mic preamp modification

[rex]

On Sat, 09 Apr 2005 02:00:27 GMT, Fred Bloggs <nospam@nospam.com> wrote:

rex wrote:

I don't have any schematic, but I opened the thing up and had a look at
what is in the preamp area. Assuming the signal flows sensibly across
the board, there are (per channel) two largish electrolytics. Then a
couple sot-23 devices (transistors I assume) marked 62Z. Then an HC4052A
analog mux. Then a JRC 2068 op amp.

I was surprized to see that 4052 so close to the input. Not sure what
exactly it is doing.


The 4052 is most probably switching all those inputs XLR and LINE into
the preamplifier chain.

I just remembered that this thing has a 20 dB mic attenuation switch.
That must be one of the functions that the 4052 is doing. There is also
a mono recording mode where the left mic input is recorded in the left
channel but a 15 dB down version of the left mic is recorded on the
right channel. I forgot about all these complexities.

I just tried flipping the 20 dB atten on and it reduces the hiss. Maybe
that is telling me that something in the cap/transistor/switch region is
the source of the noise.

The line inputs come into the board away from where this 4052 is. I
think there may have been another 4052 closer to the line inputs but I'm
not sure and don't want to dismantle it agan now.

The JRC2068 is a low noise wideband amplifier in
wide use in many designs, like a lot of Yamaha stuff- so they did not
skimp there.

I saw a discussion somewhere on the net about audio circuits where
someone replaced 2086 with AD8066 or some other amps. No one in this
thread seems to think the 2086's are a problem, so I'll accept they are
probably ok.

The 62Z SMD marking code most likely belongs to Zetex, and
if it's a simple 3-lead single transistor, it is probably a simple
emitter follower used to present a high (6.2K ohm they say) impedance to
the XLR input and a low impedance source for the JRC2068 for lowest
noise-and this does not preclude linearizing feedback from the 2068.

I have no reason to doubt that speculation on function.

I had a quick browse around the Zetex pages just now and didn't find any
SOT23 NPNs that match the marking code.

From your description, the problem sounds like it is not that the
preamplifier noise floor is so high, but that your microphone signal
amplitude is so low- at least when loaded by the preamplifier- this is
borne out by your observation that the low-level dynamic mics are coming
in at lower amplitude than you are used to..

Well as I said in another thread message, the specs aren't that good as
published and that one guy is improving S/N by 15 db with some kind of
mod. There are lots of posts various places around the net that say the
main flaw in this recorder is the mic preamps.

I do appreciate everone's comments here. It's helping me think about how
to look at it when I get to opening it up again.

 

[rex]

On Sat, 09 Apr 2005 04:50:17 +0100, Pooh Bear
<rabbitsfriendsandrelations@hotmail.com> wrote:

From browsing the web on this product, it seems to be well recognised
that the mic preamp is pretty lame. That's why Doug Oade is doing some
business by offering mods on the one's he sells. The published S/N spec
is 65 dB. He measured 80 dB after his changes. I'd be happy with that.

I'm curious about that. He may possibly be simply replacing the input transistors.
Takes a few minutes to do that !

Without seeing the unit I'm at a disadvantage. Are you familiar with the
'conventional' leaded TO-92 package ? If there aren't any in the mic amp then I
reckon I know what's going on.

Ok, this is sounding good (no pun, of course). I was just reading your
reply to Fred's message too.

The only large devices in the input area are the 4 (2 per channel I
assume) electrolytics. There are then two transistors per channel and
they are both SOT-23 form. No TO-92 sized stuff anywhere that I noticed.

Next in line after the transistors seems to be the 4052 then the 2068. I
mentioned in the other message some possible functions for the 4052 that
just occured to me.

So I await further comments on the transistors.

 

[John Woodgate]

I read in sci.electronics.design that Erik Walthinsen
<omega@pdxcolo.net> wrote (in <d37h040j2t@enews3.newsguy.com&gt about
'mic preamp modification', on Fri, 8 Apr 2005:

AFAICT there are at least 3 major classes of dimmer topologies, and I'm
assuming that the cheaper models use the least
effective/efficient/quiet.

Which three classes do you have in mind?

There is 'phase-control, turn on' - this is the oldest electronic
technology and uses triacs of thyristors as the control elements. It can
cause lamp buzz and audio interference.

There is 'phase-control, turn off', which is fairly new and uses IGBTs
as the control elements. It does not cause lamp buzz but can cause audio
interference.

There is 'sine-wave dimming', which is a PWM technique. It does not
cause lamp buzz and is claimed not to cause audio interference, but I
suspect you can't completely ignore the possibility and use the wrong
sort of audio cables and poor installation techniques.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[rex]

On Sat, 09 Apr 2005 04:32:29 GMT, rex <notat@hotmail.invalid> wrote:

On Sat, 09 Apr 2005 04:50:17 +0100, Pooh Bear
rabbitsfriendsandrelations@hotmail.com> wrote:


From browsing the web on this product, it seems to be well recognised
that the mic preamp is pretty lame. That's why Doug Oade is doing some
business by offering mods on the one's he sells. The published S/N spec
is 65 dB. He measured 80 dB after his changes. I'd be happy with that.

I'm curious about that. He may possibly be simply replacing the input transistors.
Takes a few minutes to do that !

Without seeing the unit I'm at a disadvantage. Are you familiar with the
'conventional' leaded TO-92 package ? If there aren't any in the mic amp then I
reckon I know what's going on.

Ok, this is sounding good (no pun, of course). I was just reading your
reply to Fred's message too.

The only large devices in the input area are the 4 (2 per channel I
assume) electrolytics. There are then two transistors per channel and
they are both SOT-23 form. No TO-92 sized stuff anywhere that I noticed.

Next in line after the transistors seems to be the 4052 then the 2068. I
mentioned in the other message some possible functions for the 4052 that
just occured to me.

So I await further comments on the transistors.

Nevermind about the transistors. I opened it up again and they are not
transistors. Frank's instincts were right. Two of the three device pins
are connected together so they are diodes of some kind.

I'm going to see if I can learn anything about where the mic inputs go
before I close it up, but right now I don't see any transistors at the
input. I'm beginning to wonder if there are components on the bottom of
the board.

Oh, I did get it a little further opened up and found the other
information printed on the caps. They are Elna 63V 10u.

 

[Larry Brasfield]

"Fred Bloggs" <nospam@nospam.com> wrote in
message news:425726BF.9000003@nospam.com...

Larry Brasfield wrote:

In one project I worked on, one of the most competent and
conscientious engineers I've known used the 8:1 mux from
that family for a wide, RF (2 to 13 MHz), barrel switch
where noise and distortion were both critical parameters.

You don't get any isolation at those frequencies so interchannel interference was obviously not a consideration.

There was an error budget for channel to channel
coupling, and that assembly had only a small portion
of it, and easily met its obligations in that regard.

I was one of the reviewers for that design, and that
was one of my concerns as well. But some simple
calculations from datasheet coupling performance
numbers and the circuit imedances were sufficient
to alleviate those concerns. The guy had already
done them and verified their applicability in the lab.

--
--Larry Brasfield
email: donotspam_larry_brasfield@hotmail.com
Above views may belong only to me.

 

[Fred Bloggs]

Larry Brasfield wrote:

"Fred Bloggs" <nospam@nospam.com> wrote in
message news:425726BF.9000003@nospam.com...

Larry Brasfield wrote:


In one project I worked on, one of the most competent and
conscientious engineers I've known used the 8:1 mux from
that family for a wide, RF (2 to 13 MHz), barrel switch
where noise and distortion were both critical parameters.

You don't get any isolation at those frequencies so interchannel interference was obviously not a consideration.



There was an error budget for channel to channel
coupling, and that assembly had only a small portion
of it, and easily met its obligations in that regard.

I was one of the reviewers for that design, and that
was one of my concerns as well. But some simple
calculations from datasheet coupling performance
numbers and the circuit imedances were sufficient
to alleviate those concerns. The guy had already
done them and verified their applicability in the lab.

Yeah- sure- which 4052 part are you talking about?

 

[Ted Edwards]

rex wrote:

Phantom power is not the problem, or at least not the main problem. I
connected a condenser mic, turned phantom off to disable it and cranked
up the volume. That's where I realized that there is more hiss from this
thing than I would have liked.

Terminology keeps changing! :-( What is "phantom power"?

Ted

 

[Joerg]

Hello Graham,

I think he already said it used SMT ! Maybe use a pin as a probe ?

What I do is take a through-hole cap that's slim yet big enough to fit
into my fingers. Then I leave 1/3" or so of wire on the probe side, or
whatever is enough to be able to touch the SMT part pins. I use really
old wire cutters for that which results in a sharp edge or I file it
into a point contact.

The other thing to make sure is that the outside foil is on the ground
side so the finger doesn't introduce noise. Most important is to always
discharge the cap before moving on to the next pin, especially if the
circuitry is somewhat unchartered. Else.....bzzzt... foomph...pop.

Regards, Joerg

http://www.analogconsultants.com

 

[rex]

On Sat, 09 Apr 2005 17:55:23 GMT, Ted Edwards <Ted_Espamless@telus.net>
wrote:

rex wrote:
Phantom power is not the problem, or at least not the main problem. I
connected a condenser mic, turned phantom off to disable it and cranked
up the volume. That's where I realized that there is more hiss from this
thing than I would have liked.

Terminology keeps changing! :-( What is "phantom power"?

It's not new terminology.

Good quality condenser mics contain amplifiers and require DC to
operate. Phantom power is usually about 48 Vdc and is injected common
mode down the balanced signal lines of an XLR cable. Usually limited and
isolated from the signal by running it through 6.8K to each of the
lines.

 

[Joerg]

Hello Ted,

Terminology keeps changing! :-( What is "phantom power"?

Audio speak for a voltage riding on the line going from amp to mike, to
supply electret mikes and the like. 48V most of the time, I believe, and
low current.

Regards, Joerg

http://www.analogconsultants.com

 

[rex]

On Sat, 09 Apr 2005 08:18:05 GMT, rex <notat@hotmail.invalid> wrote:

On Sat, 09 Apr 2005 04:32:29 GMT, rex <notat@hotmail.invalid> wrote:

On Sat, 09 Apr 2005 04:50:17 +0100, Pooh Bear
rabbitsfriendsandrelations@hotmail.com> wrote:


From browsing the web on this product, it seems to be well recognised
that the mic preamp is pretty lame. That's why Doug Oade is doing some
business by offering mods on the one's he sells. The published S/N spec
is 65 dB. He measured 80 dB after his changes. I'd be happy with that.

I'm curious about that. He may possibly be simply replacing the input transistors.
Takes a few minutes to do that !

Without seeing the unit I'm at a disadvantage. Are you familiar with the
'conventional' leaded TO-92 package ? If there aren't any in the mic amp then I
reckon I know what's going on.

Ok, this is sounding good (no pun, of course). I was just reading your
reply to Fred's message too.

The only large devices in the input area are the 4 (2 per channel I
assume) electrolytics. There are then two transistors per channel and
they are both SOT-23 form. No TO-92 sized stuff anywhere that I noticed.

Next in line after the transistors seems to be the 4052 then the 2068. I
mentioned in the other message some possible functions for the 4052 that
just occured to me.

So I await further comments on the transistors.


Nevermind about the transistors. I opened it up again and they are not
transistors. Frank's instincts were right. Two of the three device pins
are connected together so they are diodes of some kind.

I'm going to see if I can learn anything about where the mic inputs go
before I close it up, but right now I don't see any transistors at the
input. I'm beginning to wonder if there are components on the bottom of
the board.

Oh, I did get it a little further opened up and found the other
information printed on the caps. They are Elna 63V 10u.

I pulled my recorder almost completely apart and got the main board out.
It has lots of components on both sides. I traced the mic input circuits
for one channel and have posted it as 'mic preamp' in ABSE.
5heg51df30ubgvcvdjb4u6pfjv969b79sp@4ax.com

I think it is complete as far as I went. The first active device is one
of the op amps. The hot and cold sides of the XLR input have symetrical
circuitry so the one diagram shows both. There are some more comments in
the ABSE message.

I'd appreciate any comments on what you think of this circuit and how I
might get it working cleaner.

 

[Joerg]

Hello Rex,

....and have posted it as 'mic preamp' in ABSE.
5heg51df30ubgvcvdjb4u6pfjv969b79sp@4ax.com

Maybe it's just me but I cannot see the attached file in this ABSE post.

Regards, Joerg

http://www.analogconsultants.com

 

[Erik Walthinsen]

John Woodgate wrote:

Which three classes do you have in mind?

There is 'phase-control, turn on' - this is the oldest electronic
technology and uses triacs of thyristors as the control elements. It can
cause lamp buzz and audio interference.

There is 'phase-control, turn off', which is fairly new and uses IGBTs
as the control elements. It does not cause lamp buzz but can cause audio
interference.

There is 'sine-wave dimming', which is a PWM technique. It does not
cause lamp buzz and is claimed not to cause audio interference, but I
suspect you can't completely ignore the possibility and use the wrong
sort of audio cables and poor installation techniques.

That seems to cover it AFAICT, extrapolating from the various marketing
materials I've seen. The D2/D4DMX internally have a single inductor and
single TO-220 package, plus misc bits, per channel. Haven't checked
what the TO-220 actually is, I'll have to do that sometime soon. It
definitely causes lamp buzz however, so it's almost certainly in the
first category ;-(

The trick will be to figure out if there are any current products out
there with prices we can handle, that use sine-wave dimming. Or, we
take some steps to isolate things properly. Problem is, the current
audio runs are decent wire but are exposed runs in the attic of the
building which is the same space we'd probably be putting the dimmer
packs. Depending on what the mains and dimmer-channel installation is
we might be able to cluster the packs inside a grounded cage or two, and
with metal-clad cable as required by code, keep most of the noise contained.

 

[rex]

On Sat, 09 Apr 2005 20:57:05 GMT, rex <notat@hotmail.invalid> wrote:

I pulled my recorder almost completely apart and got the main board out.
It has lots of components on both sides. I traced the mic input circuits
for one channel and have posted it as 'mic preamp' in ABSE.
5heg51df30ubgvcvdjb4u6pfjv969b79sp@4ax.com

I think it is complete as far as I went. The first active device is one
of the op amps. The hot and cold sides of the XLR input have symetrical
circuitry so the one diagram shows both. There are some more comments in
the ABSE message.

I'd appreciate any comments on what you think of this circuit and how I
might get it working cleaner.

I suspect this design is not going to get high marks.

I was just sitting here looking at what I drew and thought, "why is that
220K resistor in parallel with the 2.4K divider?" I went back and double
checked that board in that area, and I can't see any mistake I made in
drawing it.

 

[rex]

On Sat, 09 Apr 2005 23:01:37 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hello Rex,

....and have posted it as 'mic preamp' in ABSE.
5heg51df30ubgvcvdjb4u6pfjv969b79sp@4ax.com

Maybe it's just me but I cannot see the attached file in this ABSE post.

Regards, Joerg

Sorry, when you post text and graphics there it splits it into two
messages with the same subject. I linked the text part, here is the
image part...
<infg511uht42r0d8cn2hjh3pbsof9h0d1i@4ax.com>

 

[Joerg]

Hello Rex,

Sorry, when you post text and graphics there it splits it into two
messages with the same subject. I linked the text part, here is the
image part...
infg511uht42r0d8cn2hjh3pbsof9h0d1i@4ax.com

Still can't see it. It could be my ISP though. With most posts I can see
the attached files in a.b.s.e. but then there are a few where I can't.

If you think I might be of help just email it to jsc <at> (the domain
below in my signature line).

Regards, Joerg

http://www.analogconsultants.com

 

[Joerg]

Hello Erik,

The trick will be to figure out if there are any current products out
there with prices we can handle, that use sine-wave dimming. ...

That is tough. If you walk through a hardware store they have a wide
spread of dimming products in terms of pricing. But I never saw any
hints as to technology, except for "this one buzzes less" or so.

... Or, we
take some steps to isolate things properly. Problem is, the current
audio runs are decent wire but are exposed runs in the attic of the
building which is the same space we'd probably be putting the dimmer
packs. Depending on what the mains and dimmer-channel installation is
we might be able to cluster the packs inside a grounded cage or two, and
with metal-clad cable as required by code, keep most of the noise
contained.

Did you run differentially? If not, Radio Shack has "ground loop
isolators" which seem to be little audio xfmrs at around $17 for a
two-channel. Not too bad. Muxlabs has high quality ones but they may
also cost more.

Regards, Joerg

http://www.analogconsultants.com

 

[rex]

On Sun, 10 Apr 2005 00:12:39 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hello Rex,

Sorry, when you post text and graphics there it splits it into two
messages with the same subject. I linked the text part, here is the
image part...
infg511uht42r0d8cn2hjh3pbsof9h0d1i@4ax.com

Still can't see it. It could be my ISP though. With most posts I can see
the attached files in a.b.s.e. but then there are a few where I can't.

If you think I might be of help just email it to jsc <at> (the domain
below in my signature line).

A lot of ISPs don't carry the binaries group referenced.

Here is a link to an ftp place I have with the same two text and gif
files...
ftp://ftp.sonic.net/pub/users/rexa/Preamp/

Hope that works for you and thanks for contributing.

 

[Joerg]

Hello Rex,

A lot of ISPs don't carry the binaries group referenced.

Mine does but for some reasons a few posts come without the pictures.

Here is a link to an ftp place I have with the same two text and gif
files...
ftp://ftp.sonic.net/pub/users/rexa/Preamp/

Yes, that worked. The top 10uF cap seems polarity reversed. Other than
that I don't see anything weird. Except that regular electrolytics in a
zero bias coupling function (the lower 10uF cap) aren't a great idea.
Non-polarized caps are usually better.

Regards, Joerg

http://www.analogconsultants.com

 

[rex]

On Sun, 10 Apr 2005 00:47:08 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

The top 10uF cap seems polarity reversed

You are correct. I got that wrong in my schematic.