| Author |
Message |
Scott Dorsey
Guest
|
 Posted:
Sun Oct 30, 2005 9:26 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
"Pooh Bear" <rabbitsfriendsandrelations@hotmail.com> wrote in message
| Quote: |
Why would it mean spoiling the idea ?
|
Because you can't buy a transformer with a 10M input impedance. And if
you could, it would have no top end due to the interwinding capacitance.
So, if you want a 10M input impedance, you add a FET follower in front
of a 1:1 transformer. The FET stage gives you the impedance conversion.
The transformer gives you the isolation.
This is how the classic active DI works. If you break the ground _before_
the transformer, you can no longer get phantom. If you break the ground
_after_ the transformer, you still have the induced noise on the input
unless you make an active differential stage instead of a simple follower.
| Quote: | A DI is by definition ( once ground lifted ) a differential device
anyway !
|
Right, but you can't lift that ground on the output and still get power
for the active stage.
AND, if you build an active differential stage and do without the
transformer, your maximum common mode voltage is limited to 48V at
best... which is nothing compared with some of the voltages I see on
the chassis of old guitar amps that aren't well-maintained.
| Quote: | I simply don't get what your problem is !
|
My problem is that I can't get a 1M:600 transformer out of the Digi-Key
catalogue. Because of this, we have to do a lot of active circuit
work-arounds.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
|
|
| Back to top |
|
 |
Pooh Bear
Guest
|
| Posted:
Sun Oct 30, 2005 9:31 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Richard Crowley wrote:
| Quote: | "Pooh Bear" <rabbitsfriendsandrelations@hotmail.com> wrote in message
news:4364D98F.4D9E29E2@hotmail.com...
Scott Dorsey wrote:
Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:
Scott Dorsey wrote:
Richard Crowley <rcrowley@xpr7t.net> wrote:
"Pooh Bear" wrote ...
It's perfectly possible to 'ground lift' with a phantom powered
DI.
I.e.
break the common connection between source and destination.
You'll have to demonstrate your theory with a real schematic
(or equivalent). I can't see how it is possible to run something
on phantom power without access to the ground node. Assuming
that putting a transformer on the DI box *input* (the
high-impedance
side) is even more impractical.
The guys at Radial actually manage to do with with the J48....
they
have a 555 timer in there that generates a 75 KC square wave that
goes into a little switching transformer. So they get an isolated
+/- 12V output that they regulate and use to power the front end.
I
think that is really ingenious as hell. But as far as I know, it
is
the only way to isolate the grounds with a phantom powered DI box,
and
Radial is the only company doing it.
Absolute nonsense Scott.
Stop and think where the ground actually needs to be 'lifted' !
The ground can be lifted _anywhere_. But the front end has to be
active,
which means even if you have a transformer inside there, you still
have to
get power to the active front end somehow.
Of course. Currrent sources from Pins 2&3 and returns via Pin 1
If you want to break the ground at some point where the front end can
still
get phantom power, that would mean breaking the ground _before_ the
front
end. Which either means sticking a transformer there and spoiling
the whole
idea of an active follower with a super-high-Z input, or it means a
differential front end.
Why would it mean spoiling the idea ?
A DI is by definition ( once ground lifted ) a differential device
anyway !
I simply don't get what your problem is !
And you likely don't have a real-world solution for
your hi-Z transformer scheme, either.
|
Uh ?
What do you mean.,
Ppl have been making phantom powered DIs ( with ground lift ) for almost as
long as I can remember.
The design is essentially trivial.
Do you have a point ?
| Quote: | OTOH, the in the medical biz, they like using optical
copling.
|
Isolation amplifers are entirely another kettle of fish.
Audio 'ground-lifting' does *not* require medical style isolation. Likely
the ability to deal with as little as a Volt of common mode will suffice.
Bloody well better do or ppl will be being electrocuted if the stage ground
is that bad !
| Quote: | It was even popular for video back in the days
before even cheap $60 TV receivers had video inputs.
|
Uh ?
You don't know much ( again ) do you ? I notice you're always making
smart-arse comments about stuff you clearly haven't the faintest clue about.
Graham |
|
| Back to top |
|
|
Richard Crowley
Guest
|
| Posted:
Sun Oct 30, 2005 9:58 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
"Pooh Bear" wrote ...
| Quote: | Richard Crowley wrote:
.....
And you likely don't have a real-world solution for
your hi-Z transformer scheme, either.
Uh ?
What do you mean.,
|
Read Mr. Dorsey's response.
Or you could try sitting down with a pencil and paper and
try to design a schematic for this. I'm pretty confident that
once you actually start considering real-world issues, you
will see the showstopper problem(s).
| Quote: | Ppl have been making phantom powered DIs ( with
ground lift ) for almost as long as I can remember.
The design is essentially trivial.
|
I've never seen one with BOTH ground lift AND phantom
powering. Mine will run on phantom as long as you don't
lift the ground. If you want to lift the ground, you must run
on internal battery power.
| Quote: | Do you have a point ?
OTOH, the in the medical biz, they like using optical
copling.
Isolation amplifers are entirely another kettle of fish.
Audio 'ground-lifting' does *not* require medical style isolation.
Likely
the ability to deal with as little as a Volt of common mode will
suffice.
Bloody well better do or ppl will be being electrocuted if the stage
ground
is that bad !
|
I'm suggesting that optical isolation may be a viable
substitute for a very-hi-Z isolation transformer (which
doesn't exist in the real-world with sufficient HF response)
| Quote: | It was even popular for video back in the days
before even cheap $60 TV receivers had video inputs.
Uh ?
You don't know much ( again ) do you ? I notice you're
always making smart-arse comments about stuff you
clearly haven't the faintest clue about.
|
Dunno about that. I once managed(*) a couple of med school
labs (neurology and anatomy) full of "jeeped" Sony Triniton
TV receivers with optical coupling (because they had "hot-
chassis" power design.)
My point of the video example is to show that the optical
couplers were capable of MHz (flat, low-distortion)
frequency response and are infinitely cheaper than
transformers of equivalent frequency response & distortion
specs.
(*) Where "managed" means I had to keep all the bastard
TV/monitors running (with no spares), but mostly spend an
hour before each class session going around and color-
blancing each screen. The microbiology professors were
extremely critical about the different between "pinkish-
blue" and "bluish-pink". We used an enormous 3-tube
(Plumbicon) camera with a microscope adapter so they
could show their sample slides of tissue doped with the
various stain mixtures.
I'm not trying to flame you, Graham, I appreciate your posts
and agree with you most of the time. Having a bad day? |
|
| Back to top |
|
|
Pooh Bear
Guest
|
| Posted:
Sun Oct 30, 2005 10:16 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Scott Dorsey wrote:
| Quote: | "Pooh Bear" <rabbitsfriendsandrelations@hotmail.com> wrote in message
Why would it mean spoiling the idea ?
Because you can't buy a transformer with a 10M input impedance. And if
you could, it would have no top end due to the interwinding capacitance.
So, if you want a 10M input impedance, you add a FET follower in front
of a 1:1 transformer. The FET stage gives you the impedance conversion.
The transformer gives you the isolation.
|
Can you suggest an example where 10M input Z is required ?
| Quote: | This is how the classic active DI works. If you break the ground _before_
the transformer, you can no longer get phantom. If you break the ground
_after_ the transformer, you still have the induced noise on the input
unless you make an active differential stage instead of a simple follower.
|
Differential stages are 2 a penny today. Hardly rocket science.
Sounds like an excuse for sloppy out-dated design to me.
| Quote: | A DI is by definition ( once ground lifted ) a differential device
anyway !
Right, but you can't lift that ground on the output and still get power
for the active stage.
AND, if you build an active differential stage and do without the
transformer, your maximum common mode voltage is limited to 48V at
best... which is nothing compared with some of the voltages I see on
the chassis of old guitar amps that aren't well-maintained.
|
I hope you do something about said guitar amps !
| Quote: | I simply don't get what your problem is !
My problem is that I can't get a 1M:600 transformer out of the Digi-Key
catalogue. Because of this, we have to do a lot of active circuit
work-arounds.
|
You can still do an 'instumentation amp' style front end that'll happily
present say 1M to the source.
No, it won't have unlimited common-mode input tolerance. Nor should it need it
!
'Ground lifitng' should *not* be an excuse for tolerating unsafe equipment !
It's simply a tactic to prevent common mode currents caused by indifferent
wiring methods.
Graham |
|
| Back to top |
|
|
Scott Dorsey
Guest
|
| Posted:
Sun Oct 30, 2005 11:09 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Richard Crowley <rcrowley@xpr7t.net> wrote:
| Quote: |
I've never seen one with BOTH ground lift AND phantom
powering. Mine will run on phantom as long as you don't
lift the ground. If you want to lift the ground, you must run
on internal battery power.
|
Again, Radial does this with the J48, by using a little switching supply
so they can get transformer isolation on the power line as well.
Absolutely a lifesaver at festivals where nobody remembers batteries.
| Quote: | Audio 'ground-lifting' does *not* require medical style isolation.
Likely
the ability to deal with as little as a Volt of common mode will
suffice.
|
Actually, I regularly see sixty to a hundred volts of common mode
trash coming off old clapped-out guitar amps at festivals. Proper
grounding and repairing the amp would improve this a lot, but when
you have three minutes to airtime is always when you find these
things out.
| Quote: | I'm suggesting that optical isolation may be a viable
substitute for a very-hi-Z isolation transformer (which
doesn't exist in the real-world with sufficient HF response)
|
Problem with optoisolators is that it's hard to get good linearity and
wide dynamic range with them. Great for MIDI, though, and now that we
are in a digital world, it might be the cheap solution for S-PDIF
isolation too (since most S-PDIF interfaces DO NOT meet the original
specs for isolation).
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis." |
|
| Back to top |
|
|
Scott Dorsey
Guest
|
| Posted:
Sun Oct 30, 2005 11:16 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Pooh Bear <rabbitsfriendsandrelations@hotmail.com> wrote:
| Quote: | Scott Dorsey wrote:
"Pooh Bear" <rabbitsfriendsandrelations@hotmail.com> wrote in message
Why would it mean spoiling the idea ?
Because you can't buy a transformer with a 10M input impedance. And if
you could, it would have no top end due to the interwinding capacitance.
So, if you want a 10M input impedance, you add a FET follower in front
of a 1:1 transformer. The FET stage gives you the impedance conversion.
The transformer gives you the isolation.
Can you suggest an example where 10M input Z is required ?
|
Any electric guitar or bass. 1M is minimum, and on a lot of pickups
you can tell the difference in sound between 1M and 10M. With a 1M input
on a DI, musicians will often claim that having the DI in circuit changes
the sound of the stage amp (because it halves the impedance the pickup
sees in a typical case).
| Quote: | This is how the classic active DI works. If you break the ground _before_
the transformer, you can no longer get phantom. If you break the ground
_after_ the transformer, you still have the induced noise on the input
unless you make an active differential stage instead of a simple follower.
Differential stages are 2 a penny today. Hardly rocket science.
Sounds like an excuse for sloppy out-dated design to me.
|
Give me one with a couple hundred volts rejection and we'll talk. (You
_could_ do this with a pair of 12AX7s... one as a differential pair,
the other as a couple cathode followers.... you know, that sounds like
it could be a fun project.)
| Quote: | A DI is by definition ( once ground lifted ) a differential device
anyway !
Right, but you can't lift that ground on the output and still get power
for the active stage.
AND, if you build an active differential stage and do without the
transformer, your maximum common mode voltage is limited to 48V at
best... which is nothing compared with some of the voltages I see on
the chassis of old guitar amps that aren't well-maintained.
I hope you do something about said guitar amps !
|
I normally put a red sticker on them to warn the next guy.
| Quote: | 'Ground lifitng' should *not* be an excuse for tolerating unsafe equipment !
It's simply a tactic to prevent common mode currents caused by indifferent
wiring methods.
|
That's right, but guitar rigs all have wiring methods that vary from
indifferent to unsafe. Sure, it would be nice from our standpoint if
guitars all had low-Z balanced pickups and nicely laid-out grounding
configurations. But they don't. Because they don't, we use the DI box.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis." |
|
| Back to top |
|
|
Richard Crowley
Guest
|
| Posted:
Sun Oct 30, 2005 11:21 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
"Scott Dorsey" wrote ...
| Quote: | Again, Radial does this with the J48, by using a little switching
supply
so they can get transformer isolation on the power line as well.
Absolutely a lifesaver at festivals where nobody remembers batteries.
|
Yeah, I guess for equipment the size of typical DI boxes,
it is easier to design a HF SMPS that is sufficiently
shielded/isolated/filtered so that nothing leaks into the
audio path. I keep thinking of newer, miniature devices
(like the MT2496 of recent discussion) where it is more
difficult to keep the artifacts out of the audio.
| Quote: | Problem with optoisolators is that it's hard to get good linearity
|
Typically solved by using a second isolator (or one with
dual receivers) in the feedback loop. Effectively makes the
transfer curve linear.
| Quote: | and wide dynamic range with them.
|
Yes, easier at line levels, but maybe prohibitively noisy
at mic levels.
| Quote: | Great for MIDI, though, and now that we
are in a digital world, it might be the cheap solution for S-PDIF
isolation too (since most S-PDIF interfaces DO NOT meet the original
specs for isolation).
|
TOSlink! :-) |
|
| Back to top |
|
|
hank alrich
Guest
|
| Posted:
Sun Oct 30, 2005 11:49 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Pooh Bear wrote:
| Quote: | Can you suggest an example where 10M input Z is required ?
|
For whatever reasons, the very best DI's I have used have an input
impedance in that range.
--
ha |
|
| Back to top |
|
|
Pooh Bear
Guest
|
| Posted:
Mon Oct 31, 2005 3:18 am Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
hank alrich wrote:
| Quote: | Pooh Bear wrote:
Can you suggest an example where 10M input Z is required ?
For whatever reasons, the very best DI's I have used have an input
impedance in that range.
|
Sounds to me like overkill.
I can't actually imagine where anything over 1M would be needed.
Graham |
|
| Back to top |
|
|
Kurt Albershardt
Guest
|
| Posted:
Mon Oct 31, 2005 3:51 am Post subject:
Re: S/PDIF isolation (was: 48 v phantom power to 9 v circuit |
|
|
Scott Dorsey wrote:
| Quote: |
Problem with optoisolators is that it's hard to get good linearity and
wide dynamic range with them. Great for MIDI, though, and now that we
are in a digital world, it might be the cheap solution for S-PDIF
isolation too (since most S-PDIF interfaces DO NOT meet the original
specs for isolation).
|
RME still uses pulse transformers for their S/PDIF (reconfigurable to
AES3) I/O ports. Are other manufacturers skimping on this $2 part? |
|
| Back to top |
|
|
John L Rice
Guest
|
| Posted:
Mon Oct 31, 2005 4:23 am Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
"Pooh Bear" <rabbitsfriendsandrelations@hotmail.com> wrote in message
news:436538A4.5CC69A9D@hotmail.com...
| Quote: |
hank alrich wrote:
Pooh Bear wrote:
Can you suggest an example where 10M input Z is required ?
For whatever reasons, the very best DI's I have used have an input
impedance in that range.
Sounds to me like overkill.
I can't actually imagine where anything over 1M would be needed.
Graham
|
So, you are saying your imagination is impaired?
--
John L Rice
www.DeliriumFix.com |
|
| Back to top |
|
|
Scott Dorsey
Guest
|
| Posted:
Mon Oct 31, 2005 5:09 am Post subject:
Re: S/PDIF isolation (was: 48 v phantom power to 9 v circuit |
|
|
Kurt Albershardt <kurt@nv.net> wrote:
| Quote: | Scott Dorsey wrote:
Problem with optoisolators is that it's hard to get good linearity and
wide dynamic range with them. Great for MIDI, though, and now that we
are in a digital world, it might be the cheap solution for S-PDIF
isolation too (since most S-PDIF interfaces DO NOT meet the original
specs for isolation).
RME still uses pulse transformers for their S/PDIF (reconfigurable to
AES3) I/O ports. Are other manufacturers skimping on this $2 part?
|
Yes, most are, I am sorry to report.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis." |
|
| Back to top |
|
|
Mr. DOS
Guest
|
| Posted:
Mon Oct 31, 2005 5:42 am Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Boy - I didn't mean to put this much heat into the discussion -- maybe
this needs another thread!
| Quote: | I can't actually imagine where anything over 1M would be needed.
Many finicky guitarists (myself included) claim to hear when their |
pickups are loaded by designs from engineers (which I'm not) that
"couldn't imagine." I always trust my ears in the final test, even if
I don't know/understand the reasons.
Does anyone know the current draw of an original SansAmp Bass DI :) ? |
|
| Back to top |
|
|
Gareth Magennis
Guest
|
| Posted:
Mon Oct 31, 2005 4:59 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
Hey, did you know that the BSS AR116 DI box has the option of adding a AR117
phantom power to DC converter? It takes the 24 - 48 volts phantom and, with
a CMOS oscillator and transformer circuit similar to the one Scott
describes, gives you 5 - 16.5 volts DC. It is the same size and shape, and
directoy replaces the backup 9v battery.
If anyone needs the schematic I can post it somewhere.
Gareth. |
|
| Back to top |
|
|
John L Rice
Guest
|
| Posted:
Mon Oct 31, 2005 8:12 pm Post subject:
Re: 48 v phantom power to 9 v circuit |
|
|
"Geoff@home" <geoff@nospam-paf.co.nz> wrote in message
news:Mam9f.3892$S24.244638@news.xtra.co.nz...
| Quote: |
"Paul Stamler" <pstamlerhell@pobox.com> wrote in message news:DPj9f.9833
Any electric guitar or bass. 1M is minimum, and on a lot of pickups
you can tell the difference in sound between 1M and 10M. With a 1M
input
on a DI, musicians will often claim that having the DI in circuit
changes
the sound of the stage amp (because it halves the impedance the pickup
sees in a typical case).
I'll give you another example of something that needs a 10M input
impedance:
a piezo pickup. They sound somewhat less like dog crap with a very high
load
Z. Sort of going from worse to bad.
So what devices have you seen with a 10M Z-IN ?
geoff
|
http://www.leonaudio.com.au/active.htm
http://www.lrbaggs.com/html/products/preamps_paradi.shtml
--
John L Rice
www.DeliriumFix.com |
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in