I've seen a few small amplifier circuits that included some inductor coils near the output, and I have run into an article that talks about benefits of CM chokes for Op-Amp circuits applied at both the inputs and outputs to reduce interference. Looking at the circuit diagram, below do you think the CM chokes idea has any merit for gainclone amps or is it unnecessary given the design parameters typical in these circuits?
The same document also mentioned mains filtering. Led me to wonder if anyone is using this type of filter in their amps. I notice that computer power supplies certainly do, but these are switching type, so perhaps that is the reason? I'm having a little difficulty locating X1/Y2 caps online, so far 10,000 pf is the largest value I have seen (was looking for .1mfd since that was recommended in Nuuk's PSU how-to, but will use the .01 if that's the only thing out there). I read some technical specs from Japan about how they do the certification testing for these caps, it was pretty impressive.
Anyway, the document also discusses cable interconnects between devices, proper connection of shields, etc. The electronics involved in this document are way above my hobbyist paygrade, but certainly interesting...
Design Techniques for EMC – Part 1 Circuit Design, and Choice of Components
By Eur Ing Keith Armstrong CEng MIEE MIEEE
Partner, Cherry Clough Consultants, Associate of EMC-UK
The full document is available here:
http://www.wsfamily.idv.tw/WENG/emc/emc-designtech.doc

The same document also mentioned mains filtering. Led me to wonder if anyone is using this type of filter in their amps. I notice that computer power supplies certainly do, but these are switching type, so perhaps that is the reason? I'm having a little difficulty locating X1/Y2 caps online, so far 10,000 pf is the largest value I have seen (was looking for .1mfd since that was recommended in Nuuk's PSU how-to, but will use the .01 if that's the only thing out there). I read some technical specs from Japan about how they do the certification testing for these caps, it was pretty impressive.

Anyway, the document also discusses cable interconnects between devices, proper connection of shields, etc. The electronics involved in this document are way above my hobbyist paygrade, but certainly interesting...
Design Techniques for EMC – Part 1 Circuit Design, and Choice of Components
By Eur Ing Keith Armstrong CEng MIEE MIEEE
Partner, Cherry Clough Consultants, Associate of EMC-UK
The full document is available here:
http://www.wsfamily.idv.tw/WENG/emc/emc-designtech.doc
CM chokes in the audio circuit are neither required nor recommendable for chipamps. The small inductor coils you have probably seen, are there to protect the amplifier, if loudspeakers with very high capacitance are connected.stven said:I've seen a few small amplifier circuits that included some inductor coils near the output, and I have run into an article that talks about benefits of CM chokes for Op-Amp circuits applied at both the inputs and outputs to reduce interference. Looking at the circuit diagram, below do you think the CM chokes idea has any merit for gainclone amps or is it unnecessary given the design parameters typical in these circuits?
Yes, some people do. Those filters can help, if you live in a region with a very contaminated mains supply. If your mains are clean, these filters do nothing, i. e. they do no harm to the amplifier or the audio signal. They can however produce convenience tripping of earth-leakage protectors.stven said:The same document also mentioned mains filtering. Led me to wonder if anyone is using this type of filter in their amps.
Common values for those capacitors are less than 1 µF. Are you refering to the 100 nF capacitors in the "Mains Conditioning" section? There is also an 8 µF capacitor mentioned, and a motor starter capacitor was chosen, because the X- and Y-types are not available that big.stven said:I'm having a little difficulty locating X1/Y2 caps online, so far 10,000 pf is the largest value I have seen (was looking for .1mfd since that was recommended in Nuuk's PSU how-to, but will use the .01 if that's the only thing out there).
X-1
Y-2
with the proliferation of products with radios in the home (wifi, protable phones, cell phones, bluetooth, satellites etc) I believe it is wise to design equipment with this philosophy. In fact, if you want to sell in Europe, it's the law. CE certification requires that the product is designed to work reliably with all interference, conducted or radiated from DC to 3GHz or more.
When I last checked my mains, the harmonics were just under the regulatory limits for mains distortion and a wideband measurement showed several volts of noise on the waveform. I don't have RF analyzers but I'd be surprised if the spectrum didn't continue way into the MHz.
Obviously, if your system is quiet and sounds great then no worries, you live in a "clean" location but sadly I don't think that's true for most of us.
When I last checked my mains, the harmonics were just under the regulatory limits for mains distortion and a wideband measurement showed several volts of noise on the waveform. I don't have RF analyzers but I'd be surprised if the spectrum didn't continue way into the MHz.
Obviously, if your system is quiet and sounds great then no worries, you live in a "clean" location but sadly I don't think that's true for most of us.
Your answers are most helpful. Also thanks for the links to the caps. I think I will go for the mains filtering on my next chipamp project (3886 stereo).
Steve
Steve
I was just about to make a new post about this in the power supply section, but since there's already a thread with some answers here - here goes:
About the main filtering - I couldn't exactly figure out what is the use of the Y caps that couple L and N to E? Most designs utilize a CM choke on L and N, a 0.33uF-ish X2 cap between L and N and a parallel resistor. Some even include posts for the Y caps, but they're not installed. Just wondered why some designs decide they're not important.
I'm making a small filter on a prototype board, using salvaged computer SMPS parts (where the choke will have amp ratings higher than my mains fuses) and some thermistors for power-on surges. I'd just like to know if there's a point to allocate space for the Y caps or not 🙂
About the main filtering - I couldn't exactly figure out what is the use of the Y caps that couple L and N to E? Most designs utilize a CM choke on L and N, a 0.33uF-ish X2 cap between L and N and a parallel resistor. Some even include posts for the Y caps, but they're not installed. Just wondered why some designs decide they're not important.
I'm making a small filter on a prototype board, using salvaged computer SMPS parts (where the choke will have amp ratings higher than my mains fuses) and some thermistors for power-on surges. I'd just like to know if there's a point to allocate space for the Y caps or not 🙂
Atilla said:About the main filtering - I couldn't exactly figure out what is the use of the Y caps that couple L and N to E? Most designs utilize a CM choke on L and N, a 0.33uF-ish X2 cap between L and N and a parallel resistor. Some even include posts for the Y caps, but they're not installed. Just wondered why some designs decide they're not important.
Without the Y-caps you get a simple 12 dB low pass filter. Excess energy, which distortions bring into the circuit, remains there. Some of it is converted into heat. the rest is damped, but still there.
With Y-caps excess energy is discharged into earth. That makes the filter more efficient. Trade-offs are, filters with Y-caps can lead to convenience tripping of earth-leakage detectors, and they are more expensive. When you use them, you must make sure that PE is connected. If it is not connected, the capacitors form a voltage divider across L and N, thus bringing the chassis to half mains voltage.
exactly right, PacificBlue!
Atilla,
Most manufacturers add just enough components to meet regulatory requirements for immunity and radiation. One product with a big noisy power oscillator (e.g. PC clock) will need more filtering on all ports than a small phono pre-amp (for example)
As a DIY'er we can overdesign to our hearts content and aim for as much attenuation as possible. Can't hurt eh? Just watch out for capacitor leakage currents.
edit: and make sure you use mains rated caps! There's a special grade for this duty. 🙂
Atilla,
Most manufacturers add just enough components to meet regulatory requirements for immunity and radiation. One product with a big noisy power oscillator (e.g. PC clock) will need more filtering on all ports than a small phono pre-amp (for example)
As a DIY'er we can overdesign to our hearts content and aim for as much attenuation as possible. Can't hurt eh? Just watch out for capacitor leakage currents.
edit: and make sure you use mains rated caps! There's a special grade for this duty. 🙂
Thanks guys, that was quite straightforward.
I'll make places for the Y caps, but leave them unpopulated for now. Shuko plugs can go in both grounded and ungrounded sockets and around here it is, sadly, pretty common to encounter power outlets without a ground wire and pins. Chasis on 125volts is not something I want to be touching.
I'll make places for the Y caps, but leave them unpopulated for now. Shuko plugs can go in both grounded and ungrounded sockets and around here it is, sadly, pretty common to encounter power outlets without a ground wire and pins. Chasis on 125volts is not something I want to be touching.
- Status
- Not open for further replies.
- Home
- Amplifiers
- Chip Amps
- Question about chokes