Slew Rate

[kongwee]

In layman's terms, does Class D have (dis)advantages with respect to slew rate?

No sure about slew rate, but Class D easily reach Mhz switching in MOSFET. Of course, the bandwith is already limited by output filer already mention.

 

[Cbdb2]

Now, since the power content of recorded music decrease a lot with frequency, an amplifier could be actually works fine even with not sufficient slew rate for 'full power-full frequency operation' (but not completely true for what said by DonH56 about bandwidth limitations).

You don't need full 20khz power to get slew limiting. Look at the freq. spectrum of a square wave, the fastest signal, most of the power is in the fundamental.

 

[SirPaulGerman]

I am late to this topic

I am comparing the Slew rate of a Hafler Pro5000 to other Hafler amps, and noticed that the slew rate is low

Hafler pro5000 (1991 $1500 ) yes more expensive than the XL-600
SLEW RATE: 1 kHz, 12OV p-p square wave: 40 volts
per microsecond.

Hafler Transnova 9505 ( 1991 $2200 )
150V/us

Hafler XL-600 ( 1989 $1100 )
SLEW RATE: 10 kHz, 130 volts peak-to-peak squarewave: 100 V/us


Why is the pro5000 so low compared to the other amps ?
Is that a sign of low performance ?

 

[DonH56]

I am late to this topic

I am comparing the Slew rate of a Hafler Pro5000 to other Hafler amps, and noticed that the slew rate is low

Hafler pro5000 (1991 $1500 ) yes more expensive than the XL-600
SLEW RATE: 1 kHz, 12OV p-p square wave: 40 volts
per microsecond.

Hafler Transnova 9505 ( 1991 $2200 )
150V/us

Hafler XL-600 ( 1989 $1100 )
SLEW RATE: 10 kHz, 130 volts peak-to-peak squarewave: 100 V/us


Why is the pro5000 so low compared to the other amps ?
Is that a sign of low performance ?

Slew rate is one of many design variables and the number is just one choice to make. It is impacted by other design issues so does not by itself indicate quality of performance (high or low). Higher slew rate requires higher bandwidth, which means more power, more noise, and potentially greater sensitivity to reactive loads (reduced stability). Higher bandwidth and thus higher slew rate may produce lower distortion (e.g. if higher loop gain and thus more feedback is implemented in the design). High or low slew rate is not necessarily an indicator of audio performance; noise and distortion are better metrics IMO. When I am looking at amplifiers I rarely (or barely) glance at the slew rate (if it is even specified).

A 120 Vpp sine wave at 20 kHz requires 7.54 V/us (2 * pi * 20 kHz * 60 Vpk) which is pretty much all you need for audio. Note 40 V/us is sufficient to produce a full-scale 106 kHz sine wave, much higher than we can hear. As an amp targeting professional applications, the designers may have chosen to limit the slew rate to something reasonable for achieving low enough distortion whilst ensuring greater stability (lower sensitivity to speaker loads) under demanding (professional sound reinforcement) applications.

HTH - Don

 

[SirPaulGerman]

What do you make of the pro5000 for home audio ?
Very expensive amp back in the day
Can you check the specs and topology?

Thanks for your help

 

[DonH56]

What do you make of the pro5000 for home audio ?
Very expensive amp back in the day
Can you check the specs and topology?

Thanks for your help

Sorry, I have practically no experience with the Hafler Pro amps, they came out after I was working for a store that sold them, and by 1991 I had a new baby boy to keep me busy along with a demanding new job. It has decent specs for the time, but also has a fan for cooling, that may be an issue for home use (if it activates in a quiet room).

This is a technical thread so I suggest you create a new thread to ask about the Hafler amps. A lot of folk here have experience with vintage gear and can help you out. My experience with Hafler is mostly with their earlier gear. I met David a few times but IIRC by 1991 Hafler Corp had been bought by another company (that I do not recall). Hard to believe David Hafler passed away 20+ years ago...

 

[SirPaulGerman]

The company was Rockford fosgate , and the designer was the late Jim Strickland

 

[egellings]

I am late to this topic

I am comparing the Slew rate of a Hafler Pro5000 to other Hafler amps, and noticed that the slew rate is low

Hafler pro5000 (1991 $1500 ) yes more expensive than the XL-600
SLEW RATE: 1 kHz, 12OV p-p square wave: 40 volts
per microsecond.

Hafler Transnova 9505 ( 1991 $2200 )
150V/us

Hafler XL-600 ( 1989 $1100 )
SLEW RATE: 10 kHz, 130 volts peak-to-peak squarewave: 100 V/us


Why is the pro5000 so low compared to the other amps ?
Is that a sign of low performance ?

The lowest of those rates would not have a deleterious effect on audio quality. The higher rates simply are not needed for living room audio settings.

 

[SirPaulGerman]

I noticed that the Yamaha p2200 40/us and the Crown ps-400 16/us very low.
Do studio amps don’t care about higher rates ?

 

[DonH56]

I noticed that the Yamaha p2200 40/us and the Crown ps-400 16/us very low.
Do studio amps don’t care about higher rates ?

They don't need higher slew rates. Professional amps care about reliability, not impressing marketing with numbers that do not matter in the real world. Higher slew rates require higher bandwidth which usually means higher noise, and that can be a problem for sound reinforcement systems that tend to use horn-loaded compression drivers that are more sensitive than many home speakers. The audience is not interested in hearing hiss. For consumers, higher slew rates well beyond that needed for full-power 20 kHz output is not a plus unless it supports some other metric like lower distortion, noise, etc. For consumers, in terms of audibility, all a high slew rate does is provide a bigger number for marketing and puts more tweeters at risk.

 

[Blumlein 88]

I noticed that the Yamaha p2200 40/us and the Crown ps-400 16/us very low.
Do studio amps don’t care about higher rates ?

You could always get a Spectral. 1 mhz bandwidth, and slew rate of 1000 volts/microsecond.

 

[Blumlein 88]

They don't need higher slew rates. Professional amps care about reliability, not impressing marketing with numbers that do not matter in the real world. Higher slew rates require higher bandwidth which usually means higher noise, and that can be a problem for sound reinforcement systems that tend to use horn-loaded compression drivers that are more sensitive than many home speakers. The audience is not interested in hearing hiss. For consumers, higher slew rates well beyond that needed for full-power 20 kHz output is not a plus unless it supports some other metric like lower distortion, noise, etc. For consumers, in terms of audibility, all a high slew rate does is provide a bigger number for marketing and puts more tweeters at risk.

Get a hold of yourself man. Sounds like you are claiming enough slew rate is enough or some odd idea like that.

 

[restorer-john]

For consumers, in terms of audibility, all a high slew rate does is provide a bigger number for marketing and puts more tweeters at risk.

Wise words.

The only time I ever instantly blew a pair of tweeters was with a 400V/uS power amplifier. I clipped it on a benign transient at a high level. About 30 years ago.

A direct distortion cancelling, DC-daylight, semi feed forward throw-everything-at-the-wall big Denon.

 

[ahofer]

You could always get a Spectral. 1 mhz bandwidth, and slew rate of 1000 volts/microsecond.

The Spectral I had hissed audibly. The salespeople were struck with sudden deafness when I pointed it out.

 

[restorer-john]

The salespeople were struck with sudden deafness when I pointed it out.

They had probably blown their HF hearing...

 

[Blumlein 88]

The Spectral I had hissed audibly. The salespeople were struck with sudden deafness when I pointed it out.

Kind of odd. I had a DMA 50 and it was dead silent. I knew someone with a DMA 80 with some rather efficient speakers and it also was dead silent. This was back when I my hearing was pretty extended to something like 18 khz. The Spectral amps were straight wire with gain plain and simple.

 

[voodooless]

You could always get a Spectral. 1 mhz bandwidth, and slew rate of 1000 volts/microsecond.

You could feed it a PWM signal, add some filtering, and have a Class D amp

 

[Blumlein 88]

You could feed it a PWM signal, add some filtering, and have a Class D amp

Demian Martin who designed them has posted here a few times. He said the reason for the wide bandwidth is they wanted extremely fast settling time after a signal has come and gone. I understand that is also what you want for class D. So probably would work just fine.