Post by thread:[OT]: Audio Amplifier

Please excuse the completely OT nature of this post; its just that I have seen such great talent in various branches of electronics on this list, I feel tempted to put my query here. I have a TV which came without a earphone socket. I just wired in a socket the usual way (speaker automatically gets disconnected if you plug in a headphone), and it seemed to work fine...for a few seconds. After that, there was a bang, and the amplifier chip within blew up. The chip used is Philips TDA 8945S. The TV speaker has an impedence of 8 ohms; the headphones are of course, a much larger impedence. This being the case, why would the chip blow? Do some amplifiers also get damaged if the load impedence is *higher* than the one specified? In which case, would a resistor network that still offerred about 8 ohms to the amplifier even when the headphone is connected help? I just might have inadvertantly shorted the output wires (for not more than a second or two, I'm sure). This chip is supposed to have both, thermal and short circuit protection. Could the possible short then still have caused the problem? I have replaced the chip, but before trying out the headphone again, I would like to get some advice from you folks out there. Regards, Anand Dhuru -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

A higher impedance load won't damage the output amplifier on a solid state amp. You have to remember that the ground in a TV receiver may not necessarily be earth ground. It could float as high as your line voltage if not properly polarized. If you add a jack, you could be in for some sparks. This is why you don't see many headphone jacks on most cheap TV's. Those that do have a jack have an isolation transformer isolating the chassis ground from the headphone ground. Isolating with just capacitors won't meet UL approval. Also, putting a 100 ohm in series with the output will protect the amp from possible shorts damaging the amp without loss of signal to the headphones. Rick Anand Dhuru wrote: {Quote hidden}> Please excuse the completely OT nature of this post; its just that I have seen such great talent in various branches of electronics on this list, I feel tempted to put my query here. > > I have a TV which came without a earphone socket. I just wired in a socket the usual way (speaker automatically gets disconnected if you plug in a headphone), and it seemed to work fine...for a few seconds. > > After that, there was a bang, and the amplifier chip within blew up. > > The chip used is Philips TDA 8945S. > > The TV speaker has an impedence of 8 ohms; the headphones are of course, a much larger impedence. > > This being the case, why would the chip blow? > > Do some amplifiers also get damaged if the load impedence is *higher* than the one specified? > > In which case, would a resistor network that still offerred about 8 ohms to the amplifier even when the headphone is connected help? > > I just might have inadvertantly shorted the output wires (for not more than a second or two, I'm sure). This chip is supposed to have both, thermal and short circuit protection. Could the possible short then still have caused the problem? > > I have replaced the chip, but before trying out the headphone again, I would like to get some advice from you folks out there. > > Regards, > > Anand Dhuru > > -- > http://www.piclist.com hint: PICList Posts must start with ONE topic: > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

I failed to mention why. If you had a hot chassis, and you were wearing a headphone or earphones, you would have a lethal potential right next to your head. Any leakage in the wiring would be lethal. If you added a jack and ran your signal to an amplifier that was grounded to earth ground...... POP!!!!!!! Rick "Rick C." wrote: {Quote hidden}> A higher impedance load won't damage the output amplifier on a solid state amp. You have to remember that the ground in a TV receiver may not necessarily be earth ground. It could float as high as your line voltage if not properly polarized. > If you add a jack, you could be in for some sparks. This is why you don't see many headphone jacks on most cheap TV's. Those that do have a jack have an isolation transformer isolating the chassis ground from the headphone ground. Isolating > with just capacitors won't meet UL approval. > > Also, putting a 100 ohm in series with the output will protect the amp from possible shorts damaging the amp without loss of signal to the headphones. > Rick > > Anand Dhuru wrote: > > > Please excuse the completely OT nature of this post; its just that I have seen such great talent in various branches of electronics on this list, I feel tempted to put my query here. > > > > I have a TV which came without a earphone socket. I just wired in a socket the usual way (speaker automatically gets disconnected if you plug in a headphone), and it seemed to work fine...for a few seconds. > > > > After that, there was a bang, and the amplifier chip within blew up. > > > > The chip used is Philips TDA 8945S. > > > > The TV speaker has an impedence of 8 ohms; the headphones are of course, a much larger impedence. > > > > This being the case, why would the chip blow? > > > > Do some amplifiers also get damaged if the load impedence is *higher* than the one specified? > > > > In which case, would a resistor network that still offerred about 8 ohms to the amplifier even when the headphone is connected help? > > > > I just might have inadvertantly shorted the output wires (for not more than a second or two, I'm sure). This chip is supposed to have both, thermal and short circuit protection. Could the possible short then still have caused the problem? > > > > I have replaced the chip, but before trying out the headphone again, I would like to get some advice from you folks out there. > > > > Regards, > > > > Anand Dhuru > > > > -- > > http://www.piclist.com hint: PICList Posts must start with ONE topic: > > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > -- > http://www.piclist.com hint: PICList Posts must start with ONE topic: > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

On Mon, 1 Jul 2002, Anand Dhuru wrote: >Please excuse the completely OT nature of this post; its just that I have seen such great talent in various branches of electronics on this list, I feel tempted to put my query here. > >I have a TV which came without a earphone socket. I just wired in a socket the usual way (speaker automatically gets disconnected if you plug in a headphone), and it seemed to work fine...for a few seconds. > >After that, there was a bang, and the amplifier chip within blew up. > >The chip used is Philips TDA 8945S. I dunno about that chip but if it is a bridge amplifier (I think it is) then the results you describe match the expected behavior, especially if you grounded one of the headphone wires. The other possible solution involves self oscillation caused by the unexpected low load and long wires. Normal phones sockets use a resistor in series with the socket and another in parallel with the phones. Also inexpensive headphone sockets briefly short the amp on insertion and removal. Most amps do not like this. Use a relay to switch the socket if you can't afford a good quality female socket (that is specifically designed not to short on insertion). And be careful, the usual reason for a missing phones socket is an unsafe chassis. You could electrocute someone or start a fire. Peter -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Thanks Rick, Peter, for the information. I did some checking after reciving the posts from you guys, and what I find is this: The TV has a 2 pin mains plug (no earth); neither of the two seem connected to the PCB ground (the ground on the Audio IN, Video IN plugs). Also, neither of the speaker wires seems directly connected to the PCB ground. On this basis, am I right in assuming that a workable solution might be: 1) a 100 ohm resistor in series with the speaker / headphone socket, always in circuit regardless of whether the headphones are plugged in, for short circuit protection, and 2) an audio transformer to drive the headphones; this comes into circuit only when the speakers are cut out, and would give complete isolation to the headphones? The only issue I foresee is that I still would short the circuit at the socket while inserting the headphones, but now it would be thru' the 100 ohm resistor. Would this work? And, would it be safe? {Original Message removed}

> 1) a 100 ohm resistor in series with the speaker / headphone socket, > always in circuit regardless of whether the headphones are plugged in, > for short circuit protection, and h/p will be only a few mW so you must have that resistor unloess you like your sound unusually loud for a brief time > 2) an audio transformer to drive the headphones; this comes into > circuit only when the speakers are cut out, and would give complete > isolation to the headphones? Shouldn't be necessary. A transformer is normally used for impedance matching - what you are looking for is voltage matching (or rather voltage dropping, therefore a resistor) > The only issue I foresee is that I still would short the circuit at the > socket while inserting the headphones, but now it would be thru' > the 100 ohm reistor. Would this work? And, would it be safe? The socket shouldn't short. Not in the way you think perhaps. The tip and ring contacts on the plug shouldn't be able to be shorted to the barrel of the plug via the socket contacts, nor should the tip or ring be able to short out the contacts inside the socket. However, if you think there is shorting then the 100R will really help -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Jinx wrote: > > Shouldn't be necessary. A transformer is normally used for > impedance matching - what you are looking for is voltage > matching (or rather voltage dropping, therefore a resistor) > > For most headphone impedances a 600ohm/600ohm transformer will give you the isolation you need without impedance matching problems. Jameco has them for a buck or so. You could use the phone line transformer out of an old modem card too, but you might suffer just a little high frequency loss but no so bad. Rick -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

-----Original Message----- Please excuse the completely OT nature of this post; its just that I have seen such great talent in various branches of electronics on this list, I feel tempted to put my query here. I have a TV which came without a earphone socket. I just wired in a socket the usual way (speaker automatically gets disconnected if you plug in a headphone), and it seemed to work fine...for a few seconds. After that, there was a bang, and the amplifier chip within blew up. The chip used is Philips TDA 8945S. The TV speaker has an impedence of 8 ohms; the headphones are of course, a much larger impedence. This being the case, why would the chip blow? ==================================================================== the only way you can damage the chip is to short the speaker wires for some reason it happened that. If you didn't add resistor because of your headphone is low powerful in this case first headphone will get damaged then the amp. ==================================================================== Do some amplifiers also get damaged if the load impedence is *higher* than the one specified? ================== never ================== In which case, would a resistor network that still offerred about 8 ohms to the amplifier even when the headphone is connected help? =============================== the help is to lower the power to match with your head phone =============================== I just might have inadvertantly shorted the output wires (for not more than a second or two, I'm sure). This chip is supposed to have both, thermal and short circuit protection. Could the possible short then still have caused the problem? ===================================== that is the only way you can damage the chip I do not think the chip has short circuit protection. ========================================== I have replaced the chip, but before trying out the headphone again, I would like to get some advice from you folks out there. =========================== make sure when you plug it in it is not making any short connection for a moment or so and add resistors to lower the power ============================= Andre Abelian Regards, Anand Dhuru -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

No. It's like saying, do I need a load on an opamp? A higher impeance load will just mean a lower power transfer to the load. Back in the vacuum tube days when you had a plate transformer matched to a low impedance speaker, if you took the load (speaker) off and drove it hard, huge amounts of voltage would be generated from inductive reactance that would literally arc and burn the transformer up. Rick "Richards, Justin P" wrote: {Quote hidden}> <snip> > Do some amplifiers also get damaged if the load impedence is *higher* > than the one specified? > > ================== > never > ================== > Just thought I might clarify, the post was with regards to audio amps but > thought it best to mention I *think* RF power amps like to see a load or > they can fail. > > Is that accurate? > Justin > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> <snip> > Do some amplifiers also get damaged if the load impedence is *higher* > than the one specified? > > ================== > never > ================== > Just thought I might clarify, the post was with regards to audio amps but > thought it best to mention I *think* RF power amps like to see a load or > they can fail. IIRC there are some audio tube amp designs out there that will get damaged if no speaker is attached. As for RF amps yes, quite a few of the high power ones can self destruct if they don't have an antenna attached. TTYL -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> > Do some amplifiers also get damaged if the load impedance is *higher* > > than the one specified? > No. It's like saying, do I need a load on an opamp? A higher impeance load > will just mean a lower power transfer to the load. > > Back in the vacuum tube days when you had a plate transformer matched to a > low impedance speaker, if you took the load (speaker) off and drove it hard, > huge amounts of voltage would be generated from inductive reactance that > would literally arc and burn the transformer up. While in most cases the above answer is correct, there are situations where a load is needed. (Incidentally to this, the transistor amp equivalent of a valve amp open circuit is a short circuit - valve amps tended to survive these whereas transistor amps need to be protected against them.) Some amplifiers perform 'differently" under little or no load. An LM324 opamp is a specialist example - it has an "interesting" output stage which runs in (AFAIR somewhere between class A and AB under light load and switches to full class B under heavier loads. Under certain conditions of light load it has nasty output distortion which vanishes as the amp is biased up to run in true class B. A voltage regulator is a specialised amplifier with extra circuitry added to do a particular job. SOME voltage regulators become unhappy at very light loads. An amplifier could be designed (and some are) to rely on output current draw to bias itself correctly. RM -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

>Some amplifiers perform 'differently" under little or >no load. An LM324 opamp is a specialist example - it >has an "interesting" output stage which 0runs in (AFAIR >somewhere between class A and AB under light load and >switches to full class B under heavier loads. Under >certain conditions of light load it has nasty output >distortion which vanishes as the amp is biased up to run >in true class B. Seen this happen with 741 op-amps driving 600 ohm headphones on a project we did as an exercise for apprentices. Solution was a 1k resistor from output to ground before the coupling capacitor so the op amp always had to drive current through the crossover region. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Yes, RF amps must have a load on the output and be properly matched to the load or there could be problems. Most properly designed RF amps have a "foldback" circuit to detect any mismatch by measuring the reflected power (power coming back) and thus reducing the RF drive back to a safe point to prevent self destruction. Better amps will actually shut down the output if the condition continues. I work with RF amps from 5 watts to 50,000 watts, all of which have foldback circuits and you'd be amazed at the extent the design engineers go through to provide this safety feature. Rick "Richards, Justin P" wrote: {Quote hidden}> Just thought I might clarify, the post was with regards to audio amps but > thought it best to mention I *think* RF power amps like to see a load or > they can fail. > > Is that accurate? > Justin > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

On Wed, 3 Jul 2002, Richards, Justin P wrote: ><snip> >Do some amplifiers also get damaged if the load impedence is *higher* >than the one specified? > >================== >never >================== >Just thought I might clarify, the post was with regards to audio amps but >thought it best to mention I *think* RF power amps like to see a load or >they can fail. It depends which kind of RF amp. Medium power and up certainly. Low power usually not unless it's really badly built. Peter -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

On Tue, 2 Jul 2002, Rick C. wrote: >No. It's like saying, do I need a load on an opamp? A higher impeance load >will just mean a lower power transfer to the load. > >Back in the vacuum tube days when you had a plate transformer matched to a >low impedance speaker, if you took the load (speaker) off and drove it hard, >huge amounts of voltage would be generated from inductive reactance that >would literally arc and burn the transformer up. >Rick Afair from 100Volt line adventures, removing the load allows the transformer to develop maximum undamped inductance and this transforms the amplifier into an oscillator (the feedback is usually from a secondary on the transformer) at the s.r.f. of the final transformer. This is due to the phase shift induced into the feedback loop. This is one of the reasons why tube amps work best with as low as possible feedback. Good units had a resistor across one of the transformer windings to prevent this. Feedback-less tube amps do not have this problem. Thay also have a snubber across the anode winding however. Peter -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

{Quote hidden}> -----Original Message----- > From: Russell McMahon [SMTP:spam_OUTapptechTakeThisOuTPARADISE.NET.NZ] > Sent: Wednesday, July 03, 2002 8:55 AM > To: .....PICLISTKILLspam@spam@MITVMA.MIT.EDU > Subject: Re: [OT]: Audio Amplifier > > > > Do some amplifiers also get damaged if the load impedance is *higher* > > > than the one specified? > > > No. It's like saying, do I need a load on an opamp? A higher impeance > load > > will just mean a lower power transfer to the load. > > > > Back in the vacuum tube days when you had a plate transformer matched to > a > > low impedance speaker, if you took the load (speaker) off and drove it > hard, > > huge amounts of voltage would be generated from inductive reactance that > > would literally arc and burn the transformer up. > > While in most cases the above answer is correct, there are situations > where > a load is needed. > (Incidentally to this, the transistor amp equivalent of a valve amp open > circuit is a short circuit - valve amps tended to survive these whereas > transistor amps need to be protected against them.) > > Some amplifiers perform 'differently" under little or no load. An LM324 > opamp is a specialist example - it has an "interesting" output stage which > runs in (AFAIR somewhere between class A and AB under light load and > switches to full class B under heavier loads. Under certain conditions of > light load it has nasty output distortion which vanishes as the amp is > biased up to run in true class B. > > A voltage regulator is a specialised amplifier with extra circuitry added > to > do a particular job. SOME voltage regulators become unhappy at very light > loads. > > An amplifier could be designed (and some are) to rely on output current > draw > to bias itself correctly. > > I think that may be slightly backwards, true class B has no biasing of the complimentry output stage which leads to crossover distortion. Class AB deliberataly biases both transistors into conduction which virtual eliminates crossover distortion at the expense of efficiency. IIRC the way to achieve minimal distortion with the LM324 was to use a resistor from the output to Vcc to bias it into Class A operation (again at the expense of efficiency). Regards Mike -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

>>Just thought I might clarify, the post was with regards to audio amps but >>thought it best to mention I *think* RF power amps like to see a load or >>they can fail. >It depends which kind of RF amp. Medium power and up certainly. Low power >usually not unless it's really badly built. When I was dealing with VHF power amps we were using transistors rated for 25W output (for a pair in push-pull) and they were rated for infinite VSWR on motor vehicle supplies (13.6V). -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

At 10:04 AM 7/3/02 +0100, Michael Rigby-Jones wrote: >IIRC the way to achieve minimal distortion with the LM324 was to use a >resistor from the output to Vcc to bias it into Class A operation (again at >the expense of efficiency). Actually, NS shows the resistor from the output to ground. The app notes suggest 6K2. dwayne -- Dwayne Reid <.....dwaynerKILLspam.....planet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 18 years of Engineering Innovation (1984 - 2002) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspam_OUTTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body