Thanks for the explanation and insight.
There is a lot of relevant info and experimental results on radiomuseum website, i just need to slowly read through it.
They had this little tube's filament operate down to 0.7v, and found that gain was still acceptable and quite linear for control grid voltage in -0.3v to +0.3v range (with anode and screen at only 30V). The Vgc goes up to +0.5v in that graph.
Personally, i call the Ip vs Vgc graphs that are cutoff at Vgc=0 axis, the "cripple" or "crippled" curves.
The more I read about this tube, the more I like it, so I just ordered a handful of beefier 1J24B-V version.
There is a lot of relevant info and experimental results on radiomuseum website, i just need to slowly read through it.
They had this little tube's filament operate down to 0.7v, and found that gain was still acceptable and quite linear for control grid voltage in -0.3v to +0.3v range (with anode and screen at only 30V). The Vgc goes up to +0.5v in that graph.
Personally, i call the Ip vs Vgc graphs that are cutoff at Vgc=0 axis, the "cripple" or "crippled" curves.
The more I read about this tube, the more I like it, so I just ordered a handful of beefier 1J24B-V version.
I tried them as 1st stage in a high-gain phono amp, battery heater and battery plate supply. It was devastating ...
More than 50% were extremely microphonic; when tapped with a screw driver few responded just with a muffled tick but most were ringing endlessly, a sustained 5kHz tone with very little damping over 10 seconds plus. Seems to be the mechanical resonance of rods. So bad that acoustic feedback from the speakers set in when volume was turned up. Had to wrap them in corrugated card board to suppress sustained feedback whistle.
Screen current wasn't so low either, 100uA @ 1mA, so 10% of plate current is not much better than ordinary small signal tubes.
Actual gain in triode was 10 with rsistor load, 13 with gyrator, in pentode up to 100 with gyrator but lots of distortion, too, and the best I could get from triodised pentode mode was 20.
Either way noise was huge, mostly low frequency - not hum - rather a statistcal grumble.
I gave up on there ...
Thanks for sharing your experience. First hand user information is much more valuable than that from articles, data sheets, and web sites.
In the data sheet, microphony test for these tubes was to vibrate them at 10xg acceleration, 50 Hz, and measure 50 Hz output signal due to vibration. It was 50 mV or less. Of course, this test says nothing about 5 kHz vibration.
It appears then that these tubes are not good at low signal levels. Probably, no directly heated tube is.
I still think that 1Ж24Б and its up-rated cousin 1Ж29Б are useful in audio. They seem particularly suitable for intermediate signal levels, especially for transformer-coupled preamplifier output. I don't believe there are any other tubes that have such low plate resistance at such small plate currents, which is advantageous in transformer coupling. 1Ж24Б is 15 K at 1 mA and 1Ж29Б is 4 K at 8 mA.
At intermediate signal levels, microphony does not seem to be a problem. Here is a youtube video demonstrating the sound of a two-stage amplifier using 1Ж29Б and 1П24Б, both in pentode mode (narrative in Russian, but it doesn't matter). The schematic is attached.
YouTube
Due to their small size, rod tubes are amenable to acoustic decoupling treatments. For example, a tube can be encased in sand-filled metal box suspended on air pads. Acoustic treatment may be worth it to exploit the advantages of these plentiful and super cheap (so far) tubes.
Attachments
Rod pentode screen current
Rod pentode screen current (10% of plate current) in data sheets is maximum guaranteed at zero volts on control grid, which is the worst condition for electron beam focusing. At negative g1 voltages, focusing improves, and Ig2/Ia decreases.
Here are screen currents for typical small signal audio pentodes, % of plate current:
EF86: 20%
E80F: 22%
6U8: 37%
6AU6: 42%
Rod pentode screen current (10% of plate current) in data sheets is maximum guaranteed at zero volts on control grid, which is the worst condition for electron beam focusing. At negative g1 voltages, focusing improves, and Ig2/Ia decreases.
Here are screen currents for typical small signal audio pentodes, % of plate current:
EF86: 20%
E80F: 22%
6U8: 37%
6AU6: 42%
Last edited:
Um er, maybe you had bad 1ZH24bs? Mine are not microphonic, or at least no more than any sub-miniature tube. You DO have to be careful with magnetic fields, but not too much. I built an HF receiver using them, in conjunction with a few 1ZH29bs, 1ZH37bs and 2x 1P24bs. Actually, 15 tubes in total, with an inbuilt speaker. Better than 1uV sensitivity and no microphonics! The audio PA produces about 2WRMS at 120V with 90V on G2 and about -7V on G1. Here is the PA on an early test:
YouTube
Some other things:
Positive grid voltages - in 'normal' operation, the grid is actually negatively biased by virtue of the DH cathode 'middle' being at ~0.6V. In fact, the tubes will handle about 300-400mV positive in this configuration, although the tube will be turned hard-on. My concern is people running the heaters backwards (they are polarity-sensitive), or negative, which will result in a severe deviation from the published curves!
People running these in triode configuration are missing the point. Reduce gain with a lower plate resistor and/or reduce screen voltage. That way, bandwidth is not affected and still good beyond 100MHz. Some 'trioderised' models i have seen have trouble at 16kHz =]
1ZH37b - a useful tube as an RF or audio mixer. The 2x G1 inputs are NOT differential, which so many people seem to have missed. There IS a small amount of coupling between them (<1pF), so not really well isolated at RF, but fine for AF. Parallel the 2 grids for twice the transconductance, or feed audio to both, but vary the DC voltage to one only for variable gain.
The general 'rule' for all of these pentodes (apart from the 1P24b) is to keep grid-cathode AC voltage below about 1Vpk and you cannot go wrong.
YouTube
Some other things:
Positive grid voltages - in 'normal' operation, the grid is actually negatively biased by virtue of the DH cathode 'middle' being at ~0.6V. In fact, the tubes will handle about 300-400mV positive in this configuration, although the tube will be turned hard-on. My concern is people running the heaters backwards (they are polarity-sensitive), or negative, which will result in a severe deviation from the published curves!
People running these in triode configuration are missing the point. Reduce gain with a lower plate resistor and/or reduce screen voltage. That way, bandwidth is not affected and still good beyond 100MHz. Some 'trioderised' models i have seen have trouble at 16kHz =]
1ZH37b - a useful tube as an RF or audio mixer. The 2x G1 inputs are NOT differential, which so many people seem to have missed. There IS a small amount of coupling between them (<1pF), so not really well isolated at RF, but fine for AF. Parallel the 2 grids for twice the transconductance, or feed audio to both, but vary the DC voltage to one only for variable gain.
The general 'rule' for all of these pentodes (apart from the 1P24b) is to keep grid-cathode AC voltage below about 1Vpk and you cannot go wrong.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.