Ah, just a few things from me today. 😉
It might help to:
Make the ground wires from all 4 amps to be exactly the same length. That will help eliminate differences that cause ground loops.
Put some black plastic automotive door edge guard (edit: that's "C" channel that slips over the edge) between the sawn edge of the heat spreader bar. . . and the transformer. That edge is a bit close.
Have the ground (s) never more than 1/4" away from the live, whenever you are near a signal area. Point blank, as in zip-tied or woven, is even nicer. Your photo looks fine. 😉
Possibly isolate the potentiometer (if present) from the metal case--well, just try it, sometimes less hum, sometimes more. Anyway, Signal lines to some potentiometers can make a ground loop because some potentiometers ground onto amplifier case. Also, some potentiometers "tie" the ground lines together.
OH!!! Signal lines! Any signal lines need to be "about" the same length per amplifier channel. Their grounds will not always meet together, but there is a probability.
Are the RCA jacks and speaker terminal ground lines isolated from the chassis? Those ground lines go "each to its own" but I couldn't see that part of the photo.
DC output: Did you measure? I ask, because the amp looks great, but DC output is one sure fire way to tell about the success of the wiring layout. 😉
Ah, I didn't see either "C" an input filter cap or Ci, so if you plug in a source that outputs DC, there could be many volts DC go right into your speakers. So, check to make sure that your preamplifier protects the amp from DC on the signal line.
I like that amp because its elegant.
It might help to:
Make the ground wires from all 4 amps to be exactly the same length. That will help eliminate differences that cause ground loops.
Put some black plastic automotive door edge guard (edit: that's "C" channel that slips over the edge) between the sawn edge of the heat spreader bar. . . and the transformer. That edge is a bit close.
Have the ground (s) never more than 1/4" away from the live, whenever you are near a signal area. Point blank, as in zip-tied or woven, is even nicer. Your photo looks fine. 😉
Possibly isolate the potentiometer (if present) from the metal case--well, just try it, sometimes less hum, sometimes more. Anyway, Signal lines to some potentiometers can make a ground loop because some potentiometers ground onto amplifier case. Also, some potentiometers "tie" the ground lines together.
OH!!! Signal lines! Any signal lines need to be "about" the same length per amplifier channel. Their grounds will not always meet together, but there is a probability.
Are the RCA jacks and speaker terminal ground lines isolated from the chassis? Those ground lines go "each to its own" but I couldn't see that part of the photo.
DC output: Did you measure? I ask, because the amp looks great, but DC output is one sure fire way to tell about the success of the wiring layout. 😉
Ah, I didn't see either "C" an input filter cap or Ci, so if you plug in a source that outputs DC, there could be many volts DC go right into your speakers. So, check to make sure that your preamplifier protects the amp from DC on the signal line.
I like that amp because its elegant.
Hi,
I don't know if there is a rule governing the choice of cap for use in the audio ground to Safety Earth disconnecting network.
Thinking about it, I cannot see the risk of high mains voltage appearing across this cap except in the failure scenario. During the kA failure condition the inverse diodes try to carry the fault current to Safety Earth and having an X or Y rated cap in parallel seems to me to be irrelevant.
I think that poly anything caps will do for this duty but I tend to put in a little 100nF ceramic to maximise attenuation of the highest frequencies.
I do not like the idea of directly connecting all the audio grounds to chassis.
More importantly the thought of distributing all these grounds to various locations on the same piece of aluminium seems abhorrent.
However, each channel could have it's own audio ground separate from all the other channels. Then decide (experiment with) connecting the four audio grounds directly to or via four disconnecting networks to the chassis.
The Safety Earth connection must have it's own permanent fixing to chassis, not shared with other ground wires. It is OK to fix the Safety wire under it's own permanent nut and then place all the other grounds above that and secured with a separate nut.
I don't know if there is a rule governing the choice of cap for use in the audio ground to Safety Earth disconnecting network.
Thinking about it, I cannot see the risk of high mains voltage appearing across this cap except in the failure scenario. During the kA failure condition the inverse diodes try to carry the fault current to Safety Earth and having an X or Y rated cap in parallel seems to me to be irrelevant.
I think that poly anything caps will do for this duty but I tend to put in a little 100nF ceramic to maximise attenuation of the highest frequencies.
I do not like the idea of directly connecting all the audio grounds to chassis.
More importantly the thought of distributing all these grounds to various locations on the same piece of aluminium seems abhorrent.
However, each channel could have it's own audio ground separate from all the other channels. Then decide (experiment with) connecting the four audio grounds directly to or via four disconnecting networks to the chassis.
The Safety Earth connection must have it's own permanent fixing to chassis, not shared with other ground wires. It is OK to fix the Safety wire under it's own permanent nut and then place all the other grounds above that and secured with a separate nut.
I can try the disconnecting network. I also found that simply tieing all audio grounds together seems to work just fine.
I did find the true source of the grounding problems. It has to do with my TV and cable system. The video out from the receiver to the TV has grounds common with the pre-outs. The TV has no safety ground (plastic TV) but commons the video in ground with the cable in ground. The cable ground is connected probably somewhere in the cable companies equipment down the street. I measured 130 ohms from the cable shield to ground in my house. If I unplug either the cable TV or the video to the TV the buzz goes away. Somehow in all my tests this must have happened, which is why the buzz does disappeared. So I don't think the grounding change actually did anything.
Now the question is what to do about this. Should I simply ground the cable feed at the back of my TV? Should I put some sort of isolator in somewhere?
I did find the true source of the grounding problems. It has to do with my TV and cable system. The video out from the receiver to the TV has grounds common with the pre-outs. The TV has no safety ground (plastic TV) but commons the video in ground with the cable in ground. The cable ground is connected probably somewhere in the cable companies equipment down the street. I measured 130 ohms from the cable shield to ground in my house. If I unplug either the cable TV or the video to the TV the buzz goes away. Somehow in all my tests this must have happened, which is why the buzz does disappeared. So I don't think the grounding change actually did anything.
Now the question is what to do about this. Should I simply ground the cable feed at the back of my TV? Should I put some sort of isolator in somewhere?
Daniel, could you explain further? I'm unclear on the concept. Since a 50kHz wave in copper has a wavelength on the order of three miles, I shouldn't think this would be a consideration unless one plans on building an amp somewhat larger than those usually discussed here.
craigg4c said:
Having the signal lines be the same length (per left and right channels) applies when one or more amplifier channels are in the same enclosure--its just to help prevent the problem of ground loops.
I'm not saying to put grounds together on purpose. No.
But, I am saying that your potentiometer or source could and probably will.
Oh darn. Maybe I didn't explain that at all.
But, if you make your signal lines the same lenth for left channel as for right channel. . . it probably won't hurt anything. 😉
Nice looking amp. I just finished stuffing and soldering a dual mono kit and am looking forward to catching up with you. What did you use for wire? I have choices but no particular wisdom on that topic yet.
insulated 0.6mm copper does the job for much of the internal connections carrying low to medium currents. 1 to 2sqmm for the high current conductors.
- Status
- Not open for further replies.