southern NH, 03076
Ive found many 6V audio tubes to be particulary prone to leakage and I dont mean in audiophoolery circuits either. The 6F6, 6V6 and sometimes 6L6's seem especially prone just from normal radio use.

Miniature pentodes (plus the 6BE6 converter) and a few octals used in RF and IF amps are others that my 752A reject AND cause problems in radios, especially to the AVC/AGC.

This is the main reason I rarely buy used tubes on here or from other credible sources as most tube testers just dont test this well.

Carl


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Feb Wed 01, 2012 7:56 am 
Member

Joined: Sep Thu 23, 2010 6:37 am
Posts: 11018
Location: Powell River BC Canada
Here is tests for tubes made for testing components of tubes.

Heater to cathode leakage ihk is measured +/-100 volts.
Attachment:
ASTMstuff(12AZ7)maybe.jpg
ASTMstuff(12AZ7)maybe.jpg [ 243.44 KiB | Viewed 4596 times ]

_________________
de
VE7ASO VE7ZSO
Amateur Radio Literacy Club. May we help you read better.
Steve Dow
ve7aso@rac.ca


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Feb Wed 01, 2012 5:59 pm 
Member

Joined: Oct Sat 20, 2007 3:36 am
Posts: 14393
Location: Southern NH, 03076
The grid emission test is also interesting.

Where is that document from?

Carl


Top
 Profile  
 
 Post subject: Re: Acceptable heater-cathode leakage for some common tubes?
PostPosted: Feb Wed 01, 2012 6:30 pm 
Member

Joined: Sep Thu 23, 2010 6:37 am
Posts: 11018
Location: Powell River BC Canada
1966 Book of ASTM Standards Part 8
(F8 - 1964)
..(ASTM part 8 Nov.1960)
...testing electron tube materials using reference triodes...
Appendix II... electrical characteristics..suggested methods of testing..

With these tubes perhaps leakage occurs as the end-bends flex
exposing metal which then sees the tungsten sleeve ends,
and the tungsten being brighter than the uncoated
inner surface of the cathode sleeve forms a diode, especially at
higher heater voltages. Some 12A?7s do flash bright from that
spot on power up, yet work fine.

Heater Details
Attachment:
TriodeHeatermech.jpg
TriodeHeatermech.jpg [ 104.46 KiB | Viewed 4571 times ]

Attachment:
HeaterCoating11.jpg
HeaterCoating11.jpg [ 136.42 KiB | Viewed 4571 times ]

Attachment:
Heatercoating22.jpg
Heatercoating22.jpg [ 140.9 KiB | Viewed 4571 times ]

_________________
de
VE7ASO VE7ZSO
Amateur Radio Literacy Club. May we help you read better.
Steve Dow
ve7aso@rac.ca


Top
 Profile  
 
Post New Topic Post Reply  [ 23 posts ]  Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: anchorman, dge333, dholland, Ed in SoDak, ertzc2020 and 9 guests



Search for:
Jump to:  




























Privacy Policy :: Powered by phpBB