Does anybody use coaxial cable for wiring speaker output on the amplifier circuit board to speaker terminals on the amplifier chassis?
I have some Belden 1694 that has 19AWG core and 12AWG shield, the resistance of which is low enough for 150mm length of wire between the amp board and speaker terminals.
General Description of the cable: 19 AWG stranded (7x27) bare copper conductor, gas-injected foam HDPE insulation, double tinned copper braid shield (95% coverage), PVC jacket.
I think using coaxial cable may reduce EMI due to the high current of speaker wire, or does it not?
Is the coaxial cable made of materials suitable for carrying high current?
I have some Belden 1694 that has 19AWG core and 12AWG shield, the resistance of which is low enough for 150mm length of wire between the amp board and speaker terminals.
General Description of the cable: 19 AWG stranded (7x27) bare copper conductor, gas-injected foam HDPE insulation, double tinned copper braid shield (95% coverage), PVC jacket.
I think using coaxial cable may reduce EMI due to the high current of speaker wire, or does it not?
Is the coaxial cable made of materials suitable for carrying high current?
I wrapped around the PSU wires near the amp board with clamp-on ferrite and it made the sound clearer! I love it.
I am not sure if it is the effect of the ferrite filtering the RF or the ferrite wrapping around the PSU wires helping to shield the EMI. I guess it may be the later so I think coaxial cable for speaker wires may help. I happen to have the PSU and speaker output wiring points on the PCB right in the middle of the PCB so radiation at those points could affect the sensitive signal level circuits.
I can't think of a case that coaxial cable will make it worse. The only concern is that I don't know if a coaxial cable is up to the job for high current use.
I am not sure if it is the effect of the ferrite filtering the RF or the ferrite wrapping around the PSU wires helping to shield the EMI. I guess it may be the later so I think coaxial cable for speaker wires may help. I happen to have the PSU and speaker output wiring points on the PCB right in the middle of the PCB so radiation at those points could affect the sensitive signal level circuits.
I can't think of a case that coaxial cable will make it worse. The only concern is that I don't know if a coaxial cable is up to the job for high current use.
Last edited:
19awg may start to warm up slightly when asked to pass pass continuous high currents.
It's well insulated inside that foam dielectric. The foam may start to overheat and maybe melt?
The resistance of the 150mm of wire is actually quite low so will not interfere with the performance of the speakers.
A couple of metres to the speaker may be too high a resistance and adversely affect LF performance.
It's well insulated inside that foam dielectric. The foam may start to overheat and maybe melt?
The resistance of the 150mm of wire is actually quite low so will not interfere with the performance of the speakers.
A couple of metres to the speaker may be too high a resistance and adversely affect LF performance.
Thanks, DF and Andrew.
I was using twisted pair for it as well as for the PSU wires. Wrapping around the PSU wires with a few clamp-on ferrite improved the sound by much, so it might be due to common mode RF noise on the PSU lines, instead of the EMI from the PSU line radiating into the circuit board.
If there is a risk of using coaxial cable then I have to abandon the idea.
I was using twisted pair for it as well as for the PSU wires. Wrapping around the PSU wires with a few clamp-on ferrite improved the sound by much, so it might be due to common mode RF noise on the PSU lines, instead of the EMI from the PSU line radiating into the circuit board.
If there is a risk of using coaxial cable then I have to abandon the idea.
Calculated based on the datasheet, 150mm length will have a resistance of 0.0042R at 20 degree temperature. Consider 5A constant current, the dissipation would be 5 x 5 x 0.0043 = 0.1W.
Will 0.1W melt the cable?
It is possible that the current will be less than 5A constant current for music. For 5A constant current and 6R speakers, the speaker power dissipation will be 160W. I guess that would be very loud.
If the averaged current going through speaker wires is 1A for music, then the power dissipated in the 150mm coaxial cable will be 0.0042W. If 2A current, dissipation will be 0.017W. That is possibly not going to melt the cable.
Will 0.1W melt the cable?
It is possible that the current will be less than 5A constant current for music. For 5A constant current and 6R speakers, the speaker power dissipation will be 160W. I guess that would be very loud.
If the averaged current going through speaker wires is 1A for music, then the power dissipated in the 150mm coaxial cable will be 0.0042W. If 2A current, dissipation will be 0.017W. That is possibly not going to melt the cable.
Last edited:
A good read from Analog Devices....
http://www.analog.com/library/analogDialogue/archives/43-09/EDch 12 pc issues.pdf
The requirements for an unbalanced interconnect cable are fairly simple:
good screen coverage
concentric construction
low screen resistance
not too high capacitance
Any decent coaxial cable will do this. 75ohm cables will have less capacitance than 50 ohm cable. As any decent cable will do, it is not possible that one decent cable can be much better than some other decent cable.
good screen coverage
concentric construction
low screen resistance
not too high capacitance
Any decent coaxial cable will do this. 75ohm cables will have less capacitance than 50 ohm cable. As any decent cable will do, it is not possible that one decent cable can be much better than some other decent cable.
You are right that it is not much better and the 75ohm cables will have less capacitance than 50 ohm cable.
The Belden 9913F (RG8) has tinned copper braid as well as aluminium foil for insulation, whereas the 1964 has double tinned copper braids. So the 9913 may be better in shielding from very high frequency interference.
The Belden 9913F (RG8) has tinned copper braid as well as aluminium foil for insulation, whereas the 1964 has double tinned copper braids. So the 9913 may be better in shielding from very high frequency interference.
By the way, I am not sure if capacitance and inductance are as important as the shield resistance in an unbalanced interconnect cable. I think the latter is more important. This is because the RF roll off for solid state amps is at very high frequencies and there is no impact on the sound. On the other hand, the shield is used as the ground wire so any resistance in the shield introduces errors in the signal within audio band. I chose Belden 1964 because it has the lowest shield resistance I can find from any coaxial cable at 5.6mR per meter.
Resistance in the shield does not necessarily introduce "errors". That would depend on the details of the grounding arrangements for the system.
Capacitance can be an issue if the source has too high output impedance. Unlikely with standard solid-state circuits, but possible with some DIY or silly 'high-end' stuff.
Capacitance can be an issue if the source has too high output impedance. Unlikely with standard solid-state circuits, but possible with some DIY or silly 'high-end' stuff.
I would re-order the requirements for an unbalanced interconnect cable:
1] low screen resistance (low end-to-end resistance)
2] left blank
3] good screen coverage
4] not too high capacitance
5] concentric construction
******************************
Note that low screen resistance does depend on the system. It's more important in long cables than short (over 10 feet/ 3 meters).
Resistance in the shield does not necessarily introduce "errors". That would depend on the details of the grounding arrangements for the system.
1] low screen resistance (low end-to-end resistance)
2] left blank
3] good screen coverage
4] not too high capacitance
5] concentric construction
******************************
Note that low screen resistance does depend on the system. It's more important in long cables than short (over 10 feet/ 3 meters).
Resistance in the shield does not necessarily introduce "errors". That would depend on the details of the grounding arrangements for the system.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.