If i try to remove the output coil on my NAD3020 power amplifier, could it means that the amp will blow?
I know that is is there to secure stability in a NFB amp, but what does it actually do? Is it for extreame cases and special speaker cables? All speakers do have a coil af input for the woofer.
Some amplifiers does not have this coil....Pass, Hiraga Class A, etc.....why?
I know that is is there to secure stability in a NFB amp, but what does it actually do? Is it for extreame cases and special speaker cables? All speakers do have a coil af input for the woofer.
Some amplifiers does not have this coil....Pass, Hiraga Class A, etc.....why?
Hi,
just follow me for a moment.
NAD sell very competitively priced amps.
The designer decided his amp did not need a Thiel network on the output.
The sales department said the amp would sell better if a Thiel network was added and would leave more margin for profit by allowing NAD to sell the amp at a higher price.
I think not.
just follow me for a moment.
NAD sell very competitively priced amps.
The designer decided his amp did not need a Thiel network on the output.
The sales department said the amp would sell better if a Thiel network was added and would leave more margin for profit by allowing NAD to sell the amp at a higher price.
I think not.
The way I see it, in a consumer market amp, they have to cater for all kinds of events, such as the user having 10 metre cables and such nonsense.
DIY or enthusiast amps such as Pass's amps know that the user is going to be (fairly) sensible and not do such things.
Whether there's any audible difference between the two is another matter. I usually add them "just in case".
DIY or enthusiast amps such as Pass's amps know that the user is going to be (fairly) sensible and not do such things.
Whether there's any audible difference between the two is another matter. I usually add them "just in case".
In the real world, this coil rejects the extra phase shifting caused by adding components to the output such as complex reactive loads and long strands of wire. Having said this I digress to Mike B in post 35 of this thread: http://www.diyaudio.com/forums/showthread.php?threadid=83834&perpage=10&pagenumber=4
and also rt7085 for post 46.
and also rt7085 for post 46.
Coils are expensive components. If the manufacturer of an amp was absolutely certain that the amp couldn't oscillate under any imaginable load condition, it would skip the coil. Serious manufacturers are probably cautious, though, so the cases where the coil is really needed may be rare, but on the other hand, they may not. It depends on the particular amp.
I know that they don't put unnessasary components inside, but my theory was that they try to prevent all sorts of errors and such like.
An example is the many capacitors in equipment to be able to conform with CE regulations.
I have 2x2 meters Van Den Hull Clearwater speaker cables, do not need short circuit protection and do not insert and/or remove signal cables with power on.
I understand the reason for the coil (by the way it has no parallel resistor) is to prevent unstability with large capacitive loads....from long capasitive cables.
My first thought was to remove it to try the sonics without it. It no big difference, the put it back. But if the amp burns during testing, this would be a bad idea...
An example is the many capacitors in equipment to be able to conform with CE regulations.
I have 2x2 meters Van Den Hull Clearwater speaker cables, do not need short circuit protection and do not insert and/or remove signal cables with power on.
I understand the reason for the coil (by the way it has no parallel resistor) is to prevent unstability with large capacitive loads....from long capasitive cables.
My first thought was to remove it to try the sonics without it. It no big difference, the put it back. But if the amp burns during testing, this would be a bad idea...
That coil isn't of high inductance but will have a great effect on RF or RF entering the feedback loop, but is very low impeadence and little effect, if any, to audio frequencies. Now if you insert an iron core inside it to increase the inductance, you have a different situation....a lowpass filter in the audio band. As is it still acts as a filter, but also 'blocks' some of the capacitive loading from being seen by the amp at higher frequencies. If at any time, there is phase shift of any frequency of 180 degrees inside the loop, the amp will occilate. If the amp is designed to drive a specific set of speakers(without a mile of wire) it may not need an output coil. Stability can be load dependent as well. As for me, I always use an output coil because I never know what speakers it will be driving and how much wire. If I were building 10,000 amps on an assembly line, it might make a financial difference not to have it.
As is it still acts as a filter, but also 'blocks' some of the capacitive loading from being seen by the amp at higher frequencies. If at any time, there is phase shift of any frequency of 180 degrees inside the loop, the amp will occilate. If the amp is designed to drive a specific set of speakers(without a mile of wire) it may not need an output coil. Stability can be load dependent as well. As for me, I always use an output coil because I never know what speakers it will be driving and how much wire. If I were building 10,000 amps on an assembly line, it might make a financial difference not to have it. Skorpio: Maybe I missed it, but I didn't see an answer to the question: "why do you want to remove it?"
It's power consumption is virtually non-exisitent and it might protect your amp. I have seen designs where the small resistance of that coil was part of the over current sensing loop to protect the amplifier against accidental dead shorts. Most of us are quite careful with our speaker cables, but the occasional accident does happen. Given the investment in your amp and maybe more importantly, the turn around time to send it in and get it back from repair, it just seems best to leave it alone.
If you are concerened about the kinds of metal in the signal path, I suppose you could make one out of 6 nines silver, but frankly as nice as NAD equipment is, it seems extremely unlikely you could hear a difference one way or the other.
There are many possible changes that can be made to a stereo system to improve the sound quality, this one should be very far down on the list. My advice; leave it alone.
It's power consumption is virtually non-exisitent and it might protect your amp. I have seen designs where the small resistance of that coil was part of the over current sensing loop to protect the amplifier against accidental dead shorts. Most of us are quite careful with our speaker cables, but the occasional accident does happen. Given the investment in your amp and maybe more importantly, the turn around time to send it in and get it back from repair, it just seems best to leave it alone.
If you are concerened about the kinds of metal in the signal path, I suppose you could make one out of 6 nines silver, but frankly as nice as NAD equipment is, it seems extremely unlikely you could hear a difference one way or the other.
There are many possible changes that can be made to a stereo system to improve the sound quality, this one should be very far down on the list. My advice; leave it alone.
SKORPIO
---Some amplifiers does not have this coil....Pass, Hiraga Class A, etc.....why?---
Their output impedance is high. The Hiraga so-called "darlingnots" are not voltage followers but can be considered as enhanced common emitter transistors, just like Sziklai pairs can be considered as enhanced emitter followers.
I think that even a single high current output stage of the follower type without global feedback needs a damped coil (or a resistor of small value) and a zobel network. No harm at all at audio frequencies but good to avoid any problem with capacitive loads and to reject radio frequencies.
You never use capacitve loads ? Sure ? a cable not connected at the other end to a speaker is a capacitive load... which may be more dangerous to an amplifier than a much larger one.
---Some amplifiers does not have this coil....Pass, Hiraga Class A, etc.....why?---
Their output impedance is high. The Hiraga so-called "darlingnots" are not voltage followers but can be considered as enhanced common emitter transistors, just like Sziklai pairs can be considered as enhanced emitter followers.
I think that even a single high current output stage of the follower type without global feedback needs a damped coil (or a resistor of small value) and a zobel network. No harm at all at audio frequencies but good to avoid any problem with capacitive loads and to reject radio frequencies.
You never use capacitve loads ? Sure ? a cable not connected at the other end to a speaker is a capacitive load... which may be more dangerous to an amplifier than a much larger one.
We, high enders, have found that the coil changes the sound for the worse, and we have developed ways of making the amp unconditionally stable without it. (at least most of us have). This was known, decades ago, by NAIM and Threshold first (I believe), but the rest of us caught on over the years. No power amp that I make now, or in the last 15 years has output coils. This sonic abberation made by the coil was demonstrated to me, back in the late 80's by a friend who proved it to me.
The coil is necessary in many designs, because the amp will become unstable with certain capacitive loads, in general. It has to do with the pole generated by the open loop output impedance and a capacitive load adding extra phase shift in with the dominant pole in the input or driver stage. Sometimes the worst case capacitance is a rather small value such as 0.1uf or less, so just testing with a large cap at the output will NOT necessarily bring out the problem.
The coil is necessary in many designs, because the amp will become unstable with certain capacitive loads, in general. It has to do with the pole generated by the open loop output impedance and a capacitive load adding extra phase shift in with the dominant pole in the input or driver stage. Sometimes the worst case capacitance is a rather small value such as 0.1uf or less, so just testing with a large cap at the output will NOT necessarily bring out the problem.
Presumably, John, you use a series output resistor instead. Of course it depends upon the type of control system being used, Naim relies on a series output R and a minimum length of prescription speaker cable. The NAD is a NFB design...do you think, John, that the NAD3020 will be noticeably improved by removing its output coil?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.