Simon,
Even though op-amps have a high rejection ratio, you probably won't cure the humm entirely without regulators.
For the buffers or filters with op-amps (no power amplifiers), the 78xx (positive) and 79xx (negative) series will be just fine.
They come in different types, one of which is a 100mA device (I think it has an "L" in it's name. This is plenty for a buffer, as few op-amps can supply more than 20mA to their output anyway.
The xx refers to regulated output voltage. 7809 and 7909 would give you +/- 9V. Buy a 2x12V 400mA transformer if you go for 2x9V out. These regulators are thermally stable (shut down if overheated) and short-circuit proof. The only way I can think of to destroy them, is to supply high voltage (or reverse), or fry them witha soldering iron.
As for your power supply, try to remove it from the set-up, and measure each rail voltage to ground under artificial load (power resistor or light bulb). Does one voltage fail?
(edit: Regarding star ground (which someone mentioned), it's mentioned in my notes, too if you don't know what we're talking about.)
Jennice
Even though op-amps have a high rejection ratio, you probably won't cure the humm entirely without regulators.
For the buffers or filters with op-amps (no power amplifiers), the 78xx (positive) and 79xx (negative) series will be just fine.
They come in different types, one of which is a 100mA device (I think it has an "L" in it's name. This is plenty for a buffer, as few op-amps can supply more than 20mA to their output anyway.
The xx refers to regulated output voltage. 7809 and 7909 would give you +/- 9V. Buy a 2x12V 400mA transformer if you go for 2x9V out. These regulators are thermally stable (shut down if overheated) and short-circuit proof. The only way I can think of to destroy them, is to supply high voltage (or reverse), or fry them witha soldering iron.
As for your power supply, try to remove it from the set-up, and measure each rail voltage to ground under artificial load (power resistor or light bulb). Does one voltage fail?
(edit: Regarding star ground (which someone mentioned), it's mentioned in my notes, too if you don't know what we're talking about.)
Jennice
I disagree with you Jennice on needing regs to kill the hum. He is using about 15,000uF per rail which would be enough to run a pretty powerful power amp! I've run little circuits like these off just 470uF per rail without any really noticeable hum.
I'm sure the problem is incorrectly implemented grounding, but without physically being able to examine the birds nest I can't pick out the exact fault.
SimontY, make sure that for each set of circuitry that all 0V points really are connected to ground and that they are done so only at one local star point, which should be physically in the centre of the PSU caps. Also, make your signal grounds, i.e. interconnect leads, just go direct to the circuit boards. If you have metal cases, do not ground anything to them apart from the star ground, and at this point also connect mains earth..
I'm sure the problem is incorrectly implemented grounding, but without physically being able to examine the birds nest I can't pick out the exact fault.
SimontY, make sure that for each set of circuitry that all 0V points really are connected to ground and that they are done so only at one local star point, which should be physically in the centre of the PSU caps. Also, make your signal grounds, i.e. interconnect leads, just go direct to the circuit boards. If you have metal cases, do not ground anything to them apart from the star ground, and at this point also connect mains earth..
Good afternoon,
Last night I removed one psu completely, and have each circuit sharing a supply - the one with 5,000uF / rail @ ~11v, as yet unregulated. This change brought about another considerable reduction in hum, altough there is still some audible hum coming from the subwoofer when no music is playing (it's masked by any amount of music).
The reason I have lost one psu is that I've decided, for now at least, to dedicate the buffer to the subwoofer x-over, and leave the main amp buffer-free. A buffer here is something I want to try carefully and separately. And this way the x-over can be boxed as a complete little package, all quite neat (eventually).
I will try connecting at least one of my two psu grounds to house earth. I will report back with results.
Once again, thank you so much for your kind help, I would be nowhere without it!
Making this filter has proved useful sonically, and many times more useful as an educational process.
I only have to listen to something like the Bjork track 'Headphones' to appreciate what the LM3886 and this subwoofer do remarkably well (that is simple music with deep bass content).
If only the LM3886 didn't get lost when the going gets tough....
Last night I removed one psu completely, and have each circuit sharing a supply - the one with 5,000uF / rail @ ~11v, as yet unregulated. This change brought about another considerable reduction in hum, altough there is still some audible hum coming from the subwoofer when no music is playing (it's masked by any amount of music).
The reason I have lost one psu is that I've decided, for now at least, to dedicate the buffer to the subwoofer x-over, and leave the main amp buffer-free. A buffer here is something I want to try carefully and separately. And this way the x-over can be boxed as a complete little package, all quite neat (eventually).
I will try connecting at least one of my two psu grounds to house earth. I will report back with results.
Once again, thank you so much for your kind help, I would be nowhere without it!
Making this filter has proved useful sonically, and many times more useful as an educational process.
I only have to listen to something like the Bjork track 'Headphones' to appreciate what the LM3886 and this subwoofer do remarkably well (that is simple music with deep bass content).
If only the LM3886 didn't get lost when the going gets tough....
Hmm.... In my world PSU's can't get big enough.. but we're talking somewhere around 30W rms @ 8 ohm here. (Assuming that your PSU is +/- 25V.) Don't expect it to rattle the room apart.
If you got ome at hand, try to add more capacity. Also I would check for heat. Some chips have a temperature and/or output current limiter, which makes them survive.
However, I don't know this chip, so I herewith welcome any comment from people who know it.
Jennice
If you got ome at hand, try to add more capacity. Also I would check for heat. Some chips have a temperature and/or output current limiter, which makes them survive.
However, I don't know this chip, so I herewith welcome any comment from people who know it.
Jennice
I'd like to try a bigger tx *and* more capacity. Don't think I've got any appropriate caps around, so will have to buy some.Jennice said:
It's not purely how strong the bass is when cranked though, it's also the quality at low levels. However, a big part of this may be strongly influenced by other factors...
Hi all!
Had a nightmare of a time this evening, trying to put my crossover into a box!
Broken wire here, not working for no apparent reason here... after some long hours I hot melt glued most stuff together, and not much more work remains to have it boxed up with sockets etc.
At one point (after a couple of 'Eureka' fixing moments already) I was annoyed at the buzzing the crossover was putting back into the main amp... So I thought "it works, for now I'll glue in some components... *unplugs soldering iron to plug in glue gun* and.... "wtf? the buzzing has gone!!!" Pesky soldering iron!
On a side-note, a Farnell package arrived today, with 4 more 3,300uF caps. I am going to try doubling up in the LM3886... Then I will try this hocus pocus 'snubber' when it sounds crap and dull
Here's a pic from yesterday:
Had a nightmare of a time this evening, trying to put my crossover into a box!
Broken wire here, not working for no apparent reason here... after some long hours I hot melt glued most stuff together, and not much more work remains to have it boxed up with sockets etc.
At one point (after a couple of 'Eureka' fixing moments already) I was annoyed at the buzzing the crossover was putting back into the main amp... So I thought "it works, for now I'll glue in some components... *unplugs soldering iron to plug in glue gun* and.... "wtf? the buzzing has gone!!!" Pesky soldering iron!
On a side-note, a Farnell package arrived today, with 4 more 3,300uF caps. I am going to try doubling up in the LM3886... Then I will try this hocus pocus 'snubber' when it sounds crap and dull
Here's a pic from yesterday:
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.