line level signal?

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
cant
Posts: 6
Joined: Thu Apr 16, 2009 2:48 pm
Contact:

line level signal?

hi
i was wondering if someone knew if 3.8v @ 2mA would be too high a line level signal.
i read that a nominal line level falls around 1v @1mA but i'm more curious if the numbers i initially stated would be too far out of range and would endanger devices it would be passed on to.
the idea i'm working with is to provide a signal input where a electret microphone formerly was.
i cant seem to find a way around having a voltage sent to power the former mic and so this 1/4" input jack i'm placing there would be recieving the [email protected], so i'm trying to find out whether this would damage anything i would plug into the 1/4" jack.
in a nutshell this former mic input is becoming a 1/4" input on this tape player i'm modifying.
hrm, tried to make that clear, let me know if there is anything else i could clarify.
any help is appreciated.
thanks

bodgy
Posts: 1044
Joined: Tue Dec 04, 2001 1:01 am
Location: Australia
Contact:

Re: line level signal?

If you require a parasitic power supply, you need to feed your 3v8 to the top end of a d.c choke. The other end of the choke goes to the microphone power connection and you'll need a capacitor in series to prevent the d.c getting into the line in circuitry.

so +3v8
|
L
|
|......C................. Line in
Microphone power supply +
GND

What you propose, if it doesn't 'blow' the input circuitry, you'll get massive distortion and noise.

Colin
On a clear disk you can seek forever.

bodgy
Posts: 1044
Joined: Tue Dec 04, 2001 1:01 am
Location: Australia
Contact:

Re: line level signal?

Ah just realised I misread your question. A blocking capacitor would stop the dc power voltage reaching the other side.

Colin
On a clear disk you can seek forever.

haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: line level signal?

While a 3.8vac/2mA signal probably wouldn't "blow" the next stage, it will probably result in clipping and severe distortion of the audio you reproduce. All you need is a voltage divider. 2 resistors or a potentiometer. A series cap would be needed if the new line level signal had a DC offset otherwise its optional.

2mA depends on the impedance of the next stage. If it is high impedance, much less than 2mA will flow but you will still have the 3.8V. If it is low impedance and more than 2mA wants to flow, the voltage will be reduced (collapse) to satisfy ohms law.

Input signal 3.8vac
|
|
R1
|
------- output to next stage
|
|
R2
|
|
Ground

You want 1V across R2 and 2.8V across R1 and 2mA or less flowing through R1 + R2.

Using 1mA R=V/I R=3.8/.001 = 3800 ohms=R1+R2.
for 1V across R2 R=1/.001 or 1kohm with 2.8k left over for R1

Or you could use a 3K to 5K pot (audio taper) and have a level adjust.

This assumes a high impedance (>>10k ohm) next stage. If the next stage is much lower, it will offset the previous calculation and needs to be known to get an accurate value. A pot will give you enough adjustability to compensate for most assumptions.

cant
Posts: 6
Joined: Thu Apr 16, 2009 2:48 pm
Contact:

Re: line level signal?

A blocking capacitor would stop the dc power voltage reaching the other side.
just any old capacitor will do?

haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: line level signal?

Well, not just any cap but a wide range will work likely acceptably. In general, loo large a value will have excessive cost and perform poorly at higher frequencies because they look more like an open circuit. Too small a value won't have the capacity to let the lower frequencies through. But most small ceramic or small electrolytic caps you run across are probably adequate.

To select an optimal value I would either have to dust off a text book or more simply google a few amplifier circuits, survey what those designs use and choose a convenient value close to those values. Anything from 220pF to 20uF will probably work Okay anything over 16V is good enough also 1uF seems to be a popular choice.

A lot has to do with the impedance of the stage you are decoupling. The RC product (time constant) should be long enough to accomodate your lowest frequency. Ive seen as low as 47pF on op amp inputs and as high as 47uF before a 22k to ground.

Some examples
http://www.discovercircuits.com/A/a-audioamp.htm

Your best bet is to experiment with a few vlaues and choose one that does not distort the sound much.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: line level signal?

cant wrote:
A blocking capacitor would stop the dc power voltage reaching the other side.
just any old capacitor will do?
The "blocking" capacitor term applies only to DC and can be of any value for that purpose. But this same capacitor is also a "coupling" capacitor for the AC signal it must pass on to the following circuitry. A good rule is to keep its reactance at no more than 10% of the ensuing loads input impedance.
For example:
Lowest frequency in passband - 100 Hz
Input impedance of following stage - 600 ohm
Capacitor required - 25 MFD @ sufficient breakdown voltage for your circuit.
( 25 MF = 60 ohms Xc @ 100 Hz = 10% of input Z of 600 ohm)
You can use the above as 'Normalized' ratings and scale up or down to your specific frequency and required cap value.

MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: line level signal?

Hi Robert,

I am happy to see that you included an example with your general rule so that people that
want to do this can check their work with the example. I like to see that with posts that
show different ways to do things. You can throw a bunch of theory at people but it is the
example that clinches the understanding. Thanks for taking the time to do that.

The only thing i would add is to put your basic rule into a formulitic form to make the
calculation quick. Based on your rule, here it is:

C=5/(pi*F*Z)

where
C is the capacitance in Farads
F is the frequency in Hertz
Z is the existing impedance in ohms
pi is the constant 3.14159...

Using the above, we come out with 26.5uf, so 25uf would be close enough to a more standard value.
LEDs vs Bulbs, LEDs are winning.

bodgy
Posts: 1044
Joined: Tue Dec 04, 2001 1:01 am
Location: Australia
Contact:

Re: line level signal?

You could also be more standard by using - the normal 1/(2pi * F * Z)

Double checking: Z = 1/(2pi *F * C) = 1/(6.28 * 100Hz * 26.5uF) = 60.1 ohms
MrAl wrote:

C=5/(pi*F*Z)

where
C is the capacitance in Farads
F is the frequency in Hertz
Z is the existing impedance in ohms
pi is the constant 3.14159...

Using the above, we come out with 26.5uf, so 25uf would be close enough to a more standard value.
Colin
Last edited by bodgy on Fri May 15, 2009 11:28 pm, edited 1 time in total.
Reason: Replaced R with Z for correctness
On a clear disk you can seek forever.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: line level signal?

Both your formulas have merit for newcomers or more accurate work, but beleive it or not I just quickly did this reply in my head from a few' normalized' component values that are stuck in my head from years of audio and RF amp design. Being too lazy to dig out the textbooks when I need a quick answer for interstage coupling or even Vcc decoupling values - I always revert to this method ( hence the approximation of 25 mfd and 60 ohm). The fact that I can never get the exact over the counter component values (10% or 5% at best) as a more accurate formula would give me I just generalize with the above thinking. Even the "10% reactance" rule is somewhat arbitrary in that it gives almost a purely resistive circuit with little phase shift and fractional DB attennuation at the low end frequency. However when higher accurracy counts then your formulas would definately be needed and convenient.

CeaSaR
Posts: 1759
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: line level signal?

So what you are really doing is removing a built-in condenser mic and putting a
1/4" jack in it's place, hence the need to remove the operating voltage from the
signal line. I'm pretty sure that the preamp end already has a blocking cap to
prevent the DC from entering, so find that one and use the same value. The other
option that comes to mind is to find the voltage injection trace and "break" it. One
thing you will probably find is that a standard dynamic mic could use a boost, as
they do not normally have as much output as a condenser (electret) mic. You might
need a little extra amplification to go along with it. You could use one that the guys
came up with in my earlier thread on "Mic Pre-Amp". You could also use any number
of op-amp solutions also, ranging from (the low end) LM741 into the venerable TLO72
and beyond. Most have basic circuits in their spec sheets that you can look at from
the individual vendors. That, and there are a "S"-load of schematics on-line.

What kind of tape player is it? Is the 1/4" jack just for another mic or are you going
to plug other things into this jack?

If you plug other "line-level" devices into the same jack, you'll overload the preamp
stage in the tape player. This will, at the very least, create horrible distortion and
may damage the input altogether. Just food for thought.

CeaSaR
Hey, what do I know?

MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: line level signal?

Robert Reed wrote:Both your formulas have merit for newcomers or more accurate work, but beleive it or not I just quickly did this reply in my head from a few' normalized' component values that are stuck in my head from years of audio and RF amp design. Being too lazy to dig out the textbooks when I need a quick answer for interstage coupling or even Vcc decoupling values - I always revert to this method ( hence the approximation of 25 mfd and 60 ohm). The fact that I can never get the exact over the counter component values (10% or 5% at best) as a more accurate formula would give me I just generalize with the above thinking. Even the "10% reactance" rule is somewhat arbitrary in that it gives almost a purely resistive circuit with little phase shift and fractional DB attennuation at the low end frequency. However when higher accurracy counts then your formulas would definately be needed and convenient.
Hi Robert,

Sorry, i didnt mean to imply that the formula would be more accurate in some way, just that it is
convenient to also have things like this in a nice little formula which can be pulled out now and then.
LEDs vs Bulbs, LEDs are winning.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: line level signal?

MrAl
Oh yes I agree with that. My quickie method is a little humorous and almost too embarassing to reveal, but I base it on the following:
1 PFD (1x10'-12) = 150K XC @ 1 MHz (1x10'6)
1 uH(1x10'-6) = 6.3 ohm XL @ 1 MHz (1x10'6)

When confronted with a caculation involving reactances, I just quickly pull these out of my head since they seem to be stuck there forever. Then its just a matter of quickly moving thru orders of magnitude to arrive at an answer. Not very professional, but for initially laying out a circuit scheme it has saved me a lot of time early on. Of course when its time for the rubber to hit the pavement, then some finer tuning with formulas is usually in order.

Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: line level signal?

Robert Reed wrote:Both your formulas have merit for newcomers or more accurate work, but beleive it or not I just quickly did this reply in my head from a few' normalized' component values that are stuck in my head from years of audio and RF amp design. Being too lazy to dig out the textbooks when I need a quick answer for interstage coupling or even Vcc decoupling values - I always revert to this method ( hence the approximation of 25 mfd and 60 ohm).
AHA! I sometimes use the same method, except it's 1,000uF is 8 Ohms @ 20Hz. (My long deceased Lafayette audio amplifier had 1,000uF caps in series with the outputs.) With those values I can scale just about any other values in my head for rough calculations.

Doing math in my head helps me get to sleep at night.
-=VA7KOR=- My solar system includes Pluto.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: line level signal?

As the Shakers used to say - "sometimes simple is better"

Who is online

Users browsing this forum: No registered users and 33 guests