I got my first chip amp working and have just learnt what shielding really means! DUH! 
I thought it meant the plastic casing!
Now I have some mains hum in my chip amp which is earthed. Ok so I need to change the cables in it to shielded.
Some questions:
* where is shielding needed and not needed - where is hookup wire ok?
* how shielded? ... I have heavily sheilded cable which is about 6mm outer diam (interconnect) ... I also have cheaper 2.5mm lightly shielded cable - is the latter ok inside my chip amp for input wires?
* when do speaker level cables need to be shielded?
I thought it meant the plastic casing!
Now I have some mains hum in my chip amp which is earthed. Ok so I need to change the cables in it to shielded.
Some questions:
* where is shielding needed and not needed - where is hookup wire ok?
* how shielded? ... I have heavily sheilded cable which is about 6mm outer diam (interconnect) ... I also have cheaper 2.5mm lightly shielded cable - is the latter ok inside my chip amp for input wires?
* when do speaker level cables need to be shielded?
g'day Paul,
Speaker "level" cables shouldn't need shielding. This section of your amp is operating relatively high voltage levels, and - more importantly - low inpedance levels. Depending on your amp design (personally I don't know about the chip amp), the output may be of higher impedance at HF (way beyone human hearing). while not directly audible, it bay cause demodulation at audio frequencies. One way of solving this, is to a resistor and capacitor connected in series, and hook this in parallel with the speaker. I know this circuit has a commonly used name but I can't remember it right now.
Also, your internal power wires need not be shielded.
All your line-level connections should most certainly be shielded.
Your voltages are low, and impedance levels high(-er). This is where your problems come from. (If it isn't PSU ripple that you hear in your speakers as humming...)
The thickness of your shielded cable isn't that important. What really matters is the type of shield.
Some cheap cables are using shielding that is wound one way around the insulator of the inner conductor. This can cause situations which are worse than non-shielded cables, because the one-way circling shield can act as a long coil inductor, generating error voltages over the length of the shield.
you will want to use cable with a criss-cross shielding. (Don't know what to call it on a non-native language
) It'll have a lot of fine threads, half of which run clock-wise around the inner insulator, while the rest runs counter-clock-wise. This cancels out the error-voltage-induction very effectlively.
Hope this helps.
Jennice
Speaker "level" cables shouldn't need shielding. This section of your amp is operating relatively high voltage levels, and - more importantly - low inpedance levels. Depending on your amp design (personally I don't know about the chip amp), the output may be of higher impedance at HF (way beyone human hearing). while not directly audible, it bay cause demodulation at audio frequencies. One way of solving this, is to a resistor and capacitor connected in series, and hook this in parallel with the speaker. I know this circuit has a commonly used name but I can't remember it right now.
Also, your internal power wires need not be shielded.
All your line-level connections should most certainly be shielded.
Your voltages are low, and impedance levels high(-er). This is where your problems come from. (If it isn't PSU ripple that you hear in your speakers as humming...)
The thickness of your shielded cable isn't that important. What really matters is the type of shield.
Some cheap cables are using shielding that is wound one way around the insulator of the inner conductor. This can cause situations which are worse than non-shielded cables, because the one-way circling shield can act as a long coil inductor, generating error voltages over the length of the shield.
you will want to use cable with a criss-cross shielding. (Don't know what to call it on a non-native language
) It'll have a lot of fine threads, half of which run clock-wise around the inner insulator, while the rest runs counter-clock-wise. This cancels out the error-voltage-induction very effectlively.Hope this helps.
Jennice
In general, the hum problems need not be solved by any kind of shielding. The most frequent reason is a wrong layout of the PCB, and it can't be explained by few words...
Concerning the shielding of the loudspeaker cables, they exist certain special LS cables built as coaxial, so they are shielded inherently. In general, the shielding can reduce the RF interfereces, if properly connected to the amp earth. At least, the shield should be connected to the cold end of the output. For the bridged connection there should be used a symmetrical cable, i.e. two conductors (twisted) in a common shield eventually. Then the shield should be connected to the case.
Concerning the shielding of the loudspeaker cables, they exist certain special LS cables built as coaxial, so they are shielded inherently. In general, the shielding can reduce the RF interfereces, if properly connected to the amp earth. At least, the shield should be connected to the cold end of the output. For the bridged connection there should be used a symmetrical cable, i.e. two conductors (twisted) in a common shield eventually. Then the shield should be connected to the case.
could be the type of transformer your are using and the proximity of the PCB's to the transformer. While everyone loves torroids for their compact size, weight and efficiency they radiate about equally in a plane (well, perhaps more like a donut) about the center of the torroid, a standard old iron transformer will usually radiate elliptically around the unshielded ends of the trannie. in low noise applications this should be taken into consideration -- if you take apart just about anything from Hewlett Packard, Agilent, Tektronix, Krohn-Hite, Fluke, Boonton you will see that the power transformer and power circuitry is separated from the rest of the circuitry by a shield -- you can make a somewhat effective shield with a couple scrap pieces of printed circuit board.
What else can I use to shield the transformer? (I don't have scrap pcbs) It's a toroidal tranny and is 160 VA. I will look at cable runs and see if I can do it better. Will also look at the type of shielding on the cables I have. I can probably lay things out better. PCB layout should be fine, it was done by Rod Elliott at ESP, he knows his stuff.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.