LM3886 "fullrange"

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Daniel,
A different brand of potentiometer shouldn't make a significant difference;
I have several pot or step ladder, and I ear difference..
Alps is not so good in first (besides other factors) cause low precision, with 2 pot I mesure about 5% (between left and right), with 2 Tocos precision is 1,5%.. Abd carbon track is not the same.
I'am building new passive preamp with Tocos, and see what's happen..

however, a different resistance value or different input arrangement really could. Did you want more midrange or less midrange?
Level and quality of midrange was good, I test directly at output dac..

But what values and quality change midrange sounding ?
Thank's.
Phil.
 
...............Alps is not so good in first (besides other factors) cause low precision, with 2 pot I mesure about 5% (between left and right), with 2 Tocos precision is 1,5%.. ...............
5% difference at the low end of a dual track vol pot is pretty good.
If you are at ~-50dB on the tracks, then for a 10k vol pot where the lower leg is ~32r. That leaves 9k968 in the upper leg.
If the other part of the dual is 5% lower, then the two legs are 30r4 and 9k9696
The attenuations of these two example tracks with a 5% mismatch are-49.89700dB and 50.34253dB
The channel imbalance is 0.4455dB

The channel imbalance remains substantially the same whether the 5% error is plus or minus and whether the vol pot is 10k or 20k or 50k or 100k or 1M.

A channel imbalance at -50dB of less than ½dB is good in my book.

BTW,
a 5% error @ -60dB (and for -80dB), leaves a channel to channel imbalance of 0.424dB
 
Hi,
The channel imbalance is 0.4455dB

For me it's a little too much (but I do not say this is the case for all), cause I can ear difference with 0,3/0,4 db..
I can explain why, if you want.
But yes, Andrew you are right when you said 0,44 db is good value !
With pots, generally the least difference ( between left and right channel) is near twelve o'clock.
I mesure two Tocos RV24 pots (10K) around the middle of track, twelve o'clock :
pot 1 :
Left 8,95K Right 9K and left 1K right 1,03K
pot 2 :
Left 8,685K Right 8,66K and left 1,22K right 1,24K
It seems that the pot 2 is a little better.
Phil.
 
Did you calaculate the channel to channel balance errors at that ~-20dB setting.
The "worse" one has an error of <0.19dB
The "better" one has an error of <0.15dB
A difference of 0.04dB in balance between the channels.

Are you seriously expecting us to believe you can hear the difference in matching between the better and worse pot?
 
Are you seriously expecting us to believe you can hear the difference in matching between the better and worse pot?

Yes, I'll explain it but I recognize that this is not the case for everyone, and I understand that it is difficult to believe it !

For about ten years, I suffer from tinnitus. It's like a noise made by internal ear, it's all day and night. It's not comfortable, and I hate silence, cause I ear only this noise in my head.
After a while the brain adjusts and compensates like for the view as with glasses.
I hear very well according to my doctor (up to 12000 hz and 11000 hz for left ear), but I more sensitive to the noise and music.
I hear distortion before it happens, I dont's listen metal cone or class T amp or D amp.

It may be an advantage for audio, but I would not want anyone to suffer this evil !!
Phil.
 
But what values and quality change midrange sounding?

Volume pot:
100k input may be too much noise. 50k may be slightly too much dampening, but possibly a good centerpoint figure (a good place to start exploring). 20k is probably so much dampening that you could get hard/edgy tone mids (which is worse) and possibly improve other aspects because stronger dampening can also lower noise (at the cost of a less usable tone).

See where more or less current can have an effect on the midrange sound? That can also be observed with other resistors in the system.

Another matter is that volume pots don't like to drive cables, especially not long cables.
 
Hi Daniel,
See where more or less current can have an effect on the midrange sound? That can also be observed with other resistors in the system.
Another matter is that volume pots don't like to drive cables, especially not long cables.

I agree with you, change value of pot change mid sounding..
With several test, we can observe different tones.
Next week I will build new passive preamp with Tocos 10K (have not other value) and try a pre-attenuation before pot (source is around 2 V output)..
Cables are short as possible, 50 cm.
Phil.
 
Lots of attenuation going on before amplifying? Maybe too much amplifying?

With speakers that efficient, a flea power amp might be fun to try.
For example, either LM1875 or a TDA7293, either used along with a 12+12vac transformer at the power supply. The LM1875 could use its split rail datasheet example for the gain divider, LM1875's baby sister LM675 could go to even lower gain, or the TDA7293 could be run at the minimum gain listed in the datasheet (singapore TDA7294 also usable). For any of those, I think that, you'd like to have some 220uF for amp board decoupling and a nice power board too. The DC voltage is approximately 17v per each rail. The output power is in the neighborhood of 13W.
Flea power amp: At reduced output power, you might get other features done better such as excellent imaging, high linearity, and cooler running.
 
Hi,
Today, I test with this :

An externally hosted image should be here but it was not working when we last tested it.


Amp is connected to "high", so I can fix first pot "low" and taste sound.
At his time, best compromize is when Rs 2,1K and Rp 7,8K, if I am right that give -2,6 db for pre-attenuation.

It's a passive preamp I use for bi-amping..

Next amp will be for 8 ohm monitor speaker, actually I use a LM3875 kit, good for speakers and close listening but very noisy and bad with bass..
So I'm looking for "best" schematic LM3875 with 17/18 Volts or maybe less..

Phil.
 
Feed your audio into the 10k attenuator.
Feed the output of the attenuator into a Buffer.
Feed the output of the Buffer into two output resistors.
One output resistor feeds the main signal.
The other output resistor feeds an attenuator and Buffer and an output resistor.
This feeds the "special" receiver.

Source >> interconnect >> 10k vol pot (log law) >> Master Buffer >> 100r >> interconnect >> Receiver
and
The same Master Buffer >> 100r >> 10k pot (lin law) >> Buffer >> 100r >> interconnect >> Special Receiver.
 
Last edited:
To do without a Buffer the Source must be capable of driving the interconnect cable AND the loading presented by the Receiver.

A vol pot is not at all capable of driving any high capacitance interconnect ! especially if there is an effective RF attenuator at the Receiver's Input.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.