Grey, Geoff, Sdman and John well said.
As Hugh said his design has not blown a transistor yet. It is a proven design, a great sounding amp, a great kit and that I think is what the DIY'er wants. (IMHO)
Hugh, I know that you are active on so many forums and like the gent you are you take time to answer every email and question. People who build your kits will never be stuck for an answer to a problem and you strive to improve every aspect of your product.
Phil.
As Hugh said his design has not blown a transistor yet. It is a proven design, a great sounding amp, a great kit and that I think is what the DIY'er wants. (IMHO)
Hugh, I know that you are active on so many forums and like the gent you are you take time to answer every email and question. People who build your kits will never be stuck for an answer to a problem and you strive to improve every aspect of your product.
Phil.
Hi Michel,
I can now continue this discussion with you, and others, after returning with my family from a weekend at the beach!
Thank you Michel, for your email attachment. I have it, will open and study it later tonight.
Thank you Sdman, Phil and Grey for the nice comments in support; Michel, I think Grey has a valid point, and you need to think on how your posts appear to others from your very blunt, categorical statements. Writing style is VERY important, as this is a sterile medium all too easily misunderstood. Almost anything discussed between human beings is inclined to passion, and people look for misinterpretations and the merest sniff that you are slighting others. When they perceive it, right or wrong, you are a marked man. And it is so much nicer to stay friends than to effortlessly make enemies.........
I have worked as a consultant in IT, and one thing you very quickly learn is 'NO CATEGORICAL STATEMENTS'. Truth is, someone always proves you wrong, as I may do here.
One such statement on matters technical, was this:
"SOA protection as discussed here, is totally unnecesary, indeed irrelevant in single ended, current source/sink biased output stages as designed by N. Pass for instance. "
This applies, and you might have used this proviso, ONLY if the current source lies beneath the rail and the amplifying device (AD - usually an emitter follower or CFP). If however the reverse is true, viz the AD is powered from the positive rail and the current source is in fact a sink, strapped to ground, then there is no constraint over the current the upper device may pass at all. A shorted load will pass a very large current from the rail; I know because I have designed such an amp - a Zen turned on its head, no less, and called the Glass Harmony - and while I never did blow any amplifying devices it certainly did have the potential there. Hence, SOA protection is a good thing in this context, particularly with highly reactive loads.
To resume: VI protection is two fold, as you have pointed out. One must protect against voltage and current spikes, and protection must take account of both. I agree with your premise so far; let's consider the options.
Voltage protection. This can be effected two ways. A catch diode across each output device will short any voltage spikes at the load which actually exceed the rail supply, thus protecting both output devices from dV/dt destruction. The second method I use is to use a low rail voltage in comparison with the transistor max Vceo rating. On modern bipolars (which I prefer) this means it is difficult to blow the output on voltage spikes.
Current protection. This I choose to ignore, except that current limiting is de rigeur if the emitter resistor is high and a polyphase switch on the speaker protects the bass driver. I use 0R47 emitter resistors, which limit transconductance - and hence feedback factor - of the output stage. It works very well, and gives strong local feedback which is linearising, since the ohmic voltage feedback from the emitter resistor is about the same as the non-linear junction voltage feedback, no bad thing.
If an AKSA is mischievously presented with a short circuited load, it will naturally destroy at least one output device, and sometimes both. The drivers survive every time since the output devices use 10R base stoppers, which limit current to less than the driver maximum rating. Thus only the outputs, and usually only one, will go. This certainly takes the attitude that casualties are inevitable, I agree, but the collateral damage is highly localised and relatively inexpensive to repair. It certainly does not take out the entire output stage, which can be quite expensive.
Someone suggested (was it ERGO, from Estonia??) that a chip could be used for precise SOA protection. Yes, I agree, but such beasts are not cheap (if indeed I can obtain them in the land of OZ), and an IC is tricky for many first timers to mount on a pcb, particularly if it is only available as an SMD. It adds complexity, but is definitely worth considering if I should ever uprate the AKSA to 200 watts or so since this would mean at least three matched output pairs, an expensive stage to replace.
I hope this gives some glimpse of where I am coming from. Cost and simplicity are always factors, particularly in the DIY world where the market is as limited by price barriers as the consumer product. While it leaves a nasty taste in the mouth to pronounce a beautiful, highly linear 12A 230V transistor as expendable, it is a sensible compromise, and with care during assembly and use, outputs need never be sacrificed. The upside is the sonics, which are exceptional, because despite what you say, I am not 100% convinced that current protection has absolutely no effect on sonics.
BTW, I am most impressed that you regard the best amps as the one which sound the best! How absurd would we all be if we insisted that the best amps are those which measure best???
Cheers,
Hugh R. Dean
www.printedelectronics.com
Melbourne AUSTRALIA
I can now continue this discussion with you, and others, after returning with my family from a weekend at the beach!
Thank you Michel, for your email attachment. I have it, will open and study it later tonight.
Thank you Sdman, Phil and Grey for the nice comments in support; Michel, I think Grey has a valid point, and you need to think on how your posts appear to others from your very blunt, categorical statements. Writing style is VERY important, as this is a sterile medium all too easily misunderstood. Almost anything discussed between human beings is inclined to passion, and people look for misinterpretations and the merest sniff that you are slighting others. When they perceive it, right or wrong, you are a marked man. And it is so much nicer to stay friends than to effortlessly make enemies.........
I have worked as a consultant in IT, and one thing you very quickly learn is 'NO CATEGORICAL STATEMENTS'. Truth is, someone always proves you wrong, as I may do here.
One such statement on matters technical, was this:
"SOA protection as discussed here, is totally unnecesary, indeed irrelevant in single ended, current source/sink biased output stages as designed by N. Pass for instance. "
This applies, and you might have used this proviso, ONLY if the current source lies beneath the rail and the amplifying device (AD - usually an emitter follower or CFP). If however the reverse is true, viz the AD is powered from the positive rail and the current source is in fact a sink, strapped to ground, then there is no constraint over the current the upper device may pass at all. A shorted load will pass a very large current from the rail; I know because I have designed such an amp - a Zen turned on its head, no less, and called the Glass Harmony - and while I never did blow any amplifying devices it certainly did have the potential there. Hence, SOA protection is a good thing in this context, particularly with highly reactive loads.
To resume: VI protection is two fold, as you have pointed out. One must protect against voltage and current spikes, and protection must take account of both. I agree with your premise so far; let's consider the options.
Voltage protection. This can be effected two ways. A catch diode across each output device will short any voltage spikes at the load which actually exceed the rail supply, thus protecting both output devices from dV/dt destruction. The second method I use is to use a low rail voltage in comparison with the transistor max Vceo rating. On modern bipolars (which I prefer) this means it is difficult to blow the output on voltage spikes.
Current protection. This I choose to ignore, except that current limiting is de rigeur if the emitter resistor is high and a polyphase switch on the speaker protects the bass driver. I use 0R47 emitter resistors, which limit transconductance - and hence feedback factor - of the output stage. It works very well, and gives strong local feedback which is linearising, since the ohmic voltage feedback from the emitter resistor is about the same as the non-linear junction voltage feedback, no bad thing.
If an AKSA is mischievously presented with a short circuited load, it will naturally destroy at least one output device, and sometimes both. The drivers survive every time since the output devices use 10R base stoppers, which limit current to less than the driver maximum rating. Thus only the outputs, and usually only one, will go. This certainly takes the attitude that casualties are inevitable, I agree, but the collateral damage is highly localised and relatively inexpensive to repair. It certainly does not take out the entire output stage, which can be quite expensive.
Someone suggested (was it ERGO, from Estonia??) that a chip could be used for precise SOA protection. Yes, I agree, but such beasts are not cheap (if indeed I can obtain them in the land of OZ), and an IC is tricky for many first timers to mount on a pcb, particularly if it is only available as an SMD. It adds complexity, but is definitely worth considering if I should ever uprate the AKSA to 200 watts or so since this would mean at least three matched output pairs, an expensive stage to replace.
I hope this gives some glimpse of where I am coming from. Cost and simplicity are always factors, particularly in the DIY world where the market is as limited by price barriers as the consumer product. While it leaves a nasty taste in the mouth to pronounce a beautiful, highly linear 12A 230V transistor as expendable, it is a sensible compromise, and with care during assembly and use, outputs need never be sacrificed. The upside is the sonics, which are exceptional, because despite what you say, I am not 100% convinced that current protection has absolutely no effect on sonics.
BTW, I am most impressed that you regard the best amps as the one which sound the best! How absurd would we all be if we insisted that the best amps are those which measure best???
Cheers,
Hugh R. Dean
www.printedelectronics.com
Melbourne AUSTRALIA
Hi Michel,
Thank you for your swift response. I look forward to your answers, and hopefully by that time I will have done justice to your attached email.
Sincerely,
Hugh
www.printedelectronics.com;)
Thank you for your swift response. I look forward to your answers, and hopefully by that time I will have done justice to your attached email.
Sincerely,
Hugh
www.printedelectronics.com;)
I'm Jealous!
I have nothing to contibute to the technical discussion -- Hell, I barely understand most of it! I'll make only two points:
1. Since I'm shivering here in the NE US, I'm extremely jealous of anyone who has just returned from the beach.
2. Hugh, I've been doing final testing and adjustment on my 100W AKSA. It's obviously not broken in and I haven't been listening for sound quality, but I can report that it's hum-free and that Beniamino Gigli has been singing distortion-free arias at levels that drove my wife out of the room. I've posted some pictures of the unfinished amp showing the basic layout at http://photos.yahoo.com/elektratig
Bob
I have nothing to contibute to the technical discussion -- Hell, I barely understand most of it! I'll make only two points:
1. Since I'm shivering here in the NE US, I'm extremely jealous of anyone who has just returned from the beach.
2. Hugh, I've been doing final testing and adjustment on my 100W AKSA. It's obviously not broken in and I haven't been listening for sound quality, but I can report that it's hum-free and that Beniamino Gigli has been singing distortion-free arias at levels that drove my wife out of the room. I've posted some pictures of the unfinished amp showing the basic layout at http://photos.yahoo.com/elektratig
Bob
Bob's New 100W AKSA
Hi Bob,
I visited your photos and the AKSA you have just made looks great!
All praise to you for doing such a good job on your AKSA; it's a credit to you, particularly the striking neatness of the pcbs and the white wiring on the Vbe multipliers!
I have found you can secure the power supply pcb very nicely by dabbing epoxy on the ends of the filter caps; this is anti-microphonic too, of course, and keeps the right side up for securing all the power wiring.
Delighted to see you love cats; so do I! Our cat, a very large tabby with an exquisitely self-centred nature, is fetchingly called 'spewbags'. Well you might wonder......
Struck by your wonderful collection; Laurel, Moondog, ST40, Straight Eight, Foreplay. I see you as smitten as I by this bloody virus!
Thank you for sharing it with us, Bob. Appreciate the workmanship and the gesture.
Cheers,
Hugh
www.printedelectronics.com
Hi Bob,
I visited your photos and the AKSA you have just made looks great!
All praise to you for doing such a good job on your AKSA; it's a credit to you, particularly the striking neatness of the pcbs and the white wiring on the Vbe multipliers!
I have found you can secure the power supply pcb very nicely by dabbing epoxy on the ends of the filter caps; this is anti-microphonic too, of course, and keeps the right side up for securing all the power wiring.
Delighted to see you love cats; so do I! Our cat, a very large tabby with an exquisitely self-centred nature, is fetchingly called 'spewbags'. Well you might wonder......
Struck by your wonderful collection; Laurel, Moondog, ST40, Straight Eight, Foreplay. I see you as smitten as I by this bloody virus!
Thank you for sharing it with us, Bob. Appreciate the workmanship and the gesture.
Cheers,
Hugh
www.printedelectronics.com
Hi Michael,
Yes, indeed, food for thought.
I have studied your two schematics, in particular 39, the dual slope. I find myself liking the circuit, even though there is no integrating function (no Cs at all!). I am presently playing a guessing game trying to work out the values. Now, this is a challenge!
I am frantic with workload at present, so can't give you the attention this discussion deserves. Perhaps later tonight.
Interesting comments about emitter degeneration. I find rock solid quiescent control and excellent current sharing to be the primary advantages of high degeneration resistors, and since seeing them in many pro-audio amps, which are built like the Dreadnought for obvious reasons, I have opted to use them as a little emitter inefficiency is not a problem for high fi amps. Swinging cleanly to the rail is important, however, for pro-audio.
You raise many issues, Michel, and most charmingly put, with copious use of emoticons. I will give it more time a little later.....
Clever circuit, but very tetchy values, necessary to tailor them specifically to the device and application, of course. I'm glad and grateful that I saw it...... Thank you, Michael.
Cheers,
Hugh
www.printedelectronics.com
Yes, indeed, food for thought.
I have studied your two schematics, in particular 39, the dual slope. I find myself liking the circuit, even though there is no integrating function (no Cs at all!). I am presently playing a guessing game trying to work out the values. Now, this is a challenge!
I am frantic with workload at present, so can't give you the attention this discussion deserves. Perhaps later tonight.
Interesting comments about emitter degeneration. I find rock solid quiescent control and excellent current sharing to be the primary advantages of high degeneration resistors, and since seeing them in many pro-audio amps, which are built like the Dreadnought for obvious reasons, I have opted to use them as a little emitter inefficiency is not a problem for high fi amps. Swinging cleanly to the rail is important, however, for pro-audio.
You raise many issues, Michel, and most charmingly put, with copious use of emoticons. I will give it more time a little later.....
Clever circuit, but very tetchy values, necessary to tailor them specifically to the device and application, of course. I'm glad and grateful that I saw it...... Thank you, Michael.
Cheers,
Hugh
www.printedelectronics.com
Hi Guys,
Despite all rumours to the contrary, I am actually wide awake, it is 11:50 am here, and I have enjoyed my morning cuppa. It is drizzly in Melbourne, pleasantly balmy at around 20C, and neight of my two daughters (one Boadicea, and the other Cindarella) are causing me any angst at this time, God be thanked.....
I am wrapping parcels of AKSAs for my customers, and scrabbling to get to the parts stockist for some more bloody heatshrink tubing for the RF suppression inductors!
Oh, I am listening to Madonna. Why? Long since given up on the music. It's the recording I like!
Cheers,
Hugh
Despite all rumours to the contrary, I am actually wide awake, it is 11:50 am here, and I have enjoyed my morning cuppa. It is drizzly in Melbourne, pleasantly balmy at around 20C, and neight of my two daughters (one Boadicea, and the other Cindarella) are causing me any angst at this time, God be thanked.....
I am wrapping parcels of AKSAs for my customers, and scrabbling to get to the parts stockist for some more bloody heatshrink tubing for the RF suppression inductors!
Oh, I am listening to Madonna. Why? Long since given up on the music. It's the recording I like!
Cheers,
Hugh
Michel/Hugh
Whilst this debate is interesting and informative, I do not think that it is appropriate to discuss, in open forum, schematics that are not available to all members, since it becomes impossible to follow the some of the reasoning and comments. Could I please therefore suggest that either Michel posts his schematics for all to see, and comprehend, or that you continue your discussions by email.
Geoff
Whilst this debate is interesting and informative, I do not think that it is appropriate to discuss, in open forum, schematics that are not available to all members, since it becomes impossible to follow the some of the reasoning and comments. Could I please therefore suggest that either Michel posts his schematics for all to see, and comprehend, or that you continue your discussions by email.
Geoff
From the site mission statement: "When it comes down to it, DIY Audio is all about fun. This site is about making our fun easier to find!".
Michel,
I think you have something to add to our community. I think you just started out on the wrong foot.
I think that when you started posting in this forum, you did not understand netiquette. Posting ASCII is not like having a normal conversation. <i>I please request that you will spend time researching netiquette before posting anything that might be considered sarcastic, or that might 'stir the pot', in our forum again</i>.
I am positive that you will find the people here from now on will treat you with the same respect that you will give them.
<b>We are now starting with a clean slate</b>. I have done my best to remove off topic discussion from this thread. It took me a while, because I think there is a lot of good content here.
Until now there has not been a thread that has gotten so emotional. Please don't let it happen again!
I am not trying to be big brother, but sometimes every community needs a little policing.
I will post a community code of conduct to the site in the next few days.
Sorry for the tersity of my post, but I have to get back to work.
I look forward to further discussion about this amplifer. It would be nice if Mr. AKSA could post the schematics so everyone else can see what is going on, otherwise, you might as well be communicating via email and not being a 'tease' to the drooling onlookers. In our shared goal of DIY we have abundance mentality, we're all about sharing, just ask Mr. Pass
Thanks, sorry, to the future,
Jason.
Michel,
I think you have something to add to our community. I think you just started out on the wrong foot.
I think that when you started posting in this forum, you did not understand netiquette. Posting ASCII is not like having a normal conversation. <i>I please request that you will spend time researching netiquette before posting anything that might be considered sarcastic, or that might 'stir the pot', in our forum again</i>.
I am positive that you will find the people here from now on will treat you with the same respect that you will give them.
<b>We are now starting with a clean slate</b>. I have done my best to remove off topic discussion from this thread. It took me a while, because I think there is a lot of good content here.
Until now there has not been a thread that has gotten so emotional. Please don't let it happen again!
I am not trying to be big brother, but sometimes every community needs a little policing.
I will post a community code of conduct to the site in the next few days.
Sorry for the tersity of my post, but I have to get back to work.
I look forward to further discussion about this amplifer. It would be nice if Mr. AKSA could post the schematics so everyone else can see what is going on, otherwise, you might as well be communicating via email and not being a 'tease' to the drooling onlookers. In our shared goal of DIY we have abundance mentality, we're all about sharing, just ask Mr. Pass
Thanks, sorry, to the future,
Jason.
JohnR,
There is no need for your attitude. I don't recall you contributing to the discussion apart from your opinion of the ASKA amplifer kit. We were discussing the merits of a non-evasive protection scheme for amplifiers. The people involved were well able to defend their opinions without your help.
Regards,
Jam
There is no need for your attitude. I don't recall you contributing to the discussion apart from your opinion of the ASKA amplifer kit. We were discussing the merits of a non-evasive protection scheme for amplifiers. The people involved were well able to defend their opinions without your help.
Regards,
Jam
I stuffed up the '10 minutes till posts are locked in' setting (it was turned off), which has allowed Michel to now turn this thread into swiss cheese. I have now fixed this.
Jam: You may not have seen the recent diatribe from Michel that JohnR responded to, or you may. Regardless, the issue is now over and out.
Please direct anything that should not go into this thread to me personally at jason@diyaudio.com rather than posting it.
<b>This thread is now closed</b>. Please start a new thread if you want to continue discussing amplifier protection schemes or the AKSA amp, and feel free to cut and paste from this thread to the new in order to create a basis for discussion.
If I can, I will restore Michel's posts so that you can continue discussing non evasive amplifier protection schemes and use his material as reference, but this is painful (sifting through .tgz mysql update logs).
Apologies to all concerned.
Jam: You may not have seen the recent diatribe from Michel that JohnR responded to, or you may. Regardless, the issue is now over and out.
Please direct anything that should not go into this thread to me personally at jason@diyaudio.com rather than posting it.
<b>This thread is now closed</b>. Please start a new thread if you want to continue discussing amplifier protection schemes or the AKSA amp, and feel free to cut and paste from this thread to the new in order to create a basis for discussion.
If I can, I will restore Michel's posts so that you can continue discussing non evasive amplifier protection schemes and use his material as reference, but this is painful (sifting through .tgz mysql update logs).
Apologies to all concerned.
- Status
- Not open for further replies.