If you disconnect one channel does the hum go away? If so it
could be multiple paths to ground causing current flow through the interconnect shields. This can then induce hum in the signal paths. I just went though this with a power amp build. Solution was to attach the signal grounds from both channels to the same point in the grounding scheme such that there would be no current flowing between them. For me the problem was in the power amp chassis but didn't show up until both channels were connected to my passive pre which then ties the two shield grounds together. I would think you could get the same scenario in the pre-amp chassis. I suppose it could still be a similar problem even if it occurs with only one channel connected. Idea is to make sure signal shield has nice quiet path to ground where there is no current flowing.
Good luck
could be multiple paths to ground causing current flow through the interconnect shields. This can then induce hum in the signal paths. I just went though this with a power amp build. Solution was to attach the signal grounds from both channels to the same point in the grounding scheme such that there would be no current flowing between them. For me the problem was in the power amp chassis but didn't show up until both channels were connected to my passive pre which then ties the two shield grounds together. I would think you could get the same scenario in the pre-amp chassis. I suppose it could still be a similar problem even if it occurs with only one channel connected. Idea is to make sure signal shield has nice quiet path to ground where there is no current flowing.
Good luck
Sorry Guys, I was unexpectedly whisked off to Saturday afternoon football - Michigan State vs. Central Michigan at Mt. Pleasant, Mi. (My school, CMU lost the game but both bands were a real treat. As DM, I started a post-game "circle around the tree - single file" ceremony that continues almost fifty years later.
)
The configuration has no common mains wiring. That's what makes the problem so confusing. The fact that the JC-2 is completely free of hum when connected to the powered speakers indicates the problem has to do with the pre to FE combination. I'll do some more puttering today.
)The configuration has no common mains wiring. That's what makes the problem so confusing. The fact that the JC-2 is completely free of hum when connected to the powered speakers indicates the problem has to do with the pre to FE combination. I'll do some more puttering today.
Attachments
Last edited:
Have you put a scope on the output of the amp while the hum is present? Is it 60 or 120hz? If it is read on, if it's something else just ignore my suggestions.
Is your signal ground tied to earth(safety) gound somewhere? Either directly or through a thermistor or some other method. I'm not seeing the connection in your pic anywhere. If it is, try disconecting the shield ground from the amp board and instead connect it to the lowest (closest to earth gnd) point that you have available. Just on the signal ground side of the thermistor, for example. When it's attached to the amp board it's path to ground may have a lot of current present with PS ripple. Alternately you might try connecting the shield directly to earth ground.
Phil
Is your signal ground tied to earth(safety) gound somewhere? Either directly or through a thermistor or some other method. I'm not seeing the connection in your pic anywhere. If it is, try disconecting the shield ground from the amp board and instead connect it to the lowest (closest to earth gnd) point that you have available. Just on the signal ground side of the thermistor, for example. When it's attached to the amp board it's path to ground may have a lot of current present with PS ripple. Alternately you might try connecting the shield directly to earth ground.
Phil
Unfortunately I don't own a scope yet.
This is what I'm hearing:
Noise - Flv - previewable
Noise - Wmv download
Nothing here but swapping units - no grounding changes. What is strange is the JC-2 is completely quiet with the powered speakers. I have a second JC-2 built and will wire that up to see if this one is malfunctioning - though I am getting ~ 0-1 mv on outputs in this piece.
I'll do some ground switching/jumpering later and report back.
This is what I'm hearing:
Noise - Flv - previewable
Noise - Wmv download
Nothing here but swapping units - no grounding changes. What is strange is the JC-2 is completely quiet with the powered speakers. I have a second JC-2 built and will wire that up to see if this one is malfunctioning - though I am getting ~ 0-1 mv on outputs in this piece.
I'll do some ground switching/jumpering later and report back.
you left pic clearly shows loops in the grounding. Two sets of them.
The centre and rightmost show just one loop, but the other end of that loop is not connected = no hum.
Had the other end been terminated with a stereo amplifier then that too would probably have hummed.
It's loop area that is the problem.
The centre and rightmost show just one loop, but the other end of that loop is not connected = no hum.
Had the other end been terminated with a stereo amplifier then that too would probably have hummed.
It's loop area that is the problem.
Hey Andrew. I think I see what you are describing, but the grounds on the JC-2 are all the same - two big connected planes on both sides of the PCB.
The mystery I have is why that same configuration of the JC-2 does not suffer from hum/loops with the powered speakers.
The mystery I have is why that same configuration of the JC-2 does not suffer from hum/loops with the powered speakers.
Attachments
Bob,
It's just a guess, but it is likely that the powered speakers don't have the same ground and shield connections internally. Because they seal the package together, they may take other approaches to safety ground. After all, the powered speakers probably use a wall wart and all their high voltage is constrained to the wall wart.
Jac
These are semi-pro speakers that were often used for field video production monitoring. Your description of a different grounding scheme may still be on point. I did a lot of ground jumpering this morning but some connections only made the problem worse - none eliminated it. I even tried to use the HSGND near C-1/C-2 with no effect.
I sent a request off to Stanton Tin , the JC-2 kit vendor, to see if he can spot something. Thanks for all the suggestions. It's probably something very simple that will surprise us all.
I sent a request off to Stanton Tin , the JC-2 kit vendor, to see if he can spot something. Thanks for all the suggestions. It's probably something very simple that will surprise us all.
New development - finally seeing the obvious I connected the JC-2 to a Hafler 120 power amp. No hum at all and music plays cleanly without distortion.
That, combined with the powered speakers result makes me believe the problem is the JC-2 to FE connection, placing the build on wood, or as willamp suggests a leakage of signal to ground after the RCAs on the FE build.
One of the patches I did was very interesting - and probably potentially destructive. With the FE on and one end of a jumper at point "A" (Spkr out gnd), touching point "B" (RCA in gnd) stops the hum. BUT with both ends of the jumper connected, applying power causes a quick series of static clicks before sending the LM3886 into self protection mode, or a relay shutdown - not sure which one it is. I tried it twice but it's too scary to continue - and I don't like that stinky green smoke.
Normal operation resumes with jumper removed and a fresh audio signal sent through the amp. So go figure???
That, combined with the powered speakers result makes me believe the problem is the JC-2 to FE connection, placing the build on wood, or as willamp suggests a leakage of signal to ground after the RCAs on the FE build.
One of the patches I did was very interesting - and probably potentially destructive. With the FE on and one end of a jumper at point "A" (Spkr out gnd), touching point "B" (RCA in gnd) stops the hum. BUT with both ends of the jumper connected, applying power causes a quick series of static clicks before sending the LM3886 into self protection mode, or a relay shutdown - not sure which one it is. I tried it twice but it's too scary to continue - and I don't like that stinky green smoke.
Normal operation resumes with jumper removed and a fresh audio signal sent through the amp. So go figure???
Attachments
Bob, the idea is not to tie your input signal shield of the FE to the speaker gnd but instead to tie it to some quiet ground point. And also to tie both input signal shields to this same quiet point. You don't want any current flowing in these shields, that's not what they're for.
I don't quite understand your grounding scheme in the FE. It's a dual mono design but are the two audio signal grounds from each channel tied together? (These are the grounds that would go to your speakers.) Also are they tied to earth ground in any way? Depending on the answers to these questions I could better suggest a place to attach the input shield grounds.
Phil
I don't quite understand your grounding scheme in the FE. It's a dual mono design but are the two audio signal grounds from each channel tied together? (These are the grounds that would go to your speakers.) Also are they tied to earth ground in any way? Depending on the answers to these questions I could better suggest a place to attach the input shield grounds.
Phil
Measure the resistance between Left signal ground at the RCA to the Right signal ground at the RCA.
Since both channels must be connected to PE, there must be a resistance that is measurable.
Now that you have confirmed the LOOP in the input grounds, look to see how big the LOOP AREA is. That may be the problem.
Similarly for the outputs from the Source.
The two RCA signal grounds must be connected to PE, if ClassI.
If the Source is ClassII, then neither signal ground have a PE to connect to. But they may connect to each other. Again, if you find a connection between the two source RCA signal grounds then look at where the LOOP areas are. How big are they?
Can any of the loops be made innocuous by inserting some additional resistances?
All of this is avoided by building monoblock amplifiers.
Since both channels must be connected to PE, there must be a resistance that is measurable.
Now that you have confirmed the LOOP in the input grounds, look to see how big the LOOP AREA is. That may be the problem.
Similarly for the outputs from the Source.
The two RCA signal grounds must be connected to PE, if ClassI.
If the Source is ClassII, then neither signal ground have a PE to connect to. But they may connect to each other. Again, if you find a connection between the two source RCA signal grounds then look at where the LOOP areas are. How big are they?
Can any of the loops be made innocuous by inserting some additional resistances?
All of this is avoided by building monoblock amplifiers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.