I am wondering what people think about adding pre-attenuation before a volume pot (pot in a box preamp) and any negative effect it could have on the FE. I gather the best value for a pot is 20K, I have a 20K Tocos stereo pot with decent L&R channel balance but in my system it goes very loud with only a small amount of rotation and usually reaches max volume at just under half rotation. Another down side is that the pots tracks don't end evenly so one channel cuts out completely before the other when I turn the volume down very low at night.
I was thinking about using a couple of resistors in a fixed attenuator to reduce the incoming signal and make the pot more useable range wise. Say an L pad configuration with R1 - 76K and R2 - 15K. Hopefully this would give around -20bB attenuation before the pot without causing too much instability.
I still don't understand impedance and probably never will
I was thinking about using a couple of resistors in a fixed attenuator to reduce the incoming signal and make the pot more useable range wise. Say an L pad configuration with R1 - 76K and R2 - 15K. Hopefully this would give around -20bB attenuation before the pot without causing too much instability.
I still don't understand impedance and probably never will
Hi Dario, just wondering what real difference that extra 10K would make to the amp and if I add a dropping resistor before the pot would it make things worse. I need to reduce the input signal quite a bit. I have played around with values and found that around 22K before the 20K pot gives me a much more user friendly control and it does not seem to have harmed sound quality.
I found a thread where they discuss this same issue with another type of LM3886 amp
http://www.diyaudio.com/forums/chip-amps/263289-lm3886-based-amp-too-loud.html
I found a thread where they discuss this same issue with another type of LM3886 amp
http://www.diyaudio.com/forums/chip-amps/263289-lm3886-based-amp-too-loud.html
I'm not an 'impedance expert' either but from what I know those extra 10K will make the amp less 'confortable' with its input and more prone to noise problems.
From Wikipedia (input impedance):
Audio systems
Generally in audio and hi-fi systems, amplifiers have an input impedance several orders of magnitude higher than the output impedance of the source device connected to that input. This concept is also called voltage bridging or impedance bridging. In this case,
Zload >> Zsource
The input impedance of the driven stage (load) is much larger than the output impedance of the driver stage (source).
In general, this configuration will be more resistant to noise (particularly mains hum). Also the loading effects on the driving amplifier stage are reduced. In certain circuits a voltage follower stage is used to match the source and load impedance, which results in maximum power transfer.
I am wondering what people think about adding pre-attenuation before a volume pot (pot in a box preamp) and any negative effect it could have on the FE. I gather the best value for a pot is 20K, I have a 20K Tocos stereo pot with decent L&R channel balance but in my system it goes very loud with only a small amount of rotation and usually reaches max volume at just under half rotation. Another down side is that the pots tracks don't end evenly so one channel cuts out completely before the other when I turn the volume down very low at night.
I was thinking about using a couple of resistors in a fixed attenuator to reduce the incoming signal and make the pot more useable range wise. Say an L pad configuration with R1 - 76K and R2 - 15K. Hopefully this would give around -20bB attenuation before the pot without causing too much instability.
I still don't understand impedance and probably never will
Is your 20k logaritmic? From your description behaves like a linear one.
Sent from my iPad using Tapatalk
davym,
Very strange. I just checked and my CD player, for example, also puts out about 2 Vrms on the analog channel. I have used just a plain pot for a long time and never had any problem with the volume level of the FE. Typical and comfortable music levels were at slightly more than 1/2 volume on a cheap Alps 10k pot.
One question. Do you have your pot wired so that the input signal to the pot is across the fixed resistance of the pot and the wiper is wired to the input of the FE? I believe that is the correct way to use the pot as a voltage divider and should give you a wide range of volume control. This may sound like a silly question, but I did it wrong once and it didn't work well at all.
Yup, been there too before now. As far as I can see the way you describe it is the way I wired it too. Part of the problem is my current speakers which are very easily driven to high levels by just a few watts. My old speakers were similarly rated at 6 Ohms but needed more power to reach the same output. I didn't have quite the same problem at very low volume settings with them because I had to use more of the pots travel.
I have also had this issue with these speakers with a few commercial amps, it results in way too much gain with only a small amount of pot travel.
You can buy attenuators which you can plug inline with your power amps signal input to help with this issue but I'd prefer to find my own value and fit the resistor's inside the preamp case.
I have also had this issue with these speakers with a few commercial amps, it results in way too much gain with only a small amount of pot travel.
You can buy attenuators which you can plug inline with your power amps signal input to help with this issue but I'd prefer to find my own value and fit the resistor's inside the preamp case.
Adding a resistor to the top of an attenuator pot is like limiting the stroke/rotation of the vol pot.
Start with a 10k vol pot and move it all the way to the top and the output equals the input.
Now add a 10k to the top of the pot.
At the bot the attenuator stills sends out zero signal.
Now when the pot is rotated all the way to maximum the output is half the input (-6dB).
The added resistor converts the vol pot from 10k to 20k and limits the output to half way up. The combination behaves like a 20k vol pot with a stop that prevents you turning it all the way up.
Since the resistor is attached to the vol pot there is no extra capacitance to be driven so no concerns with cable lengths, nor with HF roll off.
You could use a 1k vol pot and add a 100k resistor to the top.
The maximum output would be 1/101 = -40dB.
The 1kVR+100k behaves like a 100k vol pot with a stop that prevents you turning the volume up any higher than -40dB.
Start with a 10k vol pot and move it all the way to the top and the output equals the input.
Now add a 10k to the top of the pot.
At the bot the attenuator stills sends out zero signal.
Now when the pot is rotated all the way to maximum the output is half the input (-6dB).
The added resistor converts the vol pot from 10k to 20k and limits the output to half way up. The combination behaves like a 20k vol pot with a stop that prevents you turning it all the way up.
Since the resistor is attached to the vol pot there is no extra capacitance to be driven so no concerns with cable lengths, nor with HF roll off.
You could use a 1k vol pot and add a 100k resistor to the top.
The maximum output would be 1/101 = -40dB.
The 1kVR+100k behaves like a 100k vol pot with a stop that prevents you turning the volume up any higher than -40dB.
Last edited:
Cheers Andrew, After a hunt i found a couple of 10K TAKMAN carbon film resistors in my parts box so I am currently trying them on top of the 20K Tocos pot. It's still going very loud with only a small amount of rotation but it's better than with no resistors.
I can get a 10K Tocos pot but I suspect I'd need to add 20K to that or I'd just be back in the same position as I am now. Probably a waste of time in other words.
I can get a 10K Tocos pot but I suspect I'd need to add 20K to that or I'd just be back in the same position as I am now. Probably a waste of time in other words.
use a higher value resistor.
The only disadvantage is extra noise due to passing the signal through a high value pot.
It's always the same when too much gain is in the system, the signal needs to be excessively attenuated and then the noise of the excessive attenuation and the excessive gain add up.
It the noise is not a problem, then keep increasing the resistor.
The only disadvantage is extra noise due to passing the signal through a high value pot.
It's always the same when too much gain is in the system, the signal needs to be excessively attenuated and then the noise of the excessive attenuation and the excessive gain add up.
It the noise is not a problem, then keep increasing the resistor.
Haven't had much to write about lately, but here a few recent developments.
One of the big attractions of monoblocks, at least for me, is the opportunity to put amps very near speakers and use short speaker cables. One of the risks of doing so is that the amplifiers will be close enough to the speakers to be subjected to increased vibration, both airborne and through whatever the amp rests on. Especially with these lightweight chip amps, that can introduce distortion that a heavier amplifier might not produce. My amps rest directly on the wood floor directly behind the speakers. Originally, I incorporated some damping material into the amp enclosures to reduce airborne vibes, but the whole enclosure is free to shake about as the floor vibrates. I just built a couple of amp stands out of plywood and pieces of stone countertop, and putting them under the amps seems to have increased transparency and dynamics. The stands are nearly twice the weight of the amps alone, so it is a substantial increase in mass. The stone material is mounted onto 3/4" plywood with a thick layer of silicone adhesive to damp the stone from ringing and form a sort of CLD structure, and the plywood rests on spikes with tips ground down a bit so it doesn't wreck the floor. Seems to be eminently stable.
I also changed the leadout wires inside the amp for the third time. Some will scoff that only those few inches of wire can make a difference, but now I believe it does. Strictly subjective, so I may be fooling myself but I have no motivation to do so. I tried to make my investigation legitimate. All the wire I've used is solid core copper, OCC with teflon insulation. I have found this to be the most musical and accurate type of wire. What impresses me now is that the gauge can make such a difference in the tone of the amp. I started with 16ga, and that seemed to flatten the highs and the image, dull but very full-bodied. I switched that out for 18ga which put more life and bounce into the music, but the highs still seemed slightly recessed. Since my amps run only the highs and mids of my system, I thought 20ga would be adequate to carry the current and possibly provide even more sparkle. It did, but it seemed to reduce the body of the midrange, as well. So now I have just finished installing a parallel run of 16ga and 20ga (for each leg of the signal), and that really seems perfectly balanced between detail and fullness. I never would have expected a few inches of wire to make that much difference, but I have now demonstrated, at least to myself, that it does.
Add mass. Play with wire. There is always more to be gained.
Peace,
Tom E
One of the big attractions of monoblocks, at least for me, is the opportunity to put amps very near speakers and use short speaker cables. One of the risks of doing so is that the amplifiers will be close enough to the speakers to be subjected to increased vibration, both airborne and through whatever the amp rests on. Especially with these lightweight chip amps, that can introduce distortion that a heavier amplifier might not produce. My amps rest directly on the wood floor directly behind the speakers. Originally, I incorporated some damping material into the amp enclosures to reduce airborne vibes, but the whole enclosure is free to shake about as the floor vibrates. I just built a couple of amp stands out of plywood and pieces of stone countertop, and putting them under the amps seems to have increased transparency and dynamics. The stands are nearly twice the weight of the amps alone, so it is a substantial increase in mass. The stone material is mounted onto 3/4" plywood with a thick layer of silicone adhesive to damp the stone from ringing and form a sort of CLD structure, and the plywood rests on spikes with tips ground down a bit so it doesn't wreck the floor. Seems to be eminently stable.
I also changed the leadout wires inside the amp for the third time. Some will scoff that only those few inches of wire can make a difference, but now I believe it does. Strictly subjective, so I may be fooling myself but I have no motivation to do so. I tried to make my investigation legitimate. All the wire I've used is solid core copper, OCC with teflon insulation. I have found this to be the most musical and accurate type of wire. What impresses me now is that the gauge can make such a difference in the tone of the amp. I started with 16ga, and that seemed to flatten the highs and the image, dull but very full-bodied. I switched that out for 18ga which put more life and bounce into the music, but the highs still seemed slightly recessed. Since my amps run only the highs and mids of my system, I thought 20ga would be adequate to carry the current and possibly provide even more sparkle. It did, but it seemed to reduce the body of the midrange, as well. So now I have just finished installing a parallel run of 16ga and 20ga (for each leg of the signal), and that really seems perfectly balanced between detail and fullness. I never would have expected a few inches of wire to make that much difference, but I have now demonstrated, at least to myself, that it does.
Add mass. Play with wire. There is always more to be gained.
Peace,
Tom E
Plenty laugh about cable choice affecting sound quality but if you hear a difference which makes you happy I say why not. Maybe I should swap out that stuff I used at some point, it's silver coated copper with teflon insulation, could just be my imagination but the sound can seem a bit bright with some music. I've just ordered 4 of those Black Gate PK 220uF 4V caps from parts exchange while they are still available. I got a couple of them off ebay a while ago but I cant be certain they are authentic. I also ordered 8 ELNA RJH 220uF 50V from a polish ebay seller called euroelectronikspj, they cost in total around £6 delivered which seems cheap. Hope they aint clones, they look authentic in the picture....
I hope to order Amtrans AMCH and Mundorf Mcap Supremes from hifi collective in the next week or so. Maybe by this time next year i'll have a second set of boards with premium components to compare with the industrial BOM ones I'm currently enjoying.
Why hurry a good thing
I hope to order Amtrans AMCH and Mundorf Mcap Supremes from hifi collective in the next week or so. Maybe by this time next year i'll have a second set of boards with premium components to compare with the industrial BOM ones I'm currently enjoying.
Why hurry a good thing