• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Weird Tube Test Results

P

prww

AK Subscriber
Subscriber
I have a B+K 700 Dynamic Mutual Conductance tube tester. You look up the tube in the chart, set the dials, plug it in and test. Will test for shorts, grid emissions, and give a reading on the single meter. Above 65 is in the green up to 120. I pulled out a box of tubes I've had for decades and plugged in an RCA 12BY7. Tested at 78. Put the tube back in the box and it came out the bottom and fell 2 1/2 feet to the cement floor and bounced. S***! Plugged it back in the tester. No shorts and no grid leaks. Pushed the test button and now it read a solid 90. Unplugged it, let it cool, and tested a solid 90 again. Anyone guess what happened? Maybe I'll start a service. Send me your tubes, I'll drop them on my basement floor and return them to you in better shape. Paul.
 
Register to hide this ad
I was reading somewhere that manufacturers actually used to do this to correct some out of spec tubes, as well as to "shock seat" tubes to reduce microphonics and discover defects before they happened naturally in the field. I normally despise AI, but it is useful in such errands:


⚡ You are hitting on one of the most fascinating "secret menu" engineering notes in vintage vacuum tube history! You are likely thinking of a technical discussion surrounding Western Electric, Mullard, or Telefunken laboratory practices—or specifically, the historical "tapping" or "pulsed shock" repair methodologies detailed in old engineering forums like The National Valve Museum.
What you are describing is an extreme, factory-level extension of the common "tapping a tube to check for microphonics" technique. Instead of checking for noise, engineers used a directional kinetic strike to physically cold-form or bend warped internal components back into spec.

The "Targeted Directional Shock" Technique​

When a vacuum tube left the production line out of specification (for example, having low transconductance or a skewed grid bias), it was often because the control grid or cathode was physically off-center. Because electron flow changes exponentially based on the distance between the cathode and the control grid, even a microscopic warp on one side could ruin the tube's metrics.
To salvage high-cost or ultra-premium valves, factory technicians and advanced laboratory engineers utilized a highly specific recovery process:

  1. Electrical Mapping: The tube was placed in a dynamic analyzer to determine exactly how the parameters were skewed. If the plate current was too high on one side, it meant the grid was leaning too close to the cathode on that specific axis.
  2. Axis Alignment: The technician would calculate the exact inverse physical plane of the error. The tube would be rotated in a specialized jig so that the leaning element faced the direct trajectory of a kinetic strike.
  3. The Engineered Strike: A calibrated, spring-loaded mechanical striker (or a specialized high-G drop-shocker) would strike the base or a reinforced side of the tube.
  4. The Result: The momentum of the high-G shock would utilize the inertia of the heavier internal elements (like the plate or the grid support rods). The kinetic energy would slightly elongate or compress the microscopic mounting tabs on the mica spacers, snapping or "nudging" the grid back into a perfectly centered position.

️ Where This is Documented​

If you read this on a discussion board or a specialized archive like The National Valve Museum, it is usually tied to three specific historical contexts:

  • Frame-Grid Tube Calibration: In the late 1950s and 1960s, manufacturers introduced "frame-grid" tubes (like the famous ECC88/6DJ8 or the Telefunken CCa). These grids used incredibly thin wire wound tightly under massive tension onto a rigid metal frame. If the frame shifted a microscopic amount during factory baking, an engineered tap was the only way to re-center the incredibly tight tolerances.
  • The "Rubber Mallet" Myth vs. Reality: On vintage audio and radio restoration forums, old-timers frequently discuss "fixing" a shorted or wildly out-of-spec tube by striking it against a phone book or hitting it with a small rubber mallet at a precise angle. While it sounds crude, it is the backyard equivalent of the factory directional G-shock—attempting to knock a sagging filament or grid wire out of a short-circuit state.
  • The Bendix "Red Bank" Division: Bendix famously manufactured tubes designed to survive missile launches and atomic blasts (like the 6094 and 6106). Their elements were locked in place with ceramic spacers instead of mica. Because ceramic doesn't give or flex like mica, factory "shock-seating" on specific directional axes was used to permanently settle the rigid structures before final military certification.
 
wparks, Thanks for the detailed explanation. Amazing what we learn here. As a vintage tube hoarder (collector), I hate to throw any tube in the trash. While I have a few days until the next trash pickup to change my mind, I should be comfortable leaving it there. I have no way of knowing if I knocked something in or out of alignment. I posted this looking for an explanation and I have that now. Just prior to posting, I learned to be very careful with old tube boxes. Paul.
 
Cannot put my hands on it at this time, however RCA did say that some tubes could be restored to testing good by tapping. I have had shorted o1-As test good after tapping and when checked in radio worked.
 
kcline said:
Cannot put my hands on it at this time, however RCA did say that some tubes could be restored to testing good by tapping. I have had shorted o1-As test good after tapping and when checked in radio worked.
Click to expand...
I have an RCA 6L6GC that tests good, but lights the shorts light. Really pains me with an expensive tube like that. I think I'll play with that tube to see if the short light will go out. What would bother me is if a "tap" fixed the short, how careful do you need to be so that another accidental tap doesn't put it back where it was? I would likely not want to use that tube and risk an expensive output transformer. Paul.
 
short/leak from where to where though, and how bad is it? If its from heater to cathode, a decent bit of leakage isn't uncommon but also doesn't have a lot of effect in a power tube. The cathodes are generally grounded, or bypassed to ground so any heater hum that may get leaked in just gets dumped to ground instead of amplified.
 
I've had power tubes with severe H-K shorts burn out either the hum balancing resistors or pot, and occasionally cause blown fuses when the heater was directly grounded at one end or through a center tap. That variety of heavy short is pretty rare though.
 
gadget73 said:
short/leak from where to where though, and how bad is it? If its from heater to cathode, a decent bit of leakage isn't uncommon but also doesn't have a lot of effect in a power tube. The cathodes are generally grounded, or bypassed to ground so any heater hum that may get leaked in just gets dumped to ground instead of amplified.
Click to expand...
Good question gadget. I hooked my VOM meter to all pins on the suspect 6L6GC. No shorts between any pins on both X1 and X10k scales. The shorts test does not light the shorts light. The grid leak test shows considerable meter deflection and lights the shorts light. I opened the manual for my B+K 700 tester. The 6BN8 tube referred to is internal to the tester. This is from the manual. "There is some little evaporation of the cathode coating material on the grid of a tube. Some of this vapor tends to deposit on the grid of a tube and gives rise to grid emissions. For the grid leak test, the tube under test has normal grid to plate voltage applied but the grid is biased beyond cutoff so no plate current flows. This bias is applied through a 5.6 meg resistor. The same 5.6 meg resistor is also in the grid circuit of a 6BN8 DC amplifier that is also biased just beyond cutoff. If the tube under test is gaseous or the grid is contaminated with cathode coating, current will flow from grid to plate lifting the cut off bias across the 5.6 ohm grid resistor and causing current to flow through the 6BN8 tube and causing meter deflection, and the short light to light. The sensitivity of the grid leak emissions circuit is set so that a leakage of 25 meg. ohms will read in the grid leak reject area" I condensed this a bit. But obviously this RCA 6L6GC is shot. The grid is contaminated with cathode coating. Thanks for the question gadget. It prompted me to look this up and learn what is happening with this 6L6. It may take me a few kleenex to place this one in the trash. Paul.
 
it usually only appears with voltage applied, so not too surprised. I don't suppose your tester shows which pins it thinks are shorted? My Eico doesn't exactly have a short test, its got an ohm meter and it will just read directly for each pin, and its also got a button to push for heater-cathode leakage testing.


I did once have a heater-cathode short in a 12ax7 that nearly smoked an output tube set. My Fisher 600 uses a pair of 12ax7 heaters as the cathode resistor. I happened to be taking some voltage measurements and suddenly they stopped making sense. A little further poking and observation showed low cathode voltage and the output tubes starting to glow a dull red. One quick shut-down and lots of head scratching later I pulled the 12ax7s to test since I wasn't finding any explanation. The one closer to the ground end shorted heater to cathode on one section and that cathode goes straight to ground. Effectively the cathode resistor became 25% smaller and the amp wanted none of that. I think it was a Westinghouse branded one that said "England" but closer inspection found "Korea" etched in the glass, so I really don't know what the heck it was other than a deposit in the circular file. Oh well, it was an excuse to put something better in the phono stage. I had those plugged in just for testing purposes.
 
This is what I think I understand. The cathode has a coating that helps it emit electrons. This coating has deteriorated, turned to gas and some of that gas has attached to the grid. The grid now emits electrons (it's not supposed to), causing grid current to flow, causing a drop across the grid resistor. The bias now goes more positive causing more current to flow. Self destruct of the tube has started and it will not last long. There is no direct short in the tube. That is why I could not measure one. But current flow from grid to plate acts like one. The B+K 700 manual is available online and can explain what I left out starting on page 6. Too much for me to type. I type the Columbus method. First I discover the key, then I land on it. Paul.
 
cathode material on the grid is it's own issue, and a bad one to be sure. The tube will simply run away with current. I've had tubes that tested excessively high, and they all behave that way. I've never been quite sure of the exact failure mode, I just know that if I test a power tube and my Eico tells me 140% and puts the needle in the corner, it is not going to work correctly. Could be grid emission, it would track with the symptoms.
 
Just goes to show you that sometimes what the tube tester tells you, and what really happens in the unit can be two very different things.

3516c0d3-6784-4eb4-9501-8ce553d35876.gif
 
maybe I should propose ultrasonic treatment. Chuck crusty used tubes in the ultrasonic bath while I'm washing records and claim it relieves stresses or whatever other magical language will make people pay me a fortune for tired pulls.
 
Is that Copper on a Lead head?

Among the 32 hammers I own there is no copper one... :(

Plastic.
Lead.
Aluminum.
Leather.
Steel.
Hardened steel.
Brass.
Rubber 3 kinds.
Along with a dozen different shapes.
 
You must log in or register to reply here.
Back
Top Bottom