I am struggling with hum/buzz in a 2-channel amp that I just built. Attached is a power/grounding diagram.
The amp is quiet if no inputs are connected or if only one channel is connected to a preamp.
There is about 7mV 120Hz buzz/hum (pretty loud on 95dB speakers) if I connect both channels to the preamp.
Also, the hum is coming if I only connect left channel to the preamp and connect the preamp to power outlet. It’s quite with only right channel connected in all situations.
The hum disappears when I turn off the power from the amp.
I tried troubleshooting with:
- - changing the point of grounding the RCAs in various ways
- - adding caps to the chassis from RCAs
- - connecting RCAs together before grounding
- - raising up filament from 0V to 20V.
Nothing made any difference. I am clueless at this point and I would greatly appreciate any help.
The amp is quiet if no inputs are connected or if only one channel is connected to a preamp.
There is about 7mV 120Hz buzz/hum (pretty loud on 95dB speakers) if I connect both channels to the preamp.
Also, the hum is coming if I only connect left channel to the preamp and connect the preamp to power outlet. It’s quite with only right channel connected in all situations.
The hum disappears when I turn off the power from the amp.
I tried troubleshooting with:
- - changing the point of grounding the RCAs in various ways
- - adding caps to the chassis from RCAs
- - connecting RCAs together before grounding
- - raising up filament from 0V to 20V.
Nothing made any difference. I am clueless at this point and I would greatly appreciate any help.
Attachments
You probably have a hum loop, or are injecting charging pulses into a ground. Look for differences in the grounding of the two channels - details matter. I hope you haven't used the negative tag on the cap as the star point - if so, then it is a mystery why you don't have buzz all the time.
On a separate issue, I am unclear what the point of the common-mode choke is. It won't act as a normal smoothing choke. If the rectifier can see the big smoothing cap then it could have a very short life before it joins the big vacuum in the sky.
On a separate issue, I am unclear what the point of the common-mode choke is. It won't act as a normal smoothing choke. If the rectifier can see the big smoothing cap then it could have a very short life before it joins the big vacuum in the sky.
Very interesting. Thank you for your input. Will try rewiring the choke tonight. I did not realize it does not act as a normal choke when wired as a common-mode choke. I followed this article: 5U4G-Mesh and 5Z3-Mesh Data sheet. Emission Labs..
I have a very short wire - about an inch - from the cap to the star point. Will try rewiring it as well. Can I ask you, why would it be a source of the hum?
I have a very short wire - about an inch - from the cap to the star point. Will try rewiring it as well. Can I ask you, why would it be a source of the hum?
Sakuma grounding is very interesting and unusual. Have you been able to successfully implement it?
I did not use common-mode connection, I connected the choke as per Lundahl recommendation, "Serial connection for improved common mode rejection": http://www.lundahl.se/wp-content/uploads/datasheets/1638.pdf. I will try connecting it in a more traditional way.
I did not use common-mode connection, I connected the choke as per Lundahl recommendation, "Serial connection for improved common mode rejection": http://www.lundahl.se/wp-content/uploads/datasheets/1638.pdf. I will try connecting it in a more traditional way.
If the choke connects to the cap negative, and then a wire goes from there to the star point you should be OK. This is not what your diagram shows.
I originally wrote:
For some reason I cannot fathom, common-mode chokes are currently popular in valve PSUs. They don't work as a proper smoothing choke, which may be why they need to be followed by a huge capacitor. They are small and cheap, because they don't need a gap - maybe that is why people use them. They will reject common-mode stuff (as your link says), but this is not what you want because ripple is differential mode. I think whoever wrote that page was confusing mains noise (which a common-mode choke will reject) with mains ripple (which a common-mode choke will ignore).
I now write:
Sorry, I misunderstood. It is not a common-mode choke but a normal choke, but with two windings wired in series and arranged so one winding is in each leg. This will work fine for smoothing. How much it will aid common-mode rejection I am not clear; I suspect not very much as it may just be the leakage inductance which acts in a common-mode way. I retract my criticism of tha page you linked.
I originally wrote:
For some reason I cannot fathom, common-mode chokes are currently popular in valve PSUs. They don't work as a proper smoothing choke, which may be why they need to be followed by a huge capacitor. They are small and cheap, because they don't need a gap - maybe that is why people use them. They will reject common-mode stuff (as your link says), but this is not what you want because ripple is differential mode. I think whoever wrote that page was confusing mains noise (which a common-mode choke will reject) with mains ripple (which a common-mode choke will ignore).
I now write:
Sorry, I misunderstood. It is not a common-mode choke but a normal choke, but with two windings wired in series and arranged so one winding is in each leg. This will work fine for smoothing. How much it will aid common-mode rejection I am not clear; I suspect not very much as it may just be the leakage inductance which acts in a common-mode way. I retract my criticism of tha page you linked.
Last edited:
Hi Azazello, do you mean I should disconnect both RCAs from ground completely? I have 33k greadleak, also, there is no hum when both RCA inputs are disconnected from the source.
I rearranged the star connections on 1 inch wire from the cap according to this article http://hifisonix.com/wordpress/wp-c.../How-to-wire-up-a-Power-Amplifier_Updated.pdf. It was a significant improvement, but still a long way to go.
I greatly appreciate your help!
I rearranged the star connections on 1 inch wire from the cap according to this article http://hifisonix.com/wordpress/wp-c.../How-to-wire-up-a-Power-Amplifier_Updated.pdf. It was a significant improvement, but still a long way to go.
I greatly appreciate your help!
-"The amp is quiet if no inputs are connected or if only one channel is connected to a preamp."
Your amp is hum free when power up, it only hums when connected to preamp.
"Also, the hum is coming if I only connect left channel to the preamp and connect the preamp to power outlet. It’s quite with only right channel connected in all situations. "
There is no hum when right channel is connected, so your preamp and power amp doesn't appears to have ground loop.
You got hum from left channel alone, then check left channel of preamp. Try to swap channel see what happens.
Your amp is hum free when power up, it only hums when connected to preamp.
"Also, the hum is coming if I only connect left channel to the preamp and connect the preamp to power outlet. It’s quite with only right channel connected in all situations. "
There is no hum when right channel is connected, so your preamp and power amp doesn't appears to have ground loop.
You got hum from left channel alone, then check left channel of preamp. Try to swap channel see what happens.
@Koonw If preamp is not connected to the power, connecting any (but only one) channel does not create hum. Connecting both channels creates hum. When preamp is connected, the only difference is that left channel alone will create hum of lower level. Disconnecting ground from the chassis/earth removes that hum. However, it does not help when both channels are connected to the preamp. Swapping channels did not help.
-Swap channels did not help, you got hum when two channel plugin.
Since your preamp and power amp appear hum free or ground loop problem, the problem might be the ground loop of two RCA interconnect screen. Are both ends (output and input) of the 2 screen joint together? The test is to disconnect one of the screen at power amp input. If no more hum, connect a 1 ohm resistor between this screen and ground, thereby create a ground lift.
Last edited:
I use twisted wire. The 'ground' wire is connected to RCA on one end and the local star of its channel's input (driver) stage. The amp has two stages with 6SH7 and EL36. I tried connecting these 'ground' wires to the main star with no difference.
I just tried connecting both RCAs with a single interconnect and I got the same hum. The power transformer is between two channels. I think the interconnect maybe closing the loop around the transformer. The transformer is well shielded Hammond 372HX and it's at least 3 inches away from anything and about 5 inches away from input wires. It's also on top of the chassis. Weird.
I just tried connecting both RCAs with a single interconnect and I got the same hum. The power transformer is between two channels. I think the interconnect maybe closing the loop around the transformer. The transformer is well shielded Hammond 372HX and it's at least 3 inches away from anything and about 5 inches away from input wires. It's also on top of the chassis. Weird.
Last edited:
So it appears you have mismatched pairs when could not cancel out hum completely. the twisted pairs are sensitive to hum 1) if both wire are not equal length, any extension must be made out of shield cable. 2) ground side should as close as possible to input socket and ground to chassis with short wire, usually the short wire is replaced with a resistor each for both channel. Try a different interconnect if not solved. This interconnect hum is more common in amp with higher sensitive, so lower input level or gain of front stage level will help.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.