To all tube experts... can you name tube amplifier and preamplifier designs that feature NO capacitors or chokes (an output transformer is a choke to me) in their signal paths, except those in the power supply... Curious... thanks
Capacitor-less & choke-less tube amplifiers and preamps?
- Thread starter ack
- Start date
They can still leak some DC which will be amplified which is why Dan probably took that stance.
Actually it was one tube pre-amplifier in particular that was a popular combination with Dan's amps back then. And no, it had nothing to do with DC leakage; it had to do with the outright failure of the output coupling caps. So you had all the DC you wanted and Dan's amplifier were rated DC to infinity
You can figure out the rest I'm sure 
Futterman and other OTL amps use a servo circuit at the output to suppress D.C. I designed some preamp and power amp circuits that were d.c.-coupled but it did not improve the sound enough (if at all) to offset the decreased reliability (one glitch in the servo circuit and boom!)
A lot of tube preamps exhibited very high LF energy at the outputs as they warmed up. My old Audio Research SP3a1 does, I know (from painful experience).
A lot of tube preamps exhibited very high LF energy at the outputs as they warmed up. My old Audio Research SP3a1 does, I know (from painful experience).
All in all, no transformers, no caps in the signal path, no chokes is a very dangerous proposition.
There are not such tube preamps or amps that I am aware of.
Our amps and preamps have direct-coupled inputs and outputs FWIW. We have a patent on how the preamp does its direct-coupled outputs. Its safe with Krells too....
When you have direct coupling from input to output you have a problem with the power supply. This is sort of a design engineering rule FWIW... anyway the problem is that the power supply should have Low Frequency bandwidth that goes lower than that of the circuit that is using it. When you direct-couple, the timing constant of the circuit is lower than that of the power supply. At that point it is possible to modulate the power supply with noise from the amplifier. This creates intermodulations in the amplifier- not a desirable quality. So it is good engineering practice to limit the LF cutoff of the circuit to something higher than that of the power supply. For this reason we limit the low end of our amps and preamps to 1 or 2Hz (full power) depending on the amp or preamp in question. That way we don't modulate the supplies and we also have no phase shift in the audio passband (10x the cutoff frequency being the rule of thumb) so we can do 20Hz with no squarewave tilt- something unusual in tube circuitry in general. By eliminating the phase shift you get more apparent impact, even though the response curve at 20Hz and up is the same as another circuit with less LF bandwidth.
Sorry if this got overly technical...
Our amps and preamps have direct-coupled inputs and outputs FWIW. We have a patent on how the preamp does its direct-coupled outputs. Its safe with Krells too
....When you have direct coupling from input to output you have a problem with the power supply. This is sort of a design engineering rule FWIW... anyway the problem is that the power supply should have Low Frequency bandwidth that goes lower than that of the circuit that is using it. When you direct-couple, the timing constant of the circuit is lower than that of the power supply. At that point it is possible to modulate the power supply with noise from the amplifier. This creates intermodulations in the amplifier- not a desirable quality. So it is good engineering practice to limit the LF cutoff of the circuit to something higher than that of the power supply. For this reason we limit the low end of our amps and preamps to 1 or 2Hz (full power) depending on the amp or preamp in question. That way we don't modulate the supplies and we also have no phase shift in the audio passband (10x the cutoff frequency being the rule of thumb) so we can do 20Hz with no squarewave tilt- something unusual in tube circuitry in general. By eliminating the phase shift you get more apparent impact, even though the response curve at 20Hz and up is the same as another circuit with less LF bandwidth.
Sorry if this got overly technical...
I think your only chance is an OTL from the likes of GRAAF, which I believe also have no output capacitors...however, there might be caps between stages.
Otherwise, you need to look at a hybrid amp. I have a NAT Symbiosis SE, which is pretty unusual in the hifi world. It is single ended and hybrid. It uses tubes for gain and driver and a SINGLE big industrial MOSFET for the output stage. Naturally it runs Class A, has no feedback and also is completely direct coupled from input to output. Generates a ton of heat, has huge heat sinks, especially when you consider that it is one huge heatsink for one transistor and puts out 100 watts into 8 or 4 ohms (power supply switches not transformer taps as there are none!).
So, it is basically a Single ended OTL where the output tube has been replaced by a beastly industrial MOSFET.
Otherwise, you need to look at a hybrid amp. I have a NAT Symbiosis SE, which is pretty unusual in the hifi world. It is single ended and hybrid. It uses tubes for gain and driver and a SINGLE big industrial MOSFET for the output stage. Naturally it runs Class A, has no feedback and also is completely direct coupled from input to output. Generates a ton of heat, has huge heat sinks, especially when you consider that it is one huge heatsink for one transistor and puts out 100 watts into 8 or 4 ohms (power supply switches not transformer taps as there are none!).
So, it is basically a Single ended OTL where the output tube has been replaced by a beastly industrial MOSFET.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |