He is a great thread on Tim Mellow’s OTL.
https://www.diyaudio.com/community/threads/new-tim-mellows-otl-project.204960/
https://www.diyaudio.com/community/threads/new-tim-mellows-otl-project.204960/
I wanted to know in my initial posting if an otl tube design is a kind of transconductance amplifier as it's output impedance is usually much higher than the impedance of the speaker.
For my fullrange loudspeaker projects that would be of sonical profit.
I heard there are tubes for car Hifi working at low voltages.
Is there any project realized with them - would be ideal for beginners because technical faults won't kill.
Are there tubes up to max 60volts?
Thats the voltage barrier where below its not lethal.
Maybe something like PCC88
https://www.diyaudio.com/community/threads/has-anyone-used-6dj8-at-b-around-150v.406939/post-7545180
For my fullrange loudspeaker projects that would be of sonical profit.
I heard there are tubes for car Hifi working at low voltages.
Is there any project realized with them - would be ideal for beginners because technical faults won't kill.
Are there tubes up to max 60volts?
Thats the voltage barrier where below its not lethal.
Maybe something like PCC88
https://www.diyaudio.com/community/threads/has-anyone-used-6dj8-at-b-around-150v.406939/post-7545180
Last edited:
Agreed. Years ago I came across a bunch (25 pc.) of NOS Sylvania EL519/6KG6A tubes for comparatively low money and all of a sudden felt tempted to create a pair of OTL monoblocs, using (some of) them as finals. I'm lucky now that I didn't do it, as most OTL designs proved to be complicated, absurdly inefficient and hard to tame. Not exactly a beginner's project.
The tubes you most probably were referring to were a set developed in the mid 1950ies and intended for car radios, operating directly (without vibrators, inverters etc.) from the car battery. They were small signal AF or RF or IF tubes. No power tubes were available, as significant output power with low supply voltages would mean mean high plate currents, hence huge cathodes.
Regarding your voltage limit, please look at Burkhard Kainka's books, especially this one.
Best regards!
The 2nd half of this post:
https://www.diyaudio.com/community/...bwoofer-with-tubes.290654/page-3#post-4700824
https://www.diyaudio.com/community/...bwoofer-with-tubes.290654/page-3#post-4700824
Please do tell. If the load impedance is pretty constant, then I guess.
It depends on what you mean by “ultra”, also.
For headphones, 40 ohm into 32 ohm can be sketchy, no?
Well, speaking about the output impedance of OTL amps, my U-OTL6080 posted above measure 20 ohms (using E=U-rI, half-voltage method), making these amp usable on any conventional speaker enclosures.
T
T
I remember too , both how strange they look having magnet placed on the frontside, and also how they sounded so good placed in those old tube-TV wooden box.
ps,think that I still have couple examples in very good condition.
Bare bones, no output transformer.
Suppose you build a simple OTL amplifier . . .
Then you remove the global negative feedback, and you measure 100 Ohm output impedance;
And it is designed to output 500mA peak (0.5 Amp peak).
Then you apply 40dB of global negative feedback, now you will measure 1 Ohm output impedance.
But it still can only output 500mA peak (0.5 Amp peak).
The OTL impedance was reduced by 100 when 40dB global negative feedback was applied;
But the 40dB global negative feedback is not capable to increase the maximum output current (it still is only 0.5 Amp peak).
A sine wave of 500mA peak is 350mA rms. I = 0.35 Amps rms
Take an 8 Ohm loudspeaker, R = 8.
Power = (I rms)squared x R
Power = (0.35) squared x 8 = 0.98 Watts rms
Sounds good . . . OK it might
Efficient. . . no
Suppose you build a simple OTL amplifier . . .
Then you remove the global negative feedback, and you measure 100 Ohm output impedance;
And it is designed to output 500mA peak (0.5 Amp peak).
Then you apply 40dB of global negative feedback, now you will measure 1 Ohm output impedance.
But it still can only output 500mA peak (0.5 Amp peak).
The OTL impedance was reduced by 100 when 40dB global negative feedback was applied;
But the 40dB global negative feedback is not capable to increase the maximum output current (it still is only 0.5 Amp peak).
A sine wave of 500mA peak is 350mA rms. I = 0.35 Amps rms
Take an 8 Ohm loudspeaker, R = 8.
Power = (I rms)squared x R
Power = (0.35) squared x 8 = 0.98 Watts rms
Sounds good . . . OK it might
Efficient. . . no
One of the objections to OPTs is that you can't get SS levels of damping factor. In reality the actual damping is a factor of the native damping of the speaker and the electrical damping of the amplifier. Thus depending on the speaker a large amount of electrical damping may not be necessary. Add to that the fact that the resistance of speaker wires and connections makes the point of diminishing returns at still lower DF the "sacrifice" for using an OPT is likely less than it might seem at first glance which seems to be borne out by the many excellent sounding traditional tube amps.
For OTL amplifiers you can add gobs of feedback to lower the Zout but the real limitation is maximum output current.
I am not very familiar with headphone drivers so I am not sure how sketchy the 1.25 DF you mention would be given their rather low mass. For most dynamic loudspeakers it would be questionable unless you were intending current drive. But with low Q loudspeakers a DF of 10 might be just fine. Certainly DF of 200 is not necessary.
I have nothing against output transformers. If they are high quality, ISO-Tango, Hashimoto, Thermionics Lab, etc. Alas, the art of winding those is slowly being lost, it seems.
And those who still know the art are charging an arm and a leg for them.
For the rest, a DF > 8 is rule that has been drilled into my head long ago and I never quite questioned it.
If it’s just a matter of power transfer, then that’s good news…I just need more power.
And those who still know the art are charging an arm and a leg for them.
For the rest, a DF > 8 is rule that has been drilled into my head long ago and I never quite questioned it.
If it’s just a matter of power transfer, then that’s good news…I just need more power.
On that topic…I think gobs of feedback makes tubes sound like SS. And that’s not what I seek.
If I want low THD, I’ll go SS.
If I want those pleasant even harmonics, I’ll go tubes and no or very little feedback.
And I’m not absolute about either, sometimes I like it straight and other times on the rocks!
ElArte, I am also a mixture in a way. Sand for subwoofers and knockabout amps with tubes and highish efficiency speakers for everything else. OTL are of intellectual interest but as a practical matter my interest is limited. Perhaps as a tweeter amp in a multiamp system or possibly a current drive single driver experiment.
P.S. Correction on my above post: Obviously the DF on your headphone example was 0.8 not 1.25 🙄
P.S. Correction on my above post: Obviously the DF on your headphone example was 0.8 not 1.25 🙄
Last edited:
You all just need a SS impedance converter hybrid. Uses a current output (Hi Z) SS amplifier to minimize current drawn from a tube amplifier. You put a small sampling resistor in the ground return of the tube ampllfier output to sample load current. The SS amplifier tries to minimize the current draw on the tube amplifier by monitoring the resistor V. The two amplifiers have their outputs paralleled, but since the Hi-Z SS amplifier is Hi-Z output it does not affect the V output of the tube amplifier. The SS amplifier just carves out most of the current draw. Could mount the SS hybrid in the speaker cabinet and never know it's not really a 5K impedance speaker. Tubecad has gone over this numerous times. It's an old idea actually.
Last edited:
There are also piezo "xfmrs" that can transform impedance. But I'm not aware of any that are sufficiently broadband.
You could also just use a field coil type speaker with a ferrite magnetic yoke and DC on the voice coil. Audio goes to the field coil then. The voice coil will run hot.
You could also just use a field coil type speaker with a ferrite magnetic yoke and DC on the voice coil. Audio goes to the field coil then. The voice coil will run hot.
Last edited: