|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.17.48.137
In Reply to: RE: PSUD Playtime posted by drlowmu on May 27, 2015 at 09:03:41
850mV of B+ ripple will give ~50mV or so of ripple out of the output transformer. Through relatively efficient speakers (90+dB), most people can hear that, probably from a few rooms away.
Follow Ups:
Given that I run 99dB speakers, for the type of music I listen to that would be a definite no go!
Regards,
91
It seems the posts below mine have been locked - I can't reply to them. So I will post here.
Jeff: "Don't be so gullible. Dennis runs GPA 604s, like you, and in 2005, Six Moons reviewer gave his amps and speakers " Best Sound At The Show", or something like that. Read the Six Moons review, By S. Harrell, on DF's web site Raymond. His supply is similar to my suggestion."
Gullible? As we have discussed before, I listen to a lot of "quiet" music at low levels and I do find amp noise distracting. I can certainly hear 4mV on the output and prefer less than 3mV (currently I run between 1 - 2mA wit AC heaters). I don't believe I would hear the hum from one of Dennis' amps from rooms away - he'd not design an amp like that - but that is not what you initially were referring to. If 850mV does translate to 50mV out of the output transformer, then that would indeed be (very) audible. I vaguely recall PJ saying that ~ 300mV PS ripple would result in an audible noise when using efficient speakers.
So you initially state that 850mV is okay then cite Dennis' circa 2005 Serious Stereo 2A3 amp as an example. Just to be clear - you are suggesting that Dennis' amp produced 850mV of power supply ripple?
Jeff: "In 2015, DF will "smoke" everyone in tube amp land at the upcoming show - I WILL predict. He has been busy doing positive things. !!!"
I am sure Dennis' system is sounding fine and some folks are going to love it. He is certainly taking the paradigm as far as it can go and is investing more effort than many (most?). I'd expect that effort to reap some benefit.
In the context of the discussion however, you appear to be citing the Serious Stereo amp as an example of 850mV ripple? Or perhaps using it to support claims for the general PS topology, though this would be off-topic unless it has ~850mV ripple. Maybe you are just taking the opportunity to promote it.
Regards,
91.
I never measured Dennis' amps, nor did I ever specifically ask him about mVAC to the Finals.I know there is some ripple, higher than average, due to his "fast" power supply with low storage Ls and Cs. He, as a designer of first order, would trade off a bit of ripple versus a supply that lags from over-filtering, as ALL "dead quiet amps" do !!!
Dennis has one customer that runs Serious Stereo amps on a pair of ALE Horns, which must be 110 dB / 1 Watt/ 1 meter, and hes one happy dude with the equipment and amp.
Then there is Lakerfan, who went through fifty different SET amps, and ended up keeping just two, Larry D. Moore's "Ten Squared" and Dennis' " Serious Stereo 2A3" monoblocks.
Ripple is not a concern, re-creating the musical experience is what matters, coming to us on time, and with full dynamics. Thats what separates the men from the boys.
Dennis' new room at RMAF 2015 will be most interesting for us to sample this October, Denver.
Jeff Medwin
Edits: 05/31/15
Okay, the reference to Dennis' amp had little to nothing to do with the question.Have you run a 2A3 amp with 850mV PS ripple into an efficient pair of speakers - I am curious as to what you heard or did not hear?
Regards,
91.
Edits: 06/01/15
If you have 850mv of ripple on a 2.5k to 8ohm output transformer there will be 48.1mv across the speaker.
2500/8=312.5 is the impedance ratio of the output transformer.
The square root of 312.5 is 17.67 and that's the voltage step down ratio of the transformer. (it's also the winding ratio)
850mv / 17.67 is 48.1mv
48.1mv will produce a lot of sound depending on the sensitivity of the speaker.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
That isn't exactly correct.
The 2.5K output is in series with the Rp of the tube so you need to use ohms law to figure out the AC across the transformer. Assuming a 2K5 load and a 750R tube it works out to:
.850V/(2500+750)= 262uA of current. 262uA of current through 2500 ohms is 654mV of ripple showing up across the transformer that divided by 17.68 nets 27mv of ripple at the speaker. (still way to much)
dave
So a quick glance at the math tells me that 100-150mV of ripple is about the limit to obtain ~ 4-5mv to the speakers?
.
Have Fun and Enjoy the Music
"Still Working the Problem"
"He, as a designer of first order, would trade off a bit of ripple versus a supply that lags from over-filtering, as ALL "dead quiet amps" do !!!"A designer of the first order would be perfectly capable of building an amplifier with vanishingly small (i.e. inaudible) hum, and without any loss of performance from the amplifier. It does require having a basic understanding of the physical principles involved in the functioning of capacitors, inductors, etc., and the nature of the load presented by the amplifier.
Maybe a bit of hum is what some people like to hear as part of their audio experience? There is nothing wrong with that, if it gives them enjoyment. But it cannot be promoted as the ultimate in sound reproduction.
Chris
Edits: 06/01/15
The 60/120Hz hum/buzz would probably cause the driver's diaphragm some excursion but make little noise through the horns . I have no doubt that 250mV ripple would cause some nasty distortion .
I spent a great deal of effort to get a dedicated amp quiet enough on my TAD TD4001 which incidentally also have 110dB sensitivity . Even 1mV of ripple at the output terminals was unbearable . 250mV would be horrendous !
Al
You are confusing mVAC at the output transformer's secondary, with the primary. The mVAC on the primary can be much higher as it is stepped down by the turns ratio of the output trannie, primary to secondary.
Also, with a DHT, there is a certain amount of mVAC from the DHT's filament when we AC excite a 2A3's Ef.
Jeff
Ok , maybe I got confused with that but what you suggest , even at the output termianls would be horrible on 110dB drivers . I aim for 1mV maximum at the output terminals with 97dB speakers . I couldn't manage that even with DC filaments with that amount of ripple . How quiet are your amps ? What about pentode SE ? What sort of maximum ripple would you suggest is allowable ?
Al
Hiah Al,Generally, I think building a supply for lowest hum is a mistake in a amplifier.
An amp that "plays the music to us in the music's time frame, and not the amp's time frame, with maximum possible dynamic contrasting", is the goal I seek and value the most in designs. This requirement and " lowest possible hum" are at odds with each other, design wise.
Playing back the music in music's time, is the one thing, for example, that a no-feedback amp can begin to give us.
Well, today I layed my main supply caps, and Rk bypass caps into my Type 45 amp chassis. I coated the chassis and the bottom of the caps heavily with GE Silicone II, and mated them together, so that , when cured, the DC LINK Vishay and WIMA caps will be elevated about 1/4 of an inch off the steel chassis. Caps sound better when they don't touch the chassis, and, when they don't touch each other.
BTW, that fellow Barry in London, "Smart845" that I wanted you and Andy Evans to visit and give me a report, has started a blog on audio. Check out his center tapped 45 Globes, a new design from EML.
http://smart845.blogspot.co.uk/
As for SE Pentode, I have no idea what kind of supply that needs. But I KNOW G2 has to be super stable and ideally - separate from the B+ supply, to do it right.
But, why would anyone want to build a Pentode output stage, and have two things going on at once, when simple triodes like the 2A3 and 45 exist ?? That is why I never messed with SE pentodes.
Regards and best wishes,
Jeff Medwin
Edits: 06/03/15 06/03/15
I don't know why you think that I go for minimum hum . I go for a power supply that provides acceptable hum and the best sonics using what iron and caps I have available in stock .
SE pentodes have far less immunity to power supply ripple so a power supply for a SE pentode needs to be approached slightly differently to that of a triode . As for the extra grid and feedback , so what ? If you have not experimented with SE pentodes , you should
I have told you and your creepy cheerleader this already . I don't have the time or the inclination to go and listen to someone elses amp who is a nightmare journey the other side of town .
Al
You missed my entire point !! No its NOT possible, there are audible trade offs, and YOU don't have a clue !!!!! You are clueless.You best stay with OTL amps Chris, and voltage regulators, CCSes, etc. !!!
Jeff Medwin
Edits: 06/01/15
"No its NOT possible, and YOU don't have a clue !!!!! You are clueless."
Maybe you could define what you would consider to be the properties of an "ideal" power supply? My definition would be one that gives a constant voltage, independent of the loading by the amplifier.
In practice, of course, one can never precisely attain the ideal goal of an exactly constant voltage. But one can get essentially arbitrarily close to it. And certainly, close enough for the rather undemanding purposes of a home stereo system.
Do you disagree with this ideal goal of a constant voltage? If so, you are then arguing that you want the power supply voltage to be modulated by the loading of the amplifier. Or by the oscillatory mains voltage.
Now, as I and others have said before, it may perhaps be that you actually *like* the sound effects that result from having the power supply modulated by the amplifier load or by the mains. That is your prerogative. All kinds of sound effects can be achieved by making the power supply voltage vary, either in time with the music or independently of it.
But if you want to make a sound-effects box, there are better and more controlled ways to achieve it than by means of a floppy or sagging or rippling power supply.
Chris
Chris,
You are beyond me having a conversation with in audio, at anything over a mid-fidelity level.
Go back to your text book approach, and live happy ever after !!
Jeff Medwin
"You are beyond me having a conversation with in audio, at anything over a mid-fidelity level."
I wasn't aware that you had any aspirations to listen to audio above a mid-fidelity level.
Chris
Don't be so gullible. Dennis runs GPA 604s, like you, and in 2005, Six Moons reviewer gave his amps and speakers " Best Sound At The Show", or something like that. Read the Six Moons review, By S. Harrell, on DF's web site Raymond. His supply is similar to my suggestion.
In 2015, DF will "smoke" everyone in tube amp land at the upcoming show - I WILL predict. He has been busy doing positive things. !!!
Jeff Medwin
Are you saying that Denis's amplifiers have 850mV of ripple in the power supply? The last comments I saw from him, IIRC, indicated that he wouldn't be OK with anything over 250mV.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: