Getting the parts today -- can't wait to get this going; should be ready tonight if all goes well.
I currently have a "passive attenuator" with a 10K alps split going to LM3886 and to my Polk powered sub.
It's a fairly longer cable running to the polk sub as well.
I just unplugged the sub from the passive pre and A/B'd with it plugged in.. I can hear a pretty big difference in the clarity actually.
So the B1 Buffer should resolve this issue from what I was told -- so that's why I bought it 🙂 Also, even without the sub plugged in I believe I was told it should sound better than just the 10k pot running to the LM3886 alone.
Can't wait to hear it after.
I currently have a "passive attenuator" with a 10K alps split going to LM3886 and to my Polk powered sub.
It's a fairly longer cable running to the polk sub as well.
I just unplugged the sub from the passive pre and A/B'd with it plugged in.. I can hear a pretty big difference in the clarity actually.
So the B1 Buffer should resolve this issue from what I was told -- so that's why I bought it 🙂 Also, even without the sub plugged in I believe I was told it should sound better than just the 10k pot running to the LM3886 alone.
Can't wait to hear it after.
I have a 10K and a 100k ALPS stereo pot. The schematic calls for a 25K pot but Nelson Pass says that the pot can be changed out and that the pot is the input impedance.
So would a 10K input impedance be fine for my FIIO D3 DAC or should I go with the 100K? And perhaps I need to buy a 20K ALPS to swap out later?
So would a 10K input impedance be fine for my FIIO D3 DAC or should I go with the 100K? And perhaps I need to buy a 20K ALPS to swap out later?
I got it put together. Not really noticing an improvement in sound. (Does it need along burn in?)
Just found out what the non-inverting feature of the chipamp.com LM3886 amp boards is. Basically with that, the input impedance of the amp is in the megaohms. So a 10k pot in external passive attenuator is just fine, heck even a 100k pot in external passive attenuator should be fine as well?
So I might have a B1 buffer board to sell, fully populated with all Digikey BOM parts (except for the 10uf poly film caps which are the red Audiophilers). (Everything is soldered very well and cleanly -- resistances and capacitance were tested before installing them) Along with the 24V SMPS wall wart I got that was with the BOM.
Just found out what the non-inverting feature of the chipamp.com LM3886 amp boards is. Basically with that, the input impedance of the amp is in the megaohms. So a 10k pot in external passive attenuator is just fine, heck even a 100k pot in external passive attenuator should be fine as well?
So I might have a B1 buffer board to sell, fully populated with all Digikey BOM parts (except for the 10uf poly film caps which are the red Audiophilers). (Everything is soldered very well and cleanly -- resistances and capacitance were tested before installing them) Along with the 24V SMPS wall wart I got that was with the BOM.
Last edited:
I changed out the pot from 10k to 100k and liking it better now. Also reorganized the wiring and replaced some so it looks neater. I am liking the sound of it.. sounds more clear than before. Also seem to be getting better bass. I'm definitely going to keep it. It will be especially useful I imagine for other amplifiers that aren't non-inverting. The Polk powered sub's amp might not be non-inverting as well. SO perhaps it is already helping the sub bass.
Pictures to come soon after I design some parts in OpenSCAD and print on 3D printer (standoff adapter plate and a small square backplate insert for the power).
Pictures to come soon after I design some parts in OpenSCAD and print on 3D printer (standoff adapter plate and a small square backplate insert for the power).
Your B1 Buffer is there to drive the cables and any RF attenuating capacitance.
It is not there to modify the way the chipamp conditions the input signal.
The B1 Buffer allows you to place your volume control where you need it for best and convenient operation. It could be beside your favourite listening seat, or in the kitchen for when you are cutting the vegetables.
The amplifier should be beside/behind the speakers using VERY SHORT speaker cables.
The input impedance of your chipamp is set by the components fitted to the front end to condition your input signal and the input impedance of the chipamp itself. This could be anywhere from 1k to 1M and certainly not in the Megohms range.
Last edited:
But my point is since it is non-inverting, the 10K potentiometer + the capacitance of the split cables (split to powered sub) should be handled appropriately by the amp right? The B1 buffer could of been uneccessary.
But now for some reason, I think it does sound better with the B1 buffer than it did before.. definitely isn't harming the sound -- going to keep it in the chain. Besides, it could be helping the Polk powered sub because I don't know the design of it internally.. whether it is non-inverting or not and if is inverting, what the input impedance is.. going to look it up now.
the maximum output impedance of a 10k vol pot is ~2k51
That should suit any short cable with lowish capacitance.
Any high capacitance cable will lose some of the extreme treble.
Any very high capacitance cable will roll off the treble.
The Buffer allows you to drive all reasonable cables and any RF attenuation properly.
The vol pot alone cannot do that.
When the vol pot is turned down to -20dB, it's output impedance is ~ 900r
That allows it to drive a cable that is more than twice as long for the same treble attenuation.
If you listen at -30dB, then the output impedance of your vol pot drops to ~ 300r and it is now able to drive most medium and slightly higher capacitance cables, or even multiple cables to two or three receivers.
The formula that allows you to predict the treble roll off of the filter created by the vol pot and the cable capacitance is:
F-3dB = 1 / {2 * Pi * R * C}
Decrease R by a factor of 8 and you can increase the capacitance by a factor of 8 for the same roll off frequency.
That factor of 8 is what I got by dividing the 2k51 by 300r
The vol pot when turned down a lot becomes able to drive cable capacitance.
The output impedance of the B1 is low. It does not matter where you set the vol pot level. The B1 drives the cable.
That should suit any short cable with lowish capacitance.
Any high capacitance cable will lose some of the extreme treble.
Any very high capacitance cable will roll off the treble.
The Buffer allows you to drive all reasonable cables and any RF attenuation properly.
The vol pot alone cannot do that.
When the vol pot is turned down to -20dB, it's output impedance is ~ 900r
That allows it to drive a cable that is more than twice as long for the same treble attenuation.
If you listen at -30dB, then the output impedance of your vol pot drops to ~ 300r and it is now able to drive most medium and slightly higher capacitance cables, or even multiple cables to two or three receivers.
The formula that allows you to predict the treble roll off of the filter created by the vol pot and the cable capacitance is:
F-3dB = 1 / {2 * Pi * R * C}
Decrease R by a factor of 8 and you can increase the capacitance by a factor of 8 for the same roll off frequency.
That factor of 8 is what I got by dividing the 2k51 by 300r
The vol pot when turned down a lot becomes able to drive cable capacitance.
The output impedance of the B1 is low. It does not matter where you set the vol pot level. The B1 drives the cable.
Last edited:
I've built a couple of B1 buffers over the years and I've never been impressed with their sound despite trying different input/output coupling caps and power supplies.
Psychoacoustics at work here?? After all you changed the volume pot from 10K to 100K. The circuit works well with <25K ohm pot. I wouldn't use 100K ohm, but that's me.
Perhaps you're getting used to the sound of your B1 in the chain.🙂
Have fun!
Here is the inside of my 2205 case, housing the B1 Buffer.
I still need to design & print a plastic baseplate with integrated standoffs for the board -- then I can remove the temporary chipboard under the pcb.
I also need to design and print of small panel to secure power input connector to the backplate -- there was an IEC Ac inlet there before.
There are two rca inputs, to the back left -- need to add two more as my greyhill switch up front is a 4 position switch.
Should look really nice when it's done.
EDIT: those 10,000uF Nichicon electrolytics are low profile at 37mm tall, leaving plenty of room to clear top plate.
I still need to design & print a plastic baseplate with integrated standoffs for the board -- then I can remove the temporary chipboard under the pcb.
I also need to design and print of small panel to secure power input connector to the backplate -- there was an IEC Ac inlet there before.
There are two rca inputs, to the back left -- need to add two more as my greyhill switch up front is a 4 position switch.
Should look really nice when it's done.
EDIT: those 10,000uF Nichicon electrolytics are low profile at 37mm tall, leaving plenty of room to clear top plate.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
The loops at the terminals could be made a lot smaller.
Note the black+white at lower right. Compare to black+red at bottom right or to white+green at top left.
The big hole needs a metal, or conductive, panel secured electrically around it's perimeter to attenuate external interference as it passes through the hole.
All else looks good.
Note the black+white at lower right. Compare to black+red at bottom right or to white+green at top left.
The big hole needs a metal, or conductive, panel secured electrically around it's perimeter to attenuate external interference as it passes through the hole.
All else looks good.
Last edited:
Thanks. I can tighten up those loops no problem.
Regarding the big hole. How about I design it in CAD and print on 3D printer in PLA plastic, then put 2 layers of aluminum foil tape on the backside of it? I don't have a lot of tools for metal working -- so much easier to do plastic.
Also, what do you think about the 100k potentiometer I am using for the B1 Buffer -- is that too much? I notice Nelson Pass uses 25k in the schematic. He said the pot sets the input impedance. It'd only make it easier for the source to drive it right?
The aluminium tape needs to be electrically connected to the sides of the hole. and preferably continuously along the whole perimeter.
Any thin (insulated) gaps let in interference.
The smaller the holes the less the interference.
The shorter the slot the less the interference.
Small holes and short slots let in UHF interference and attenuate LF & MF & HF interference.
I don't recall seeing the answer, so that's why I asked again (if I asked before -- I think of might of). I'll have to try and find if it's been answered.
Last edited:
What would tighter loops do for the buffer? Take a look at Nelson Pass's B1 buffer he sells for $1000. Much bigger loops:
Okay I see a comment from Ammel that says he wouldn't use a 100k pot. Didn't notice that post until now.
- Status
- Not open for further replies.