oh! your post seem to be interesting but i can't understand . Can you write it in way that is easy to understand ?
perhaps I shouldn't believe in degen resistor too much .
It is a 2- face-coin !
thanks !
perhaps I shouldn't believe in degen resistor too much .
It is a 2- face-coin !
thanks !
Thahn
You are trying to correct all the amp's problems just by changing the resistors.
You have to linearise all the stages one by one and make sure there is enought gain for at least 40dB feedback @20Khz for a really good high feedback amp.
You are trying to correct all the amp's problems just by changing the resistors.
You have to linearise all the stages one by one and make sure there is enought gain for at least 40dB feedback @20Khz for a really good high feedback amp.
Hi Thanh,
Think this way. If you degen a local stage you are taking global FB and applying it locally to the stage. So if another stagew is creating the bulk of the distortion it will be less rediced by FB.
Degen is good but controlled. Work out your effect as Re = 26/Ie in mA. 26 ohms will halve your gain given low Rs.
You will decrease NFB by the same amount so expect THD to rise.
Cheers,
Greg
Think this way. If you degen a local stage you are taking global FB and applying it locally to the stage. So if another stagew is creating the bulk of the distortion it will be less rediced by FB.
Degen is good but controlled. Work out your effect as Re = 26/Ie in mA. 26 ohms will halve your gain given low Rs.
You will decrease NFB by the same amount so expect THD to rise.
Cheers,
Greg
Do you means "reduce by NFB" ?
Generally , I haven't been able to reduce TIM/THD by adding degen resistor yet .
Degen resistor often is uesed because there is no complete similar
I haven't succeed in buiding a amp with current mirror actually yet . Have you listened the sound of TDA7294 ? many diyer like it . Probadly it is noisier than hi-end amp . Its sound is not cleaner than hi-end amp but what about another characteristic ?
they said that sound of DA7294 is quite sweet . TDA7294 also use current mirror
Hi thanh !
In which way have you not been successfull in making an amp with
currentmirror ? What kind of problems did you experience ?
I tried some and never had problems.
I also tried to reduce THD/TIM in sims by degeneration, but i had
always the "NFB cures everything" effect.
I have collected some ideas what to change to get amp better
sounding, but never was able to reproduce any errors in sims.
Also, i did not observe explicit IMD in sims yet, only the unavoidable
resultant IMD caused by THD. Can anyone show me how to do that ?
Mike
In which way have you not been successfull in making an amp with
currentmirror ? What kind of problems did you experience ?
I tried some and never had problems.
I also tried to reduce THD/TIM in sims by degeneration, but i had
always the "NFB cures everything" effect.
I have collected some ideas what to change to get amp better
sounding, but never was able to reproduce any errors in sims.
Also, i did not observe explicit IMD in sims yet, only the unavoidable
resultant IMD caused by THD. Can anyone show me how to do that ?
Mike
Hi Jorge !
Yes, i am familiar with this, THD on complex signals results in IMD,
that's why i was looking for "explicit" IMD, but it seems not to exist ?
But, then why is there a TIM-measuring ? Is it the type of distortion
that creates different amounts of TIM ?
I was told that IMD is the reason for bright sounding and can be
caused by not flat OLBW, but i can't reproduce in sims.
(In realworld, yes, but is it really IMD ?)
Increasing OLBW by degeneration only result in less NFB, causing
IMD to rise. (In realworld again, bright sound goes away)
Mike
Yes, i am familiar with this, THD on complex signals results in IMD,
that's why i was looking for "explicit" IMD, but it seems not to exist ?
But, then why is there a TIM-measuring ? Is it the type of distortion
that creates different amounts of TIM ?
I was told that IMD is the reason for bright sounding and can be
caused by not flat OLBW, but i can't reproduce in sims.
(In realworld, yes, but is it really IMD ?)
Increasing OLBW by degeneration only result in less NFB, causing
IMD to rise. (In realworld again, bright sound goes away)
Mike
The purpose of degeneration is not to increase OLBW, per se, but to better the linearity of the input to larger differential voltages which can cause the stage to overload and slew. However, as this reduces the OL gain of the first stage, it reduces the available loop gain available for global feedback, as stated in previous posts. Any change in degeneration will involve necessary changes in the VAS compensation capacitor to maintain the GBP of the amplifier and the slew rate, as well as adjustments in the input stage current. Just changing the degeneration resistors is not a one-variable phenomenon, so you just can't change these resistors alone without making adjustments elsewhere, to compare what degeneration alone will do. They are all interelated.
Hello, Mike
TIM is just a distortion like any other - IMD/THD but generated when one drives the output - so it's related to output interface distortion.
The main difference between output interface distortion and TIM is that TIM is slew rate induced.
Otala, many yrs ago, at the time amps were designed for numbers, found some amps had TIM.
A well designed am will not show any measurable TIM - that's why it's not specified by sellers!
To design a good amp in my opinion one has to take all factors in account - stage by stage linearization, output stage bias, slew rate, feedback factor @20kHz, phase margin, PSRR and, just maybe, OLBW...
TIM is just a distortion like any other - IMD/THD but generated when one drives the output - so it's related to output interface distortion.
The main difference between output interface distortion and TIM is that TIM is slew rate induced.
Otala, many yrs ago, at the time amps were designed for numbers, found some amps had TIM.
A well designed am will not show any measurable TIM - that's why it's not specified by sellers!
To design a good amp in my opinion one has to take all factors in account - stage by stage linearization, output stage bias, slew rate, feedback factor @20kHz, phase margin, PSRR and, just maybe, OLBW...
Hmm, i thought TIM is TotalInterModulation ?
Is TIM only measurable, or can it be reproduced in sims ?
Mike
Is TIM only measurable, or can it be reproduced in sims ?
Mike
TIM is Transient InterModulation. As I stated, very much related to slew rate.
Any amp with a fast enough slew rate (say, 1V/us per output peak volt) will not present TIM.
And re slew rate, once the minimum is reached, there will be no more improvements.
You may find this link interesting:
http://home.online.no/~tsandstr/OtalaStory.htm
Any amp with a fast enough slew rate (say, 1V/us per output peak volt) will not present TIM.
And re slew rate, once the minimum is reached, there will be no more improvements.
You may find this link interesting:
http://home.online.no/~tsandstr/OtalaStory.htm
Thanks, damn, i always tend to shuffle abbreviations... 
PIM TIM IMD THD LEAD LAG VAS LTP CFP CFB EF CCS NFB GFB SR SNR CMRR PSRR ... grr...
Mike
PIM TIM IMD THD LEAD LAG VAS LTP CFP CFB EF CCS NFB GFB SR SNR CMRR PSRR ... grr...
Mike
V1 is 1.2V in sim
but it is about 1V and not stable in the real circuit .If I adjust VR to increase V1 , V1 will be more than 1.2V or be less than 1.2V for 5 minutes
Thanks pooge ! I also think that Degen resistor is a goog method to reduce TIM/THD . I dont' know another VAS except common emitter . It is always low
P/s: The Admin ! Can you correct my thread "I thought I mastered TIM " .Thank you !
Attachments
I'm sorry ! a mistake
I don't think that Degen resistor is a good method to reduce TIM/THD 😀 .
Leach said that degen resistor can help decreasing TIM
but in the sim the more degen resistor is big , the more TIM is produced even with amp which has use current mirror at input stage
I use Darlinton VAS for asym amp
and use it for sym amp http://www.diyaudio.com/forums/attachment.php?s=&postid=697310&stamp=1123231580
I don't think that Degen resistor is a good method to reduce TIM/THD 😀 .
Leach said that degen resistor can help decreasing TIM
but in the sim the more degen resistor is big , the more TIM is produced even with amp which has use current mirror at input stage
I use Darlinton VAS for asym amp
and use it for sym amp http://www.diyaudio.com/forums/attachment.php?s=&postid=697310&stamp=1123231580
Attachments
Hi Thanh,
I think you are confusing TIM which is really SID or slewing induced distortion which is THD generated as a particular stage reaches it's high frequency signal handling limit. This occurs mostly in the Vas when it runs out of current to charge/discharge it's Ccomp or at the input diff'l pair as they reach their undegenerated 52mV .
Adding small emitter degen of < 26/I mA can linearise this stage prior to overload without losing gain significantly. An alternative way to reduce diff'l voltage at the emitters is to increase the gain of the stage by a CM.
Hope that helps.
😉
Greg
I think you are confusing TIM which is really SID or slewing induced distortion which is THD generated as a particular stage reaches it's high frequency signal handling limit. This occurs mostly in the Vas when it runs out of current to charge/discharge it's Ccomp or at the input diff'l pair as they reach their undegenerated 52mV .
Adding small emitter degen of < 26/I mA can linearise this stage prior to overload without losing gain significantly. An alternative way to reduce diff'l voltage at the emitters is to increase the gain of the stage by a CM.
Hope that helps.
😉
Greg
Thanks ! I will think of your comments
Orcad always say that Degen resistor increase TIM/THD although Leach is right 🙂
Degen resistor decrease CMRR
Orcad always say that Degen resistor increase TIM/THD although Leach is right 🙂
Degen resistor decrease CMRR
- Status
- Not open for further replies.