Shortcut protection
John W
I think, you not saw every Jakas tekst regard me onto this forum. Jaka is writen ofensive a few times before and i even not answer on it.
I know what you mean, but look on the first question on this threade and my ansver(s) on. If not, its impossible to understand the point. I'll write again: It's the simplest way to made shortcut protection (regard on general question in this threade). With resistor such this 1 (maybe more Ohms) Ohm in location like on picture (but for general purpose (look the question) , od course). You don't agree with?
Jaka
You suggest the fuses (like you saw in original motorola application - copie again) like the answer on first question in this threade?
Yor calculation seems like manipulation, to calculate what you want. It has nothing with themas in this threade. Seems like exchanges themas and questions with yourself. Like DC on the input,...
This 1 Ohm is in use in prakse for some audiophiles (also in Slovenia, like producer Malboret (ex Pear audio), with in this case a more (two pairs of) transistors on the output. I think he use 2 Ohms.... Its the question of his integrated amplifier known as TABLA (also saled in Austria, Germany and Slovenia). His (Malboret) question to HL was similar like here in this threade. In real life, this aplications of 1 Ohm (or more) resistors works (you leave in Slovenia, you know for Malboret (look on the internet and finf e mail of him) - manufacturer of audio support equipment and some amps) and ask him how (is or isn't) work in reality.
I'm sure it's not optimal and more clever way to made protection, but IT WORKS and you have aprooved this in reality and than you have to write on this forum what you see (in reality).
p.s. Malboret produced this amp in the past and have traubles with shortcuts (even has fusses before). He come to us, with the same question (how to do this, but without additional electronic). He made this (its a bit influence on soun, but works enough good for him.....) and it works.
This 1 Ohm (maybe 2) works at him, but for some kind of amplifiers will not (it will but with different values) , so a wrote that 1 or more Ohms like the answer on primar question of threade. I put a picture on and write: "look on schema and on resistors on output". Not: "look on picture and thing how good it works in this schematics I shown".
It is impossible to understand even a sympliest answer, if somebody dont want to undestand. If somebody have troubles if see logical answer on the question. Than even the simplest thing became complicated, like a lot of calculations like (Jakas) answer on the questions no on put on this forum.
John W
I think, you not saw every Jakas tekst regard me onto this forum. Jaka is writen ofensive a few times before and i even not answer on it.
I know what you mean, but look on the first question on this threade and my ansver(s) on. If not, its impossible to understand the point. I'll write again: It's the simplest way to made shortcut protection (regard on general question in this threade). With resistor such this 1 (maybe more Ohms) Ohm in location like on picture (but for general purpose (look the question) , od course). You don't agree with?
Jaka
You suggest the fuses (like you saw in original motorola application - copie again) like the answer on first question in this threade?
Yor calculation seems like manipulation, to calculate what you want. It has nothing with themas in this threade. Seems like exchanges themas and questions with yourself. Like DC on the input,...
This 1 Ohm is in use in prakse for some audiophiles (also in Slovenia, like producer Malboret (ex Pear audio), with in this case a more (two pairs of) transistors on the output. I think he use 2 Ohms.... Its the question of his integrated amplifier known as TABLA (also saled in Austria, Germany and Slovenia). His (Malboret) question to HL was similar like here in this threade. In real life, this aplications of 1 Ohm (or more) resistors works (you leave in Slovenia, you know for Malboret (look on the internet and finf e mail of him) - manufacturer of audio support equipment and some amps) and ask him how (is or isn't) work in reality.
I'm sure it's not optimal and more clever way to made protection, but IT WORKS and you have aprooved this in reality and than you have to write on this forum what you see (in reality).
p.s. Malboret produced this amp in the past and have traubles with shortcuts (even has fusses before). He come to us, with the same question (how to do this, but without additional electronic). He made this (its a bit influence on soun, but works enough good for him.....) and it works.
This 1 Ohm (maybe 2) works at him, but for some kind of amplifiers will not (it will but with different values) , so a wrote that 1 or more Ohms like the answer on primar question of threade. I put a picture on and write: "look on schema and on resistors on output". Not: "look on picture and thing how good it works in this schematics I shown".
It is impossible to understand even a sympliest answer, if somebody dont want to undestand. If somebody have troubles if see logical answer on the question. Than even the simplest thing became complicated, like a lot of calculations like (Jakas) answer on the questions no on put on this forum.
Bojan,
I welcome a relative decrease of hostility in your post, now it seems that you are almost willing to defend your point of view ad rem instead ad personam. You seem to be obsessed by this Motorola AP note which I even don't have. You were asked before to at least post it's number. I would really like to know where all my ideas come from. Power calculation procedure must have been from there. The following will also be from there. Post the number, and I will not have to type a single word again.
I see you still think that those resistors are there to limit the current. If 1R (or 2R) resistor would be used to limit the current, then current would be limited to 29A (or 14.5A) in saturation with corresponding 841W (or 210W) power dissipation on the resistor. In reality current is limited only by choice of Darlington's base resistor and it's hfe. Since we all know how wildly hfe changes (10 to 1 would not be uncommon) in Darlingtons, this is a very bad choice. But those resistors actually have some protection function. By using them, output transistors can operate within the bounds of their SOA (safe operating area). But that only means that the undetermined part of power dissipation in short circuit condition is transferred from output transistors to resistors which have to be dimensioned accordingly for worst case condition (maximum possible hfe of Darlington.). Analysis that I have posted assumed that amplifiers works in B class. Since you specified your circuit to work also in class A, you can guess how already low efficiency and output power of class A is diminished by those resistors and how much power would be dissipated on them. And complaining about DC on the input. If your amplifier is DC coupled, then expect DC at the input.
I work in a field where every design is subject to rigorous safety agency testing. There design philosophy "I have built it and it works" gets you nowhere. In my designs safety agency has the right to make a short or open circuit or reverse connection or something similarly harmful in every single point of the circuit, and the product must remain safe under this condition. So risk analysis in such designs sometimes takes weeks to accomplish. Risk analysis includes some of the simple techniques that were shown in the previous post. And BTW, safety agencies are fond of fuses.
It is now similar with audio equipment. If you want to do audio design professionally, then talk with the guys in the safety agency, see what the requirements are, read them, apply them and submit your products to the agency for testing. That will keep you from legal troubles. I hope you understand that professionalism also means responsibility.
For DIY designs, builder is responsible for his own health and property. But unsafe design techniques are not welcome even on this forum.
Best regards,
Jaka Racman
I welcome a relative decrease of hostility in your post, now it seems that you are almost willing to defend your point of view ad rem instead ad personam. You seem to be obsessed by this Motorola AP note which I even don't have. You were asked before to at least post it's number. I would really like to know where all my ideas come from. Power calculation procedure must have been from there. The following will also be from there. Post the number, and I will not have to type a single word again.
I see you still think that those resistors are there to limit the current. If 1R (or 2R) resistor would be used to limit the current, then current would be limited to 29A (or 14.5A) in saturation with corresponding 841W (or 210W) power dissipation on the resistor. In reality current is limited only by choice of Darlington's base resistor and it's hfe. Since we all know how wildly hfe changes (10 to 1 would not be uncommon) in Darlingtons, this is a very bad choice. But those resistors actually have some protection function. By using them, output transistors can operate within the bounds of their SOA (safe operating area). But that only means that the undetermined part of power dissipation in short circuit condition is transferred from output transistors to resistors which have to be dimensioned accordingly for worst case condition (maximum possible hfe of Darlington.). Analysis that I have posted assumed that amplifiers works in B class. Since you specified your circuit to work also in class A, you can guess how already low efficiency and output power of class A is diminished by those resistors and how much power would be dissipated on them. And complaining about DC on the input. If your amplifier is DC coupled, then expect DC at the input.
I work in a field where every design is subject to rigorous safety agency testing. There design philosophy "I have built it and it works" gets you nowhere. In my designs safety agency has the right to make a short or open circuit or reverse connection or something similarly harmful in every single point of the circuit, and the product must remain safe under this condition. So risk analysis in such designs sometimes takes weeks to accomplish. Risk analysis includes some of the simple techniques that were shown in the previous post. And BTW, safety agencies are fond of fuses.
It is now similar with audio equipment. If you want to do audio design professionally, then talk with the guys in the safety agency, see what the requirements are, read them, apply them and submit your products to the agency for testing. That will keep you from legal troubles. I hope you understand that professionalism also means responsibility.
For DIY designs, builder is responsible for his own health and property. But unsafe design techniques are not welcome even on this forum.
Best regards,
Jaka Racman
Jaka,
(Dunno if this is the right thread, but as you are in to it......)
Your consern regarding electrical safety is a good thing and safety
is always No1.
Maybe, putting together some "essence" from the IEC60950
regarding the requirements of the primary circuit would be nice.
Anyone eager to do this?
EMI (EMC) we do not have to care to much for as long as the
neighbour can still see the boll against the grass on the TV.
/ Regards / Mattias
(Dunno if this is the right thread, but as you are in to it......)
Your consern regarding electrical safety is a good thing and safety
is always No1.
Maybe, putting together some "essence" from the IEC60950
regarding the requirements of the primary circuit would be nice.
Anyone eager to do this?
EMI (EMC) we do not have to care to much for as long as the
neighbour can still see the boll against the grass on the TV.
/ Regards / Mattias
Risk
With every amplifier is the risk of AC (DC) Voltage and to risk to burn something, or to "grill" somebody.... . Oscilations are even risk for the ears....
(Those warnings are because to protect owner of the site.... )
I havn't seen any warning with a lot of circuits here. If you have function here to warn and protect readers, for this site, its OK.
But if such explanations are only interesting info you have for this site, .......
With every amplifier is the risk of AC (DC) Voltage and to risk to burn something, or to "grill" somebody.... . Oscilations are even risk for the ears....
(Those warnings are because to protect owner of the site.... )
I havn't seen any warning with a lot of circuits here. If you have function here to warn and protect readers, for this site, its OK.
But if such explanations are only interesting info you have for this site, .......
There are two answers.
1/ Dont play into under rated speakers. Cheapest solution.
2/ Add a protection circuit using a comparator or even be clever and use a PIC which can have a programmable time delay or even just cut out the amp altogether until power is removed.
Personally I use a PIC then I can not only look for a short but also look for a DC fault on the output.
This must have saved me hundreds of pounds in speakers.
can you share your PIC protect... the input section of the pic if posible.
thanks,
coolet
Greetings to all.
Until recently, i have not payed much attention on the short circuit protection on the amps i have constructed, i always thought i'd be verry carefull not to short them out and instructed who ever i gave an amp of mine to do the same, but as the inevitable happened and one of my amps got one channel shorted at the spk jack the FET's on that channel gave up, now it would be as simple as changing them, and check for any other problems then the amp would be just fine, but i got thinking that still a protection would be needed indeed, searching the net, searching here on this topik i have made an ideea of what needs to be done, i will present the schematick of my amp with a schort circuit protection circuit added to it and marked by a green rectangle:
This too would be simple to add to my amp, it would just need another 2 small PCB's and a few wires, easy job, so then you might be asking what is my problem? well it is simple this circuit like most other limits the current/drive signal on the output devices, but still they would have to cope with large power dissipation, and so it is only efficient for a verry short while, maybe a few seconds, but let's say that trough some case, some way, the short circuit last more than that, then how do you protect the output devices? i am trying to find a more relyable way to protect my amp and would verry much appreciate your imput.
I thout about a combination of both a current limitter for fast protection and some long term option, say like some relay that would either cut off the speaker or short the fuse it's self to blow it up so the power is cut off from the amp, but both the cases would i think cause dammage to that relay, i know there is much debating on this subject, manny opinions, so are the options, but i need to find something both relyable and not too complicated, i do not want to spend more on the protection than on the amp it's self, i would desighn this to be adaptable to any amplifier i built, be it more or less powerfull, BJT's or FET's, and so on.
I would appreciate especially any cind of schematic that can help inspire me in my quest, and thank's for u're time.
All the best, Marian.
Until recently, i have not payed much attention on the short circuit protection on the amps i have constructed, i always thought i'd be verry carefull not to short them out and instructed who ever i gave an amp of mine to do the same, but as the inevitable happened and one of my amps got one channel shorted at the spk jack the FET's on that channel gave up, now it would be as simple as changing them, and check for any other problems then the amp would be just fine, but i got thinking that still a protection would be needed indeed, searching the net, searching here on this topik i have made an ideea of what needs to be done, i will present the schematick of my amp with a schort circuit protection circuit added to it and marked by a green rectangle:
An externally hosted image should be here but it was not working when we last tested it.
This too would be simple to add to my amp, it would just need another 2 small PCB's and a few wires, easy job, so then you might be asking what is my problem? well it is simple this circuit like most other limits the current/drive signal on the output devices, but still they would have to cope with large power dissipation, and so it is only efficient for a verry short while, maybe a few seconds, but let's say that trough some case, some way, the short circuit last more than that, then how do you protect the output devices? i am trying to find a more relyable way to protect my amp and would verry much appreciate your imput.
I thout about a combination of both a current limitter for fast protection and some long term option, say like some relay that would either cut off the speaker or short the fuse it's self to blow it up so the power is cut off from the amp, but both the cases would i think cause dammage to that relay, i know there is much debating on this subject, manny opinions, so are the options, but i need to find something both relyable and not too complicated, i do not want to spend more on the protection than on the amp it's self, i would desighn this to be adaptable to any amplifier i built, be it more or less powerfull, BJT's or FET's, and so on.
I would appreciate especially any cind of schematic that can help inspire me in my quest, and thank's for u're time.
All the best, Marian.
Just in case I am working in something like your are looking also. I found that using the ASC712 20 amps hall effect current sensor the speaker can be protect from high current. You can use it without the micro just using a comparator to disconnect the speaker when the speaker current reach a certain limit.
See my thread 28 link:
http://www.diyaudio.com/forums/chip...ng-speaker-output-using-microprocessor-3.html
See my thread 28 link:
http://www.diyaudio.com/forums/chip...ng-speaker-output-using-microprocessor-3.html
My goul is not to desighn a circuit that would perotect the speakers, for that i have a CD-Detect that works just fine, what i need is a good way to protect the output devices from accidental shortcircuit across the load, i need to protect the power transistors ( be they BJT's or FET's ).
As i sayd the limitter is not bulletproof, it only protects for a very short while, for long time protection the output devices would have to be more in number and a massive cooling system, so it is not cost effective.
That current senzor would be ok if it would be easyer to find, but it can be replaced by a tranzistor that cuts open when a certain current flows through the source rezistors, when that current would be high enough it would generate let's say 0,7V across the resistor so an bjt would open driving a relay to disconect the speaker, but i fear that high current would dammage if not destroy the relay contacts, more so if the relay would short up the rails so the fuses would blow ( still i think maybe the last one is a dumb ideea from my part
But anyway, is it possible to do what i say or is it just rubbish?
As i sayd the limitter is not bulletproof, it only protects for a very short while, for long time protection the output devices would have to be more in number and a massive cooling system, so it is not cost effective.
That current senzor would be ok if it would be easyer to find, but it can be replaced by a tranzistor that cuts open when a certain current flows through the source rezistors, when that current would be high enough it would generate let's say 0,7V across the resistor so an bjt would open driving a relay to disconect the speaker, but i fear that high current would dammage if not destroy the relay contacts, more so if the relay would short up the rails so the fuses would blow ( still i think maybe the last one is a dumb ideea from my part
But anyway, is it possible to do what i say or is it just rubbish?
Yes , your right but you do not want to install something that will interferes
with the quality of the sound. You have to be careful how you do it. I was going to use the emitter resistors but then I need to use an op amp that require more components. My system with save the amplifier output transistors and also the speaker from a short. Your are looking at the current going out of the amplifier.
with the quality of the sound. You have to be careful how you do it. I was going to use the emitter resistors but then I need to use an op amp that require more components. My system with save the amplifier output transistors and also the speaker from a short. Your are looking at the current going out of the amplifier.
You can use a power supply, capable to protect output transistors against short circuit and in same time with a loudspiekers protection too, something like this: http://www.diyaudio.com/forums/powe...ply-cap-multiplier-electronic-protection.html
dado
dado
I do not trust too much current limiting cus the output devices would still have to cope with lots of power dissipation, i woul rather prefer something like a simulated thirystor that takes down the signal at the gates, without the signal there would be no power dissipated so the FET's would be safe, but the circuit would have to stay on after it got started by the short circuit cus otherwise as soone as the signal goes of the cuurent on the sensor would go down and...and you know what i'm takin about.
To be transparent the protection must allow all audio signals to pass unaltered.
That becomes a dominant requirement of any good protection scheme.
The amplifier should always pass any valid audio signal to any valid load.
If the protection scheme interferes with the audio signal when there is nothing wrong with the signal level and nothing wrong with the load, then the protection is crippling a necessary characteristic of the amplifier. Why else do we use amplifiers? I suggest to hear the music.
Now to a few questions:
What is the maximum continuous DC current that the amplifier can pass without damaging itself?
What is the maximum medium term current that the amplifier can pass without damaging itself?
What is the maximum short term current that the amplifier can pass without damaging itself?
Can a protection scheme that works well for all three of those conditions be met by one sensing system?
I would suggest very strongly that the first can be met by fusing the supply rail and/or using Dadod's suggested rail current limiter. BUT, that rail current limiter must be able to pass short term and medium term currents that are valid audio signals. Fuses will certainly meet this additional requirement.
That becomes a dominant requirement of any good protection scheme.
The amplifier should always pass any valid audio signal to any valid load.
If the protection scheme interferes with the audio signal when there is nothing wrong with the signal level and nothing wrong with the load, then the protection is crippling a necessary characteristic of the amplifier. Why else do we use amplifiers? I suggest to hear the music.
Now to a few questions:
What is the maximum continuous DC current that the amplifier can pass without damaging itself?
What is the maximum medium term current that the amplifier can pass without damaging itself?
What is the maximum short term current that the amplifier can pass without damaging itself?
Can a protection scheme that works well for all three of those conditions be met by one sensing system?
I would suggest very strongly that the first can be met by fusing the supply rail and/or using Dadod's suggested rail current limiter. BUT, that rail current limiter must be able to pass short term and medium term currents that are valid audio signals. Fuses will certainly meet this additional requirement.
Last edited:
I did a lot of work on this subject a few months ago for my new amp ('e-Amp'). I found that my Ovation 250 amp sounded much better without SOA protection and when I took a closer look, the SOA was loading the VAS stage, so I removed it and got a much better sound. I looked into 3 slope SOA with time delays using LTspice, but still could not guarantee the protection did not interfere with the amplifer. Multislope SOA protection works well on resistive loads, but highly reactive loads are very problematic. My solution in the e-Amp was to increase the number of output devices in the amplifer and use fast, simple current limiting using an opto across the output emitter degen resistors. Bob Cordell is 100% right when he says the best protection is no protection. A Hall sensor looks like a good option as well - as long as it does not introduce non-linearities.
I think there has been some real progress on this subject over the last year or so on DIYAudio -there's been a move away from relays to mosfet based SSR's (with some great practical examples), and hopefully further discussion on protection in general of the output stage will lead to further progress.
I think there has been some real progress on this subject over the last year or so on DIYAudio -there's been a move away from relays to mosfet based SSR's (with some great practical examples), and hopefully further discussion on protection in general of the output stage will lead to further progress.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.