Ok, I apologise for asking this very basic question, but I haven't found a simple answer anywhere.
I have a simple MOSFET amp (Maplin kit) using Exicon MOSFETS (I am ordering PCBs for the Aleph-X!). When I run the amp through a dummy load, I can hear the TO3 transistors "singing" to the input, not a resonant mode, but complete reproduction. Is this an example of magnetostriction, are the silicon dies lose or is there a perfectly simple explanation?
Once again, I am sorry if this is a rediculous question.
Andrew
I have a simple MOSFET amp (Maplin kit) using Exicon MOSFETS (I am ordering PCBs for the Aleph-X!). When I run the amp through a dummy load, I can hear the TO3 transistors "singing" to the input, not a resonant mode, but complete reproduction. Is this an example of magnetostriction, are the silicon dies lose or is there a perfectly simple explanation?
Once again, I am sorry if this is a rediculous question.
Andrew
Do you mean the TO3's are mechanically vibrating? If so that's too wierd to think about.
If you just mean that the input network seems to picking up signal from the output, then it is "just" a matter of shielding the input wiring and keeping the small signal stage away from the output stages. This is a common problem with a simple (but not always easy to implement) solution.
Or do you mean something else -- if so try restating a bit more descriptively and precisely.
If you just mean that the input network seems to picking up signal from the output, then it is "just" a matter of shielding the input wiring and keeping the small signal stage away from the output stages. This is a common problem with a simple (but not always easy to implement) solution.
Or do you mean something else -- if so try restating a bit more descriptively and precisely.
I've also heard this..not heard of someone hearing it, but actually hearing it myself.
Within the package is a pair of wirebonds. The source will be anywhere from a 5 to 20 mil thick aluminum wire...the gate can be anything from .7 mil gold up to 5 mil aluminum.
If the source wire is vibrating, I'd be afraid of the reliability of the device, as the wire would tend to fatigue. (But, perhaps that is only an ultrasonic resonance issue, one I've been bitten by.)
My gut feeling is that it's not an internal thing, but the wires connecting the source or drain moving in response to the current within, and the to-3 is larger surface area to radiate it.
Neat, isn't it? who needs speakers??
BTW..if the silicon were loose, it would blow in less than a second as there would be no heat path keeping it from vaporizing.
Cheers, John
PS...it's not a ridiculous question..
Within the package is a pair of wirebonds. The source will be anywhere from a 5 to 20 mil thick aluminum wire...the gate can be anything from .7 mil gold up to 5 mil aluminum.
If the source wire is vibrating, I'd be afraid of the reliability of the device, as the wire would tend to fatigue. (But, perhaps that is only an ultrasonic resonance issue, one I've been bitten by.)
My gut feeling is that it's not an internal thing, but the wires connecting the source or drain moving in response to the current within, and the to-3 is larger surface area to radiate it.
Neat, isn't it? who needs speakers??
BTW..if the silicon were loose, it would blow in less than a second as there would be no heat path keeping it from vaporizing.
Cheers, John
PS...it's not a ridiculous question..
thank you for the replies. It would appear that it is not an especially common occurance then. Both of my amp channels exhibit this behaviour, and in both N and P channel devices.
The only thing that I can think of, is that these are "lateral power devices", would this have any bearing on the noise output? I don't have the circuit diagram to hand here, but I have included the details of the tranistor below (the N and P are identical in these figures).
ECF10P16
Ptot(max) 125W
VDSmax 160V
VGSmax ±14V
IDmax 8A
IDdiode 8A
I/P cap 500nF
IDSSmax 10ma
One last thought is that these MOSFETS have some kind of diode inside them, could it be this that is producing the audible noise? It is "quite" loud, in that at about 100W from the amplifier (dummy load) with the lid off, I can hear it about 1m away.
I am not too worried though, a quality amplifier will follw (I hope).
Thank you for all your help,
Andrew
The only thing that I can think of, is that these are "lateral power devices", would this have any bearing on the noise output? I don't have the circuit diagram to hand here, but I have included the details of the tranistor below (the N and P are identical in these figures).
ECF10P16
Ptot(max) 125W
VDSmax 160V
VGSmax ±14V
IDmax 8A
IDdiode 8A
I/P cap 500nF
IDSSmax 10ma
One last thought is that these MOSFETS have some kind of diode inside them, could it be this that is producing the audible noise? It is "quite" loud, in that at about 100W from the amplifier (dummy load) with the lid off, I can hear it about 1m away.
I am not too worried though, a quality amplifier will follw (I hope).
Thank you for all your help,
Andrew
Oh, and sam9, it is a vibration from inside the TO3. I wasn't going to put my ear on the case because there are a few volts flying around, but you can localise it to them by getting close and shielding the noise with your hand. There is nothing else close enough to produce the noise.
As long as it isn't a normal occurance that simply isn't mentioned because it is so obvious, I am not worried. See no evil, hear only the music!
Andrew
As long as it isn't a normal occurance that simply isn't mentioned because it is so obvious, I am not worried. See no evil, hear only the music!
Andrew
Andrew,
I have a very hard time believing this, however you can test this out with out putting your ear even close to the device that is making this noise. You need say about a 26mm x 26mm x 400mm about piece of wood and press it against the part, then press the other end against you ear, if you can physically get to the part. The sound energy will conduct vibrations through the wood or some other dense material to your ear.
The bonding wires being so short could only vibrate at some high frequency. In addition, if were doing such a crazy thing the bond would loosen over time causing a failure.
I have a very hard time believing this, however you can test this out with out putting your ear even close to the device that is making this noise. You need say about a 26mm x 26mm x 400mm about piece of wood and press it against the part, then press the other end against you ear, if you can physically get to the part. The sound energy will conduct vibrations through the wood or some other dense material to your ear.
The bonding wires being so short could only vibrate at some high frequency. In addition, if were doing such a crazy thing the bond would loosen over time causing a failure.
Actually this is a pretty common phenomonon and I used to hear this noise all the time when I used to repair Phase Linears, ESS, and Crown amplifiers back in the mid 70's. The higher power the amp the more you would hear it sing. I believe the noise is caused by current induction in the wiring itself... somewhat similar to the buzzing noise that is heard in a power transformer. I've found that almost all high power amps sing to some extent...what they sing just depends on what you feed them. I've even been able to feed an amp from an FM tuner and actually hear dialog eminating from amplifiers by running an amp near wide open into a static 8 ohm load.
Perhaps someone like Nelson Pass would be able to better explain it.
Mark
Perhaps someone like Nelson Pass would be able to better explain it.
Mark
Mark Gulbrandsen,
What have you been smoking, send me some. I've worked on a truck load of amps even some of those crapy Phase Linears, MAC's Crown DC300 to Ampzillas and never hear this. Maybe its the amps crying not singing, I could understand a Carver amp doing that.
Do you hear this under load or no load?
What have you been smoking, send me some. I've worked on a truck load of amps even some of those crapy Phase Linears, MAC's Crown DC300 to Ampzillas and never hear this. Maybe its the amps crying not singing, I could understand a Carver amp doing that.
Do you hear this under load or no load?
From inside the TO3. It sounds wierd but since I used to broadcast AM and CB on my turntable (*HONEST!*) back when I still had one I shouldn't be sketical. Anyway, if I was getting sound out of my TO3s I would soon consider goint to TO-247s if I couldn't find the cause. After fighting ground loop hums, buzzing transformers and some other annoyances in the past I'd be inclined to go to work on the TO3s with a hammer!
I can confirm this phenomenon as well, though I did never bother to find out where it comes from (even though I was wondering).
AFAIK I even heard it without load connected on a receiver. We just turned the volume up with inter-station noise and one yould clearly hear the sound emerging from the area of the output stages. And it wasn't restricted to amps with TO3 devices.
Regards
Charles
AFAIK I even heard it without load connected on a receiver. We just turned the volume up with inter-station noise and one yould clearly hear the sound emerging from the area of the output stages. And it wasn't restricted to amps with TO3 devices.
Regards
Charles
I've Always Thought That This Is Just Normal.....
Hi Andrew, your question is not even slightly ridiculous, and top marks to you for being observant.
A couple of posters have stated they have heard this effect, as have I also.
The first time I encountered this was in the first amplifier that I built - a bridging output 13.8V powered IC amplifier.
The thin U-shaped heatsink mounted on top the IC's sang in sympathy to the music with the amplifier unloaded.
The sound was thin and tizzy, but was recognisable as music.
Damping the heatsinks by squeezing them reduced the level, but did not eliminate this sonic emission.
I did not get around to finding the root cause, but I don't think that it is related to bonding wires - these IC's were fully encapsulated so there was no internal room for the bonding wires to move.
I also regularly observe this effect in standard domestic amplifiers driven into dummy loads or unloaded, and the level and nature of this sonic emission substantially changes when the amp is driven into clipping.
Perhaps there is a differential themal expansion or piezo electric effect in the device chips at work here - at present time I do not know the mechanism responsible but be comforted that you are not the only one to observe this effect.
Eric.
"That is totally off in the outer space region of our Galaxy. "
Jim W. - And you call yourself an engineer ?????.
Hi Andrew, your question is not even slightly ridiculous, and top marks to you for being observant.
A couple of posters have stated they have heard this effect, as have I also.
The first time I encountered this was in the first amplifier that I built - a bridging output 13.8V powered IC amplifier.
The thin U-shaped heatsink mounted on top the IC's sang in sympathy to the music with the amplifier unloaded.
The sound was thin and tizzy, but was recognisable as music.
Damping the heatsinks by squeezing them reduced the level, but did not eliminate this sonic emission.
I did not get around to finding the root cause, but I don't think that it is related to bonding wires - these IC's were fully encapsulated so there was no internal room for the bonding wires to move.
I also regularly observe this effect in standard domestic amplifiers driven into dummy loads or unloaded, and the level and nature of this sonic emission substantially changes when the amp is driven into clipping.
Perhaps there is a differential themal expansion or piezo electric effect in the device chips at work here - at present time I do not know the mechanism responsible but be comforted that you are not the only one to observe this effect.
Eric.
"That is totally off in the outer space region of our Galaxy. "
Jim W. - And you call yourself an engineer ?????.
A singing contest...
Hi
Me to !! I also have observed this "singing" that become louder as the load impedance become lower...that means that is a current dependent issue.!!
The some happen with output transformers in valve amplifiers when conected to dummy loads...
The vibrational hum of a mains transformer is also the "singing" of the transformer...in this case with a only "note" 50 Hz here in Europe.
Cheers
Hi
Me to !! I also have observed this "singing" that become louder as the load impedance become lower...that means that is a current dependent issue.!!
The some happen with output transformers in valve amplifiers when conected to dummy loads...
The vibrational hum of a mains transformer is also the "singing" of the transformer...in this case with a only "note" 50 Hz here in Europe.
Cheers
Black arts
Richard,
thant is exactly what I have been wondering. I have always dismissed the possibility of microphony in modern solid state electronics, and still regard it with some suspicion ("I tried this £10,000 stand and it made my system sound twice as clear" sort of comments I doubt, but then, I have never listened to a decent commercial system and am willing to be proved wrong).
I was going to put the sound down to magentostriction with the package, but if it also occurs in IC amps, and plastic devices, I don't know. Purely out of interest, I wonder if it would occur with a couplig material (to the heatsink) that wasn't metallic. For instance would using a Peltier, or TE, device reduce this noise? Not something I can try with TO3's. When I have my equipment back at home over the summer I will try experimenting. I am at university at the moment, without the tools or space!
Richard,
thant is exactly what I have been wondering. I have always dismissed the possibility of microphony in modern solid state electronics, and still regard it with some suspicion ("I tried this £10,000 stand and it made my system sound twice as clear" sort of comments I doubt, but then, I have never listened to a decent commercial system and am willing to be proved wrong).
I was going to put the sound down to magentostriction with the package, but if it also occurs in IC amps, and plastic devices, I don't know. Purely out of interest, I wonder if it would occur with a couplig material (to the heatsink) that wasn't metallic. For instance would using a Peltier, or TE, device reduce this noise? Not something I can try with TO3's. When I have my equipment back at home over the summer I will try experimenting. I am at university at the moment, without the tools or space!
- Status
- Not open for further replies.