What is your opinion on this schematic for OPA-541

[100db]

Hi everyone,

What do you think about this sch. ?
It was applied from Inverted LM-3875 sch.

[lineup]

I see no reason why this shouldnt work.
Using OPA548, OPA549 and OPA541.

See for example this website with Gainclones:
http://www.freewebs.com/matttcatttweb/audio.htm

A search with google generates many hits:
OPA541 gainclone


As told by Kuei Yang Wang, with good knowledge about inverted gainclones,
R1 potentiometer should be a linear pot
lm3875 inverted layout with 100k pot, not enough volume?

[100db]

Hello lineup, thanks for a nice link.

I just tried it out with no Rcl at pin 8 ( not connected, just leave it ) and use 100k pot. 24-0-24 transformer with UR1620/1620R give +/-35 v to the chip.

I'm facing the opposite problem, sound come out too loud.
dont' know why
When listening at volume ~10% I feed lack of bass, need to turn the volume about 25% way to get sufficient bass but over all sound come out too loud.

I'm not sure if the gain is too high or not ?

[100db]

Get it fixed by changing pot to steped atten'
sound quality doesn't beat the old 3875 yet, will give it some time to break in.

[Bazukaz]

datasheet shows 0.1 % thd+n for OPA at 1 khz. Pretty high.

[lineup]

Originally posted by 100db
Hello lineup, thanks for a nice link.

I just tried it out with no Rcl at pin 8 ( not connected, just leave it ) and use 100k pot.
24-0-24 transformer with UR1620/1620R give +/-35 v to the chip.

I'm facing the opposite problem, sound come out too loud.
dont' know why
When listening at volume ~10% I feed lack of bass, need to turn the volume about 25% way to get sufficient bass but over all sound come out too loud.

I'm not sure if the gain is too high or not ?

Here is an app note for OPA541, about how using Current Limit.
http://www.ti.com/sc/docs/psheets/ab...ps/sboa057.htm

The pin 8, should be conneted to output, if you do not use Rcl, resistor for Current Limit.

If you use 4.7uF - 10 kOhm at input, there shouldnt be a problem with bass.
And gain of 22 should be a very good value with signal from CD-player.

You should add a zoble network from output(pin5+7) to ground.
One resistor 4.7 Ohm (2/5 Watt) in series with 100nF capacitor will be good.
Will make your amp a bit more safe.
And avoid high frequency oscillation, which not always you can discover and know is there.

Looking at THD+noise diagram, and at 1 kHZ
0.07 % at 50 Watt is very good!
0.11 % at 5 Watt is strange, but still okay.

Normally I would expect lower distortion at lower power output.
Maybe somebody can explain ?

[odysseybmx414]

you should only use gain up to 10 with opa541 and opa549 they dont sound too well with higher gain but sound awesome at gain of 10. lower the gain and u should hear beautiful music out of your opa541



also use lower value resistors (1K and 10K for gain setting works great and sounds great but 470 ohm and 4.7K resistors on my 2 ch opa541 chipamp sounds the best so far)

[100db]

Every one,
thanks for your input. I try to correct my sch. follow your advice, is this seems to be ok ?

After changing from a cheap pot to a stepped atten', output that used to be too loud become normal. I have to turn the volume up to 50-60 % at normal listening level.

I found this opa-541 (20 hrs used ) give a tighter bass but the old 3875 ( 200 hrs used) with a basic sch. as above give better detail.

What is your experience on this 2 chips ?

[lineup]

The 4.7uF input cap and 10k resistor sets the rolloff of low frequency.
It is a high pass RC filter.
The filter has a roll off frequency, and passes frequencies higher than this roll off frequency.
If we call this freq 'f'

Formula:
f = 1 / (2 x pi x R x C )

pi = 3.14
Using 10k and 4.7uF we get:
R = 10.000
C = 0.0000047

f = 3.38 Hz
At this freq bass is -3dB


10 x 4.7 = 47 (kOhm x uF)
If you keep 4.7uF cap and makes 10k resistor 1k, we get
1 x 4.7 = 4.7 (kOhm x uF)

As this value is 10 times lower, the frequency,
according to formula f will be 10 times higher.
So you get roll off at 33.8 Hz.
This filter will pass frequencies higher than 33.8 HZ
but begin to block lower freq.

I personally can accept a roll off at f=10 Hertz.
If RC =
15k x 1uF = 15 (kOhm x uF)
This makes f = 10 Hz, almost
1.5k x 10uF = 15 gives the same.


To keep your f, now 3.38 Hz
and use 470 Ohm resistor, you will have to increase Cap
to keep RxC the same.
As R is 20 times smaller ( 10.000/470 )
your cap need to be 100uF. (20x 4.7)

This is the drawback of using too small resistor, instead of 10k.
You need a very high value cap.
And as we want to use film cap (polypropylene, 9/10 people agree film cap should be used in input)
this will be a little problem.

10uF you might find (or parallel two 4.7uF) and if you make resistor 1.5k
you will have roll off: f=10 Hz

I would not recommend lower feedback resistors,
than 1k-2.2k and the other 10 times higher
for example 2.2k and 22k

[Nordic]

I have been wondering about that, this week...
Why do we attenuate the amp, beloew the range of what our speakers can produce? Are we not makeng the amp do unneccesary work?