In general, there is not a one to one correspondence. Most audio amps have a capacitance that dominates the frequency response. Slew rate limiting occurs for large signals, because the current available to charge and discharge this cap is limited. Thus the name slew rate limiting. The small signal response is limited by the same cap, but for small signals the response is controlled by an RC time constant.
Slew Rate Vs Frequency Response
sorry, I'm wrong on typing "colleration", its Correlation.
so what is the correlation (equation?)?
Mr. Self says that SR 2.1V/us can produce 20kHz.
http://www.dself.dsl.pipex.com/ampins/webbop/opamp.htm
Mr. Elliott on his Project68 500W PA have a SR about 3V/us and can produce ~30kHz (not at full power)
How about LM324 that is only have 0,4V/us SR?
is it can only produce less than 4kHz?
sorry, I'm wrong on typing "colleration", its Correlation.
so what is the correlation (equation?)?
Mr. Self says that SR 2.1V/us can produce 20kHz.
http://www.dself.dsl.pipex.com/ampins/webbop/opamp.htm
Mr. Elliott on his Project68 500W PA have a SR about 3V/us and can produce ~30kHz (not at full power)
How about LM324 that is only have 0,4V/us SR?
is it can only produce less than 4kHz?
The conventional formula used for large signal slew rate limited response is F = Sr/2*pi*Vp. So for a slew rate of 7 V/uS (7*10 ^6 V/S) and voltage peak of 38 V the maximum frequency before triangulation occurs is around 29.3 KHz. Note that outright triangulation of a sine wave represents a rather severe state of affairs, and many prefer a slew rate that does not allow an amp to even come close to this condition.
I think you mean full-power bandwidth (rather than frequency response).
If you have a sinewave at an amplitude of A [volts] and frequency f [Hz], then its highest slew rate will be where it crosses zero volts and will be
2*pi*f*A [Volts/sec].
Divide by 1,000,000 to get it in Volts/usec.
So, if you have a 100Watt power amplifier (40volts peak into 8 ohms), then for it to do an undistorted full scale sinewave of 40V peak at 20kHz, it needs to have a slew rate of at least
2*pi*20,000*40/1,000,000 = 5.03 V/usec
If you have a sinewave at an amplitude of A [volts] and frequency f [Hz], then its highest slew rate will be where it crosses zero volts and will be
2*pi*f*A [Volts/sec].
Divide by 1,000,000 to get it in Volts/usec.
So, if you have a 100Watt power amplifier (40volts peak into 8 ohms), then for it to do an undistorted full scale sinewave of 40V peak at 20kHz, it needs to have a slew rate of at least
2*pi*20,000*40/1,000,000 = 5.03 V/usec
kroto said:So, is anyone know here the colleration (equation?) about Slew Rate and Frequency response that is produced?
Sorry, not necessarily any correlation, if speaking in small signal terms.
large signal SR =Iq/Cc
where Cc=miller cap at VAS and Iq= total collector current in half of long tail pair. ie max current to charge Cc.
small signal fbw =SR/(2*PI*Vp)
small signal limit, solve for Vp given Fbw=closed loop BW and SR
Full power BW
fFPBW is defined as the highest frequency at which the op amp can put out a sinewave with a peak voltage equal to Vclip. It is given by fFPBW =SR/(2*PI*Vp) where Vp= Vcliping
reference http://users.ece.gatech.edu/~mleach/ece4435/tutorial.pdf
where Cc=miller cap at VAS and Iq= total collector current in half of long tail pair. ie max current to charge Cc.
small signal fbw =SR/(2*PI*Vp)
small signal limit, solve for Vp given Fbw=closed loop BW and SR
Full power BW
fFPBW is defined as the highest frequency at which the op amp can put out a sinewave with a peak voltage equal to Vclip. It is given by fFPBW =SR/(2*PI*Vp) where Vp= Vcliping
reference http://users.ece.gatech.edu/~mleach/ece4435/tutorial.pdf
The correct measure for square-wave:
http://www.falco-systems.com/high_voltage_amplifiers.html
- Rise Time must be (us), is the voltage peak to peak in (10%-90%).
What I would like to know what is valid to specify.......with or without filters( input RC and coil zobel ) ?
http://www.falco-systems.com/high_voltage_amplifiers.html
- Rise Time must be (us), is the voltage peak to peak in (10%-90%).
What I would like to know what is valid to specify.......with or without filters( input RC and coil zobel ) ?
What I would like to know what is valid to specify.......with or without filters( input RC and coil zobel ) ?
usually for both with (given if they are reasonable BW)
Re: Re: Slew Rate Vs Frequency response
@bwaslo: yes I mean fFPBW.
if fmax=SR/(2*p*Vp), then fmax will go higher if the Vp in small value. what will hapen in LM324 case? distorted?
@PMA: what is that mean? in small signal like OpAmp?
@infinia: if SR=Iq/Cc what is the term (unit)? Ampere/Farad = V/us?
thx folks for the attention.
sorry for my dumb question, just a newbie.
bwaslo said:I think you mean full-power bandwidth (rather than frequency response).
If you have a sinewave at an amplitude of A [volts] and frequency f [Hz], then its highest slew rate will be where it crosses zero volts and will be
2*pi*f*A [Volts/sec].
Divide by 1,000,000 to get it in Volts/usec.
So, if you have a 100Watt power amplifier (40volts peak into 8 ohms), then for it to do an undistorted full scale sinewave of 40V peak at 20kHz, it needs to have a slew rate of at least
2*pi*20,000*40/1,000,000 = 5.03 V/usec
PMA said:
Sorry, not necessarily any correlation, if speaking in small signal terms.
@bwaslo: yes I mean fFPBW.
if fmax=SR/(2*p*Vp), then fmax will go higher if the Vp in small value. what will hapen in LM324 case? distorted?
@PMA: what is that mean? in small signal like OpAmp?
@infinia: if SR=Iq/Cc what is the term (unit)? Ampere/Farad = V/us?
thx folks for the attention.
sorry for my dumb question, just a newbie.
Re: Re: Re: Slew Rate Vs Frequency response
Derived from the classic equation I = C dV/dT , then relating SR to dV/dT, units will be Volts/seconds using Amps and Farads.
Most will usually measure SR (ie the slope on the O-scope), unless you design. V/usec is the most common way to express SR.
@infinia: if SR=Iq/Cc what is the term (unit)? Ampere/Farad = V/us?
Derived from the classic equation I = C dV/dT , then relating SR to dV/dT, units will be Volts/seconds using Amps and Farads.
Most will usually measure SR (ie the slope on the O-scope), unless you design. V/usec is the most common way to express SR.
PMA,
Don´t be sorry. There`s a direct and fundamental correlation, speaking in any terms. High slew rate is not possible without high bandwidth and vice versa.Sorry, not necessarily any correlation, if speaking in small signal terms.
PMA,
Don´t be sorry. There`s a direct and fundamental correlation, speaking in any terms. High slew rate is not possible without high bandwidth and vice versa.
You probably do not know what you speak about. AD797 is a good example, it has GBW = 110MHz (at gain 1000) at only 20V/us slew rate. It would give HF small amplitudes well, until slew rate limit is reached. For sine wave
SR = max. dv/dt = 2 x Vp x pi x F
Vp ... peak voltage of a sine wave
F ... frequency
pi ... 3.14159.....
You probably do not know what you speak about. AD797 is a good example, it has GBW = 110MHz (at gain 1000) at only 20V/us slew rate. It would give HF small amplitudes well, until slew rate limit is reached. For sine wave
SR = max. dv/dt = 2 x Vp x pi x F
Vp ... peak voltage of a sine wave
F ... frequency
pi ... 3.14159.....
so to continue using your example
solving for Vp with Av =1000 gives Fbw~280KHz from AD797 OLG graph
using SR=20V/us which is typically worst case
Vp=11.4V or close to the rails or close enough for low distortion
Conclusion the maximum signal levels are not slew rate limited in this case
In fact this is an optimum design choice, when SR and small signal BW are essentially equivalent, that is when Fbw = full power BW
Last edited:
PMA,
110MHz at gain 1000... not bad at all... even if it is only 8MHz at gain 10, since bandwidth goes up as gain goes up...dependable information provided in fancy, glossy brochures by sales department white-collar workers.
Once more: slew rate is not about gain and frequency, it is about time and voltage!
Slew rate
That's correct but we would also like to know what are the max frequencies and levels are, so we don't trigger slew rate limits ie non-linear distortion (bad) and to keep the system within small signal parameters (good). That is why preintegration filters exist on ADCs and other systems that use digital signals (square waves) feeding analog circuits.
The equations presented above are setting the limits of such, thus the relationship exists in the minds of some engineers. Read the Leach link for further understanding.
Once more: slew rate is not about gain and frequency, it is about time and voltage!
That's correct but we would also like to know what are the max frequencies and levels are, so we don't trigger slew rate limits ie non-linear distortion (bad) and to keep the system within small signal parameters (good). That is why preintegration filters exist on ADCs and other systems that use digital signals (square waves) feeding analog circuits.
The equations presented above are setting the limits of such, thus the relationship exists in the minds of some engineers. Read the Leach link for further understanding.
Last edited:
Wavebourn,
Right. (Have I stated otherwise?)Once more: slew rate is not about gain and frequency, it is about time and voltage!
Thank you, for further understanding I graduated TIASUR in 1981, have a degree of Master of Electronics Engineering; specialty is Design and Manufacturing of Radio and Electronics Equipment.That's correct but we would also like to know what are the max frequencies and levels are, so we don't trigger slew rate limits ie non-linear distortion (bad) and to keep the system within small signal parameters (good). That is why preintegration filters exist on ADCs and other systems that use digital signals (square waves) feeding analog circuits.
The equations presented above are setting the limits of such, thus the relationship exists in the minds of some engineers. Read the Leach link for further understanding.
Last edited:
LOL My mistake WaveB, I thought you disagreed in some way. I guess misunderstanding happens when you ambiguously repeat the obvious esp w/o any acknowledgment of the previous information presented.
How high should SR be if the previous signal source is BW limited. CD's have FIR and anti alias filters. Pro preamps/poweramps have RF filters. Preamps rarely need to put out more than 2Vp-p. Maybe SR is all about specmanship and bragging rights? Square wave testing is only used for large signal testing ie stabilty, over/under shoot, and SR varification.
How high should SR be if the previous signal source is BW limited. CD's have FIR and anti alias filters. Pro preamps/poweramps have RF filters. Preamps rarely need to put out more than 2Vp-p. Maybe SR is all about specmanship and bragging rights? Square wave testing is only used for large signal testing ie stabilty, over/under shoot, and SR varification.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Slew Rate Vs Frequency response