Searching \ for '[EE:] More opamp questions' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=more+opamp+questions
Search entire site for: 'More opamp questions'.

Exact match. Not showing close matches.
PICList Thread
'[EE:] More opamp questions'
2004\08\15@211354 by Josh Koffman

face picon face
Ok, I've been reading up and planning. Just a bit of a recap, I'm
trying to design a compact, single supply audio mixer with one
balanced microphone and one line level input. I'll duplicate the
circuit so I end up with a stereo mixer. This isn't super hi-fi. It'll
be used for some live recordings, but nothing for an album, mainly
just for archiving. I can tolerate some noise, as long as it isn't
overly noticable.

I've broken the project down to be able to deal with things a bit
easier. I'm workin on the balanced microphone input. I was looking at
the SSM2135, and in the datasheet
(http://www.analog.com/UploadedFiles/Data_Sheets/527742802SSM2135_0.pdf)
there is an example circuit of a balanced line receiver on page 10.
I've been looking at it more, and it mentions a line impedence of 20K.
Standard balanced microphones are more along the lines of 600 ohms.
So, I wonder if it's possible to modify this circuit for operation at
that lower impedence.

Another option is to use an SSM2135 as a standard differential
amplifier (like fig 4.18 page 185 in Art of Electronics). The circuit
in the datasheet mentions something about input impedance changing
with drive levels in a standard differential amp. Is this something I
need to worry about?

Sorry for all the questions, I really wish I knew more about analog.

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\08\15@215829 by David Duffy

flavicon
face
Josh Koffman wrote:

{Quote hidden}

Just go and buy a Behringer UB502 mixer for about USD$50 !
David...

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\08\16@100704 by Josh Koffman

face picon face
A valid suggestion I guess. Here are my reasons not to.

First off, the UB502 doesn't meet my specs (only one mic input).
Second, my little mixer project is to be integrated into larger
project consisting of an ADC/DAC combination. I'm trying to keep it
small, so integrating a mixer into my work is preferable. Third, how
am I supposed to learn if I just go out and buy whatever I need? Sure
in this case there are products out there that do most of what I want.
That may not be true in the future, and I know I have a deficit in the
analog arena. It's time for me to learn.

Thanks though :)

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Mon, 16 Aug 2004 11:57:36 +1000, David Duffy
<spam_OUTdavidTakeThisOuTspamaudiovisualdevices.com.au> wrote:
> Just go and buy a Behringer UB502 mixer for about USD$50 !

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@113534 by Russell McMahon

face
flavicon
face
> > Just go and buy a Behringer UB502 mixer for about USD$50 !

Sell a man a mixer and he will mix for a lifetime
Teach a man to make mixers and he will ... ?
Oh, never mind.
:-)

       RM

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@113948 by Bob Blick

face picon face
> I've been looking at it more, and it mentions a line impedence of 20K.
> Standard balanced microphones are more along the lines of 600 ohms.
> So, I wonder if it's possible to modify this circuit for operation at
> that lower impedence.

Hi Josh,

You can just add a load resistor. Actually lots of mics are happier loaded
somewhat above 600 ohms, say 1K.

But of course since the amount of power generated by a mic is tiny, and
your circuit will be throwing most of it away, it will be sub-optimum and
noisier than it would be if you could use all the mic's power. But a lot
of commercial preamps suffer from that little problem so you won't be
alone.

Cheerful regards,

Bob

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@114611 by Josh Koffman

face picon face
On Tue, 17 Aug 2004 03:03:35 +1200, Russell McMahon
<.....apptechKILLspamspam@spam@paradise.net.nz> wrote:
> Sell a man a mixer and he will mix for a lifetime
> Teach a man to make mixers and he will ... ?

...make wild stabs into projects that aren't worth the time and will
never be commercially successful (assuming they are ever even
functionally successful)?

It's times like these I wish I knew more. Actually, that's most of the time :)

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@125613 by Josh Koffman

face picon face
Thanks for the reply Bob. I must admit I don't fully understand
though. By load resistor, do you mean a resistor across the true and
complement pins? If so, how would that affect the current 20K input
impedance of the circuit. Problem is, I can't quite figure out how
they're setting the input impedance in the example. Too many 20K
resistors.

However, I've been thinking more about just using an SSM2135 as a
standard differential amp. I guess I just don't understand the
downsides to the simpler solution.

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Mon, 16 Aug 2004 08:39:37 -0700, Bob Blick <bblickspamKILLspamsonic.net> wrote:
> You can just add a load resistor. Actually lots of mics are happier loaded
> somewhat above 600 ohms, say 1K.
>
> But of course since the amount of power generated by a mic is tiny, and
> your circuit will be throwing most of it away, it will be sub-optimum and
> noisier than it would be if you could use all the mic's power. But a lot
> of commercial preamps suffer from that little problem so you won't be
> alone.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@140925 by Bob Blick

face picon face
> Thanks for the reply Bob. I must admit I don't fully understand
> though. By load resistor, do you mean a resistor across the true and
> complement pins? If so, how would that affect the current 20K input
> impedance of the circuit. Problem is, I can't quite figure out how
> they're setting the input impedance in the example. Too many 20K
> resistors.

Hi Josh,

There are three classic designs for opamp differential amps - one opamp,
two opamp, and three opamp. The three opamp one is best and also the
quietest.

The load resistor is for the mic, so it attaches to the mic terminals
where they connect to your amp.

Cheerful regards,

Bob

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@214409 by Josh Koffman

face picon face
Ok, I've been reading and searching more. This website
(http://www.dself.dsl.pipex.com/ampins/balanced/balanced.htm) has some
cool information. In particular, figures 9 and 11, a bit more than
halfway down the stage. Actually, here are the direct links:
http://www.dself.dsl.pipex.com/ampins/balanced/balfig9.gif and
www.dself.dsl.pipex.com/ampins/balanced/balfig11.gif
I'm looking at figure 11 mostly. It seems to be a nice variable gain
balanced to unbalanced job. Now, if I use the SSM2135, I can do each
input with one chip. However, do I need to worry about a capacitor on
the input? A co-worker mentioned the possibility of instability if my
opamp has too wide an input range.

Ideas? Comments?

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Mon, 16 Aug 2004 11:09:04 -0700, Bob Blick <.....bblickKILLspamspam.....sonic.net> wrote:
> There are three classic designs for opamp differential amps - one opamp,
> two opamp, and three opamp. The three opamp one is best and also the
> quietest.
>
> The load resistor is for the mic, so it attaches to the mic terminals
> where they connect to your amp.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@222306 by Bob Blick

face picon face
On 16 Aug 2004 at 21:44, Josh Koffman wrote:

> Ok, I've been reading and searching more. This website
> (http://www.dself.dsl.pipex.com/ampins/balanced/balanced.htm) has some
> cool information. In particular, figures 9 and 11, a bit more than
> halfway down the stage. Actually, here are the direct links:
> http://www.dself.dsl.pipex.com/ampins/balanced/balfig9.gif and
> www.dself.dsl.pipex.com/ampins/balanced/balfig11.gif
> I'm looking at figure 11 mostly. It seems to be a nice variable gain
> balanced to unbalanced job. Now, if I use the SSM2135, I can do each
> input with one chip. However, do I need to worry about a capacitor on
> the input? A co-worker mentioned the possibility of instability if my
> opamp has too wide an input range.

Hi Josh,

Using the feedback loop to control gain, while it looks great on paper, is
a step away from high fidelity. You end up with a circuit where almost
every characteristic varies along with the gain. Not to say it hasn't been
done, it's just a compromise.

As far as the basic one-amp design, your microphone becomes part of
the gain structure. Now with large (15k as shown) input resistors, the
mic's impedance is not dominant, but you get their noise right at the
front end, the gain will be pretty low since the mic signal is attenuated.
When people make mic preamps like this they usually make the
resistors very low value (say 75 to 150 ohms). Now you get a fair
amount of gain, but the gain is dependent on your mic's impedance,
which varies with frequency.

Cheerful regards,

Bob

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\16@230045 by Russell McMahon

face
flavicon
face
> Ok, I've been reading and searching more. This website
> (http://www.dself.dsl.pipex.com/ampins/balanced/balanced.htm) has some

just skimmed it, but that appears to be an excellent discussion.

But do have a look at the classic 3-amplifier "instrumentation amplifier"
circuit. It has some advantages and you can also vary gain with a single
resistor. Generally speaking, it is easier to play with gain, passband
shaping etc after the input stage and optimise things in the input stage
which are unable to be well addressed elsewhere.

Note that the 2 amplifier version usually has inferior common mode rejection
performance across frequency due to unbalanced  delays in the two signal
paths. The 3 amplifier version does not have that particular problem.

Here's a good introductory tutorial - this plus the next few pages take you
from resistive summers to the traditional 3-amplifier 'instrumentation
amplifier". looks like the rest of the course may be worthwhile as well.

       http://www.allaboutcircuits.com/vol_3/chpt_8/8.html

Note that varying Rgain on the cct on page
www.allaboutcircuits.com/vol_3/chpt_8/10.html
gives you single resistor gain control. Remove Rgain and the overall
amplifier gain is unity - which can be useful. The circuit is in fact often
drawn without Rgain present At that stage it becomes VERY familiar - it's
the one amplifier instrumentation amplifier with two input buffers - obvious
but brilliant :-). (Note that it can have gain with Rgain removed if the
resistors per side are unequal - just as the single amplifier cct has gain)

Also note that the input buffers remove the dependence of gain on input
impedance.



Here's another take on the same thing with SPICE emphasis
(Get FREE SPICE from eg Linear Technology)

       http://www.ecircuitcenter.com/Circuits/instamp1/instamp1.htm

Here's a 1973 (!) app note that you won't want to copy directly but which
may jar your mental filters and give you an idea or two.

       http://www.national.com/ms/LB/LB-21.pdf



       Russell McMahon

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\19@093928 by Josh Koffman

face picon face
...more reading and thinking.

I like the circuit at
http://www.allaboutcircuits.com/vol_3/chpt_8/10.html. Good links (as
usual) Russell. I think the best idea is to go for a 3 opamp
differential instrumentation amplifier. I have some SSM2135 on hand
that I could use, or would it make more sense to use a TLC232 as the
actual differential amp and the SSM2135 as the input buffers?

Now my problem becomes gain. I honestly don't quite understand how
much I need to amplify the mic signal in order to get it to the right
level to be mixed with a consumer line level signal. One of the other
sites I've been reading mentions an input gain range of +20 to -10dB
on their mic input stage. I don't think I'll ever need the -10dB
portion, so I don't have to worry about attenuating the input. If I go
with 10K resistors for all the R in
http://sub.allaboutcircuits.com/images/03046.png would I be ok using a
10K pot for the Rgain control? My read of the formulas is that as
Rgain approaches 0, the gain increases. For 10K resistors and a 10K
pot, minimum gain would be 2, and maximum would just go higher as I
decrease the pot. Does that sound right? Will that be enough gain?

Also, I'm very confused about input caps. If I just follow the
schematic I linked to above, do I need input caps? If so, how do I
calculate the value, etc?

Thanks!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Tue, 17 Aug 2004 14:58:55 +1200, Russell McMahon
<EraseMEapptechspam_OUTspamTakeThisOuTparadise.net.nz> wrote:
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2004\08\19@105310 by Nigel Orr

flavicon
face
> Now my problem becomes gain. I honestly don't quite understand how
> much I need to amplify the mic signal in order to get it to the right
> level to be mixed with a consumer line level signal.

It depends on the mic and the source volume :-)

Typically, for close-miked instruments or vocals, you'll need >40dB gain
(x100).
If you're recording gentle classical music, you might need 70dB (I wouldn't
expect it to be any higher, just because none of the professional recording
mixers I've used go any higher :-) ).
If it's rock'n'roll and you're just in front of the speaker stack, you
could probably skip the preamp altogether :-)

> sites I've been reading mentions an input gain range of +20 to -10dB
> on their mic input stage.

That sounds more like line in, or possibly a high impedance mic.  More
expensive mixers usually have a 'pad' switch to give up to 30dB
attenuation, usually used for kick drum mics etc, but mic gain would
probably be 20-70dB

> Also, I'm very confused about input caps. If I just follow the
> schematic I linked to above, do I need input caps? If so, how do I
> calculate the value, etc?

Are you using phantom power (do you normally need the '+48V' switch on for
these microphones)?  Are you planning a transformer balanced input or
direct connection?

I would recommend small (22pF) capacitors across all the opamps from + to -
inputs, if you're not using phantom power or are using a transformer
balanced input, you could get away without series caps.

Nigel
--
This e-mail and any files transmitted with it ("E-mail") is intended solely
for the addressee(s) and may contain confidential and/or legally privileged
information.  If you are not the addressee(s), any disclosure,
reproduction, copying, distribution or other use of the E-mail is
prohibited.  If you have received this E-mail in error, please delete it
and notify the sender immediately via our switchboard or return e-mail.

Neither the company nor any individual sending this E-mail accepts any
liability in respect of the content (including errors and omissions) and
timeliness of the E-mail which arise as a result of transmission.  If
verification is required, please request a hard copy version

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2004\08\19@121143 by Bob Blick

face picon face
If I go
> with 10K resistors for all the R in
> http://sub.allaboutcircuits.com/images/03046.png would I be ok using a
> 10K pot for the Rgain control? My read of the formulas is that as
> Rgain approaches 0, the gain increases. For 10K resistors and a 10K
> pot, minimum gain would be 2, and maximum would just go higher as I
> decrease the pot. Does that sound right? Will that be enough gain?

Typical gain you will need is 20 to 40 db.

Do not use a potentiometer where you suggested. Run your differential amp
at 20 db fixed gain, put the pot afterwards as a voltage divider, and
follow that with a single opamp with 20 db more gain.

Using input coupling capacitors gives you a small measure of protection
against DC into your opamps.

You do need to set a DC level on the inputs of the opamps - pull them down
with resistors. 10K is fine.

And load the mic with 1 or 2k.

Cheerful regards,

Bob

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2004\08\23@113629 by Josh Koffman

face picon face
So, just to confirm,  here is what you suggest. The total gain of the
differential amp (three opamps total) should be 20db. Then a voltage
divider, and a simple single opamp set at 20db gain. Should it be
inverting or non inverting? I could do this with two SSM2135 chips as
they are dual opamps per chip. I'll put 10K pulldown resistors on each
input of the ultimate input. Do I need them on subsequent stages as
well? I'll put 22pF caps from the true input to the complement input.
If I use DC blocking series caps, how do I figure out the value? I
will also have a 2K resistor across the put (in parallel with the 22pF
cap).

How does that sound? I'm going to try to draw this up so I can post an
image. And just out of curiousity, why is varying the Rgain resistor
to change the gain a bad idea?

Thanks,

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Thu, 19 Aug 2004 09:11:40 -0700, Bob Blick <bblickspamspam_OUTsonic.net> wrote:
{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
@spam@piclist-unsubscribe-requestKILLspamspammitvma.mit.edu

2004\08\24@041355 by Nigel Orr

flavicon
face
pic microcontroller discussion list <> wrote:
> So, just to confirm,  here is what you suggest. The total gain of the
> differential amp (three opamps total) should be 20db. Then a voltage
> divider, and a simple single opamp set at 20db gain.

You still haven't said what sort of mics you are using and exactly what you
are expecting to record, but I don't think that's enough gain, unless they
are super-cheap high impedance or super-cheap electret mics.  Can you
connect one into a 'scope and see what the typical signals are?  Or test it
through an existing mixer?

If these are cheap high-output mics, or you only need to get up to -10dB
consumer audio recording level, and the source is fairly loud, 40dB may be
adequate.

> Should it be
> inverting or non inverting?

If you're not worried about phase with respect to other mics, it doesn't
really matter.  (As it's a true differential input, you can arbitrarily set
the input pins as '+' and '-')

> If I use DC blocking series caps, how do I figure out the value? I
> will also have a 2K resistor across the put (in parallel with the
> 22pF cap).

The 2 C's go along with the two 10K resistors (ignoring mic impedance) to
make a filter with rolloff frequency (3dB down) at
f = 1/(2*pi*R*C)

So a 1uF cap would set a rolloff at about 8Hz.

Nigel
--
Nigel Orr, Design Engineer                 KILLspamnigelKILLspamspamaxoninstruments.co.uk
Axon Instruments Ltd., Wardes Road,Inverurie,Aberdeenshire,UK,AB51 3TT
              Tel:+44 1467 622332 Fax:+44 1467 625235
                  http://www.axoninstruments.co.uk

This e-mail and any files transmitted with it ("E-mail") is intended
solely for the addressee(s) and may contain confidential and/or legally
privileged information.  If you are not the addressee(s), any
disclosure, reproduction, copying, distribution or other use of the
E-mail is prohibited.  If you have received this E-mail in error,
please delete it and notify the sender immediately via our switchboard
or return e-mail.

Neither the company nor any individual sending this E-mail accepts any
liability in respect of the content (including errors and omissions)
and timeliness of the E-mail which arise as a result of transmission.
If verification is required, please request a hard copy version

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2004\08\25@082331 by Josh Koffman

face picon face
Ok, I've posted a PDF of my current efforts at
http://joshy.redmartian.org/opamp.pdf

I'd love any and all comments on my design choices, component values,
etc. Some things in particular I'm unsure of: I believe I've set both
amplification stages to 10x. A signal running through both stages
would be amplified by 100x then, correct? I haven't added the input
series caps. Nigel's calculations show a value of 1uF. Would I want
this to be polarized or non polarized? Do the 22pF caps across the
opamps need to be present on all the opamps, or just the two leftmost
ones? Does the mic loading resistor look ok?

In regards to the question about my input source, this will vary. I
would guess that for the most part I will be using dynamic mics,
professional ones, but low output level. There is the chance for a
higher output model at some point in the future. The mics I will
likely be using won't neccesarily be designed for ambient micing, so I
might have to boost their signal quite a bit. Would it make sense to
bump up the gain a bit more? I was thinking of taking the two stages
up to about 12x each.

Thanks!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Tue, 24 Aug 2004 09:13:03 +0100, Nigel Orr
<RemoveMEnigelTakeThisOuTspamaxoninstruments.co.uk> wrote:
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\08\26@081822 by Josh Koffman

face picon face
Ok, no responses so far. No, I'm not complaining, I don't demand help
everytime I ask :)

That said, would anyone prefer my schematic snippet in a different
file format? I posted a link to a pdf, but I will happily convert it
to whatever you need if you think you might be able to offer some
advice.

The current link is: http://joshy.redmartian.org/opamp.pdf and my
questions are below.

Thanks!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Wed, 25 Aug 2004 08:22:07 -0400, Josh Koffman <spamBeGonejoshybearspamBeGonespamgmail.com> wrote:
{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\26@124446 by Bob Blick

face picon face
> Ok, no responses so far. No, I'm not complaining, I don't demand help
> everytime I ask :)

Hi Josh,

Just well over 1.0 load for a few days :-)

You can run more gain if you like. If you are recording quiet things you
will need it.

I usually like to put something in to protect the inputs from static
discharge, and also something to roll off RF.

Someone posted a PAIA mic preamp schematic a few days ago that had a
couple of caps and four diodes. That's the kind of thing you need to put
in.

I notice you have a 22 pf cap from inverting to non-inverting input on two
opamps. Unless your opamp needs it for stability, stay away from stuff
like that.

Cheerful regards,

Bob

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\26@154353 by Dwayne Reid

flavicon
face
At 06:18 AM 8/26/2004, Josh Koffman wrote:
>Ok, no responses so far. No, I'm not complaining, I don't demand help
>everytime I ask :)
>
>That said, would anyone prefer my schematic snippet in a different
>file format? I posted a link to a pdf, but I will happily convert it
>to whatever you need if you think you might be able to offer some
>advice.
>
>The current link is: http://joshy.redmartian.org/opamp.pdf and my
>questions are below.

Hi there, Josh.

1) Why include R52 & R67?  They only degrade your CMRR and I don't see any
gain from including them.  Input stage gain is (R37+R39)/R38.

2) I'm not real happy with C22 & C23 - they have the potential for causing
you real grief.  If the data sheet says to include them, then do
so.  Otherwise, remove them.

3) RFI suppression - small caps from both inputs to gnd.

4) ESD protection - 2 diodes per input to Vee & Vcc.

5) input capacitors - count on having someone hook this up to a hard-wire
mic split that has 48V phantom on it.  Might be years before it happens . . .

6) input termination - 1K seems just a tad low.  I'm used to seeing between
1.2K to 2K.  This matters only if you wind up connected to a transformer
mic split - this input impedance forms part of the damping network that
flattens the response of the transformer.  Too low of a load impedance
results in somewhat reduced HF response from the transformer iso.

7) R43 might want to be a 1K trimpot in series with 9K53 resistor - short
both inputs together and apply 1K tone.  Adjust trimpot for minimum
output.  Means that you don't have to match R40, 41, 42, 43 to within .01%
for decent CMRR.

Hope this helps.

dwayne

--
Dwayne Reid   <TakeThisOuTdwaynerEraseMEspamspam_OUTplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax

Celebrating 20 years of Engineering Innovation (1984 - 2004)
 .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-.   .-
    `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'   `-'
Do NOT send unsolicited commercial email to this email address.
This message neither grants consent to receive unsolicited
commercial email nor is intended to solicit commercial email.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\26@164004 by Josh Koffman

face picon face
Ok. Thanks for the note about the caps. I believe it was Nigel or
someone else who suggested that. Maybe a good plan would be to leave
them in the schematic, and just not populate them? That way the pads
are there should I need them.

My plan for the inputs is to put some capacitors on them. I just
haven't figured out if they should be polarized or not.

And thanks for taking the time to reply :)

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

On Thu, 26 Aug 2004 09:45:38 -0700, Bob Blick <RemoveMEbblickspamTakeThisOuTsonic.net> wrote:
{Quote hidden}

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\26@164726 by Josh Koffman

face picon face
Thanks for the response Dwayne. Lots of good comments...my replies within.

On Thu, 26 Aug 2004 13:42:14 -0600, Dwayne Reid <dwaynerEraseMEspam.....planet.eon.net> wrote:
> 1) Why include R52 & R67?  They only degrade your CMRR and I don't see any
> gain from including them.  Input stage gain is (R37+R39)/R38.

Well, they were there to bias the inputs. They were suggested earlier,
but I was unclear if they were needed on both inverting and non
inputs, or on just the actual inputs from the mic.

> 2) I'm not real happy with C22 & C23 - they have the potential for causing
> you real grief.  If the data sheet says to include them, then do
> so.  Otherwise, remove them.

As I mentioned in another email, this was suggested by someone else.
I'm going to leave the pads on the board but not populate unless I
have to.

> 3) RFI suppression - small caps from both inputs to gnd.

Perhaps I misunderstood and put C22 and C23 in instead of these caps.
How small a value should I use? Would they go from the inputs of each
opamp to ground, or from the microphone input to ground?

> 4) ESD protection - 2 diodes per input to Vee & Vcc.

Just regular signal diodes ok? 1N4148?

> 5) input capacitors - count on having someone hook this up to a hard-wire
> mic split that has 48V phantom on it.  Might be years before it happens . . .

I'm looking at 1uF caps. Should I use polarized or non?

> 6) input termination - 1K seems just a tad low.  I'm used to seeing between
> 1.2K to 2K.  This matters only if you wind up connected to a transformer
> mic split - this input impedance forms part of the damping network that
> flattens the response of the transformer.  Too low of a load impedance
> results in somewhat reduced HF response from the transformer iso.

Ok, sounds good.

> 7) R43 might want to be a 1K trimpot in series with 9K53 resistor - short
> both inputs together and apply 1K tone.  Adjust trimpot for minimum
> output.  Means that you don't have to match R40, 41, 42, 43 to within .01%
> for decent CMRR.

If I short the inputs together, am I applying the tone between the
inputs and ground? I'm not going to be using long input cables, and I
will likely have noise from elsewhere. I'm trying to stick to surface
mount, and a trimpot is a bit big. I guess I can try to fit this in
though.

> Hope this helps.

You bet it does! Thank you!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
       -Douglas Adams

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\08\27@052404 by Nigel Orr

flavicon
face
pic microcontroller discussion list <> wrote:
> Ok. Thanks for the note about the caps. I believe it was Nigel or
> someone else who suggested that.

It was me!  I started doing it routinely a few years ago, after it was
recommended somewhere or by someone.  It's certainly never caused problems
yet, and has definitely resolved some EMC issues, where the caps were
omitted but the tracking was there for them, they were fitted after EMC
testing gave borderline immunity results.

I think it was recommended in a book on EMC design or the UK EMC journal,
and seemed like a useful technique at the time.  I could well be wrong, or
it could have been for a particular opamp- any other info on the practice
would be welcome!

From the two postings, I guess they aren't all beneficial (nothing ever
is!)- what problems would you expect them to cause?  HF instability?  I can
see that caps from each input to ground would be better, but I have enough
trouble trying to convince people to put any extra components around a
circuit to cover ESD/EMC, two caps would definitely be pushing my luck :-)

Other than that and the other comments posted, the circuit looks fine.
With my 'sound engineer' hat on, I'd still recommend an extra 20dB of gain
in reserve.  You (Josh) suggested upping the gain of each stage to x12.
That's not really significant, 21.6dB per stage against 20dB per stage.
2-3dB is the minimum gain step that most people can perceive as being
louder or quieter, it won't make a big difference in real usage.  Note that
it doesn't make the circuit any 'better'- there's no reason not to build
it, try it, then adjust the gain for what you find you need in your
application.

Like the others, I've not had time to reply to Josh sooner, hope it was
worth waiting for :-)

Nigel
--
This e-mail and any files transmitted with it ("E-mail") is intended
solely for the addressee(s) and may contain confidential and/or legally
privileged information.  If you are not the addressee(s), any
disclosure, reproduction, copying, distribution or other use of the
E-mail is prohibited.  If you have received this E-mail in error,
please delete it and notify the sender immediately via our switchboard
or return e-mail.

Neither the company nor any individual sending this E-mail accepts any
liability in respect of the content (including errors and omissions)
and timeliness of the E-mail which arise as a result of transmission.
If verification is required, please request a hard copy version

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email EraseMElistservspammitvma.mit.edu with SET PICList DIGEST in the body

2004\08\27@053235 by Nigel Orr

flavicon
face
pic microcontroller discussion list <> wrote:
> this to be polarized or non polarized?

Non polarised, ideally.  Have a search on the web for recommended capacitor
types for audio applications, electrolytics in the signal path are usually
best avoided!

> would guess that for the most part I will be using dynamic mics,
> professional ones, but low output level.

If you don't want to go through the cycle of build it, test it, change the
gain that I suggest in my other mail, you can try working out the gain for
yourself.

You can probably guess the SPL at the recording position to within 10dB,
even if you don't have an SPL meter.  The mics will specify sensitivity so
you can work out the corresponding electrical output from the mic.  Then
you can figure out how much gain you need to get that up to the input level
for your recording device.

e.g. if the SPL is 100dB, and the mic sensitivity is 1mV @ 100dB, the
output voltage will be 1mV.  If you need 100mV (-10dBV) output, you need
x100, 40dB gain.  More likely the specs will be in mV/Pa and dBm, so you
will need to fight your way through the numbers.

It would probably be more interesting to just build one and try it :-)

Nigel
--
This e-mail and any files transmitted with it ("E-mail") is intended
solely for the addressee(s) and may contain confidential and/or legally
privileged information.  If you are not the addressee(s), any
disclosure, reproduction, copying, distribution or other use of the
E-mail is prohibited.  If you have received this E-mail in error,
please delete it and notify the sender immediately via our switchboard
or return e-mail.

Neither the company nor any individual sending this E-mail accepts any
liability in respect of the content (including errors and omissions)
and timeliness of the E-mail which arise as a result of transmission.
If verification is required, please request a hard copy version

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservEraseMEspamEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

2004\08\27@053649 by Russell McMahon

face
flavicon
face
I approve of reasonable message trimming.
BUT it would be nice to know what eg IT is here.
You can read the whole post with no certainty about the subject, whereas
adding perhaps one more line of quoting would have made it clear.
(It's got something to do with capacitor placement - but it's not clear
exactly what).

       RM

> pic microcontroller discussion list <> wrote:
> > Ok. Thanks for the note about the caps. I believe it was Nigel or
> > someone else who suggested that.

&

> It was me!  I started doing it routinely a few years ago, after it was
> recommended somewhere or by someone.  It's certainly never caused problems

&

> I think it was recommended in a book on EMC design or the UK EMC journal,

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservspam_OUTspamKILLspammitvma.mit.edu with SET PICList DIGEST in the body

2004\08\27@064611 by Nigel Orr

flavicon
face
Russell wrote:
> I approve of reasonable message trimming.
> BUT it would be nice to know what eg IT is here.

Sorry, I usually manage to quote _some_ context! :-)

'IT' refers to putting a small cap (22pF) between the inverting and
non-inverting inputs of an opamp, specifically for a mic preamp.  I've
reposted my email below so it's all in context :-)

Nigel


Josh <> wrote:
> Ok. Thanks for the note about the caps. I believe it was Nigel or
> someone else who suggested that.

It was me!  I started doing it routinely a few years ago, after it was
recommended somewhere or by someone.  It's certainly never caused problems
yet, and has definitely resolved some EMC issues, where the caps were
omitted but the tracking was there for them, they were fitted after EMC
testing gave borderline immunity results.

I think it was recommended in a book on EMC design or the UK EMC journal,
and seemed like a useful technique at the time.  I could well be wrong, or
it could have been for a particular opamp- any other info on the practice
would be welcome!

From the two postings, I guess they aren't all beneficial (nothing ever
is!)- what problems would you expect them to cause?  HF instability?  I can
see that caps from each input to ground would be better, but I have enough
trouble trying to convince people to put any extra components around a
circuit to cover ESD/EMC, two caps would definitely be pushing my luck :-)

Other than that and the other comments posted, the circuit looks fine.
With my 'sound engineer' hat on, I'd still recommend an extra 20dB of gain
in reserve.  You (Josh) suggested upping the gain of each stage to x12.
That's not really significant, 21.6dB per stage against 20dB per stage.
2-3dB is the minimum gain step that most people can perceive as being
louder or quieter, it won't make a big difference in real usage.  Note that
it doesn't make the circuit any 'better'- there's no reason not to build
it, try it, then adjust the gain for what you find you need in your
application.

Like the others, I've not had time to reply to Josh sooner, hope it was
worth waiting for :-)

Nigel
--
This e-mail and any files transmitted with it ("E-mail") is intended
solely for the addressee(s) and may contain confidential and/or legally
privileged information.  If you are not the addressee(s), any
disclosure, reproduction, copying, distribution or other use of the
E-mail is prohibited.  If you have received this E-mail in error,
please delete it and notify the sender immediately via our switchboard
or return e-mail.

Neither the company nor any individual sending this E-mail accepts any
liability in respect of the content (including errors and omissions)
and timeliness of the E-mail which arise as a result of transmission.
If verification is required, please request a hard copy version

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservTakeThisOuTspamspammitvma.mit.edu with SET PICList DIGEST in the body

More... (looser matching)
- Last day of these posts
- In 2004 , 2005 only
- Today
- New search...