Life expectancy of tubes vs heater voltage.

Help Support GroupDIY:

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,543
Location
UK.
Lester Smith takes care of the mics at Abbey Road Studios where they still have lots of the Neumann and AKG mics they bought in the 1950's. He does have spare VF14's still in the box for the 47/48s but he's been able to successfully 'regenerate' a couple that went 'rogue'.
 

My3gger

Well-known member
Joined
Nov 2, 2012
Messages
441
Location
EU
Lester and i spent some time writing back and forth about tube preamp EMI made under BREDD, it was for my studies about their circuits. He is great and very passionate guy, i will ask him about tube "regeneration".
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,673
Location
Marcelland
What I was referring to was emission characteristics due to the pulsing of AC, which although brief, affects emission, under strain. The reversing of polarity might also have something to do with it, although these are very slight effects, compared to voltage variations, I imagine.
OK, I completely misread your sentence.
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
93
Location
Ventura, CA
I'd imagine that "emission regeneration" is similar to the commercial "CRT rejuvenator" that I used back in the early 1960s when I was a tech in a large TV repair shop (not found in internet search but controls similar to EICO 633). Because back in the day, "your picture tube is weak" was the most dreaded diagnosis when the symptom was a dim picture, there was a big need for such a device. And it did work quite well. The "control grid" (closest to the cathode) in a TV CRT is a cylinder surrounding the cathode but having a small hole at the end toward the screen. This, along with the other downstream grids, formed the so-called "electron gun." The normal beam current in a CRT is under 1 mA, even with "brightness" maxed out (with 20 kV, that's about 20 watts to light the screen). What the rejuvenator did was first crank up the heater voltage from 6.3 to about 9 volts (scary in itself) and then briefly (as long as you held a button down) apply about +150 V to the control grid, presumably "stripping away some of the outer surface oxide coating on the cathode (you can see this whitish powder on the surface of most vacuum tubes that use indirectly-heated cathodes). But the resulting grid current would also cause the grid to get very hot - and, if you held the button down more than a few seconds, the edge of the little hole would glow orange and begin to melt. There was a test mode in the machine as well, which I think measured the screen (G2) and other beam-forming grid currents under zero bias on the control grid. So generally, you'd measure emission before starting the procedure, thenpress the button for a few seconds, then switch back to test mode to see (on the analog meter) how much emission had improved (after returning heater voltage to 6.3 of course). Although it often took several "hits," The results were very often impressive, giving an old tired CRT a few more years of life.

For small-signal tubes, I'd imagine this technique would work ... but modified to be a bit gentler on the grid, since most grids are obviously more fragile - most are a coil of tiny wire welded to a couple support pins. But the same emission testing (the method used by the least-expensive, drug-store variety tube testers) is pretty easy to do - consisting of a low-voltage AC source and a mA meter using the tube as a rectifier (tying all the non-cathode elements together to form the anode). If I desperately needed to "rejuvenate" a small-signal tube, this would have a good chance of working I think. If anybody has the time and inclination to try it, I'll offer to supply advice on building an using one.
Bill Whitlock
[email protected]
 

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,543
Location
UK.
Interesting read Bill, thanks.
When the tube is a VF14 for your Neumann U47, and a replacement is unobtanium or $3,000, then that seems like a very good proposal.
 

gyraf

Well-known member
Staff member
Moderator
Joined
Jun 4, 2004
Messages
10,217
Location
Aarhus, Denmark
For use in microphone front-ends, underheating a tube often helps achieving the needed very-very low grid bias current. So VF14 is underheated heavily in the U47/48/mm2 - specified at 55V, heated only to 36V in those. Have one that's still on it's first tube, another where we replaced the original one in 1989. Both seeing near-daily use for many decades now

How well a tube reacts to underheating in a mic circuit seem to depend on the cathode structure somehow - I have found that across all manufacturers, the PF86 reacts much better to underheating than its sibling the EF86, only difference being heater structure and material (300mA series heat in stead of 6.3V parallel)

Was a common thing to do in electrometer (stupid-high-impedance) amplifiers too

/Jakob E.

edit: Re-tubing old classics?
 

Tubetec

Well-known member
Joined
Nov 18, 2015
Messages
3,006
The VF14 in a U47 only has 34 volts at the anode , yet 36v at the cathode, I dont understand how the anode can be less positive than the cathode and still attract electrons .
VF14 spec sheet shows 200 volts max at the anode, so its massively undervolted in the U47. I read a few mentions of undervoting heaters while starvtion conditions at the anode , it does seem to make sense as the less high the anode voltage the less it attracts emission from the cathode the less electrons you need there ,

Heres a quote from the Phaedrus audio site ,

The VF14M valve acts as a voltage amplifier circuit and drives the output transformer which drops most of the voltage gain of the valve, but translates the output impedance to drive the long microphone cable. As explained above, there is only one supply which is used for the HT and for the heater supply for the valve (under-run in this application at about 35V). The heater supply is simply dropped by the 1800Ω resistor which dissipates about 2.6 watts in the process.
The VF14 valve is strapped as a triode, with the additional, unusual feature that the suppressor grid is strapped to the anode, rather than the more usual cathode. More unusual still, the valve is operated with a very low anode voltage (circa 35V).
These conditions are well outside the normal operating and design envelope of the VF14 device and, in the majority of cases, this anode voltage is not enough to pull the valve out of grid-current. (This, along with low-noise and lack of microphony, was one of the principal conditions of the Neumann selection.) This also explains why the valves are operated at 63% of their specified heater voltage. By reducing the temperature of the cathode, the emitted electrons have a lower statistical kinetic-energy and the available space-charge of electrons is shrunk to that required to sustain the modest anode current. In this way, there are less electrons “all charged up with nowhere to go” which could contribute to a grid current.
 

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,543
Location
UK.
The VF14 in a U47 only has 34 volts at the anode , yet 36v at the cathode, I dont understand how the anode can be less positive than the cathode and still attract electrons .

That's 36V on the heater, not cathode. The cathode is certainly tied to the low side of the heater, but that side sits, if memory serves, at a little over 1V.
 

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,543
Location
UK.
Was just looking at one of the BBC Eng. documents for an early 1950's mic amp and, by chance, noticed that the heater voltage for the first EF37a tube, besides being DC rather than AC like the other EF37a, is also run at 5V rather than 6V1.
 

Attachments

  • BBC pre power .png
    BBC pre power .png
    32.3 KB · Views: 7

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,673
Location
Marcelland
Was just looking at one of the BBC Eng. documents for an early 1950's mic amp and, by chance, noticed that the heater voltage for the first EF37a tube, besides being DC rather than AC like the other EF37a, is also run at 5V rather than 6V1.
How would they do taht? Separate winding or series resistor?
 

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,543
Location
UK.
How would they do taht? Separate winding or series resistor?


Same winding. 6v1 AC is used for valves 2 & 3 heaters. Also across the winding is a rectifier (obviously with a high drop), then a smoothing cap and the resultant 5V DC to valve no. 1 heaters.

This is outside broadcast equipment (OBA/9) so provision is made to switch over and run from batteries if necessary, in which case it looks like all heaters are on 6V DC


 

mjrippe

Well-known member
Joined
Dec 27, 2011
Messages
2,520
Location
Brooklyn, NY
Same winding. 6v1 AC is used for valves 2 & 3 heaters. Also across the winding is a rectifier (obviously with a high drop), then a smoothing cap and the resultant 5V DC to valve no. 1 heaters.

This is outside broadcast equipment (OBA/9) so provision is made to switch over and run from batteries if necessary, in which case it looks like all heaters are on 6V DC


Actually, if you read the paragraph right below the chart you posted, it states that the meter reading for V1 is about 83% of the battery voltage. Sounds to me like that diode and cap are still in circuit ;)
 

Attachments

  • Screen Shot 2021-06-17 at 3.09.15 PM.png
    Screen Shot 2021-06-17 at 3.09.15 PM.png
    65.4 KB · Views: 5

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,543
Location
UK.
Actually, if you read the paragraph right below the chart you posted, it states that the meter reading for V1 is about 83% of the battery voltage. Sounds to me like that diode and cap are still in circuit ;)


Ah OK, thanks for checking that out :) It actually kind of answers a question I had.

As per the blurb, it states that DC heaters for V1 enabled a less stringent testing for suitably quiet valves in that spot.
I wondered if the 5V rather than 6V1 was considered a negative side effect of tapping off the same winding, or if it was considered beneficial.
Since it would have been easy to run the 1st EF37a on 6V1 when operating from batteries, I suppose we can assume they, at least, thought it wasn't harmful running at 5V.

Other than that, I'd never seen this amp schematic before and it threw me for a loop for a while as I was figuring out what the hell was going on with the V2 to V3 coupling. It's just drawn in a way that looks strange 🤪
 

Tubetec

Well-known member
Joined
Nov 18, 2015
Messages
3,006
Anode voltage comes out at just over 100 on the first Ef37a , in or around what you'd be wanting to undervolt the heater safely.

I love the circuit descriptions included with the BBC stuff , written by techs for techs , a million miles from the usual (hogwash) sales blurb associated with audio gear nowadays.
The other thing is , imagine the cost of a metal oxide bridge rectifier back then , it most certainly wasnt a decision taken lightly to under volt the first valves heater , but one based on solid feedback from the Bee'bs worker bees in the field.

Theres a link here somewhere to all of the old BBC designs from the tube era , PRR unearthed it as far as I remember , BBCeng

Pumping vast amounts of tax payers cash in the direction of a national broadcaster does not(and never did) mean the news feed is unbiased , but it did ,at least back in the old day mean they were at cutting edge technically .

beats the album version hands down ,
 
Last edited:

Latest posts

Top