I think industry standard for amplifier gain is 30x (happens to be 30dB). So 1V in gives 30V out. You find the gain by looking at the feedback network. Feedback resistor divided by resistor to gnd, plus 1, is the gain. Unless the amplifier is without global FB loop, then look at each stage's local gain and sum them up. Very few are without global FB. The choice in transformer rating is up to u but at least 2x output power per ch in VA. So if 100W per ch at least 400VA transformer. This is not a hard rule u can go less but expect sag on the rails at high power and u can go gigantic and cause all kinds of rectification harmonics in the neighborhood. Up to debate in the endless... Capacitance is usually chosen for max allowed ripple on the voltage rails. U live in Denmark so u have 50Hz AC coming in. This gives u 100Hz ripple after the fullbridge rectifier. That means every 10ms the caps are charged up. How much are they charged each time? Depends on how much the voltage drops between each 10ms interval. The drop depends on the capacitance and the idle current of the circuit being fed by the caps. We use the all important Ic=C×dv/dt. Lets say u have 10000uF and the idle is 100mA. The ripple is Vr= 100mA×10ms/10000uF = 0.1V. The ripple amplitude is independent on voltage so if u have 10V or 100V rails the ripple is only the current, capacitance, and the time. The amp's PSRR gives the output noise from rail ripple. Usually the capacitance chosen for ripple is enough to satisfy the rail voltage drop caused by current drawn into the speakers. But u can calculate that too. Say u have a sudden 5A draw caused by a music peak. Say the pulse lasts for 2ms. The voltage rail with 10000uF will drop 1V. (voltagedrop=5A×2ms/10000uF). If you have a small amp with 10V rails that may be a big drop but if u have 100V rails the 1V drop isnt to care about. The energy stored in capacitance is the square of the voltage × C/2. So if u double the voltage u quadruple the energy. In other words if u half the voltage u have 1/4 the energy. So u actually need more capacitance for a lower pwr amp if i want the same voltage droop to output load ratio. Notice how tube amps have capacitance in tens and hundreds while low voltage SS amp have in thousands.
Sorry Im on my phone and the tiny keyboard is getying to me.. hope this helps a little. U are sure to get tons more shortly...
Sorry Im on my phone and the tiny keyboard is getying to me.. hope this helps a little. U are sure to get tons more shortly...
Thanks a lot for both your answers!
I know that all depends on the speakers, but the truth is that I do not know wich speakers I end up with, properly some home build ons.
My hope is to build an amplifier that can handle all but the most extreme speakers, a foundation for playing
SemperFi, I am impressed over that you wrote all this from your mobile!
This circuit is without GFB so it is not so easy and I am a noob of the noobs!
So have no idea on how to detect what is what and ware.
In fact I am even not sure where the input and output are for the sound, expect that it hase something to do with We and Wy?
I know that all depends on the speakers, but the truth is that I do not know wich speakers I end up with, properly some home build ons.
My hope is to build an amplifier that can handle all but the most extreme speakers, a foundation for playing
SemperFi, I am impressed over that you wrote all this from your mobile!
This circuit is without GFB so it is not so easy and I am a noob of the noobs!
So have no idea on how to detect what is what and ware.
In fact I am even not sure where the input and output are for the sound, expect that it hase something to do with We and Wy?
The chances are the surprise will be a nice one if you can maintain the positive vibesI need some components and if he means they are better and they are of good quality, maybe the result surprises me!
If the circuit is so complicated that you cant calculate the gain, try doing a simulation.
One way of estimating the power needed is to look at the efficiency of each part.
Let's say that a class AB amp is 50% and the PSU/transformer is 60-80% efficient.
A 2x50W amp will need a 200W PSU and a 300W transformer.
Choosing the voltage and current is partly dependant on the amp and its load. If the PSU voltage is too high then the amp will overheat with 4ohm speakers or clip if the PSU Voltage is too low when used with 8ohm speakers.
One way of estimating the power needed is to look at the efficiency of each part.
Let's say that a class AB amp is 50% and the PSU/transformer is 60-80% efficient.
A 2x50W amp will need a 200W PSU and a 300W transformer.
Choosing the voltage and current is partly dependant on the amp and its load. If the PSU voltage is too high then the amp will overheat with 4ohm speakers or clip if the PSU Voltage is too low when used with 8ohm speakers.
@FredMule - it's against the Forum Rules to start multiple threads on the same topic/project, your threads have been merged.
Sorry, and thanks for telling me, I was not thinking and therefore not aware that it was the same topic. But sorry and thanks again!
The chances are the surprise will be a nice one if you can maintain the positive vibes
I think that all here are only eager to help, and do there best to answer correct!
Even different meaning / knowledge comes from a good hart so I can only be happy and enjoy the good vibes!
Sorry Chester,
Yes, my comments are based on my opinions, and my experience - we each have our own version of BS - that's diyAudio
I love Elvee's comment: the circuit is worthless crap, so articulate and helpful
Fried,
Notwithstanding some of the negative comments about the possibly over-complicated circuitry and doubtful benefits about high-quality components, if you like what you've heard and want to build a similar circuit, I can't see any drawbacks to your project - it does depend on your budget as some of these components aren't cheap and any amplifier is a balance between cost and sound quality and your own expectations
All the best in diyA ...
Yes, my comments are based on my opinions, and my experience - we each have our own version of BS - that's diyAudio
I love Elvee's comment: the circuit is worthless crap, so articulate and helpful
Fried,
Notwithstanding some of the negative comments about the possibly over-complicated circuitry and doubtful benefits about high-quality components, if you like what you've heard and want to build a similar circuit, I can't see any drawbacks to your project - it does depend on your budget as some of these components aren't cheap and any amplifier is a balance between cost and sound quality and your own expectations
All the best in diyA ...
jameshillj You are right, I did love the sound of it but if i can get better sound still, then why not!
I am not afraid of trying and I have to bye components, no matter what. So as long we are not talking 68 dollars for a single resistor ore 500 dollars for a foot of cable, but are near the earth in prices, I am game. And I can always bye a little at the time, then it do not feel so wild
So please do contribute all you feel can help,and the same to all other!
I am not afraid of trying and I have to bye components, no matter what. So as long we are not talking 68 dollars for a single resistor ore 500 dollars for a foot of cable, but are near the earth in prices, I am game. And I can always bye a little at the time, then it do not feel so wild
So please do contribute all you feel can help,and the same to all other!
Well, one other thing would be to consider getting the power transformer from a company in Poland called "Toroidy" - they are very good and excellent prices
Those resistors that I mentioned (Isabellenhutte PBH) that are available from Elfa may not be the price of the 'naked Vishays' (TC2575) but are still pretty expensive (about 37SEK ea, 27Dkk each) plus the high tax - the Caddock MP 'kool pak' series are also pretty good and not so expensive
Is there a website &/or a thread about the design of the amplifier - if so, it'll help you with building your amp and the awkward problem of sourcing the case
Those resistors that I mentioned (Isabellenhutte PBH) that are available from Elfa may not be the price of the 'naked Vishays' (TC2575) but are still pretty expensive (about 37SEK ea, 27Dkk each) plus the high tax - the Caddock MP 'kool pak' series are also pretty good and not so expensive
Is there a website &/or a thread about the design of the amplifier - if so, it'll help you with building your amp and the awkward problem of sourcing the case
Did you care to read further? For example, the "clean" supply for the noble stages is in fact half-wave rectified (and two diode drops lower than the main supply). This cannot be mistaken for a stroke a genius, it is just plain technical incompetency. PRR also gave some hints. I have no problem acknowledging geniuses when I see them (Veysel, Eva, etc.), but crap is crap, no matter how you look at it. Of course, everyone is completely free to think and do whatever he/she likes, but if you ask forum members about something, you have to be prepared for a sometimes unpleasant truth, otherwise you should only frequent "true" audiophiles forums where you are certain to receive positive opinions regardless of the technical validity of what you submit
Use the Volume knob!!
Any useful hi-fi MUST be able to bring soft passages up pretty loud. All hi-fi systems should have "too much gain". The Volume knob adjusts the gain to the desired setting, to avoid clipping, deafness, angry neighbors.
The old standard hi-fi was a phono stage which would give 0.25V on a "normal" LP (they varied widely), line stage with gain of 10, power amp with gain of 20. Full-up a normal LP would try to make 0.25V*10*20= 50 Volts at the speaker. Which is 300 Watts. But these were 30 Watt power amps!
We "never" turned the Volume knob over half-way for "normal" listening. A good volume pot at halfway cuts signal to 1/10th. That makes 5 Volts or 3 Watts at the speaker, a good loud level.
Full-up would never be used. 3/4 up only for critical listening to soft passages in gently-cut LPs.
With today's more predictable levels ("all" records are highly compressed and everything comes from a DAC with 2V output) we don't need that much gain. But we still use the Volume knob appropriately to avoid overload.
If, when you try it out, you find that you can not ever turn-up past 2 or 3 without getting TOO DARN LOUD!, then take out some gain. Gain trim is the basic audio skill. Mostly resistor choices. Getting less gain is always easier than finding more gain.
Hi FriedMule,
I'm a little reluctant to step in here, with PRR, Elvee, Eva, et al contributing thoughtfully to this, but here's one observation:
This design leaves unbalanced the Sourcing CCS, and the VAS. The transistors are a good match, 2N5551 and 2N5401, but the CCS is clamped at 2.1 mA, while the CCS for the input differential pair emitters is designed to source 13.7 mA. The result is that it will leave the zener hungry (under it's 4.7V rating), while the CCS that's supposed to pull the driver bases up is also starved.
I recommend removing the 2N5401 (component designations would be useful) whose emitter is the positive rail, the base is the CCS emitter, and the collector is the CCS base, because it clamps the sourcing CCS current at 2.1 mA -- too low.
The input diff pair and VAS are designed well, allowing twice the current out of the VAS as the common emitters of the diff pair -- works out 'just right'.
But then, the CCS can't provide enough current to balance that, 'cause it's 'hard clamped' by the base-emitter junction of the 2N5401 - that needs to be removed!
Looks like a good-sounding design, hope you have fun building and listening to it!
-Rick
I'm a little reluctant to step in here, with PRR, Elvee, Eva, et al contributing thoughtfully to this, but here's one observation:
This design leaves unbalanced the Sourcing CCS, and the VAS. The transistors are a good match, 2N5551 and 2N5401, but the CCS is clamped at 2.1 mA, while the CCS for the input differential pair emitters is designed to source 13.7 mA. The result is that it will leave the zener hungry (under it's 4.7V rating), while the CCS that's supposed to pull the driver bases up is also starved.
I recommend removing the 2N5401 (component designations would be useful) whose emitter is the positive rail, the base is the CCS emitter, and the collector is the CCS base, because it clamps the sourcing CCS current at 2.1 mA -- too low.
The input diff pair and VAS are designed well, allowing twice the current out of the VAS as the common emitters of the diff pair -- works out 'just right'.
But then, the CCS can't provide enough current to balance that, 'cause it's 'hard clamped' by the base-emitter junction of the 2N5401 - that needs to be removed!
Looks like a good-sounding design, hope you have fun building and listening to it!
-Rick
No . . sorry, I don't . . .
The CCS (the pair of transistors providing sourcing current for the VAS) is a classic-enough, and effective, design -- but it fights with the zener for control of those base voltages. It tops out at 2.1 mA, while the transistor next to the 4.7 V zener wants to send 13.7 mA to the input differential pair.
It looks to me like an 'oopsie doodle' . .
But then, considering the PSU shortcomings as mentioned by PRR and others, I sorta suspect a sales or marketing influence.
Surprisingly, it'll probably work O.K. -- but it will have a few tens-of-millivolts offset (w/the offset adjustment centered - the pot has the range to adjust it out), be slower slewing positive than negative, and the zener will be starved all the while. And, it'll vary over temperature.
While we're at it, you may get slightly better bass by increasing the value of the 100 uF Elna Silmic II in the feedback signal path -- say . . 470 uF. But having no global feedback does hamper the amp's ability to deliver 'tight' bass, even with the triple EF output.
Congrats on your side advancing in World Cup! They sure look strong to me.
- Rick
The CCS (the pair of transistors providing sourcing current for the VAS) is a classic-enough, and effective, design -- but it fights with the zener for control of those base voltages. It tops out at 2.1 mA, while the transistor next to the 4.7 V zener wants to send 13.7 mA to the input differential pair.
It looks to me like an 'oopsie doodle' . .
But then, considering the PSU shortcomings as mentioned by PRR and others, I sorta suspect a sales or marketing influence.
Surprisingly, it'll probably work O.K. -- but it will have a few tens-of-millivolts offset (w/the offset adjustment centered - the pot has the range to adjust it out), be slower slewing positive than negative, and the zener will be starved all the while. And, it'll vary over temperature.
While we're at it, you may get slightly better bass by increasing the value of the 100 uF Elna Silmic II in the feedback signal path -- say . . 470 uF. But having no global feedback does hamper the amp's ability to deliver 'tight' bass, even with the triple EF output.
Congrats on your side advancing in World Cup! They sure look strong to me.
- Rick
Rick PA Stadel your comment are soo helpful! I am wary happy for your help!
Does the offset influence the sound in any way?
About the PSU part, this circuit should have a gain of 25, right? Does the PSU not deliver enough voltage to headroom and current to persuade a 4 ohm speaker?
And thanks, yes they seems to do pretty well
Does the offset influence the sound in any way?
About the PSU part, this circuit should have a gain of 25, right? Does the PSU not deliver enough voltage to headroom and current to persuade a 4 ohm speaker?
And thanks, yes they seems to do pretty well
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.