• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

LM3875 (AudioSector) Build

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
hum and hiss

Actually, both channels are humming, but it's barely audible, and not with stuff playing.

Your physical layout likely produces the hum and hiss. You have the signal running past the power supply and transformer to the amp and then back past all of that to the speaker posts. Anywhere you run unshielded signal wires in close proximity to AC you will likely pick up AC noise on the audio circuit.

2 things will likely help if you want to keep this layout.
First, all your AC wires should be twisted. This will cancel out common mode noise.
Second, be sure to keep as much separation from the signal wires away from the AC. If there is still some noise, use shielded wire for the small signal wire and keep lots of separation from the speaker wire and the AC.

I have a Peter Daniel amp that is so quiet you can literally put your ear to the speaker at full volume and hear nothing. The quieter you make this amp, the better the music will sound. I actually used a block of thick copper for the heat sink and laid it out such that the signal section was on one side of the copper and the AC on the other. Worked perfectly for past 5 years making very good music.
 
Your physical layout likely produces the hum and hiss. You have the signal running past the power supply and transformer to the amp and then back past all of that to the speaker posts. Anywhere you run unshielded signal wires in close proximity to AC you will likely pick up AC noise on the audio circuit.

2 things will likely help if you want to keep this layout.
First, all your AC wires should be twisted. This will cancel out common mode noise.
Second, be sure to keep as much separation from the signal wires away from the AC. If there is still some noise, use shielded wire for the small signal wire and keep lots of separation from the speaker wire and the AC.

I have a Peter Daniel amp that is so quiet you can literally put your ear to the speaker at full volume and hear nothing. The quieter you make this amp, the better the music will sound. I actually used a block of thick copper for the heat sink and laid it out such that the signal section was on one side of the copper and the AC on the other. Worked perfectly for past 5 years making very good music.

Thank you :) I will perhaps try something later on when I install the power switch.
 
I put the AC socket on the front of the amplifier to keep it away from the signal.

Twisting wires doesn't actually stop common mode noise, but it lowers peak common mode noise which is the most offensive (subject to phase shifting etc). Twisting wires in power cords even for regular 60hz is desirable for the same reason, common mode noise or not. There is a canceling affect of the shifted signal.

That power button in the link. Wire the incoming positive to C1, and then the out going positive to NC1. NC1 stands for normally closed (off), but you can turn it on. Then if your transformer puts out 24v or less, you can attach V+ and PG+ to the LED.
 
I made a power cord described in TNT audio. Basically heavy gauge wire twisted, inside a tube with safety ground outside twisted in opposite direction. Easy to see it makes a strong improvement on this amp. Still I had missed the twisting of AC wires from IEC socket to transformer and on to rectifiers. This too made an easy to hear improvement.
 
If your signal input and signal input ground wires are not shielded, then they too should be tightly twisted, ALL the way from the input jack to wherever they connect internally.

Actually, all conductor pairs should be twisted, or in some way kept as close together as possible. Otherwise, if they have any open area between them, they are an antenna!

(Faraday's Law is, roughly, that a time-varying magnetic field will induce a corresponding time-varying current in any conductive loop, and, a time-varying current in a conductive loop will induce a corresponding time-varying magnetic field. So, they can both receive and transmit.)

Your AC wiring (primaries, secondaries/rectifiers) is usually the worst "transmitter". And the input signal and ground pair is the worst "receiver". (They will also receive things other than just your internal AC, such as RF.) But all other pairs will also receive and/or transmit, if there is any area between their two conductors. So twist all wire pairs, if possible, including speaker pairs, and DC power and ground pairs.

If you want to shield the input pairs, instead of twisting them (since shielding is even better than twisting), a good type of cable to use is "shielded twisted pair". Using that, the signal and the ground wires are both inside of the shield. The shield is then NOT connected to the input jack's ground (signal ground). The shield should be connected to the chassis or chassis ground, and only at one end (probably the input end is better), with the other end of the shield connected to nothing.
 
Don't forget a current limiting resistor for the led in the switch.

LED calculator for single LEDs

They are supposed to be non LED, thats why I ordered them...

All my signal wires are twisted, and the boards are in the front end of the chassis, but the transformer is so big it makes a tight layout. inout and output wires are very close, but the transformer is shielded, or atleast supposed to be. I ordered custom from Toroidy.

The hum is barely audible, and the hiss, I blame my passive alps for that, I will wait to see if it gets more quiet at loud volumes when I finish my B1, waiting for the final parts to arrive any moment.

Thank you guys, you have been most awesome!

I don't know if I am to try and install the power switch, I really don't understand wiring when written and explained, I have to see pictures :)

Happy holy days :)
 
The power switch shows a diagram of the actual switch... I'll make you a picture later maybe.

PLEASE photo your amplifier and let us see it.

By the way mine has RF in it despite capacitors are certain pins. I can't figure out why at all. It doesn't come from my RCA's because volume doesn't change it. The good part is it never bothers my listening experience. It is silly though that it happens. Wish I knew why. (have tried to figure it out for years, Peter tried to help, no avail with any changes in grounding etc) Just saying, love the thing anyway.
 
The power switch shows a diagram of the actual switch... I'll make you a picture later maybe.

PLEASE photo your amplifier and let us see it.

By the way mine has RF in it despite capacitors are certain pins. I can't figure out why at all. It doesn't come from my RCA's because volume doesn't change it. The good part is it never bothers my listening experience. It is silly though that it happens. Wish I knew why. (have tried to figure it out for years, Peter tried to help, no avail with any changes in grounding etc) Just saying, love the thing anyway.

Interesting. What are the symptoms/effects or measurements that indicate it is RF?

Do you have RF filtering on the inputs? Do you have RF filtering on the outputs? (Everything is an input, for RF.). Do you have RF filtering on the DC supply line inputs to all chips? Can you describe what you have tried, so far?

If you use RCA input cables with shields, the shields might be picking up RF. There are two-wire RF filter topologies that should work for that. Look in chapter 6 or 7 of Walt Jung's book "Op Amp Applications", which is a free download at analog.com . Also discussed there are other methods for defeating RF, and for filtering power amp outputs and dc power. Twisting speaker wires comes to mind, as does installing a 5 uH or so high-current RF CHOKE in each of the speaker lines, like the bourns/j.w. miller toroidal types that can be seen at mouser.com, which are very reasonably priced. I guess those might also work in other places. Never tried them anywhere else. But my Magnepan planar speakers seem to make wonderful antennas.
 
Thank you all for the reply.

The hissing is my passive pre's fault, its more noisy with the volume just above half, more quiet when all the way full, and less so at lower leves. I don't care about that since it will go away with my new B1.

The RF noise, is only somewhat audible when the room is silent, nothing heard when I play music.

The on/off switch will go in later when I buy a drillbit big enough.

It has five connections, three in the middle and to flanging them.

I have heard some pretty expensive (by my accounts) amps from Marantz and Denon, Rotel and Luxman in the pricerange of about $500 to $3000 and this just kicks their butts...so good...

Thank you Petter Daniels for making this possible.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.