This is my 3-channel gainclone. I get a slight hum in all of the channels, but it's strongest in the left-most channel and weakest in the right most channel (left to right from the picture below). I'm guessing this is because of interference from either the AC mains lines or the transformer? Any way that I can lessen or eliminate this, or did my design simply place everything too close together?
An externally hosted image should be here but it was not working when we last tested it.
Although you have independent PSUs per channel, you are joining the grounds to earth, and with different lengths of cable.
If I'm looking correctly, the channel where you have more hum is exactly the one where you have the longest earth cable.
Make a test: disconnect the earth from all modules.
If I'm looking correctly, the channel where you have more hum is exactly the one where you have the longest earth cable.
Make a test: disconnect the earth from all modules.
I don't know if I'm understanding you correctly, I should try running the amp with none of the channels connected to earth (or "chassis") ground? Is this safe for the amp? If it is safe, why is there an earth ground connector on the PCBs?
I think it's too close together. Are the boards touching? No chance a ground plain is shorting between the boards? They almost look stacked on top of each other.
Try mounting the trans on its side and away from the heatsinks.
Try mounting the trans on its side and away from the heatsinks.
vdi_nenna said:I think it's too close together. Are the boards touching? No chance a ground plain is touching between the boards?
Try mounting the trans on its side and away from the heatsinks.
The boards are very close together, but they aren't electrically connected (well, other than all of the grounds go to the same place, and all of them get their AC from the same place). Moving the transformer at this point isn't an option for me, if thats the problem I'll just deal with it.
m0tion said:I don't know if I'm understanding you correctly, I should try running the amp with none of the channels connected to earth (or "chassis") ground? Is this safe for the amp? If it is safe, why is there an earth ground connector on the PCBs?
Yes, it's safe.
That's the problem you have there.
Unscrew those wires, separate them, check out if they are not touching anything and power on the amp.
What do you hear now?
Can you slide aluminum flashing around the edge of the transformer to shield it from the circuits? I can't tell if you have the space or not.
You can find some thin aluminum flashing at a hardware store.
If you do this, ground the shield to earth. Don't slice any wires!!
You can find some thin aluminum flashing at a hardware store.
If you do this, ground the shield to earth. Don't slice any wires!!
And for more symmetrical looks, place the heatsink that differs from the others in the middle, lookes nicer that way 😉
The hum in the left challen is most likely from the ac mains lines running right pass the left amplifier, try to twist the hot and neutral togerther. That will cancel out most of the fields radiated by the wires.
Very nice looks btw, well done!!🙂
The hum in the left challen is most likely from the ac mains lines running right pass the left amplifier, try to twist the hot and neutral togerther. That will cancel out most of the fields radiated by the wires.
Very nice looks btw, well done!!🙂
Ok, tried a few things...
Carlos:
I don't know if this test is exactly the same, but it was a little easier and I'm lazy. I disconnected the ground wire that went to one of the channels and did a before/after test on that one particular channel. I didn't really notice any difference in the amplitude of the hum, sounded about the same to me.
vdi_nenna:
I was hoping that the barrier of heatsinks in between the transformer and the amp circuits would help shield them, do you really think a thin layer of flashing will help where the heatsinks do not?
Tekko:
Thanks for the compliment. I tried twisting those wires together and I think the amplitude of the hum may have decreased slightly, but it wasn't overwhelming.
The hum isn't really THAT loud to begin with, I have to put my ear about 6 inches away from the tweeter to hear it. It's probably even more noticable because this amp is being used in an active loudspeaker so the drivers are connected directly to the amp. I've decided to use the right-most channel for the tweeter and the left-most for the woofer because the hum is much more noticable on the tweeter. Thanks for all the suggestions so far and please keep them coming if you guys think of anything. At the bottom are two other pictures of the amp. I'm building another identical one for the other speaker in the active setup.
Pic 1
Pic 2
Carlos:
I don't know if this test is exactly the same, but it was a little easier and I'm lazy. I disconnected the ground wire that went to one of the channels and did a before/after test on that one particular channel. I didn't really notice any difference in the amplitude of the hum, sounded about the same to me.
vdi_nenna:
I was hoping that the barrier of heatsinks in between the transformer and the amp circuits would help shield them, do you really think a thin layer of flashing will help where the heatsinks do not?
Tekko:
Thanks for the compliment. I tried twisting those wires together and I think the amplitude of the hum may have decreased slightly, but it wasn't overwhelming.
The hum isn't really THAT loud to begin with, I have to put my ear about 6 inches away from the tweeter to hear it. It's probably even more noticable because this amp is being used in an active loudspeaker so the drivers are connected directly to the amp. I've decided to use the right-most channel for the tweeter and the left-most for the woofer because the hum is much more noticable on the tweeter. Thanks for all the suggestions so far and please keep them coming if you guys think of anything. At the bottom are two other pictures of the amp. I'm building another identical one for the other speaker in the active setup.
Pic 1
Pic 2
Tekko:
Yeah, I've thought about that, but doesn't it introduce a 90 degree phase shift? Anywho, lets say I was going to get one, what value would you recommend? The tweeter is a Seas 27TDFC ( fs = ~550Hz).
Yeah, I've thought about that, but doesn't it introduce a 90 degree phase shift? Anywho, lets say I was going to get one, what value would you recommend? The tweeter is a Seas 27TDFC ( fs = ~550Hz).
Tekko:
Well, this is obviously not true because I've been doing it and my tweeter isn't destroyed =).
Well, this is obviously not true because I've been doing it and my tweeter isn't destroyed =).
m0tion said:The hum isn't really THAT loud to begin with, I have to put my ear about 6 inches away from the tweeter to hear it. It's probably even more noticable because this amp is being used in an active loudspeaker so the drivers are connected directly to the amp. I've decided to use the right-most channel for the tweeter and the left-most for the woofer because the hum is much more noticable on the tweeter.[/URL]
WOW, to hear hum on a tweeter it must really be LOUD.
Hey, waitaminute...😱 look at your pic.
The signal and speaker cables are too close to each other on your leftmost channel, on top of the PSU caps.
Not good.
Give them some space, separate them.
Hi m0tion..
Have you tried only running one amplifier-board with the transformer??
I may be wrong in this, so be gentle with me (I still consider myself new to DIY audio)
-but last week I experimented a bit with 2 of my my BrianGT 3875 GainClones, and everytime i hooked them up to the same transformer I also got hum in my speakers..
Dunno if this is due to a too small transformer (a mere 160VA) or if this is an effect of grounding problems..
It is my belief that this is not a problem with the positioning of the transformer, since I also tried this to get rid of the hum..
I'll follow this thread closely, hopefully one of the more experienced guys have a solution...
Cheers..
Have you tried only running one amplifier-board with the transformer??
I may be wrong in this, so be gentle with me (I still consider myself new to DIY audio)
-but last week I experimented a bit with 2 of my my BrianGT 3875 GainClones, and everytime i hooked them up to the same transformer I also got hum in my speakers..
Dunno if this is due to a too small transformer (a mere 160VA) or if this is an effect of grounding problems..
It is my belief that this is not a problem with the positioning of the transformer, since I also tried this to get rid of the hum..
I'll follow this thread closely, hopefully one of the more experienced guys have a solution...
Cheers..
The Fs doesn't mean that much..
Your only concern is to block DC from getting to the tweeters voice coil.. (could kill it!! ) A cap of 1uF should do the trick...
Of course, you could select a smaller cap and get a block of higer frequencies, but thats not a problem in my opinion...
I've never blown a tweeter directly connected to an amp, but I know it could happen... (depends on the amplifiers turn-on and turn-off characteristics..)
Your only concern is to block DC from getting to the tweeters voice coil.. (could kill it!! ) A cap of 1uF should do the trick...
Of course, you could select a smaller cap and get a block of higer frequencies, but thats not a problem in my opinion...
I've never blown a tweeter directly connected to an amp, but I know it could happen... (depends on the amplifiers turn-on and turn-off characteristics..)
check mains wiring
The mains wiring is too close to the left channel. Try using shielded cable or a faraday shield for the IEC to tranny wiring.
Good luck,
m
The mains wiring is too close to the left channel. Try using shielded cable or a faraday shield for the IEC to tranny wiring.
Good luck,
m
PS layout
You should try rectifying and filtering behind the heatsink. Then have DC going out to the boards. Have separate ground returns for signal and power decoupling caps, they should go to a star grounding point on a tee away from the center of the cap connections, away from the bridge. The way you have it laid out you have a loop of AC circling your chassis, not good at all. I'm quite surprised the hum is low level.
Good Luck.
You should try rectifying and filtering behind the heatsink. Then have DC going out to the boards. Have separate ground returns for signal and power decoupling caps, they should go to a star grounding point on a tee away from the center of the cap connections, away from the bridge. The way you have it laid out you have a loop of AC circling your chassis, not good at all. I'm quite surprised the hum is low level.
Good Luck.
Dragging this thread back out because I've started to troubleshoot this amp again. With nothing connected to the input if I disconnect the ground lead from one of the amp modules and move it around the hum gets louder and softer (actually, it seems to get louder as I move it toward the leads going to the RCA input connector). With an input connected (whether it is attached to another device or not) I definitely get an FM radio station (not sure which) as well as a good deal of static/hum. Does this help? Any ideas?
Hey m0tion,
I found that when using multiple (2 or more) BrianGT chipamp boards and rectifier boards with a single transformer there was interaction between the boards through the transformer windings that caused the hum. Shorter cables certainly helped, especially the ground wire, which fits with your findings.
I had tried a lot of the things sugggested in your thread with little sucess, however, I did find that using one rectifier board for all the amp boards or multiple transformers solved the problem. Remember that these kits boards were designed to be used as dual mono, all the way back the the mains, i.e. one transformer per board if you take it literally.
here's my thread which goes into more detail:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=95451&highlight=
I found that when using multiple (2 or more) BrianGT chipamp boards and rectifier boards with a single transformer there was interaction between the boards through the transformer windings that caused the hum. Shorter cables certainly helped, especially the ground wire, which fits with your findings.
I had tried a lot of the things sugggested in your thread with little sucess, however, I did find that using one rectifier board for all the amp boards or multiple transformers solved the problem. Remember that these kits boards were designed to be used as dual mono, all the way back the the mains, i.e. one transformer per board if you take it literally.
here's my thread which goes into more detail:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=95451&highlight=
- Status
- Not open for further replies.
- Home
- Amplifiers
- Chip Amps
- LM3886 BrianGT amp complete, slight hum