No surprises there, it's the way lead batteries sound.
Twenty+ years ago both my solid state and tube moving coil stepup devices were powered from Ni-MH batteries. Loved the speed and resolution. Much better than even my AN-6 silver wire transformer.
More recently, having read about the virtues of A123 lithium, i powered one of my SS preamps with +/-18v of A123 power in lieu of an R-core with LT4320 rectification and huge nos Cerafines.
The result was not a very pleasant sound. Lightning fast but too light and much too bright. The addition of large electrolytics and a couple of Lundahl chokes didn't succeed in making it more enjoyable.
Not sure if my perception and expectations have changed, but introducing battery power today will only be possible after a complete revoicing of the system.
Nonsense.
With the last three posts we tuned thread course to subjective "I hear it" and "you can´t prove me wrong" type.
Waste of time and an offense to reason.
The famous DIY Audio curse
Many times replacing the typical "tube" power supply with batteries or even a better DC solution will produce a different sound. This is measurable with decent FFT software and a PC audio card.
Often extreme effort is applied in amp design and construction to reduce or eliminate line frequency hum because it is audible and doesn't belong in the music.
One must also look at the IMD products created when line frequency energy intermodulates with the audio energy. Often one can see line frequency products on either side of a 1 KHz tone (940 Hz and 1060 Hz or even 820, 880, 1120 and 1180 Hz for a 60 Hz line) in an FFT display.
These products may be at a low enough level to be considered inaudible, but when present on every musical tone in a complex piece of music these products contribute to listener fatigue, or the "veil" that audiophiles often speak about. They do not affect the "blackness" of the background since they are not present without a signal.
This is why every DHT amp I have built uses DC heating. you can "null" the hum, but not the odd order IMD products of that hum.
Often extreme effort is applied in amp design and construction to reduce or eliminate line frequency hum because it is audible and doesn't belong in the music.
One must also look at the IMD products created when line frequency energy intermodulates with the audio energy. Often one can see line frequency products on either side of a 1 KHz tone (940 Hz and 1060 Hz or even 820, 880, 1120 and 1180 Hz for a 60 Hz line) in an FFT display.
These products may be at a low enough level to be considered inaudible, but when present on every musical tone in a complex piece of music these products contribute to listener fatigue, or the "veil" that audiophiles often speak about. They do not affect the "blackness" of the background since they are not present without a signal.
This is why every DHT amp I have built uses DC heating. you can "null" the hum, but not the odd order IMD products of that hum.
I am using a regulator off of a voltage tripler setup to deliver 115v dc to 4 6922 triodes. Ripple noise is below the noise generated by the triodes.
If you think you can hear it, can you hear it with a true "blind test". If not, I will call confirmation bias on you. 🙂 (Saying you are sure you would hear it in a true blind test does not count!)
You can also see the effects of IMD on the B+ of a regulator without sufficient gain in the error amplifier.
I guess the maxim should be: "Regulate when you have to, not when you can."
Ps batteries also have dynamic (frequency dependent) impedance.
I wasn't really aware of this. Can you point me to some more details there?
If we are looking at unregulated, regulated and battery supplies, what are the relevant factors to consider in which scenarios?
Which varies with the chemistry and state of charge.
It can be crudely measured by resistively loading the battery with the expected current then forcing AC across the battery from a low impedance source like the output of a SS amp, coupled to the battery through a suitable cap and 1 ohm resistor in series. Measure the drop across the resistor to calculate the dynamic resistance of the battery at that frequency.
Regulators are also capable of generating harmonic energy since they are essentially a feedback system and the loop gain falls off with frequency.
Their output impedance also increases with frequency. Sometimes there is a graph of this in the chip's data sheet.
A preamp is a class A system whose current drain should be constant for all normal operating conditions. It should be relatively immune to minor variations in power supply impedance.
I tend to use open loop mosfet followers for screen grid supplies on tube amps to avoid a possibly non linear feedback system attached to the grid of a tube that can draw huge spikes of screen current when driven hard.
No, not 90,000 mAh.
45 cells 3.6 volt-2000 mAh connected series will give you one battery of
162 volt 2000 mAh.
Just a few of my opinions:
Look up Gary Pimm and his various battery powered amplifiers.
All DC power supplies have an impedance.
It may vary versus frequency.
It may vary versus load current.
If the load is single ended (not class A push pull) and there is no capacitor across the DC battery supply, then the varying single ended signal load current may cause the battery impedance to vary as the signal current demands increase and decrease (non-linear impedance versus signal demands).
Sound?
The sound of one hand clapping in a double blindfold test.
Look up Gary Pimm and his various battery powered amplifiers.
All DC power supplies have an impedance.
It may vary versus frequency.
It may vary versus load current.
If the load is single ended (not class A push pull) and there is no capacitor across the DC battery supply, then the varying single ended signal load current may cause the battery impedance to vary as the signal current demands increase and decrease (non-linear impedance versus signal demands).
Sound?
The sound of one hand clapping in a double blindfold test.
Just to make sure I understand what you are saying:
All DC power supplies have an impedance.
It may vary versus frequency.
It may vary versus load current.
If the load is single ended (not class A push pull) and there is no capacitor across the DC battery supply, then the varying single ended signal load current may cause the battery impedance to vary as the signal current demands increase and decrease (non-linear impedance versus signal demands).
... is valid for any power supply and any battery and any combination, right?
How much such an issue manifests itself depends on the design and implementation, but none are free from it.
Jan
All DC power supplies have an impedance.
It may vary versus frequency.
It may vary versus load current.
If the load is single ended (not class A push pull) and there is no capacitor across the DC battery supply, then the varying single ended signal load current may cause the battery impedance to vary as the signal current demands increase and decrease (non-linear impedance versus signal demands).
... is valid for any power supply and any battery and any combination, right?
How much such an issue manifests itself depends on the design and implementation, but none are free from it.
Jan
jan.didden,
Yes, true.
Any power supply has those issues.
A wire has both DCR, and inductive reactance, no matter how small.
A capacitor has both ESR and capacitive reactance, no matter how small, or how large, respectively.
A choke has DCR.
A solid state diode has resistance.
A tube rectifier has resistance.
A power transformer has DCR (both the secondary; and do not forget the DCR of the primary that has the DCR power loss passed to the secondary, which changes the regulation under load).
A power transformer has non-linearity too.
Batteries have various internal resistance and reactance.
The AC power mains has DCR, non-linearity (mostly 3rd harmonic distortion of the mains fundamental frequency, and and series inductive reactance. Noise is also present, and filament AC supplies contain the noise and the 3rd harmonic of the fundamental.
Please do not loose any sleep over those items above.
A good design pays attention to those factors, and deals with them appropriately, starting with the most glaring factors, in order to make a good working amplifier.
Just listen and Enjoy the Music.
"All power systems are equal, just some are more equal than others."
That is a very bad mis-quote from the novel "Animal Farm" by George Orwell.
Yes, true.
Any power supply has those issues.
A wire has both DCR, and inductive reactance, no matter how small.
A capacitor has both ESR and capacitive reactance, no matter how small, or how large, respectively.
A choke has DCR.
A solid state diode has resistance.
A tube rectifier has resistance.
A power transformer has DCR (both the secondary; and do not forget the DCR of the primary that has the DCR power loss passed to the secondary, which changes the regulation under load).
A power transformer has non-linearity too.
Batteries have various internal resistance and reactance.
The AC power mains has DCR, non-linearity (mostly 3rd harmonic distortion of the mains fundamental frequency, and and series inductive reactance. Noise is also present, and filament AC supplies contain the noise and the 3rd harmonic of the fundamental.
Please do not loose any sleep over those items above.
A good design pays attention to those factors, and deals with them appropriately, starting with the most glaring factors, in order to make a good working amplifier.
Just listen and Enjoy the Music.
"All power systems are equal, just some are more equal than others."
That is a very bad mis-quote from the novel "Animal Farm" by George Orwell.
Last edited:
I don't understand any of this long debate.
I use old UPS for HT power supplies run through proper VR for the screen supply and a swinging choke for the main HT when off the toroid.
There is NO measurable difference whatsoever in the output when running off the 12V-460V convertor than running off the internal (oversized) toroid supply.
A US 115V invertor yields 150-160V DC, and the EU version makes an additional 320-325V, so once chained give the screen supply of 325V and the final HT of 485V peak when not under load.
They are designed to run easy 200-250VA, even on the smaller ones, so valve amps are peanuts compared with say running our IT systems when sudden power interruptions take place.
Why would anyone would ever want to run anything off series connected batteries, when you get ultra cheap invertors running off car battery supplies - 12V DC which then supply the DC heater chains also?
I use old UPS for HT power supplies run through proper VR for the screen supply and a swinging choke for the main HT when off the toroid.
There is NO measurable difference whatsoever in the output when running off the 12V-460V convertor than running off the internal (oversized) toroid supply.
A US 115V invertor yields 150-160V DC, and the EU version makes an additional 320-325V, so once chained give the screen supply of 325V and the final HT of 485V peak when not under load.
They are designed to run easy 200-250VA, even on the smaller ones, so valve amps are peanuts compared with say running our IT systems when sudden power interruptions take place.
Why would anyone would ever want to run anything off series connected batteries, when you get ultra cheap invertors running off car battery supplies - 12V DC which then supply the DC heater chains also?
Attachments
Last edited:
Any Engineering problem has at least 100 solutions, of which at least 3 will work . . .
. . . If they are implemented properly.
Pick wisely, and then pay full attention to the details.
Remember the Original Tacoma Narrows Bridge?
. . . If they are implemented properly.
Pick wisely, and then pay full attention to the details.
Remember the Original Tacoma Narrows Bridge?
I am thinking along these lines for preamp or phono-pre. They have 36V versions as well. I was thinking especially for the phono preamps to use the battery for series heaters as well. Should be exceedingly quiet. And 12AY7s and ECC88s should be able to work reasonably well at 36V B+ especially if local FB is used.
Yeah those are 9AH too, fully managed and protected. Radio builders of the 1920's would be amazed such a small battery packs 60v and 9AH.
https://www.dewalt.com/products/acc...ax-flexvolt-oilresistant-90ah-battery/dcb609g
Some site used to have noise measurements for different battery chemistries but google isn't being too helpful. Anyone else recall it?
