Interview with Dan D'Agostino

[Sir Triode]

Interesting interview with Dan D'Agostino - Founder of Krell and Momentum. 

 

He talks about his design beliefs and why he designs and builds his amps the ways he does.

 

 

[Guest yamaha_man]

I enjoyed that interview, thanks for posting.

[CraigC]

Me too, how pissed would you be at being kicked out of your own company

Edited August 17, 2013 by CraigC

[Guest yamaha_man]

True, that's so wrong!

[Guest yamaha_man]

His views on high end cables was/is interesting also.

[Addicted to music]

I just watched 1st 1/2 of this, cant wait to see the rest, gotta go to work!   many thanks to T12 for bringing this to the table

 

I remember coressponding to another SNA member here when I 1st joined SNA 24-36mth, Dan was here in Melb in the last Hifi show held in Melb promoting his Brand name amps.This SNA member discussed at great lenghts on the components that are  available and how SMD shouldnt  be used in a power amp application.

 

A great way to learn on building great amps, but then again WTF would I know when it comes to DIY amp building!...

 

All jokes a side, Dan is a legend    

Edited August 18, 2013 by pchan

[Sir Triode]

Dan actually used SMT components for his latter amps like the FPB series but as you already know, he has discovered that traditional through hole components are better not only in the reliability stakes but sonically as well - because less is more.

 

So everyone who is fawning over the latest Boulder amps with their SMT components gotta have their heads examined. Paying so much for amps that will not be easily repairable in the future is an expensive problem waiting to happen.

Edited August 18, 2013 by triode12

[metal beat]

Guys who can afford Boulder amps will never need to have their's repaired.  They will be long gone and upgraded to something else

[georgehifi]

If you read the Stereophile review on the Momentums, it seems the heatsinks are a compromise between look and functionality, as they couldn't take 1/3 power (100watts) into 8ohms for too long before going into thermal shutdown, and blowing the fuses.   

He is a design legend, along with guys like Nelson Pass, John Curl, Jean Hiraga, Peter Stein, Thorsten Loesch etc etc and I like the way they openly treat the Class D debate, he touches on it in this video, but here is a quote from him in a Stereophile discussion.

 

"In the designer's forum, Dan D'Agostino from Krell objected to class D because the waveform produced was furry and spiked. Class D has a lot of distortion that needs to be filtered out. And as with any filter some bad stuff may get through and other stuff not pass through. The best application is a subwoofer amp."

Cheers George

[metal beat]

 

If you read the Stereophile review on the Momentums, it seems the heatsinks are a compromise between look and functionality, as they couldn't take 1/3 power (100watts) into 8ohms for too long before going into thermal shutdown, and blowing the fuses.   

Cheers George

 

 

  George

 

Your statement above is incorrect. Dan answered this question when the Stereophile review was being viewed as a amp meltdown on another forum by the popcorn crowd.  http://forum.audiogon.com/cgi-bin/fr.pl?aamps&1358359891&read&keyw&zzmomentum 

I know of no owner who has had a single issue with the Momentum amps.

 

Reviewers curse I guess

 

 

The Momentum amplifiers normal operating temperature is 42-44c(107-111f)the temperature rise to 48c (118f) is inconsequential. The amplifier in question failed from a non related component failure in the power on circuit.

 

The Momentum amplifier is designed to shut down at 70c (158f). The unit under test was not close to that temperature.Many amplifiers operate temperatures at 70c(158f) and above. If John had continued his test the second amplifier would have passed without any problems.

 

The Momentum amplifier has more than enough thermal head room for any load or speaker. I was unable to give comment until the unit was returned to me from testing. Thank you for your concern.

 

Dan D'Agostino

Edited August 19, 2013 by turntable

[Addicted to music]

Oviously not enough

 

 

If you read the Stereophile review on the Momentums, it seems the heatsinks are a compromise between look and functionality, as they couldn't take 1/3 power (100watts) into 8ohms for too long before going into thermal shutdown, and blowing the fuses.   

He is a design legend, along with guys like Nelson Pass, John Curl, Jean Hiraga, Peter Stein, Thorsten Loesch etc etc and I like the way they openly treat the Class D debate, he touches on it in this video, but here is a quote from him in a Stereophile discussion.

 

"In the designer's forum, Dan D'Agostino from Krell objected to class D because the waveform produced was furry and spiked. Class D has a lot of distortion that needs to be filtered out. And as with any filter some bad stuff may get through and other stuff not pass through. The best application is a subwoofer amp."

Cheers George

 

 

 

hehehe....

 

Obviously not enough real estate to cater for residual heat George,  but what would i know about heat transfer and maintaining silicon die integrity and stability, Im just a dumb a...se who speculates!!!    

 

 

Anyway here is a shot of DIY project.  Its a class A/B 240W mosfet, note the size of the heat sink, I have and still use these to drive 1.8ohm ribbons and 2X 11 inch eaton drivers.  I built and completed this 1994 and its still in operation.  I have never had a fault or a cook output stage ever, the caps are the same era, put in 1994, still in operation to this day as we speak.  I measured the offset the other day and it still is the same as when I set Biasing back then.  With a Fluke infared it actually reaches 54 degree C on a hot day 30-33 degree C.  it reaches 35-45 when its around 25, and this is volume crack fairly high.

2ndly I prefer bulk aluminium for cooling rather than use active fans.  Because by the time the fan kicks in, it really is too late..  But then again what would I know!

 

Fully agree with all you guys on class D

[Addicted to music]

Guys who can afford Boulder amps will never need to have their's repaired.  They will be long gone and upgraded to something else

 

I need to find someone who can donate it to me.......

[Sir Triode]

Guys who can afford Boulder amps will never need to have their's repaired.  They will be long gone and upgraded to something else

 

True - more money than sense.

[Zaphod Beeblebrox]

 

2ndly I prefer bulk aluminium for cooling rather than use active fans.  Because by the time the fan kicks in, it really is too late..  But then again what would I know!

 

Fully agree with all you guys on class D

 

There is more than one way to use a fan. At least one manufacturer uses a system which constantly monitors the temperature of the heat sink/s and adjusts fan speed to ensure a constant operating temperature. With such a system the fan never 'kicks in', as it operates continuously.

[Addicted to music]

There is more than one way to use a fan. At least one manufacturer uses a system which constantly monitors the temperature of the heat sink/s and adjusts fan speed to ensure a constant operating temperature. With such a system the fan never 'kicks in', as it operates continuously.

Fans are great as a back up but only at the last resort, I dont like fans for numerous reasons,  I fully agree that fans should never kick in, but in real life it is also the environment the product is used. 

Edited August 19, 2013 by pchan

[Zaphod Beeblebrox]

Fans are great as a back up but only at the last resort, I dont like fans for numerous reasons,  I fully agree that fans should never kick in, but in real life it is also the environment the product is used. 

 

Nope. Fans are (almost) ideal. MUCH better and more efficient than convection cooling. Properly implemented, fans: 

 

* Allow MUCH less heat sink mass to be used (around 1/5th the amount of aluminium).

* Allow for amplifier orientation to be far more flexible (amplifier can be placed up, down, sideways, behind cabinets, etc).

* Allow for close temperature maintenance, independent of ambient conditions.

* Allow for hot heat sinks to be kept away from small, sensitive, fingers and hands.

* Allow for a far more compact output stage (thus reducing wiring inductance, improving stability and increasing overall performance).

* Due to less mass and a properly implemented control system, fast warm-up capability is easily implemented (around 20 mins vs. 4-5 hours for convection cooled amps).

 

 

Cheap fans (120mm) have a life-span in the order of 150,000 hours. That equates to 17 YEARS or 24/7 operation. Quality fans have a life-span of considerably more . In light of the fact that a PROPERLY engineered fan cooling system runs the fan at far less than the fan's full speed (say: nearer 20%) then the life-span will be commensurately much greater. And, regardless of how any cooling system is implemented (fan, water, convection, et al), some kind of back-up should be employed to ensure that temperatures can never exceed a pre-set level (say - 80 ~ 90 degrees C).

Edited August 19, 2013 by Zaphod Beeblebrox

[lusk]

 

If you read the Stereophile review on the Momentums, it seems the heatsinks are a compromise between look and functionality, as they couldn't take 1/3 power (100watts) into 8ohms for too long before going into thermal shutdown, and blowing the fuses.   

He is a design legend, along with guys like Nelson Pass, John Curl, Jean Hiraga, Peter Stein, Thorsten Loesch etc etc and I like the way they openly treat the Class D debate, he touches on it in this video, but here is a quote from him in a Stereophile discussion.

 

"In the designer's forum, Dan D'Agostino from Krell objected to class D because the waveform produced was furry and spiked. Class D has a lot of distortion that needs to be filtered out. And as with any filter some bad stuff may get through and other stuff not pass through. The best application is a subwoofer amp."

Cheers George

 

 

That amp tested had a component failure.

 

http://forum.audiogon.com/cgi-bin/fr.pl?aamps&1358359891&openflup&29&4#29

 

"I am sorry that you interpret this as a melt down. The Momentum amplifiers normal operating temperature is 42-44c(107-111f)the temperature rise to 48c (118f) is inconsequential. The amplifier in question failed from a non related component failure in the power on circuit.The Momentum amplifier is designed to shut down at 70c (158f). The unit under test was not close to that temperature.Many amplifiers at higher temperatures than that.If John had continued his test the second amplifier would have passed witout any problems. The Momentum amplifier has more than enough thermal head room for any load or speaker. I was unable to give comment until the unit was returned to me from testing. Thank you for your concern."

Dan D'Agostino

Edited August 19, 2013 by lusk

[Addicted to music]

ZB,

 

thanks for bringing the advantages up,

 

Im not disagree with you there but....

 

here is my experience with fans.....

 

are you aware of how many fans that are used for cooling fail on  the equipment I work on that are quoted should last the life of the machine? 

 

The equipment also requires mantenance as fans introduce outside dust that accumulated and block up.  Yuk!!!  Failure of fans are not due to complete failure, but for some, the bearings cave in and seize and become noisy, hence the client can no longer tolerate.  Most of the fans I replace are on a feedback that sends a active pulses to the system and if that signal is not there it prevents the whole m/c from working! 

Admittingly these are on 24/7 unlikely in an amp enviroment. 

Fans are an important device and I dont think many equipment can do without them

I have never measured it but how much electrical noise can a fan have on the input of an amp thats close by!

[Sir Triode]

Nope. Fans are (almost) ideal. MUCH better and more efficient than convection cooling. Properly implemented, fans: 

 

* Allow MUCH less heat sink mass to be used (around 1/5th the amount of aluminium).

* Allow for amplifier orientation to be far more flexible (amplifier can be placed up, down, sideways, behind cabinets, etc).

* Allow for close temperature maintenance, independent of ambient conditions.

* Allow for hot heat sinks to be kept away from small, sensitive, fingers and hands.

* Allow for a far more compact output stage (thus reducing wiring inductance, improving stability and increasing overall performance).

* Due to less mass and a properly implemented control system, fast warm-up capability is easily implemented (around 20 mins vs. 4-5 hours for convection cooled amps).

 

 

Cheap fans (120mm) have a life-span in the order of 150,000 hours. That equates to 17 YEARS or 24/7 operation. Quality fans have a life-span of considerably more . In light of the fact that a PROPERLY engineered fan cooling system runs the fan at far less than the fan's full speed (say: nearer 20%) then the life-span will be commensurately much greater. And, regardless of how any cooling system is implemented (fan, water, convection, et al), some kind of back-up should be employed to ensure that temperatures can never exceed a pre-set level (say - 80 ~ 90 degrees C).

 

The main problem with fans is the collection of dust in the unit either by suction or blowing - depending on the use of the fan. So use of filters and regular maintenance is a must. 

 

With oversized heatsinks, this is less of an issue. 

Edited August 19, 2013 by triode12

[metal beat]

That amp tested had a component failure.

 

http://forum.audiogon.com/cgi-bin/fr.pl?aamps&1358359891&openflup&29&4#29

 

"I am sorry that you interpret this as a melt down. The Momentum amplifiers normal operating temperature is 42-44c(107-111f)the temperature rise to 48c (118f) is inconsequential. The amplifier in question failed from a non related component failure in the power on circuit.The Momentum amplifier is designed to shut down at 70c (158f). The unit under test was not close to that temperature.Many amplifiers at higher temperatures than that.If John had continued his test the second amplifier would have passed witout any problems. The Momentum amplifier has more than enough thermal head room for any load or speaker. I was unable to give comment until the unit was returned to me from testing. Thank you for your concern."

Dan D'Agostino

 

I hear an echo

[Zaphod Beeblebrox]

ZB,

 

thanks for bringing the advantages up,

 

Im not disagree with you there but....

 

here is my experience with fans.....

 

are you aware of how many fans that are used for cooling fail on  the equipment I work on that are quoted should last the life of the machine? 

 

The equipment also requires mantenance as fans introduce outside dust that accumulated and block up.  Yuk!!!  Failure of fans are not due to complete failure, but for some, the bearings cave in and seize and become noisy, hence the client can no longer tolerate.  Most of the fans I replace are on a feedback that sends a active pulses to the system and if that signal is not there it prevents the whole m/c from working! 

Admittingly these are on 24/7 unlikely in an amp enviroment. 

Fans are an important device and I dont think many equipment can do without them

I have never measured it but how much electrical noise can a fan have on the input of an amp thats close by!

 

I don't know how many poor quality fans you've seen in equipment you deal with. OTOH, I can tell you that in 40 years of servicing audio equipment, I've had to replace less than a dozen fans that had either failed or developed noisy bearings. Computers are VERY different. Fans used in PCs are the cheapest available. I've seen many fail and/or go noisy.

 

For a PROPERLY designed amplifier, electrical fan noise is simply not an issue. Nor is mechanical noise for that matter.

 

AND, given the irresistible advantages conveyed by using PROPERLY designed fan cooling systems, convection cooling is a big drawback.

[Zaphod Beeblebrox]

The main problem with fans is the collection of dust in the unit either by suction or blowing - depending on the use of the fan. So use of filters and regular maintenance is a must. 

 

With oversized heatsinks, this is less of an issue. 

 

Quite so. None of which is under dispute. Once every year or two, filters should be cleaned. No big deal. Most manufacturers ensure that this is a quick and painless exercise.

[Addicted to music]

Quite so. None of which is under dispute. Once every year or two, filters should be cleaned. No big deal. Most manufacturers ensure that this is a quick and painless exercise.

 

hehehe...

 

I think you should call me the Fan man ZB,

mate, you are just a baby in nappies when it comes to replacing fans and blowers.... :lol:

 

When I was working for Kodak, there was a new you bewt high tech product that was released,  approx $300K and above cost to us to bring it into down under.  There was a high tech DC blower that was considered to make us techs redundant.  Within 3 months of having these on clients sites, the world supply that was there on shelf as back up had been exhausted, the next production run is 6mths away.  We had to modify the machines so it can used the old AC blowers.  I was replacing the new DC ones at a rate of 2 a day!  Mind you the cost of these blowers were $400US,  When the Aussie dollar was @ 1/2 a $US, we copped some ear blasting!

 

Manufacturers are in tune with the love of fans and blowers just like you have outline.  My requirement for an amp or any Hifi equipment is to never employ active cooling but to have it as a back up in case something falls or exposed to hash environments.           

[Addicted to music]

T12,

 

just finished the rest of it

 

F)*ing loved it

[CraigC]

When the interviewer attempted to have him define classs A operation I was interested to hear that he didn't say the devices were always biased on, he carefully chose to say there was no switching notch. Are they actually a parallel single devices in class A or differential output devices  being biased on just prior to the signal arriving?