Hi
I have a field recording to do this weekend and I'm not happy with the
quality of the preamp that I have available.

I'm recording unto a Sony DAT TCD8 with a couple of Schoeps collette
mikes.
My questions:

Can I use a Nagra IVS ( my dad's store in my closet) as a phanton power
and as a preamp ?

Do I need to have tape running though it in order to be able to monitor
the sound?

other than the weight issue are there any problems with this idea?

Is there a a battery operated preamps in the market that match the
quality of the Nagra

I would use the Nagra alone but I could find tape on time for the
event.

Thank you in advanc for your help.

stillness wrote:
Hi
I have a field recording to do this weekend and I'm not happy with the
quality of the preamp that I have available.

I'm recording unto a Sony DAT TCD8 with a couple of Schoeps collette
mikes.

That is... well... quite a combination. One of the best mikes with one
of the worst recorders....

My questions:

Can I use a Nagra IVS ( my dad's store in my closet) as a phanton power
and as a preamp ?

You can use the preamp section on it, but it may or may not have phantom
power depending on how it was ordered. Most of them were set up without
phantom power.

Do I need to have tape running though it in order to be able to monitor
the sound?

No, although if you do run tape through it, you may find it sounds a whole
lot better than the D-8.

other than the weight issue are there any problems with this idea?

Is there a a battery operated preamps in the market that match the
quality of the Nagra

Yes, but you won't like what Sonosax and Cooper charge.

I would use the Nagra alone but I could find tape on time for the
event.

Where are you located?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Stillness, while every Nagra IV-S has 48 Volt phantom power available
for its mike preamps, there are four different models of amplifier
(body) for the Schoeps Colette series and some of them are not the best
match for the preamps in an unmodified stereo Nagra.

I used to use CMC 3-- bodies with my Nagra IV-S, with its preamps set
to the +12 Volt phantom setting. That setting will also work nicely
with the CMC 6-- series of Colette amplifiers. And if you have a pair
of CMC 4-- amplifier, then use the "T" power setting on the Nagra. But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

In addition, the Nagra's preamps are somewhat vulnerable to overload if
you record sounds that are very loud close up--the microphones will
handle 130+ dB sound pressure levels, but since Schoeps microphones are
a bit more sensitive than the Nagra's preamps were really designed for,
those preamps will clip well before the microphones do. The modulometer
on the Nagra won't indicate this problem (because it can't) and you'd
have to put a pair of in-line balanced pads (maybe 20 dB) at the inputs
to the recorder.

On the other hand if the powering and level problems can be worked out,
the Nagra's mike preamps are among the quietest available, so I
definitely see the point of trying to make the combination work.

--best regards

stillness wrote:
Thanks Scott,
I'm in Australia at the moment. I checked the Sonosax SX-M2 and at
$1200 it is $400 over the budget. I use the DAT because I need to end
up digital. I also have a the latest Sony Minidisc which for the first
time records in uncompressed WAV format. Would this be a better
choice than the DAT?

Probably not. You'll have to deal with the lossy compression with the MD.

You can get something like the Denecke outboard A/D box which has a small
internal preamp. Combine that with an outboard phantom power supply and
you can get surprisingly good sound with the D-8. Unfortunately it will
not make the D-8 reliable.

If I managed to find tape for the Nagra what A/D would you recommend?

Whatever the best you can borrow is.

Is going SoundDevices preusb to Ibook a better final product?

Dunno, I never listened to it.

Ultimately the recording will end up on CD for distribution- hopefully
with as little processing as possible- beginner purist. :-)

Why not rent a good kit, then? Check your local film rental outfit.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

David Satz wrote:
Stillness, while every Nagra IV-S has 48 Volt phantom power available
for its mike preamps, there are four different models of amplifier
(body) for the Schoeps Colette series and some of them are not the best
match for the preamps in an unmodified stereo Nagra.

I used to use CMC 3-- bodies with my Nagra IV-S, with its preamps set
to the +12 Volt phantom setting. That setting will also work nicely
with the CMC 6-- series of Colette amplifiers. And if you have a pair
of CMC 4-- amplifier, then use the "T" power setting on the Nagra. But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

In addition, the Nagra's preamps are somewhat vulnerable to overload if
you record sounds that are very loud close up--the microphones will
handle 130+ dB sound pressure levels, but since Schoeps microphones are
a bit more sensitive than the Nagra's preamps were really designed for,
those preamps will clip well before the microphones do. The modulometer
on the Nagra won't indicate this problem (because it can't) and you'd
have to put a pair of in-line balanced pads (maybe 20 dB) at the inputs
to the recorder.

On the other hand if the powering and level problems can be worked out,
the Nagra's mike preamps are among the quietest available, so I
definitely see the point of trying to make the combination work.

--best regards

Thanks, David, that's good info.

I thought that at 4.5mA with 48V phantom , the CM preamps were well
within Phantom spec ( 48V/6800 Ohm = 7 mA)?
Do you know how much current the Nagra preamps put out (couldn't find it
in the Nagra manual)?

And what have I been missing (other than the finer nuances of my crappy
room) by using Schoeps CM-5 into an unmodified Nagra IV-S? (on a
not-so-loud source)?

And Scott, much as I love it otherwise, it's at times like these
(googling for nagra ) that your sig really drives me crazy...

Hans

--




This is a non-profit organization;
we didn't plan it that way, but it is

=====================================


(remove uppercase trap, and double the number to reply)

The most common preamp fitted to Nagra IV-s recorders in the USA was
the
"universal" model, which could be switched through several different
types of mic powering, incl. 48v phantom and 12v "T" power. I used
mine for years
with Schoeps CMC4 type mics and it worked great. The Nagra preamps
sound very good, but are a little noisey by today's standards. To use
the Nagra as a preamp, you only have to have the unit in its "TEST"
position,
on but not rolling. To get a signal out of the Nagra you will need
Tuchel adapter cables, wired according to the diagram on the side of
the Nagra (what a great idea--no one does this anymore!). A Tuchel
to mini adapter cable (to connect to your DAT) is a very oddball
thing--you'll either have to make it, have it made or fit together a
number of adapter cables to make it work. (Tuchel to 2 XLR males was
pretty common, 2 XLR females to stereo mini isn't that unusual). The
rig will be heavy--remember that a Nagra runs on 12 "D" batteries, and
has some serious weight on its own. You can run it on AC if you have
the Nagra "ATN" AC power supply.

Philip Perkins

Scott Dorsey wrote:
stillness wrote:

Thanks Scott,
I'm in Australia at the moment. I checked the Sonosax SX-M2 and at
$1200 it is $400 over the budget. I use the DAT because I need to end
up digital. I also have a the latest Sony Minidisc which for the first
time records in uncompressed WAV format. Would this be a better
choice than the DAT?


Probably not. You'll have to deal with the lossy compression with the MD.

Not with the new Hi-MD. Not sure if any models have S/PDIF
in, however, and I'm waiting for Sony to yield me control of
my own recordings to buy one for detailed testing of it's
front end.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Hans van Dongen wrote:

And what have I been missing (other than the finer nuances of my crappy
room) by using Schoeps CM-5 into an unmodified Nagra IV-S? (on a
not-so-loud source)?

I dunno, I think that's a fine combination.

And Scott, much as I love it otherwise, it's at times like these
(googling for nagra ) that your sig really drives me crazy...

So refine your search to avoid it. Try a search like
"nagra -"tres -precis" group:rec.audio.pro" which will find all
instances of Nagra but discard all messages with my .signature.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Hans, the standard for 48 Volt phantom powering has been through two
fundamental revisions since the Nagra IV-S was introduced. When this
machine was new, about 0.7 mA per microphone was the usual value for a
microphone designed for 48-Volt phantom powering, although there were
certainly some exceptions--the AKG C 451, for example, drew
considerably more.

I don't know the absolute maximum for an unmodified Nagra IV-S but
note, for example, that in 1989 Neumann put out information on how to
modify a stereo Nagra to support its KM 100 series, which draws only
about 2 mA per microphone on a continuing basis (see
http://www.neumann.com/infopool/down...i=info0002.pdf)

Since the two 6.8 kOhm resistors are effectively in parallel the total
current that they deliver into a short circuit is 14 mA (not 7), and
the present-day standard allows 10 mA per microphone to be drawn.

The Schoeps CMC series amplifiers (as well as Neumann's KM 100 and KM
180 series) use DC/DC converters to obtain the polarization voltage for
the capsule. If the necessary current is not available, the microphone
will have a much lower maximum SPL capability. It is not directly
proportional to the input voltage--I've seen the graph of supply
voltage vs. maximum SPL, and the dropoff is considerably more drastic
once the actual minimum amount is missed. That is certainly not the
type of problem which you want to find only in the middle of a live
recording.

With the present-day Neumann KMs the oscillator in the DC/DC converter
drops to a lower frequency if the supply is too far from adequate; it
can even create whistles or heterodynes in the audio frequency range.

My frank impression is that there is a cultural issue here. When
designing these circuits, the (mainly German) design engineers could
hardly imagine that anyone would economize or cut corners on power
supplies for professional condenser microphones. They themselves built
very sturdy and correct supplies, and took it for granted that
third-party supplies would be correct as well. Unfortunately the way
some cost-oriented designers have worked--and I say this from having
spoken with several of them here in the U.S.--this is not a conjectural
statement--is to grab the nearest microphone (an AKG electret, perhaps)
and plug it in to test the supply. If sound came out, the supply was
OK.

--best regards

Scott Dorsey wrote:

So refine your search to avoid it. Try a search like
"nagra -"tres -precis" group:rec.audio.pro" which will find all
instances of Nagra but discard all messages with my .signature.

Aaaaah, tres precis.

--
ha

If yu're plannng to feed a laptop then the Sound Devices USBPRE is what
you want. About $550 usd. The USBPRE is passive and gets it's power from
the laptop.

If you're going to feed a recorder then the Sound Devices Mix Pre at
$665usd is the choice. You can check the specs on both at:
www.sounddevices.com

Eric

Thank you David for a very thorough explanation.
I know that my father used the Nagra with the mikes for film production
so I am assuming that they are matched - he was a good geek - I am
recording a group women singing in the bush far from any power sources.
One of the peculiarities of their singing is the fact that they pause
while singing and listen to the night sounds. I was hoping to have a
very clean and transparent background - free of artifacts and component
generated noise.

an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

Do you think it would be worth for me to bring along the Schoeps
phantom power boxes ( i have 2 that take 9 volt batteries - they were
purchased with the mikes but I don't recall him ever using them with
the Nagra I'll have to trek 12 k with the gear I was hoping to lighten
my load.

Cheers

Thanks Philip,
I did try them last night and it works as you say -I do have the Tuchel
to XLR I'll have to buy connectors to reduce the XLR to 1/8 mini.
Would there be a mismatch from balanced to unbalanced line into the
DAT?

Cheers

stillness wrote:

Do you think it would be worth for me to bring along the Schoeps
phantom power boxes ( i have 2 that take 9 volt batteries - they were
purchased with the mikes but I don't recall him ever using them with
the Nagra I'll have to trek 12 k with the gear I was hoping to lighten
my load.

If you really want to lighten the load, you can make some XLR-1/8" cables
to go from the output of the phantom power boxes directly into the unbalanced
mike input on the TCD-D8. Of course, now you're stuck using the preamps
on the D8, but give it a listen and try it because it won't cost anything.

You may decide to rent an hhb.
Cheers
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

David Satz wrote:
But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

What keeps it from delivering 4.5 mA? Does the 48 V supply
crowbar at a lower current or burn out or something?


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Of course you're hoping to lighten your load, plus you want the setup
to be as simple as possible so that it will be reliable and manageable.

I'm trying to find information on the old Schoeps battery supplies. I
have a pair of them also (from the 1970s), but they are the stereo
versions and may not be wired like what you have. From the schematic I
have here, it looks as if they didn't block DC from their outputs by
default, but that there was space on the circuit board to install
blocking capacitors. That can be an important issue depending on what
"front end" equipment (preamp and possibly A/D converter) you decide to
use. The Nagra doesn't care--it has transformers at its inputs. But the
Zefiro InBox, which is a low-cost yet reasonable-quality highly
portable mike preamp with built-in A/D converters, will not work if DC
is present at its inputs. (Denecke's own PS-2 phantom power supply
blocks DC so that it can work with the InBox, and its cabling scheme is
ideal for what you're doing.)

On the other hand, if you use external phantom power supplies with the
Nagra, the only way to switch off the phantom powering from the Nagra
itself is to change its preamps over to the DYN 200 or DYN 50
settings--and those settings again have higher sensitivity and lower
overload thresholds, so you will be in trouble all over again unless
either [a] the power supplies are the BZ 51 Un version with the 20 dB
output pads built in, [b] you use in-line pads between the phantom
supplies and the Nagra, or [c] you leave the phantom powering switched
on in the Nagra--which ordinarily is OK to do (phantom powering is
designed so that two supplies can be applied to the same microphone at
the same time for "failsafe" operation).

However, by the time you've done all that, you will have in each
channel: Microphone - cable - outboard phantom supply - cable - pad
- Nagra - cable - analog line input of the DAT recorder. That's one
heck of a mess to carry around and set up in a hurry in the field.

I am no great fan of the TCD-D8 but when I use one, I either use a
Zefiro "InBox" (actually made by Denecke and now sold by them directly)
or for stuff that really matters, a Grace Lunatec V3, which has
first-rate mike preamps and phantom powering built in plus first-rate
A/D converters, metering, and useful, switchable low-cut filters
(though nothing else I've ever used is more useful than the flexible
low-cut filter arrangement of the Nagra IV-S!). However, that's an
expensive piece.

A direct digital cable is available from Sony to go from the RCA to the
TCD-D7 or -D8; with a simple mini-plug to RCA adapter, that cable works
well with the InBox or the Lunatec.

Are you aware of the problems with the TCD-D8? For one thing, its
meters under-read some peaks by as much as 4 to 5 dB. For live
recording do not let them reach 0 dB by any means; if the OVER
indicators come on even occasionally, you must lower your recording
levels and you may find that these indicators may come on occasionally
even when the level meters haven't shown anything higher than a -2 or
-3. Again: believe the OVER indicators if that happens--not the meters.

Next, do not under ANY circumstances try to use ultra thin data backup
DATs in the TCD-D8. The tape will fail to maintain good contact with
the head and the recording will very likely be useless. Stay with
ordinary, professional quality audio DAT blanks, and by all means
record at the standard tape speed (44.1 or 48 kHz), not the "LP" speed
(32 kHz).

Finally, be aware that the tiny (half-size even for a DAT recorder)
drum design of this series of portable DAT means that if the one and
only tape head gets clogged by oxide particles, you don't have the
alternate head there to make up the difference. So head cleaning is
more important on this type of unit, and in some cases you'll get lower
dropout rates if you play your recordings back on a full-sized Sony DAT
deck.

Sony's "pro" DAT machines (TCD-D10 Pro and TCD-D10 Pro II) have serious
design defects of their own, but at least they have full-sized head
drums and somewhat more reliable peak metering.

Whatever setup you do decide upon, make sure to do a "dry run" or two
before you pack up to leave--what you're doing is not easy.

--best regards

Bob, the Nagra recorders are designed to run from 12 D cells, and use a
switching power supply to generate the 48 Volts. I'm not familiar with
the circuit but could look it up in the service manual; my recollection
is that it would drop out of regulation and put out a lower, much
noisier supply voltage if you placed too heavy a load on it. It was
designed only for microphones such as the Neumann FET 80 series or the
Schoeps CMT 5-- series, which used single-stage FET impedance
converters and output transformers.

I used a Nagra IV-S (which is sitting right behind me) and Schoeps CMC
microphones for many years to record concerts and recitals. Even though
I had my Nagra's preamps modified for better resistance to overload, I
specifically used the 12-Volt Schoeps CMC 3-- preamp versions rather
than the 48-Volt CMC 5--s. The microphone circuit is the same except
for a single bridge, however; a user who's good with a soldering iron
can easily convert one to the other.

And I understand from the chief engineer at Schoeps that there is a
circuit modification (which I simply didn't know about at the time)
which does give satisfactory results with the CMC 5 and possibly also
the CMC 6--though whether and how it may differ from the modification
described in the Neumann app note that I referenced, I don't know.

--best regards

David Satz wrote:
Bob, the Nagra recorders are designed to run from 12 D cells, and use a
switching power supply to generate the 48 Volts. I'm not familiar with
the circuit but could look it up in the service manual; my recollection
is that it would drop out of regulation and put out a lower, much
noisier supply voltage if you placed too heavy a load on it.

Got it, thanks. I've always wondered about that.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote:
David Satz wrote:
But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

What keeps it from delivering 4.5 mA? Does the 48 V supply
crowbar at a lower current or burn out or something?

No, it just has too much series resistance. You put too much of a load
on it and it sags.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

David Satz wrote:
Hans, the standard for 48 Volt phantom powering has been through two
fundamental revisions since the Nagra IV-S was introduced. When this
machine was new, about 0.7 mA per microphone was the usual value for a
microphone designed for 48-Volt phantom powering, although there were
certainly some exceptions--the AKG C 451, for example, drew
considerably more.

I don't know the absolute maximum for an unmodified Nagra IV-S but
note, for example, that in 1989 Neumann put out information on how to
modify a stereo Nagra to support its KM 100 series, which draws only
about 2 mA per microphone on a continuing basis (see
http://www.neumann.com/infopool/down...i=info0002.pdf)

Since the two 6.8 kOhm resistors are effectively in parallel the total
current that they deliver into a short circuit is 14 mA (not 7), and
the present-day standard allows 10 mA per microphone to be drawn.

The Schoeps CMC series amplifiers (as well as Neumann's KM 100 and KM
180 series) use DC/DC converters to obtain the polarization voltage for
the capsule. If the necessary current is not available, the microphone
will have a much lower maximum SPL capability. It is not directly
proportional to the input voltage--I've seen the graph of supply
voltage vs. maximum SPL, and the dropoff is considerably more drastic
once the actual minimum amount is missed. That is certainly not the
type of problem which you want to find only in the middle of a live
recording.

With the present-day Neumann KMs the oscillator in the DC/DC converter
drops to a lower frequency if the supply is too far from adequate; it
can even create whistles or heterodynes in the audio frequency range.

My frank impression is that there is a cultural issue here. When
designing these circuits, the (mainly German) design engineers could
hardly imagine that anyone would economize or cut corners on power
supplies for professional condenser microphones. They themselves built
very sturdy and correct supplies, and took it for granted that
third-party supplies would be correct as well. Unfortunately the way
some cost-oriented designers have worked--and I say this from having
spoken with several of them here in the U.S.--this is not a conjectural
statement--is to grab the nearest microphone (an AKG electret, perhaps)
and plug it in to test the supply. If sound came out, the supply was
OK.

--best regards


Once again, thanks David for taking the time to respond so thoroughly.
I had a look at both the Neumann document you mentioned and the Nagra 4S
service manual, and it seems Kudelski made this change in their mic
power supply - maybe to keep up with revisions in phantom spec?- as
early as 1974 (leaving owners of later machines, and certainly the
Time-code 4S-TC "in the clear", as it were). The change consists of
changing the two resistors that supply the chopping transistors for the
switching power supply from 10 to 6.8 Ohms. Whether his is enough to
deliver 4.5 mA remains a mystery (I don't know enough about electronics
to calculate that from the diagram)

I did extensive googling on the subject ( -"tres precis" :-) and
couldn't find anything, so I suspect that if this was really an issue,
it would have come up on places like rec.arts.movies.production.sound or
here on rap. Did you ever run into trouble using phantom-powered Schoeps
collete mics, or was your use of the CM-3 instead of the CM-5 preamps a
precaution?

I agree completely on your "cultural issue" observation, although I
think one could hardly call Kudelski a "cost-oriented" company. (what
with the Nagras light-temporizer circuit and all)

Best regards,

Hans
--




This is a non-profit organization;
we didn't plan it that way, but it is

=====================================


(remove uppercase trap, and double the number to reply)

Scott Dorsey wrote:
Bob Cain wrote:

David Satz wrote:

But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

What keeps it from delivering 4.5 mA? Does the 48 V supply
crowbar at a lower current or burn out or something?


No, it just has too much series resistance. You put too much of a load
on it and it sags.
--scott

Just checking :if I measured this at the mic input with just a (cheap)
multimeter - inserting two appropriate resistors to get at the 4.5 mA (
as in :
1/2(6800 + R) = 48/0.0045)
- I'm thinking 15 kOhm so if I measure 33V across the resistor I'd be
okay (taking into account voltage drop across the 6.8k resistors)
Hey, I might even try 2.8k to see if it's up to spec.

Or would I have to do something more involved?

Hans
--




This is a non-profit organization;
we didn't plan it that way, but it is

=====================================


(remove uppercase trap, and double the number to reply)

Hans van Dongen wrote:

Just checking :if I measured this at the mic input with just a (cheap)
multimeter - inserting two appropriate resistors to get at the 4.5 mA (
as in :
1/2(6800 + R) = 48/0.0045)
- I'm thinking 15 kOhm so if I measure 33V across the resistor I'd be
okay (taking into account voltage drop across the 6.8k resistors)
Hey, I might even try 2.8k to see if it's up to spec.

Or would I have to do something more involved?

No, that sounds about right.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Many thanks to all of you who replied with your technical knowledge and
suggestions.

As of last night, my backpack weighted 26kg. 3 times my weight.
Fortunately the rainy season has began, it is cool and I don't need to
carry so much water, but I have to insulate all the gear.

I have
1) Nagra
1) Ibook
1) usb pre
I) TCD 8
2) boxes of tapes and minidiscs
2 Schoeps mics
4 capsules
2) Sennheiser mics
2) mic stands -
6) wind screens
2) One Schoeps Stereo phantom box
I) Hi-MD minidisc
2) torch lights
12) metres of cable
18 adaptors - voltage meter- tool kit
food, sleeping, gear - No beer - something valuable had to give.


Nearly 8 kilos worth of batteries- there's got to be a better way. I
hope the ipodlinux guys finish their project of a 24/96 ipod recorder
by next year so that I can lighten my load.


Redundancy is the order of the day - I managed to find 4 reels of
Maxell 1.5 mil mastering tape which I'll bring. I tested all the gear
and it works well -too many adptors for my taste. A neighbour was nice
enough to play the violin for an hour outdoors while I tried every
possible combination.

Just one last question.

Given a choice of preamps, which would you take?

between the

Sonosax or the Lunatec V2?

and between

The Denecke and the Core sound

Just in case you are wondering, I am trying to produce a
poor/beginner's version of the work of Kavichandran Alexander of water
Lilly acoustics and Australia's own Kostas Metaxas have done. - one can
dream no?

Again, many thanks

"stillness" wrote in message

As of last night, my backpack weighted 26kg. 3 times my weight.

You weigh 8-2/3 kg? Are you a fox terrier, by chance?

Peace,
Paul

Paul
haha , you found me out, a rough coat with a black ear at that.
Let's try again 1/3 my weight.

Scott Dorsey wrote:
Bob Cain wrote:

David Satz wrote:

But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

What keeps it from delivering 4.5 mA? Does the 48 V supply
crowbar at a lower current or burn out or something?


No, it just has too much series resistance. You put too much of a load
on it and it sags.

Where is that resistance? Internal impedence of the 48V
supply? I thought the internal DC impedence as seen from
the terminals was only supposed to be the two usual phantom
resistors. If it were otherwise, wouldn't mic circuit
designers be unable to pick a design point because they'd
never know what voltage they might get for a given current.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote:

Scott Dorsey wrote:
Bob Cain wrote:

David Satz wrote:

But
an unmodified Nagra doesn't quite have enough current available to
supply a pair of Schoeps CMC microphones in the 48-Volt phantom
setting, since 4.5 mA per channel is required.

What keeps it from delivering 4.5 mA? Does the 48 V supply
crowbar at a lower current or burn out or something?

No, it just has too much series resistance. You put too much of a load
on it and it sags.

Where is that resistance? Internal impedence of the 48V
supply? I thought the internal DC impedence as seen from
the terminals was only supposed to be the two usual phantom
resistors.

Internal impedance of the 48V supply. In this case mostly the result of
small capacitors chosen for the voltage multiplier ladder.

You don't get any perfect voltage sources in the real world.

If it were otherwise, wouldn't mic circuit
designers be unable to pick a design point because they'd
never know what voltage they might get for a given current.

No, because the supply should be much lower impedance than the 6.81K
resistors plus the mike input resistance. Unfortunately the AES spec
for minimum mike input resistance has changed over the years.

THEN we get things like the Mackie, where one phantom supply drives
40 pairs of resistors driving 40 mikes. If you put ten phantom powered
miked on the console, it's fine... if you fill all 40 channels up
the supply sags horribly.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Scott Dorsey wrote:

Where is that resistance? Internal impedence of the 48V
supply? I thought the internal DC impedence as seen from
the terminals was only supposed to be the two usual phantom
resistors.


Internal impedance of the 48V supply. In this case mostly the result of
small capacitors chosen for the voltage multiplier ladder.

You don't get any perfect voltage sources in the real world.

You seem to be assuming an unregulated supply, correct?
Until a supply drops out of regulation, it's internal
impedence _is_ zero as I understand things.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote:

You seem to be assuming an unregulated supply, correct? Until a supply
drops out of regulation, it's internal impedence _is_ zero as I
understand things.

A regulated supply has a very low source impedance which hopefully can
be considered near enough zero for practical purposes. How close to zero
depend on how the regulation is done, bust almost any simple electronic
regulator should have much less that the 6.8k/2 deliberately placed in
series with the output.

I don't know whether the substandard P48 supplies in question drop
linearly with current drain or fall suddenly out of regulation. I can't
understand why designers don't bother with getting it right - it's
hardly rocket science. Even a 48 channel mixer to provide full spec
(10mA /channel) phantom power on all channels only needs 480 mA at 48V.
How much cost savings do you make by catering for no more than a handful
of inputs, drawing about 2 mA each, even if that is a more typical usage
in real life?

It's not a difficult power supply to design - no short circuit
protection needed.

--
Anahata
-+- http://www.treewind.co.uk
Home: 01638 720444 Mob: 07976 263827

In article writes:

hardly rocket science. Even a 48 channel mixer to provide full spec
(10mA /channel) phantom power on all channels only needs 480 mA at 48V.
How much cost savings do you make by catering for no more than a handful
of inputs, drawing about 2 mA each, even if that is a more typical usage
in real life?

Actually, it's a substantial cost saving. To make a real 48V power
supply in a unit that's full of ICs that run on +/- 15-18V, you need
either a separate power transformer or separate winding on the power
transformer. With half a watt for each channel, that's a 50 watt power
supply for a large format mixer. That probably costs around $50 to the
manufacturer when all is done, and taking Paul's conservative 6:1
cost/selling price ratio, that adds $300 to the cost of the mixer.
You'd expect this in a "professoinal" 48 channel console, and you
usually get it.

But when it's necessary to cut costs (like for a $1,000 24-channel
board) there will be a skimpy 48V winding added to the custom power
transformer or a voltage tripler. They make the assumption that it's
not going to get a condenser mic on every channel. That was reasonable
in the days when condenser mics cost $1,000 or more (and many of those
drew only a couple of mA) but with $100 condenser mics (many of which
draw 4-5 mA) that assumption no longer holds.

Most manufacturers have figured this out and on designs from the past
couple of years, provide sufficient current from their phantom
supplies (there are other places to cut costs), but older designs are
still being sold (some new, many re-sold used) so the problem hasn't
gone away yet.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

Bob Cain wrote:
Scott Dorsey wrote:

Where is that resistance? Internal impedence of the 48V
supply? I thought the internal DC impedence as seen from
the terminals was only supposed to be the two usual phantom
resistors.

Internal impedance of the 48V supply. In this case mostly the result of
small capacitors chosen for the voltage multiplier ladder.

You don't get any perfect voltage sources in the real world.

You seem to be assuming an unregulated supply, correct?
Until a supply drops out of regulation, it's internal
impedence _is_ zero as I understand things.

Right.

There are _some_ consoles out there which use regulated supplies where
the thing drops out of regulation if you plug too many mikes in.

I recall the IV using a DC-DC converter that didn't have real regulation...
it was just an oscillator going into a voltage multiplier. I don't have
any of the docs here to verify that, though.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Scott, I've just dug up my Nagra IV-S service manual. The circuit is an
oscillator that drives the primary of a small transformer with squarish
waves at 10 V peak. Its secondary feeds [a] a full-wave rectifier with
ca. +17.5 Volt output, which in turn feeds the regulated +12 Volt
phantom supply circuit, plus [b] a voltage multiplier which develops
+66 V internally. This is then brought down via three transistors, one
of which has its bias set through a voltage divider that includes a
trim pot for setting the output of the whole shebang to 48 Volts.

The result is sent to the microphones through pairs of 6.8 kOhm 2%
resistors--the pair matching ought to be better than that, but their
absolute value is standard and I see no additional series resistance in
the line.

The mod that Neumann proposed for the KM 100 series looks as if it
would raise the limit on current going through the transformer primary
by about 50%. Unfortunately for a pair of Schoeps CMC 5 amplifiers, the
limit would need to be increased another 100% to 125% beyond that. So
whatever the modification was which the gentleman at Schoeps referred
to, it must have been at least somewhat different.

--I just now powered up my Nagra IV-S for the first time in about two
years. Back in the mid-1970s I had its mike preamps modified to avoid
overload with Schoeps CMC 3 microphones (same signal voltages as the
CMC 5, but +12 Volt phantom powered). But I don't think that the mike
power supply circuit (which is on a whole other board) was ever
modified. In any case, using a pair of Schoeps PHS 48 testers, I found
that this Nagra's 48 Volt phantom supply has enough current for one but
NOT two CMC 5 microphones.

--best regards