If you really want to measure the current and use that as a means of tripping a relay then swap the position of the sense resistor and speaker in the circuit.
As it is you have the possibility of applying signals outside the 741's range (it's supplies). Swapping them round still gives the same volt drop to work with, so design a comparator etc to work with it.
Another idea (and these are just ideas for you... I wouldn't go down this line of protection) is to use opto isolators across the emmiter resistors in the amp. Excess current causes LED part to light. Then you use that as a control signal to a relay driver etc.
The trouble with this idea is that any DC current below the trip level is classed by your circuit as acceptable... and it's not... a few amps (much less in many cases) will destroy a speaker, but your circuit would say that was OK and a valid current.
But, will it warn at -3 dB, detect clipping of either polarity and stretch brief clips to make them visible?
A version with fewer components from Rod Elliott:
Power Amplifier Clipping Indicator
There was the "Bilateral Clipping Indicator" in Audio Amateur (1975); that lacked a pulse stretcher, but it did use very few components.
Short circuit
Its Clear that he needs a MAJOR thing in that protection circuit, witch is the SHORT CIRCUIT PROTECTION.
I have no idea, if any one from all of you, DEVELOPED a protection circuit, THAT Disconnects the (Load) in case of COMPLETE short circuit at the amplifiers output, while NOT blowing up the amplifier. even for 1 hour
FOX
Its Clear that he needs a MAJOR thing in that protection circuit, witch is the SHORT CIRCUIT PROTECTION.
I have no idea, if any one from all of you, DEVELOPED a protection circuit, THAT Disconnects the (Load) in case of COMPLETE short circuit at the amplifiers output, while NOT blowing up the amplifier. even for 1 hour
FOX
A small contribution...
http://www.diyaudio.com/forums/soli...tegrated-247-output-devices-7.html#post649777
http://www.diyaudio.com/forums/soli...tegrated-247-output-devices-7.html#post649777
Protection circuits and pcbs you can find here:
http://www.diyaudio.com/forums/solid-state/164093-100w-ultimate-fidelity-amplifier-10.html
Regards
http://www.diyaudio.com/forums/solid-state/164093-100w-ultimate-fidelity-amplifier-10.html
Regards
series resistor protection
I am not qualified to criticize the more complicated protection circuits, but I can contribute to Phoenix87's circuit. The measurement of DC overcurrent is interesting but much less sophisticated than the VI limit implemented with microprocessors discussed in the early part of this thread. Transistors have a "Safe Operating Area" that allows them to put out a lot of voltage for a little while, or a lot of current for a little while, but not both simultaneously. If your music source is the cannon shot of 1812 Orchestra (see music thread how to test your system recently) then you want to allow that current - which is a very short time. If your music source is solo Japanese flute, you have the transistor putting out almost no current while holding off the full rail voltage. Both conditions are okay. But if your speaker wire is shorted or crossover cap blown, then there is too much current for too much time. The mathmatics of the SOA model is so complicated that it is best implemented in software on a microprocessor, but MP development systems, programming software, and PROM burners are expensive accessories for a one off home project. Instead of the series resistor of Phoenix87's diagram, I would vote for current transformers (of which I have a couple in blown up motor drives) or as previously suggested, linear transfer opto isolators. However, once you know the output current, you have to integrate it over time and compare against a history curve of what has happened previously - mathematics that is difficult to do with discretes or op amps. Then if the math says bad, what are you going to do? Mr Sakis superrepairman says opening the output relays causes problems -I am suspicious myself of sudden changes in load impedance. Mr. Peavey in the CS800 has opted to crowbar the output with a triac and let the power supply fuse blow. That probably limits speaker damage if you use his 500 watt music power speakers, and limits damage to the amp, but is a one time event. I was hoping an early responder would point out to me a microprocessor chip and development system including software and PROM blower and integrated A/D function that is under $300 these days. But that is probably a yellow brick road type of situation.
ApexAudio, I see you are monitoring heat sink temperature to turn on the fan and then pull the speaker relays. DJoffe & I did something similar on the ST120 modification he came up with, only we cut the output tsistor bias current in half instead of pulling the speaker relays. I wonder if heat sink temperature lags the real problem, which is transistor die temperature (predicted by SOA math). I don't have the experience to know if the heat sink mostly heats up before the output transistors blow.
I am not qualified to criticize the more complicated protection circuits, but I can contribute to Phoenix87's circuit. The measurement of DC overcurrent is interesting but much less sophisticated than the VI limit implemented with microprocessors discussed in the early part of this thread. Transistors have a "Safe Operating Area" that allows them to put out a lot of voltage for a little while, or a lot of current for a little while, but not both simultaneously. If your music source is the cannon shot of 1812 Orchestra (see music thread how to test your system recently) then you want to allow that current - which is a very short time. If your music source is solo Japanese flute, you have the transistor putting out almost no current while holding off the full rail voltage. Both conditions are okay. But if your speaker wire is shorted or crossover cap blown, then there is too much current for too much time. The mathmatics of the SOA model is so complicated that it is best implemented in software on a microprocessor, but MP development systems, programming software, and PROM burners are expensive accessories for a one off home project. Instead of the series resistor of Phoenix87's diagram, I would vote for current transformers (of which I have a couple in blown up motor drives) or as previously suggested, linear transfer opto isolators. However, once you know the output current, you have to integrate it over time and compare against a history curve of what has happened previously - mathematics that is difficult to do with discretes or op amps. Then if the math says bad, what are you going to do? Mr Sakis superrepairman says opening the output relays causes problems -I am suspicious myself of sudden changes in load impedance. Mr. Peavey in the CS800 has opted to crowbar the output with a triac and let the power supply fuse blow. That probably limits speaker damage if you use his 500 watt music power speakers, and limits damage to the amp, but is a one time event. I was hoping an early responder would point out to me a microprocessor chip and development system including software and PROM blower and integrated A/D function that is under $300 these days. But that is probably a yellow brick road type of situation.
ApexAudio, I see you are monitoring heat sink temperature to turn on the fan and then pull the speaker relays. DJoffe & I did something similar on the ST120 modification he came up with, only we cut the output tsistor bias current in half instead of pulling the speaker relays. I wonder if heat sink temperature lags the real problem, which is transistor die temperature (predicted by SOA math). I don't have the experience to know if the heat sink mostly heats up before the output transistors blow.
Last edited:
I'm pretty sure I spotted a chip that had some on-chip debugging interface (like JTAG) while browsing the Digikey catalog. Probably an Atmel or PIC. JTAG interfaces are cheap to make or buy.
I thought I'd seen an ad for some chip that provided SOA protection, but I guess I was thinking of NS's SPiKe circuit that's used in the chip amps. AN-898
I thought I'd seen an ad for some chip that provided SOA protection, but I guess I was thinking of NS's SPiKe circuit that's used in the chip amps. AN-898
sakis my dear Greek friend
Concider a modified McIntosh protection circuit......
When max output power is reached, the circuit aut. lowers the input to the VAS stage. Nice and easy and no distortion
Crown Audio's ODEP works that way, well in a more elaborate way
Here is something I knocked together that works but can be refined. It measures the maximum voltage swing that you may want to allow across the speaker. The voltage would be more or less proportional to the current although the current may lag or lead the voltage somewhat, but you can set it to a level that is meaningful.
I used two op-amps in a threshold detection mode. The one op-amp triggers from a negative half cycle and the other from a positive. The potentiometer would set both op amps thresholds to the same but opposite value, thus they would be triggered symmetrically.
The outputs of the op-amps are OR by the two diodes D5 & D5. If the transistor Q1 is triggered by either op-amps it conducts and turns the relay on and disconnects the speakers, while Q2 latches Q1 and to reset the protection circuit the power has to be turned off.
Obviously this will not protect the amp against short circuit but will protect your speakers from a fault in the amp.
One could toy around with the previous idea of measuring the current through a series resistor and thereby disconnect the load if the amp is short circuit.
Anyway my ten cents worth.
Regards
Nico
I used two op-amps in a threshold detection mode. The one op-amp triggers from a negative half cycle and the other from a positive. The potentiometer would set both op amps thresholds to the same but opposite value, thus they would be triggered symmetrically.
The outputs of the op-amps are OR by the two diodes D5 & D5. If the transistor Q1 is triggered by either op-amps it conducts and turns the relay on and disconnects the speakers, while Q2 latches Q1 and to reset the protection circuit the power has to be turned off.
Obviously this will not protect the amp against short circuit but will protect your speakers from a fault in the amp.
One could toy around with the previous idea of measuring the current through a series resistor and thereby disconnect the load if the amp is short circuit.
Anyway my ten cents worth.
Regards
Nico
I understand what Sakis means, pro audio amplifiers should work even on a "partially" shorted load, well as long there is still a usable winding on the speaker's voice coil. You cannot afford an amplifier cutting in and out of protection/muting on a live gig, that would be embarrassing.
Some pro audio amplifier uses thermistor mounted on the main heatsink to reduce the bias level of the output stage, probably from class AB level to class B level. Some designs employs a thermistor on the bias network of the transistor that "clamps" the VAS stage to reduce the drive when the heatsink is hot.
Some pro audio amplifier uses thermistor mounted on the main heatsink to reduce the bias level of the output stage, probably from class AB level to class B level. Some designs employs a thermistor on the bias network of the transistor that "clamps" the VAS stage to reduce the drive when the heatsink is hot.
serious pro audio
For serious pro audio where the show must go an reguardless of cost, Peavey has a PA management system that integrates multiple amps and multiple speakers, with a console where the sound man can send the sound from different amps to different speakers if there is a problem. Thousands of dollars. Uses stakon connectors, probably more transistor friendly than 1/4" phono plug that shorts almost every time it is inserted or tripped over and pulled out. Presumably the performance speakers are tested out before the amp is ever connected to them. As a home user, I'm just trying to avoid blowing a $250 set of speaker drivers if a $10 transistor goes out. Or avoid blowing up my amp if the dog yanks on the speaker wire and shorts the spade lug to the chassis.
For serious pro audio where the show must go an reguardless of cost, Peavey has a PA management system that integrates multiple amps and multiple speakers, with a console where the sound man can send the sound from different amps to different speakers if there is a problem. Thousands of dollars. Uses stakon connectors, probably more transistor friendly than 1/4" phono plug that shorts almost every time it is inserted or tripped over and pulled out. Presumably the performance speakers are tested out before the amp is ever connected to them. As a home user, I'm just trying to avoid blowing a $250 set of speaker drivers if a $10 transistor goes out. Or avoid blowing up my amp if the dog yanks on the speaker wire and shorts the spade lug to the chassis.
we are neither on the one nor the other side ....i haven seen amps in my place the latest 25 years that use 1/4" output plugs .... all of them uses speakon ...
then again non of us is ever going to come up with a multi thousand dollar sound management thingy ... including me that actually been looking to something more "earthy " protection enough without too much messing up with the signal ....
then again non of us is ever going to come up with a multi thousand dollar sound management thingy ... including me that actually been looking to something more "earthy " protection enough without too much messing up with the signal ....
time marches on
Uh, The Peavey CS800S amp I have that I bought but I've been afraid to use was built in 1998. The schematic of the Peavey CS800 I have been referring to in this thread, that I got by searching diyaudio, is a revision B dated 1976. The revision B has a diac-triac speaker voltage clamp, and no speaker fuses or disconnect relays. The CS800S that I bought has a "DDT" switch on the front panel that is not on the schematic. I bet one "D" stands for dynamic. Does anybody have a schematic for a Peavey CS800S? I might have a protection circuit as good as the Crown whose schematic was posted on a previous protection thread as being worth of emulation. (The crown amp had a PIC16C57 IC deciding what to do with the temperature and other inputs, that I assumed was a microprocessor because it had a clock input).
Uh, The Peavey CS800S amp I have that I bought but I've been afraid to use was built in 1998. The schematic of the Peavey CS800 I have been referring to in this thread, that I got by searching diyaudio, is a revision B dated 1976. The revision B has a diac-triac speaker voltage clamp, and no speaker fuses or disconnect relays. The CS800S that I bought has a "DDT" switch on the front panel that is not on the schematic. I bet one "D" stands for dynamic. Does anybody have a schematic for a Peavey CS800S? I might have a protection circuit as good as the Crown whose schematic was posted on a previous protection thread as being worth of emulation. (The crown amp had a PIC16C57 IC deciding what to do with the temperature and other inputs, that I assumed was a microprocessor because it had a clock input).
Last edited:
here is some thing i found out have a look everybody
regards
aniket
An externally hosted image should be here but it was not working when we last tested it.
regards
aniket
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.