I use DHTs like 45 and 300B myself, and they indeed sound better than indirect triodes of the same power class (only EL84 is a match for DHTs in terms of transparency). The specs of DHTs look inferior, and they hum unless you use DC heaters, which adds complexity.
What is the theoretical advantage of direct heated triodes? The specs look inferior.
What is the theoretical advantage of direct heated triodes? The specs look inferior.
The 45 is probably my favorite DHT, but only 2 watts or so output. I don't consider the specs inferior (but what is your comparison and/or benchmark?).
With a good quality 45, AC filaments can better 80dB s/n referenced to 1-watt (with a good design and proper layout and parts selection). The WE300b, no way... DC is required due to it's center-tapped filament.
Complexity varies with design topology and style choices of the "designer", while adding DC filament for a DHT will add another power supply section. I've done a couple clean sheet designs for the 45/2A3 and they're not complex... quite simple actually.
As for being microphonic, I'd say it depends heavily on your specific DHTs, both brand/internal structure and condition. Having a large collection of 45s and 2A3s, most of the better ST glass versions (Sylvania being preferred) are dead quiet and near mechanically immune, sans whacking directly on the tube. I have others (includes all early Globe types, some ST versions and new EML 45 solid plates) which are extremely sensitive. Playing a few trumpet notes (real trumpet, not a recording) within 10 feet literally blew the tubes about... audible when hooked to a speaker. A characteristic specific to either a bad or poor quality tube IMHO.
As with anything else, a clean simple design that has been implemented properly (parts selection, layout, construction and tube fitting) will have good measured performance, low distortion, wide bandwidth and a fairly neutral presentation. With proper parts election, they should run consistently and reliably for decades (sans a few tubes).
Regards, KM
With a good quality 45, AC filaments can better 80dB s/n referenced to 1-watt (with a good design and proper layout and parts selection). The WE300b, no way... DC is required due to it's center-tapped filament.
Complexity varies with design topology and style choices of the "designer", while adding DC filament for a DHT will add another power supply section. I've done a couple clean sheet designs for the 45/2A3 and they're not complex... quite simple actually.
As for being microphonic, I'd say it depends heavily on your specific DHTs, both brand/internal structure and condition. Having a large collection of 45s and 2A3s, most of the better ST glass versions (Sylvania being preferred) are dead quiet and near mechanically immune, sans whacking directly on the tube. I have others (includes all early Globe types, some ST versions and new EML 45 solid plates) which are extremely sensitive. Playing a few trumpet notes (real trumpet, not a recording) within 10 feet literally blew the tubes about... audible when hooked to a speaker. A characteristic specific to either a bad or poor quality tube IMHO.
As with anything else, a clean simple design that has been implemented properly (parts selection, layout, construction and tube fitting) will have good measured performance, low distortion, wide bandwidth and a fairly neutral presentation. With proper parts election, they should run consistently and reliably for decades (sans a few tubes).
Regards, KM
If there's any scientific advantage, it (of course) will be from direct heating of cathod.
1. DHT's cathod gets order of magnitude higher currents (a few amps) than indirect ones (a few tens to hundreds mA);
2. And it gets slightly hotter than indirect triodes.
Both may act to give better emission characteristics of electrons from cathod, but I am not sure.
What else?
1. DHT's cathod gets order of magnitude higher currents (a few amps) than indirect ones (a few tens to hundreds mA);
2. And it gets slightly hotter than indirect triodes.
Both may act to give better emission characteristics of electrons from cathod, but I am not sure.
What else?
Member
Joined 2009
Paid Member
When they were designed, the best minds in science and engineering were working on them, and they are designed for the audio range.
The designers had to work with materials and (much more important in my mind) geometry to get them distortion free.
I would have thought that the filament being heated with a few volts of AC would cause problems with 60Hz artifacts getting into the audio. Various nonlinearities not cancelling out even if you manage to balance the hum out.
There's some filament tubes intended for portable battery operated radios, like the 3S4 or 3V4, that were designed for DC on the filaments, and the filament pins were labeled as to which gets the positive filament supply and which gets the negative side.
There's some filament tubes intended for portable battery operated radios, like the 3S4 or 3V4, that were designed for DC on the filaments, and the filament pins were labeled as to which gets the positive filament supply and which gets the negative side.
Nah.
Indirectly heated cathodes were not invented yet.
The DHTs were designed for RF down to even DC operation.
The higher quality tubes (although some question that) that were DHT, like the 300B, came after the thoriated tungsten cathode was invented. Tubes like the 300B were intended for *telephone* service by Bell Labs, working for AT&T which at the time had the phone service monopoly.
Screen grids naturally came after triodes were created.
I think this is pretty much correct.
_-_-bear
Last edited:
They're older, perform more poorly, and cost much, much more.
Normally, those would be disadvantages. Go figure.
Linearity of lower-gain tubes, like 300B and 6F6, isn't too bad, and I think that's because the cathode-grid distance is large relative to the wire diameters and grid spacings. These parameters yield low gain and perveance (a 300B takes, what, about the same filament power as a 6L6, but the 6L6 has higher gain and much higher peak current). The distortion is very near the theoretical value of the Child-Langmuir model, whereas tubes with "tighter" construction have more pronounced nonlinearity.
It's noteworthy that, although high gain tubes (e.g., frame grid pentodes) have higher distortion, the ratio of distortion to gain is much more favorable than these old, low gain, relatively linear tubes. If you want to make a low distortion amplifier, high gain is absolutely always the correct answer.
Tim
Normally, those would be disadvantages. Go figure.
Linearity of lower-gain tubes, like 300B and 6F6, isn't too bad, and I think that's because the cathode-grid distance is large relative to the wire diameters and grid spacings. These parameters yield low gain and perveance (a 300B takes, what, about the same filament power as a 6L6, but the 6L6 has higher gain and much higher peak current). The distortion is very near the theoretical value of the Child-Langmuir model, whereas tubes with "tighter" construction have more pronounced nonlinearity.
It's noteworthy that, although high gain tubes (e.g., frame grid pentodes) have higher distortion, the ratio of distortion to gain is much more favorable than these old, low gain, relatively linear tubes. If you want to make a low distortion amplifier, high gain is absolutely always the correct answer.
Tim
DHTs were originally meant to be powered by lead acid batteries, and therefore designed to be DC powered. You're right, AC powering generates distortion even if hum is cancelled out. I have a switch on my amp to switch between AC and DC, and the AC is slightly warmer while the DC is slightly cleaner, but I'm not sure I would be able to tell the difference in a blind test. It's subtle.
A DHT like 300B to me sounds better than KT88 in both tone and detail, even though the KT88 has better specs. I wonder why this is so.
The harmonics are different, that's why.
Also the KT88 is a screen grid type tube, which makes it a pentode, and that means a different set of curves, which means a different spectrum of distortion. That and the aforementioned "microphonics" of any DHT.
Also, you probably will never want to run a pentode without feedback... which is a very different animal than ZFB.
_-_-bear
Also the KT88 is a screen grid type tube, which makes it a pentode, and that means a different set of curves, which means a different spectrum of distortion. That and the aforementioned "microphonics" of any DHT.
Also, you probably will never want to run a pentode without feedback... which is a very different animal than ZFB.
_-_-bear
No, in general they're not, compared with unipotential cathode triodes (or triode-connected pentodes). And in the case of power tubes, the drive requirements pretty much guarantee that you'll lose linearity elsewhere as well. I'll adopt Nelson Pass's term for Lowther loudspeakers- "fetish objects."
It is. My first experience with DHTs was a high powered RF amp with four V70D/8005s in parallel. Great performer on the 10 meter band.
Not true... the 45 triode was specifically designed for large output power in AC operated receivers per the original spec sheet (all manufacturers) and all specified ratings specifically state that an AC filament is used for the specs... same holds true for the 2A3 and the WE 300b, albeit WE notes bias changes for operating the filament on DC.
You do have some AC induced artifacts from the filament, but good (45) tubes and good design will minimize it. I've got sets of 45 triodes that manage upwards of 90dB s/n (ref 1-watt) with AC filaments. Granted, not all DHTs can be quiet enough with AC, but even that seems more subjective as some folks appear elated with ~1mv of output noise.
Your comments on AC vs DC via switching your amp makes no sense to me, if you can clearly articulate the difference between them, how can you claim you couldn't tell in a blind test??
Regards, KM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.