Picture of one Mono-amp
Hi,
until now i build diferent versions of JLH.
The best one is created with Vishay s102k Resistors, matched 13003`s of ON, Mundorf silver-oil cap as coupling capacitor, Blackgate and Elna Cerafine Cpacitors in Cap-Mult, other transistors matched, too.
Thers the need to say, that JLH is sensitive to parts quality.
With all experiences i and friends of me made with other amps from Pass, Gainclone, Hiraga...i can tell, that this amp is hardly to beat.
The Pic is the amp we build with a friend.
Ralf
Hi,
until now i build diferent versions of JLH.
The best one is created with Vishay s102k Resistors, matched 13003`s of ON, Mundorf silver-oil cap as coupling capacitor, Blackgate and Elna Cerafine Cpacitors in Cap-Mult, other transistors matched, too.
Thers the need to say, that JLH is sensitive to parts quality.
With all experiences i and friends of me made with other amps from Pass, Gainclone, Hiraga...i can tell, that this amp is hardly to beat.
The Pic is the amp we build with a friend.
Ralf
Grey,
You commented with skepticism on a SE Class AB.
Not so silly as it seems.
The JLH is, by one viewpoint, a Single Ended Push Pull. That is, one output device controls voltage at the load, while the other device supplies current on the other half cycle.
If you can imagine a Zen configuration with the current amp device on top, the current sink below, then if the current amp controls voltage (and current delivery on the positive half cycle), and the CCS is made variable (on load demand for negative half cycles), and both are configured with a current metering system such that neither output device ever turns off, then you have a Class AB doing Class A tricks, bit like a gasoline engine doing diesel tricks (cf. direct injection.)
If this can be done, you have all the benefits of SE with the efficiency of Class AB, providing you can maintain linearity of the current amplifier, which is tricky - but possible.
SciFi, huh? Would you mind emailing me with details of your nom de plume/titles - I love science fiction.
Cheers,
Hugh
You commented with skepticism on a SE Class AB.
Not so silly as it seems.
The JLH is, by one viewpoint, a Single Ended Push Pull. That is, one output device controls voltage at the load, while the other device supplies current on the other half cycle.
If you can imagine a Zen configuration with the current amp device on top, the current sink below, then if the current amp controls voltage (and current delivery on the positive half cycle), and the CCS is made variable (on load demand for negative half cycles), and both are configured with a current metering system such that neither output device ever turns off, then you have a Class AB doing Class A tricks, bit like a gasoline engine doing diesel tricks (cf. direct injection.)
If this can be done, you have all the benefits of SE with the efficiency of Class AB, providing you can maintain linearity of the current amplifier, which is tricky - but possible.
SciFi, huh? Would you mind emailing me with details of your nom de plume/titles - I love science fiction.
Cheers,
Hugh
Jumbo Shrimp
I don't know if I can call something that is actually push pull single ended. If you have a follower biased by a common source (or common emitter) current source (or sink), it's push pull. For example a quasi-complementary amp, which I never heard refered to as single ended. I wonder if most of the confusion is caused by marketing buzzwords of Single Ended to try and capitalize on the success of single end tube amps. Sort of like when every one started calling just
about any bias current "Class A"
I don't know if I can call something that is actually push pull single ended. If you have a follower biased by a common source (or common emitter) current source (or sink), it's push pull. For example a quasi-complementary amp, which I never heard refered to as single ended. I wonder if most of the confusion is caused by marketing buzzwords of Single Ended to try and capitalize on the success of single end tube amps. Sort of like when every one started calling just
about any bias current "Class A"
Fred,
Good point. It certainly is push pull, but there's more........
If we have a follower atop a CCS, then voltage throughout the entire half cycle is controlled by the follower, which also controls current delivery on the positive half cycle.
On the negative half cycle, the current is clearly supplied by the CCS, which blindly extracts current from the follower, or, if the voltage falls below zero point, from the load (I assume here that the earthy end of the load is connected to ground, which corresponds to AC crossover point). It is clear the follower is dominating the voltages throughout, and that we can assume the sonics are similarly dominated by this circuit element; the CCS simply supplies a current draw through its voltage-indeterminate collector (or drain).
In the sense that a SE amp uses just one element to control voltage at the load throughout the entire cycle, this mode, technically called push pull, is apeing SE. The upper device, the follower, definitely controls current delivery to the load, but even though the lower device extracts current from the load on negative half cycle, the voltage at which this takes place is solely controlled by the follower. This raises some interesting possibilities on sliding bias, which certainly NP has investigated, but I'm pretty sure there's more in this idea......
Here is a most interesting article on this concept implemented back in 1960 by an Australian EE Professor named Murray:
Murray 1960 SEPP Amplifier
The concept is quite tricky to understand, but the website gives solid information on this topology.
Thanks Wim de Haan for this great information!
Cheers,
Hugh
Good point. It certainly is push pull, but there's more........
If we have a follower atop a CCS, then voltage throughout the entire half cycle is controlled by the follower, which also controls current delivery on the positive half cycle.
On the negative half cycle, the current is clearly supplied by the CCS, which blindly extracts current from the follower, or, if the voltage falls below zero point, from the load (I assume here that the earthy end of the load is connected to ground, which corresponds to AC crossover point). It is clear the follower is dominating the voltages throughout, and that we can assume the sonics are similarly dominated by this circuit element; the CCS simply supplies a current draw through its voltage-indeterminate collector (or drain).
In the sense that a SE amp uses just one element to control voltage at the load throughout the entire cycle, this mode, technically called push pull, is apeing SE. The upper device, the follower, definitely controls current delivery to the load, but even though the lower device extracts current from the load on negative half cycle, the voltage at which this takes place is solely controlled by the follower. This raises some interesting possibilities on sliding bias, which certainly NP has investigated, but I'm pretty sure there's more in this idea......
Here is a most interesting article on this concept implemented back in 1960 by an Australian EE Professor named Murray:
Murray 1960 SEPP Amplifier
The concept is quite tricky to understand, but the website gives solid information on this topology.
Thanks Wim de Haan for this great information!
Cheers,
Hugh
Flying blind
"On the negative half cycle, the current is clearly supplied by the CCS, which blindly extracts current from the follower"
On the postive cycles the emitter follower provides enough current to supply the current demanded by the current source (a constant value) plus the current though the speaker (vout divided by the speaker impedance) On the negative cycle the emitter follower provides enough current to supply the current demanded by the current source minus the current through the speaker. When the voltage goes negative enough current from the loudspeaker becomes greater than the current from the emitter follower and the amp goes in to clipping. The fact that speaker gets all it's current from the emitter follower (except during clipping) is because the emitter follower has a very low impedance compared to the current source.
'Here is a most interesting article on this concept implemented back in 1960 by an Australian EE Professor named Murray:"
This amp is not single ended and resembles a quasi-complimentary output stage amp.
"On the negative half cycle, the current is clearly supplied by the CCS, which blindly extracts current from the follower"
On the postive cycles the emitter follower provides enough current to supply the current demanded by the current source (a constant value) plus the current though the speaker (vout divided by the speaker impedance) On the negative cycle the emitter follower provides enough current to supply the current demanded by the current source minus the current through the speaker. When the voltage goes negative enough current from the loudspeaker becomes greater than the current from the emitter follower and the amp goes in to clipping. The fact that speaker gets all it's current from the emitter follower (except during clipping) is because the emitter follower has a very low impedance compared to the current source.
'Here is a most interesting article on this concept implemented back in 1960 by an Australian EE Professor named Murray:"
This amp is not single ended and resembles a quasi-complimentary output stage amp.
Re: Flying blind
I think this thread already said that..
http://www.diyaudio.com/forums/showthread.php?postid=290721#post290721
Chris
Fred Dieckmann said:
I think this thread already said that..
http://www.diyaudio.com/forums/showthread.php?postid=290721#post290721
Chris
Fred,
We are not in disagreement; the amp is certainly not single ended, for that would strictly require one device handling the entire cycle of the signal.
However, the voltage supplied to the load is dominated by just one device, the follower, while the CCS (until it clips, granted) merely draws up the rear, supplying current set by voltage/load impedance. Since this one device determines solely the voltage at the load, and since this is in fact a characteristic of SE, a feverish mind has dubbed it 'SEPP'.
We could argue about semantics here, and you'd be right, it is semantics, but is this not potentially a non-switching Class AB, thus qualifying as a gasoline engine doing diesel engine tricks?
As such, and I don't argue the semantics, only the mode of operation, this is an extremely interesting topology.
Cheers,
Hugh
We are not in disagreement; the amp is certainly not single ended, for that would strictly require one device handling the entire cycle of the signal.
However, the voltage supplied to the load is dominated by just one device, the follower, while the CCS (until it clips, granted) merely draws up the rear, supplying current set by voltage/load impedance. Since this one device determines solely the voltage at the load, and since this is in fact a characteristic of SE, a feverish mind has dubbed it 'SEPP'.
We could argue about semantics here, and you'd be right, it is semantics, but is this not potentially a non-switching Class AB, thus qualifying as a gasoline engine doing diesel engine tricks?
As such, and I don't argue the semantics, only the mode of operation, this is an extremely interesting topology.
Cheers,
Hugh
Just to inform, i love when experienced people disagree
Yes, please, do not stop....must convince to stop... this way i will be having a very special class.
Your expertise will appear when you make the defense of some idea.
I wich someday i can challenge you related to something, i will learn a lot. If not will be bad, will show me that's nothing more to learn.
Please, go on.
Carlos
Yes, please, do not stop....must convince to stop... this way i will be having a very special class.
Your expertise will appear when you make the defense of some idea.
I wich someday i can challenge you related to something, i will learn a lot. If not will be bad, will show me that's nothing more to learn.
Please, go on.
Carlos
Hugh,
For all the pixels spilled betwixt Fred and myself, he seems to be looking at this this the way I do.
In theory, Nelson's Alephs go AB if you push them hard enough...but do you really want to do so? The whole point of a single-ended circuit is that it carries all the signal, all the time (assuming "normal" operation), and as such is class A. To get to AB operation, you are by definition going to some sort of push-pull...by definition no longer single-ended.
Semantics, sales lingo, etc. etc. etc. Don't want to derail the thread with semantic arguments. Enough other threads have fallen into that trap.
My feeling is that if you want AB, just do it honestly and design the circuit as push-pull from the beginning.
My stories are published under my own name, but the problem is that I don't believe they're available down under. Nearly everything I've published has been in Analog magazine, but I'm not sure whether they even have an Australian edition, much less where the stories to fill it might come from.
jewilson,
As far as I know, single-ended predates anything that we would call a modern recording industry. Some of the oldest tube circuits are single-ended.
Grey
For all the pixels spilled betwixt Fred and myself, he seems to be looking at this this the way I do.
In theory, Nelson's Alephs go AB if you push them hard enough...but do you really want to do so? The whole point of a single-ended circuit is that it carries all the signal, all the time (assuming "normal" operation), and as such is class A. To get to AB operation, you are by definition going to some sort of push-pull...by definition no longer single-ended.
Semantics, sales lingo, etc. etc. etc. Don't want to derail the thread with semantic arguments. Enough other threads have fallen into that trap.
My feeling is that if you want AB, just do it honestly and design the circuit as push-pull from the beginning.
My stories are published under my own name, but the problem is that I don't believe they're available down under. Nearly everything I've published has been in Analog magazine, but I'm not sure whether they even have an Australian edition, much less where the stories to fill it might come from.
jewilson,
As far as I know, single-ended predates anything that we would call a modern recording industry. Some of the oldest tube circuits are single-ended.
Grey
Being that there is no formal definition of single-ended
Class A, we have quite a bit of wiggle room.
Usually I define single-ended in an output stage as
having either the plus or the minus device driven by
the input signal and not the other. This allows a fairly
broad definition, including resistors, inductors, constant
current sources, varying current sources, and combinations
of these.
In this regard, the Murray circuit would not be single-ended
in the output stage because both devices see input signal.
Class A, we have quite a bit of wiggle room.
Usually I define single-ended in an output stage as
having either the plus or the minus device driven by
the input signal and not the other. This allows a fairly
broad definition, including resistors, inductors, constant
current sources, varying current sources, and combinations
of these.
In this regard, the Murray circuit would not be single-ended
in the output stage because both devices see input signal.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.