I was using a signal generator to listen to my amp's freq range and determine where the bass seems to roll off. It uses a capacitor coupled output, not a transformer. I am looking to get a few Hz lower rolloff point. I noticed that when down around 5Hz, inaudible, I know, I could see the coronas on the power tubes flicker in time to the cycles, but there was no subsonic movement in the speakers, at the output. Does this mean the rolloff is in the output coupling caps, but that the power tubes are seeing this low frequency? Also, can these ELFs damage the output stage? It seemed that they were making the amp work extra hard, and I think I even noticed a little plate glow, and thought I smelled ozone. Is it possible to produce ozone? I have built a Tesla coil and know well what it smells like. How could a power tube produce this?
When you speak of “coronas”, do you mean the blue glow on the inside of the glass? That is voltage dependent, and a very low frequency is the voltage moving up and down at a few Hz. The eye can pick it up if it’s slow enough. You can see this flickering to the music if you drive the amp hard enough - especially if the power supply voltage sags significantly on each kick drum hit.
Short answer, probably. As asked, the value of the cap, but also the Z of the speaker? With that, you can easily find Fc.
If it's not at least 10mF or so, it'll be surprising to see 5Hz in a speaker... (based on 4R speakers and -3 of ~4Hz).
http://sim.okawa-denshi.jp/en/CRhikeisan.htm
If it's not at least 10mF or so, it'll be surprising to see 5Hz in a speaker... (based on 4R speakers and -3 of ~4Hz).
http://sim.okawa-denshi.jp/en/CRhikeisan.htm
At 5 hz, the power tubes will be mostly dealing with a lack of inductance in the transformer and will be driving a fairly low impedance load.
The flickering of the blue glow would be normal in this case.
The flickering of the blue glow would be normal in this case.
The real question is: The signal generator or the amp has a cap coupled output? I assumed the amp was cap coupled for some reason.
1. What is the DC resistance of your speaker?
That is almost certainly also the impedance of the speaker at 5Hz.
2. Is this an OTL amplifier (no output transformer, just an output capacitor)?
If there is not enough capacitance, the output tubes voltage swing may be "from rail to rail".
At some low frequency, the capacitive reactance is very high impedance, and so the output tubes will not be loaded.
That is almost certainly also the impedance of the speaker at 5Hz.
2. Is this an OTL amplifier (no output transformer, just an output capacitor)?
If there is not enough capacitance, the output tubes voltage swing may be "from rail to rail".
At some low frequency, the capacitive reactance is very high impedance, and so the output tubes will not be loaded.
If you are transformer coupling and running below the low frequency cutoff, the load impedance will be very low because of the insufficient inductance. You’ll get low power (little cone movement) and the current in the output tubes will shoot up. For short periods of time (seconds, not minutes or hours) this won’t hurt anything. If you’re cap coupling, load impedance goes high below cutoff. The tube current will be very low at low frequency - or at least not go higher than quiescent. Plate dissipation will be safe, but screen current can go through the roof unless your output stage is triode or triode-strapped (which limits how low the plate voltage gets pulled).
Frequency response for a tube power amplifier should be measured at 1 Watt output.
Wrong.
Nobody listens at 1w.
Good tube amplifiers make full power at 5 hz. Problem is most tube amplifiers arent that good. The people who say fr should be measured at 1w are trying to hide or obscure the fact that it cant make full power bandwidth.
The 1w test is rigged to make it harder to tell the good ones from crap.
I listen to reggae.
Punk rock.
Classical.
1w output isn't going to tell me anything useful.
Making full power at 30Hz is a good real-world compromise, as is making full power at 15k.
“Frequency response” should be measured at a sufficiently low level not to encounter power bandwidth limitations. Power bandwidth is a separate measurement. Both are necessary to fully characterize an amplifier.
“Frequency response” should be measured at a sufficiently low level not to encounter power bandwidth limitations. Power bandwidth is a separate measurement. Both are necessary to fully characterize an amplifier.
On what planet?
This one.
I've got at least 2 examples which I can prove make full rated power at 5 hz.
Waste of power. Of course.
We're dealing with vacuum tubes, they waste gobs of power. Nevertheless it can be done, has been done, and is the standard by which the best amplifiers compete.
This one.
I've got at least 2 examples which I can prove make full rated power at 5 hz.
Waste of power. Of course.
We're dealing with vacuum tubes, they waste gobs of power. Nevertheless it can be done, has been done, and is the standard by which the best amplifiers compete.
I must admit I did not design nor build them myself, but I've verified 42 watts at 5 hz from my MCINTOSH Mc30s
My 50 watt rated Acrosound Ultralinear 2s also deliver right at 50 watts, at 5 cycles under half a percent.
They wont do full power at 4 hz.
They do it at 5.
Not many can pull that off.
Sure is fun finding out that information though.
My 50 watt rated Acrosound Ultralinear 2s also deliver right at 50 watts, at 5 cycles under half a percent.
They wont do full power at 4 hz.
They do it at 5.
Not many can pull that off.
Sure is fun finding out that information though.
I have seen this on my SVT. It is visually evident on the big hits, like you said. Its also cool to watch!