Difference between revisions of "Electronics project notes/Device voltage and impedance, audio and otherwise"

From Helpful
Jump to: navigation, search
m (Analog audio voltage levels)
m (Plugs's relation to balanced/unbalanced, voltage levels, etc.)
(38 intermediate revisions by the same user not shown)
Line 3: Line 3:
  
  
=Impedance and interconnection=
+
=Theory: Impedance when connecting two things=
 
<!--
 
<!--
  
Line 34: Line 34:
 
-->
 
-->
  
==Connecting two things==
+
 
 
<!--
 
<!--
Once the transistor existed, signals such as analog audio and a lot of digital communication, is now typically done with voltage signals.
+
 
{{comment|(Before then things were messier. And the oldest microphones and studios worked differently, being based on pre-transistor telephony)}}
+
Before the transistor, a number of things were difficult or expensive to and involved more expensive components.
 +
 
 +
This was mostly relevant in telephony, though early studios were modeled directly on pre-transistor telephony, sol old the oldest microphones and studios worked differently from how they do now.
 +
 
 +
 
 +
Once the transistor existed a lot of signals became much easier and cheaper to communicate,
 +
particularly over shorter distances.
 +
Analog audio and a lot of digital communication is now typically done with voltage signals.
  
  
Voltage signalling puts up two major constraints on what you do:
+
Voltage signalling puts up two major constraints of its own:
 
* voltages should be similar
 
* voltages should be similar
* the relation between the source's output impedance, the load's input impedance, and the cable's characteristic impedance for the signal you put on there
+
* implications of the relation between  
 +
** the source's output impedance
 +
** the load's input impedance
 +
** the cable's characteristic impedance for the signal you put on there {{comment|(less relevant than the above until you get to a few hundred kHZ and/or deal with longer distances)}}
 +
 
  
 
Voltage signals are usually sensed with higher impedance than the thing that produces that signal has - impedance bridging (sometimes a.k.a. voltage matching) in the list below.
 
Voltage signals are usually sensed with higher impedance than the thing that produces that signal has - impedance bridging (sometimes a.k.a. voltage matching) in the list below.
Line 50: Line 61:
 
There is a rule of thumb that the ratio between these two impedances is on the order of 1:10
 
There is a rule of thumb that the ratio between these two impedances is on the order of 1:10
 
This is a tradeoff between
 
This is a tradeoff between
 +
  
 
This is a choice. The options are:
 
This is a choice. The options are:
Line 65: Line 77:
 
===Impedance matching===
 
===Impedance matching===
 
<!--
 
<!--
[https://en.wikipedia.org/wiki/Impedance_matching Impedance matching] setups means both impedances are roughly equal.
+
[https://en.wikipedia.org/wiki/Impedance_matching Impedance matching] connections setups means both impedances are roughly equal.
  
  
 
Impedance matching can be done for a number of motivations
 
Impedance matching can be done for a number of motivations
* minimize reflections - when wavelengths of the signal are on the same order as the transmission line,
+
* minimize reflections - when wavelengths of the signal are on the same order as the transmission line  
* maximum power transfer  
+
* maximum power transfer {{verify}}
* avoid losses due to transmission line effects, or other causes
+
* avoid losses due to transmission line effects {{verify}}
 +
 
  
It is still still relevant for  
+
Impedance matching is relevant for long distances,  
long-distance applications,
+
 
high-frequency applications,
 
high-frequency applications,
 
and anything that by its nature has reactive components.
 
and anything that by its nature has reactive components.
  
For example, some final-amplification steps, such as some RF circuits like antenna outputs.
+
For example,  
 +
even in devices that mostly communicate with impedance-bridged setups,
 +
some final amplification, RF circuits such as antenna output,
 +
makes more sense in impedance-matching designs.
  
  
  
Impedance matching was a more generally sensible  
+
Impedance matching was ''more generally'' sensible
when voltage amplification was hard and amplification could not easily be separated from impedances.
+
when voltage amplification was hard,
 +
and amplification could not easily be separated from impedances.
  
 
In the pre-transistor days it was basically unavoidable,  
 
In the pre-transistor days it was basically unavoidable,  
 
because loads and drivers were often necessarily reactive,  
 
because loads and drivers were often necessarily reactive,  
which implied the best way to get a signal to play nice over a longer line was to adhere to the [https://en.wikipedia.org/wiki/Maximum_power_transfer_theorem maximum power transfer theorem], which basically says that the optimal amount of power transfer happens when impedances are matched.
+
which implied the best way to get a signal to play nice, particularly over a longer line,
 +
was to adhere to the [https://en.wikipedia.org/wiki/Maximum_power_transfer_theorem maximum power transfer theorem], which basically says that the optimal amount of power transfer happens when impedances are matched.
  
 
In that situation, maximum power and signal fidelity are a tradeoff.
 
In that situation, maximum power and signal fidelity are a tradeoff.
Line 171: Line 188:
 
How suitable mostly depends on the combination of frequency and cable length.  
 
How suitable mostly depends on the combination of frequency and cable length.  
 
(In case you're wondering: for audio, you're good up to kilometers{{verify}})
 
(In case you're wondering: for audio, you're good up to kilometers{{verify}})
 +
 +
-->
 +
 +
=Impedance mismatches=
 +
<!--
 +
 +
Having an interconnection that mixes impedances creates issues for higher-frequency content.
 +
 +
That issue is mostly signal reflection.
 +
 +
 +
'High frequency' means when wavelengths of the signal are on the same ''order'' as the transmission line's length.
 +
 +
at 100kHz or so that's the scale of a house,
 +
at MHzes it's the scale of your desk.
 +
 +
 +
https://en.wikipedia.org/wiki/Impedance_matching
 +
  
 
-->
 
-->
Line 233: Line 269:
 
I'm avoiding using dBV and dBu ''because'' they have different references: 0dBV is 1V, 0dBu is 0.7746V.
 
I'm avoiding using dBV and dBu ''because'' they have different references: 0dBV is 1V, 0dBu is 0.7746V.
 
which means that consumer's 0.3V is -10 dBV and professional's 1.2V is 4dBu and those two are not directly comparable, because there's a 2.2dB difference.
 
which means that consumer's 0.3V is -10 dBV and professional's 1.2V is 4dBu and those two are not directly comparable, because there's a 2.2dB difference.
-->
+
 
 
(TODO: sort out peak versus RMS values)
 
(TODO: sort out peak versus RMS values)
 +
 +
-->
  
  
  
* '''phono input'''  
+
* '''phono input''' <!--(typically uses RCA plugs)-->
 
: on the order of a milliVolt, and seems to often be the not-yet-amplified output of [http://en.wikipedia.org/wiki/Magnetic_cartridge phono cartridges]
 
: on the order of a milliVolt, and seems to often be the not-yet-amplified output of [http://en.wikipedia.org/wiki/Magnetic_cartridge phono cartridges]
 
: There are two common types:
 
: There are two common types:
Line 277: Line 315:
  
 
* '''professional microphone level'''
 
* '''professional microphone level'''
: order of 10mV.
+
: order of 10mV, but can vary:
 
:: Can be ~1mV, can be ~200mV (in theory more but this is atypical)
 
:: Can be ~1mV, can be ~200mV (in theory more but this is atypical)
 
:: more varied designs, and possible amplification at the mic, means more variation with design ''and'' per use
 
:: more varied designs, and possible amplification at the mic, means more variation with design ''and'' per use
Line 283: Line 321:
 
:: ...so you ''will'' need that gain knob
 
:: ...so you ''will'' need that gain knob
 
: impedance:
 
: impedance:
:: mic output: most are in the 50..200 Ohm range, with deviations (see more notes around here)
+
:: mic output: most are (higher) in the 50..200 Ohm range, with deviations (see more notes around here)
 
:: mic preamp/mixer input: order of 1..2kOhm
 
:: mic preamp/mixer input: order of 1..2kOhm
  
Line 298: Line 336:
  
  
* '''instrument level''' has no standard  
+
* '''instrument level''' has no standard, though is generally quite predictable:
: ...but is often somewhere between mic and line level
+
: voltage is often somewhere between mic and line level
 
+
 
: output impedance
 
: output impedance
 
:: pickup impedance is often quite high (see also notes below on pickup impedance)
 
:: pickup impedance is often quite high (see also notes below on pickup impedance)
:: input to a mixer will typically need a [[direct box]] (a.k.a. [[DI]]) to convert to typical impedance
+
:: input to a mixer will typically need a [[direct box]] (a.k.a. [[DI]]) to convert most things to typical XLR
::: the exception is guitar amps, which ''expect'' high impedance, basically because they expect directly connected pickups
+
:: output to guitar amps is unchanged - they ''expect'' high impedance from a directly connected pickup (DI boxes tend to have a thru on the input side so that you could do both)
 
+
  
  
Line 311: Line 347:
 
* '''headphone level'''  
 
* '''headphone level'''  
 
: ''roughly'' commercial line level, but less of a standard - can be higher.  
 
: ''roughly'' commercial line level, but less of a standard - can be higher.  
 +
: headphone amps tend to aim to power a ~30-60Ohm headphone (with a few milliamps{{verify}}
 +
:: there are ~4Ohm headphones, but you really woudn't plug those into everything (likely to distort)
 +
:: there are 250Ohm-600Ohm headphones, but these need their own preamp (the idea is that you can design for slightly better THD with less load on the amp)
 +
<!--
 
: using headphone out for line out is safe (because the voltage is the same, and line ins have higher impedance)
 
: using headphone out for line out is safe (because the voltage is the same, and line ins have higher impedance)
: driven by amps with an impedance of roughly 4Ohm in that it has enough power enough to drive a ~30-60Ohm headphone (with a few milliamps{{verify}}
+
-->
: headphones are often around 30-60 Ohm
+
:: there are ~4Ohm, but you reakky woudn't plug those into everything (likely to distort)
+
:: there are 250Ohm-600Ohm headphones, but these need their own preamp (the idea here being that you can design for slightly better THD with less load on the amp)
+
  
 
* '''Car audio''' tends to be on the order of 2V, sometimes 4V {{verify}}
 
* '''Car audio''' tends to be on the order of 2V, sometimes 4V {{verify}}
Line 329: Line 366:
  
 
* pro speaker wires
 
* pro speaker wires
: not really a thing. Most speaker cables are either
+
: not really a thing. Most speakers are connected by one of:
:: XLR-connected: carrying line signals to active speakers
+
:: XLR: carrying typical XLR line signals to active speakers
 
:: [https://en.wikipedia.org/wiki/Speakon_connector Speakon]-connected: already-amplified signal to a passive speaker
 
:: [https://en.wikipedia.org/wiki/Speakon_connector Speakon]-connected: already-amplified signal to a passive speaker
:: TS: already-amplified signal to a passive speaker. (Sometimes avoided to avoid smoky mixups, then typically Speakon instead)
+
:: 6.3mm TS: already-amplified signal to a passive speaker (Sometimes avoided to avoid smoky mixups)
::: Note these are different from TS instrument cables, basically in that instrument cables use a thinner core-and-shield and these are beefer and not shielded (just 2-lead stranded{{verify}})
+
::: Note these cables are different from TS instrument cables, basically in that instrument cables use a thinner core-and-shield and these are beefer and not shielded (just 2-lead stranded{{verify}})
  
 
<!--
 
<!--
Line 342: Line 379:
 
See also:
 
See also:
 
* https://en.wikipedia.org/wiki/Line_level#Nominal_levels
 
* https://en.wikipedia.org/wiki/Line_level#Nominal_levels
 +
 +
====dBV and dBu====
 +
<!--
 +
I'm avoiding using dBV and dBu ''because'' they have different references:
 +
: 0dBV is 1V
 +
: 0dBu is 0.7746V
 +
 +
which means that
 +
: consumer's 0.3V is -10 dBV
 +
: professional's 1.2V is 4dBu
 +
: there's a 2.2dB difference between the two standards
 +
 +
 +
-->
  
 
====Audio device differences====
 
====Audio device differences====
Line 406: Line 457:
 
====Balanced audio / pro audio====
 
====Balanced audio / pro audio====
 
<!--
 
<!--
 +
tl;dr:
  
Pro audio isn't about higher sound quality (pro and consumer have been comparable for a loong time),
+
* for longer distances, and/or when interconnecting many devices, [[differential mode]] signalling is a good idea
it'a about setting up communication to not ''lose'' it on longer distances and/or when combining many devices.
+
* differential signalling (should) imply on balanced lines
 +
* this makes "balanced" and "differential" near-synonyms in this context {{comment|(though the difference does occasionally matter)}}
  
So is very relevant on most stage setups.
+
* interconnecting many devices, differential is a good idea
 +
* Both pro and consumer can use (often selective) isolation to help solve other issues.
  
  
Pro audio usually implies balanced audio, which really means "differential signalling on balanced-impedance lines". See the notes on differential versus balanced above.
+
Pro audio is not about carrying higher sound quality (pro and consumer have been comparable for a loong time),
 +
it's about not losing it as quickly over longer distances and when combining many devices.
  
This helps two goals we have:
 
* Avoid ground
 
:: because common mode issues ''would'' otherwise eventually give you large headaches
 
  
* Lowering induced noise
+
It does this mainly using balanced audio,
:: because that means less noise -- and running longer cables
+
which really means "differential signalling on transmission lines with (ideally) equal impedance" {{coment|(See the notes on the relation between differential and balanced)}}
  
Impedance and all that is secondary, and more related to historical habits
 
that have stick around in modern designs. Because that's what habits do.
 
  
 +
In practical terms, this helps
 +
* lower the result of environmental EM inducing onto cabling, which lets us run longer analog lines
  
 +
* avoid the inevitability of ground loops (by avoiding ground in interconnections)
 +
: this is basically the same point as the first, except that ground loops are better at it, and inevitable with common mode signalling.
  
For context, the main difference between pro and consumer analog audio is that consumer is all '''[[common mode]]''' signalling, meaning a signal is sent along referenced to a signal common.
+
* (and a few other details)
  
That signal common is frequently also tied to [[ground]], which invites a whole bunch of issues - that shared reference is easily polluted (more so with more devices), and can invite pickup of magnetic fields by effectively creating big loops with all the ground wires involved (more so with more devices and longer distances), and some other muck. Much of that is solvable, but it's bothersome.
 
 
Pro audio avoids some of the most obvious noise issues you would see in common-mode audio,
 
by avoiding sharing generic signal references.
 
 
 
tl;dr:
 
* for longer distances, balanced is a good idea
 
* interconnecting many devices, differential is a good idea
 
* Not all pro audio is balanced, not all audio needs to be balanced
 
 
* Both pro and consumer can use (often selective) isolation to help solve other issues.
 
  
  
 +
Differential is the main thing that makes you call pro gear that (regardless of quality, because that's always varied).
  
 +
It is contrasted with [[common mode]], also because consumer gear uses that. Common mode references to a signal common -- which almost always ends up being ground and '''that''' is too easily polluted when you've got more than two devices around.
 +
And some other muck. Much of that is actually solvable, but it's just more bothersome.
  
 +
Most ''other'' details, to both consumer and pro, tend to be less important, and come from historical habits that stuck around.
  
Here's a fun one: ''' 'balanced' is a bad name'''.
 
  
Ninety percent of the time, balanced means "circuits where the ''receiving end'' is designed to do differential interpretation, which avoids noise, and does so better over impedance-balanced lines".
 
 
And, in practice, those lines are also shielded because that further lowers noise - but also completely distinct from the balanced and differential concepts.
 
  
  
Line 456: Line 498:
  
 
Differential-and-balanced means:
 
Differential-and-balanced means:
* sending side has a signal, and also creates its inverse
+
* sending side has a signal. It also creates its inverse
 
* sending side puts those two things on a pair of wires (a balanced pair, meaning equal impedance)
 
* sending side puts those two things on a pair of wires (a balanced pair, meaning equal impedance)
* receiving side subtracts one signal from the other
+
* receiving side subtracts one signal from the other.
  
  
 +
Many articles will explain that the trick of balanced is the differential part:
 +
noise that made its way onto both wires will make its way onto both wires equally,
 +
so be subtracted from its copy at the receiving end.
  
Now, many articles will explain that the trick of balanced is the differential part:
+
The noise rejection happens not because the signal is carried so well,
noise that made its way onto the cables will make its way onto both wires equally,
+
but because the noise arrives on both lines equally.  
so be mostly subtracted from itself at the other end.
+
  
What many ''don't'' mention is that the sending side creating that inverted-copy is completely optional.
+
Or rather, ''when'' it is, which is why you want to carry such a signal on two wires of equal impedance
It's a good idea in the why-not sense, but it's the least important bit.
+
(~= balanced lines), and not, say, coax.
  
  
The noise rejection happens when the receiver indeed gets the noise on both lines as exactly the same voltage signal at the receiver end.
+
What many ''don't'' mention is that the sending side creating that inverted-copy is
 +
completely optional.  
  
This itself happens mostly by merit of using two wires of equal impedance (~= balanced lines).
+
It's a good idea in the why-not sense, but it's the ''least important part'',
 +
because it's the way you've connected things that does all the clever noise rejection.
  
  
For example many microphones aren't differential signals,
+
Sure, under the same voltage-on-the-line restrictions, having an inverted copy
but for the noise rejection, you want to connect them to a differential receiver with proper balanced cables.
+
rather than silence on the second wire gives you twice the signal (so a few dB more)
 +
after the subtraction,
 +
and in theory that's good for noise floor reasons, but in a lot of practice it doesn't matter
 +
(for a handful of device-specific footnotes. I should write them out sometime).
  
And basically no guitar pickups are differential either, so you want to convert to balanced with a DI whenever cables travel any distance (...also because of impedance reasons, but ignore that for now).
 
  
A DI also does isolation, so avoids active instruments (say a powered keyboard) being able to accidentally introduce ground issues
 
  
 +
More to the point, it is perfectly valid to put a single-sided signal on a differential line
 +
{{comment|(as long as that ''doesn't'' mean you just directly tied in a shared ground)}}.
  
 +
In fact, this is pretty common in pro gear design, particularly the already-amplified part
 +
(because in general it's only the first amplification in a chain that matters in terms of getting above the noise).
  
And it is perfectly valid to put a single-sided signal on a differential line,
 
as long as that ''doesn't'' mean you directly tie in an earthed/shared ground
 
  
The only direct effect is a factor two (-6dB) difference in level, which you can lessen/avoid with a bit of amplification. (Stay in specs to avoid clipping, though)
+
That doesn't mean you can put common-mode cabling in.
  
 +
and you want to avoid combining consumer audio.
 +
Signal-wise it's fine, but it's ''way'' too easy to introduce shield-or-ground-related mistakes.
  
  
That doesn't mean you can put common-mode cabling in,
+
...Unless a particular input is made to accept it, meaning it has isolation.
and you want to avoid combining consumer audio all the same.
+
 
+
Signal-wise it's fine (signal-0DC = signal, though with that -6dB level difference),
+
but it's ''way'' too easy to introduce shield-or-ground-related mistakes.
+
  
 
Mixers that say "bal/unbal" on their inputs
 
Mixers that say "bal/unbal" on their inputs
: tend to use an isolation transformer (think DI without the impedance change).
+
: tend to have put an isolation transformer on that input (think DI without the impedance change).
: usually only on 6.35 (1/4 jack).
+
 
 +
: usually only on 6.35mm (1/4 jack).
 +
 
 
: It would work just the same on XLR as well, but that invites the [[pin 1 problem]]{{verify}} (putting your shield into your signal) that took so long to eradicate.
 
: It would work just the same on XLR as well, but that invites the [[pin 1 problem]]{{verify}} (putting your shield into your signal) that took so long to eradicate.
 
:: actually, those XLR-to-3.5mmjack cables do exactly that, but ''when'' they are only used to plug a microphone into a PC that's fine.
 
:: actually, those XLR-to-3.5mmjack cables do exactly that, but ''when'' they are only used to plug a microphone into a PC that's fine.
Line 505: Line 554:
  
  
More practically:
 
  
  
 +
 +
More practically:
 
* Mixers often mark things (particularly inputs) 'balanced' or 'unbalanced'. '''''At''''' a mixer  
 
* Mixers often mark things (particularly inputs) 'balanced' or 'unbalanced'. '''''At''''' a mixer  
 
:: 'balanced input' basically means 'differential interpretation'
 
:: 'balanced input' basically means 'differential interpretation'
Line 531: Line 581:
  
 
-->
 
-->
 +
=====Balanced in electrical terms=====
 +
<!--
  
 +
Historically, balanced cabling was a result of using transformers, basically in a balun-pair style:
 +
one at each end, and either end could be ground-referenced but the connection would not be.
 +
 +
The net effect is also that both wires carry half the signal (-6dB relative to the full original),
 +
and implies the signal symmetry seen in
 +
 +
 +
Upside:
 +
* fairly neat solution to not sharing ground
 +
 +
Downside:
 +
* a pair of transformers is a little more expensive than a solid-state solution
 +
* passive, meaning that if you wanted to amplify (e.g. to be further away from the environmental noise floor), that would be a separate step before
 +
 +
 +
 +
These days, you can use active components (mainly op amps) to build both line drivers and differential receivers, that sets up a very similar differential transmission.
 +
 +
Upside:
 +
* cheaper
 +
* often just as good
 +
 +
Downside:
 +
* No impedance control {{verify}}
 +
 +
 +
 +
Note that quite a few of these active line drivers don't use two line drivers
 +
save an op amp by actually transmitting single-sided (and often amplified x2 to output the same levels).
 +
 +
This requires a resistor on the other line{{verify}} which sometimes gets called impedance balancing (...extra confusingly).
 +
 +
Upside:
 +
* slightly cheaper yet
 +
* does not affect the noise rejection
 +
 +
Downside:
 +
 +
 +
 +
 +
 +
 +
https://ww2.minicircuits.com/app/AN20-002.pdf
 +
 +
 +
 +
Also related is that a transformer can be easily used to go between
 +
ground-referenced signal and a differential one. See [[balun]].
 +
There are some footnotes to this, though.
 +
 +
 +
 +
 +
 +
-->
 +
 +
=====Some other terms you see=====
 +
<!--
 +
 +
"symmetric balanced output" versus "asymmetric balanced output" seems to be purely whether the line drive puts out a single-ended signal or also an inverted one.
 +
: ...which has no effect on the noise.
 +
 +
 +
"Impedance balancing" is a term mostly repeated within audio circles, with the reverence of the semi-unknown, but really just means making sure you use balanced lines.
 +
 +
It can refer to a few different things
 +
* connecting different things designed for different impedances
 +
: this is sometimes a bad idea - e.g. if the connection is made to be impedance-bridged
 +
: this is sometimes irrelevant -
 +
: this is sometimes necessary  -
 +
 +
* using a balanced line, i.e. a wire with a pair of equal resistance/impedance, for
 +
: this is sometimes a very good idea - e.g. when using a differential decode on the other end, this means better noise rejection
 +
: this is sometimes somewhat relevant - e.g. using
 +
: this is sometimes irrelevant -
 +
 +
* adding a resistor to a differental line driver that puts out a single-ended signal
 +
: required in that design{{verify}}
 +
: also badly/confusingly named
 +
 +
 +
 +
-->
  
 
====Plugs's relation to balanced/unbalanced, voltage levels, etc.====
 
====Plugs's relation to balanced/unbalanced, voltage levels, etc.====
Line 538: Line 674:
 
* '''XLR3''' is '''pro mic level''', always balanced/differential, always mono.
 
* '''XLR3''' is '''pro mic level''', always balanced/differential, always mono.
 
: mono, because one signal requires a differential ''pair''.
 
: mono, because one signal requires a differential ''pair''.
: if you want to carry stereo over XLR, use two cables (in practice, stereo is often about inter-device, and may well be two balanced 6.3mm TRS instead).  
+
: if you want to carry stereo over XLR, use two cables (in practice, stereo is often about inter-device, and may well be two balanced 6.35mm TRS instead).  
  
* '''6.3 mm''' is (typically) '''pro line level'''  {{comment|(6.35mm but people are lazy typers)}}
+
* '''6.35 mm''' is (typically) '''pro line level'''  {{comment|(6.35mm but people are lazy typers)}}
: '''6.3mm TRS''' is balanced mono, see notes on that above.
+
: '''6.35mm TRS''' is balanced mono, see notes on that above.
 
:: or, sometimes, unbalanced stereo. This is an exception and ''will'' be noted.
 
:: or, sometimes, unbalanced stereo. This is an exception and ''will'' be noted.
: '''6.3mm TR''' is unbalanced mono, often instruments, which is also often lower voltage levels (but close enough{{verify}})
+
: '''6.35mm TR''' is unbalanced mono, often instruments, which is also often lower voltage levels (but close enough{{verify}})
: when a device instead uses 6.3mm these for mic in, aux, or controllers like pedals, or stereo, they will be marked as such (or switchable)
+
: when a device instead uses 6.35mm these for mic in, aux, or controllers like pedals, or stereo, they will be marked as such (or switchable)
  
* mixer outputs are often two 6.3mm jacks (balanced, TRS)
+
* mixer outputs are often two 6.35mm jacks (balanced, TRS)
  
 
* RCA on a mixer are typically only used for phono in, or consumer in (aux)
 
* RCA on a mixer are typically only used for phono in, or consumer in (aux)
Line 553: Line 689:
  
  
On 6.3 TS versus TRS
+
On 6.35 TS versus TRS:
* 6.3mm (1/4") TRS is most typically mono and balanced/differential. A mixer input will often mark this as "balanced"
+
* 6.35mm (1/4") TRS is most typically balanced/differential mono. A mixer input will often mark this as "balanced"
 
: Tip and Ring is the pair, Sleeve is shield.
 
: Tip and Ring is the pair, Sleeve is shield.
: No, it's not stereo. And using a 3.5mm-to-6.3mm converter to plug in consumer line level will do Weird Things.
+
: No, it's not stereo. And using a 3.5mm-to-6.35mm converter to plug in consumer line level will easily do weird things (depending on the case).
  
* 6.3mm (1/4") TS - instrument cable
+
* 6.35mm (1/4") TS - instrument cable
: is mono, and ''not'' differential  
+
: is mono, and ''not'' differential
 +
: from actual instruments it won't
 
: Tip is signal, Sleeve is shield,
 
: Tip is signal, Sleeve is shield,
: mixers tend to accept both TRS balanced and TS unbalanced. If on the same socket they usually mark it (e.g. "bal/unbal")
+
 
:: Note that unbalanced inputs are not always isolated, so connecting unbalanced things other than instuments ''could'' create common mode issues.
+
* mixers tend to accept both TRS balanced and TS unbalanced. If on the same socket they usually mark it (e.g. "bal/unbal")
 +
:: Note that unbalanced inputs are not always isolated, so connecting unbalanced things (other than floating instuments) ''could'' create common mode issues.
 
: contrast with...
 
: contrast with...
  
Line 571: Line 709:
  
  
'''Things to avoid'''
+
'''Things that won't work / things to avoid'''
 
<!--
 
<!--
 
* TS and TS: confusing speaker cable and instrument cable
 
* TS and TS: confusing speaker cable and instrument cable
Line 577: Line 715:
 
: using instrument cable to connect speakers will probably burn that cable
 
: using instrument cable to connect speakers will probably burn that cable
 
: plugging speaker out to mixer in will fry that mixer
 
: plugging speaker out to mixer in will fry that mixer
 +
 +
 +
* consumer 3.5mm mic to 6.35mm input (via adapter)
 +
: it doesn't provide plug-in power, so won't work {{comment|(with a few sort-of exceptions, like battery'd lavs)}}
  
  
 
* TS and TRS: Plugging a TS output (often an instrument) into balanced (TRS) input  
 
* TS and TRS: Plugging a TS output (often an instrument) into balanced (TRS) input  
 
: the plug will short between sleeve and ring, i.e. tie one of the signal lines to shield.
 
: the plug will short between sleeve and ring, i.e. tie one of the signal lines to shield.
:: which on the 'cheaty' single-ended balanced driver design could be fine
+
:: which on the 'cheaty' single-ended balanced driver design could be fine,
:: on a halfway proper one can cause damage  
+
:: on a halfway proper one can cause damage
:: a fancier proper one is probably protected against this case (but don't assume it)
+
:: while particularly fancy ones may be protected against this case (but don't assume it)
: may introduce common mode from the mixer itself. {{verify}}
+
: might introduce common mode from the mixer itself. {{verify}}
  
 
* TS and TRS: Plugging a TRS output (balanced) into a unbalanced-only TS input
 
* TS and TRS: Plugging a TRS output (balanced) into a unbalanced-only TS input
Line 590: Line 732:
  
  
* TRS and XLR: 6.3mm TRS to XLR male
+
* TRS and XLR: 6.35mm TRS to XLR male
 
: if used to plug balanced gear into a mic in, this works (attenuates the signal though, because of)
 
: if used to plug balanced gear into a mic in, this works (attenuates the signal though, because of)
 
: trying to use it for instrument will probably work too.
 
: trying to use it for instrument will probably work too.
Line 596: Line 738:
  
  
* Adapters: XLR female (output) to 6.3mm input adapters
+
* Adapters: XLR female (output) to 6.35mm input adapters
 
: to unbalanced (TS): does not convert
 
: to unbalanced (TS): does not convert
 
: to balanced (TRS):
 
: to balanced (TRS):
Line 658: Line 800:
  
 
Low-impedance pickups exist - e.g. order of 400 Ohm, - mean you can plug it directly into a mixer without a DI.
 
Low-impedance pickups exist - e.g. order of 400 Ohm, - mean you can plug it directly into a mixer without a DI.
 
See e.g. Les Paul's
 
 
  
  
Line 739: Line 878:
  
 
=====On microphone impedance=====
 
=====On microphone impedance=====
 +
{{stub}}
 
<!--
 
<!--
 +
Consumer mics used to all be high impedance, because that's cheaper to make and noise isn't much of an issue on a one-or-two-meter cable.
  
Consumer mics used to all be high impedance,
 
which is fewer components, so cheaper,
 
but the impedance means it's more susceptible to noise with cables longer than a meter or two.
 
  
 +
This stuck around in some design habits, including PC sound cards, to this day.
 +
Electret mics are usually ~1kOhm, bridging with 5-10kOhm (varied, so roughly!)
 +
(and powered by intentional DC bias on, well, many things with 3.5mm inputs).
  
  
 +
''Most'' modern pro-audio mics have an output impedance in the 150 to 250 Ohm range, so that they're impedance-bridging (factor 5 to 10) with a typical mic amp, which often have an input impedance of around 1.2kOhm and almost always in the 1kOhm to 3kOhm range.
  
''Most'' modern pro mics have an output impedance in the 150..250 Ohm range,
 
so that they're impedance-bridging with a mic amp, which typically have an input impedance in the 1.2k .. 3kOhm range.
 
  
The low impedance also means you can use longer cables with less bother
+
Bridging, in general, makes it easier to avoid distortion, and of longer cables with less bother.
: ...if [[balanced cables]] and interpreted differential-style. Which with XLR wiring to XLR mic inputs is basically a given).
+
...at the cost of being more noise-sensitive,
 +
which is why pro mics the balanced+differential-style wiring (implied by XLR) is necessary, and universal.
  
  
For higher-impedance mic elements, it also implies adding a transformer in the mic to get to expected impedance.
+
Compare this to e.g.
e.g. most dynamic mics have transformers, apparently also largely to increase the voltage to mic levels (while not making the impedance higher than practical).{{verify}}
+
: Cheap electret mics, which is higher impedance so you rarely get much cable
  
 +
: Guitar wiring, which is higher impedance and loses high frequencies faster, which is why these are always fairly short cables.
  
 +
: Most dynamic microphones's actual capsule is also higher-impedance, which is why they'll often have a transformer to get typical impedance (also to get higher voltage levels)
  
While the more typical split is into  
+
 
: low (50-600) and  
+
 
 +
For pro mics, the typical split is into  
 +
: low (50-600 and usually under 200 or so)
 
: high (10k+)
 
: high (10k+)
  
Line 769: Line 914:
 
: high impedance (15,000 ohms rating)
 
: high impedance (15,000 ohms rating)
  
...because 600 ohm is a bit of a weird one (given typical ~1.5kOhm-input mic amps).
+
...because 600 ohm is a bit of a weird one (given typical ~1.2kOhm-input mic amps).
  
  
 
Notes:
 
Notes:
* These are "the nearest category" style ratings, not exact figures. It wily vary per microphone
+
* Specs are often approximate figures, sometimes "nearest category" style, and not exact figures.
 +
: It will vary per microphone design
  
* impedance varying with frequency somewhat is part of its frequency response
+
* and also vary over its frequency response
  
* not using typical bridge-style impedance ratios will alter its frequency response
+
* not using typical bridge-style impedance ratios will ''alter'' its frequency response
: this is strangely common - strange in that most people do not quite understand how/why, just that.
+
: this is strangely common - strange in that most people do not quite understand why it does or how much, just that.
  
* 50 Ohm is basically an obsolete standard
+
* some new preamps have variable input impedance.
 +
: for most neutral response, use the highest setting.
 +
: lower will load the mic more.
 +
:: A little more load changes only the sound, and can be useful to do intentionally.
 +
:: a lot more will distort more easily, and may lower the max level
 +
:: not ideal for long runs
  
 +
* 50 Ohm is basically an obsolete standard. As is 600, mostly.
 +
: So low is often around 200.
  
 +
* impedance matters to very long cables, which you won't see in studios, but may in e.g. orchestras, and to setting up live things.
  
  
There are also some modern ~600-ohm mics out there.
 
  
It seems that microphones may also be marked with their designed load impedance.
+
600 Ohm seems to originate in old studios, though there ''are'' some modern ~600-ohm mics out there.
600 Ohm suggests this{{verify}} and this seems to be some of the confusion.
+
  
 +
It seems that some microphones are marked with their designed load impedance.
 +
600 Ohm is sometimes this{{verify}} and this seems to be some of the confusion.
  
  
  
 +
Modern mics (pro and consumer) are intended to be loaded impedance-bridge style, so with an impedance rougly ten times as high
  
 +
...more roughly than in bridging in general - loading e.g. a dynamic microphone more than that (e.g. factor five) also seems common enough, despite that this loads the microphone itself enough that it will e.g. drop out the highs.
  
Modern mics are intended to be loaded impedance-bridge style, so with an impedance rougly ten times as high (more roughly than in bridging in general).
+
And regularly ''because'' of that: it makes for a warmer/darker sound.
  
Loading it more than that (i.e. with lower impedance) also seems common enough,
+
However, loading it more yet tends to also drop the lows, which just sounds tinny. {{comment|(also levels decrease, but you can often gain that back up)}}
despite that this loads the microphone itself that it will drop out the highs.  
+
And often ''because'' of that: it makes for a warmer/darker sound.
+
  
However, loading it more yet tends to also drop the lows,
 
which just sounds tinny.
 
 
If you like the frequency response, can accept a somewhat lower signal, and don't blow up anything, have fun with whatever amount of loading.
 
If you like the frequency response, can accept a somewhat lower signal, and don't blow up anything, have fun with whatever amount of loading.
(but know that it's not quite the intent. Also if you have the choice of using an EQ if you have it anyway, this may be better for SNR{{verify}})
+
(note that if you have an EQ on your mixer, that may be easier to control, and slighly better for overall SNR{{verify}})
  
 +
Don't expect the same from cheaper mics - e.g. the transformers in cheap dynamics tend to be smaller, so react differently{{verify}}.
  
For some condenser mics, higher-than-1:10-loading will start distorting (loaded FET?{{verify}}).
 
Which is sometimes done intentionally too.
 
  
 +
Condenser mics react differently, because you're dealing with the FET and not the capsure. Loading them more will more easily distort the signal -- which is also sometimes entirely intentional.
  
In all cases the exact response varies per microphone type, and components/design so can vary with model.
 
Also don't expect the same from cheaper mics - e.g. the transformers in cheap dynamics tend to be smaller, so react differently{{verify}}.
 
  
 +
In all cases the exact response varies per microphone type, and ''can'' vary with model due to the components/design used.
  
  
Some mic amps allow you to select their input impedance.
+
 
 +
Some mic amplifers allow you to select their input impedance.
  
 
This in part to work with a wider range of mic impedances,
 
This in part to work with a wider range of mic impedances,
 
yet many people will probably consider it a dark/bright sound setting.
 
yet many people will probably consider it a dark/bright sound setting.
 
 
  
  
Line 938: Line 1,088:
  
 
===DI===
 
===DI===
<!--
+
{{stub}}
Direct box, DI box, DI unit, DI. {{comment|(People argue over whether it stands for direct input, direct injection, direct induction, or direct interface)}}.
+
Direct box, DI box, DI unit, DI. {{comment|(people argue over whether it stands for direct input, direct injection, direct induction, or direct interface)}}.
  
  
 
'''Functionally'''
 
'''Functionally'''
  
Takes a high-impedance pro-line-level (possibly-unbalanced-but-may-be-balanced) signal, e.g. from a passive instrument, or amp's thru.
+
Takes a high-impedance, possibly-unbalanced signal (often roughly pro line level),
: Usually on a TS plug
+
e.g. from a passive instrument - probably most frequently electric guitars and electric basses.
: there's typically a second TS socket (marked link or thru), wired directly to the first. Meant to use the DI as a splitter of a (dry) signal towards the mixer and something stage-side, e.g. pedal or stage monitor.
+
  
Outputs a low-impedance pro-mic-level balanced signal.
+
Outputs a low-impedance pro-mic-level balanced signal, usually on an XLR plug.
: Usually on an XLR plug
+
  
 +
 +
In other words, usually plugs a high-impedance instrument into a (line that ends up at the) mixer.
 +
 +
<!--
  
  
Line 957: Line 1,109:
 
Most designs do ground isolation / floating designs, to avoid common mode issues like ground loops.
 
Most designs do ground isolation / floating designs, to avoid common mode issues like ground loops.
  
That, the lower impedance, and the XLR/balanced may let you run longer lines / lower losses and less interference / avoid common mode issues
+
That, the lower impedance, and the XLR/balanced may let you run longer lines / lower losses and less interference / avoid common mode issues (...than you could with the cable type you connect on the DI's input).
(...than you could with the cable type you connect on the DI's input).
+
 
  
  
Line 966: Line 1,118:
 
* [[ground lift]] switch
 
* [[ground lift]] switch
  
* RCA input (to connect laptops or whatnot)
+
* RCA input (to connect laptops or whatnot, while avoiding ground issues)
  
 
* XLR input, in case you somehow have a ground-referenced XLR cable you need to avoid using the ground of. {{verify}}
 
* XLR input, in case you somehow have a ground-referenced XLR cable you need to avoid using the ground of. {{verify}}
  
 +
* a second TS socket, marked link or thru, wired directly to the first.
 +
: Meant to use the DI as a simple splitter, e.g. so you can send both dry via this DI, and feed a it through a few pedals as well. (Or maybe a cab, or stage monitor, or something else stage-side).
  
  
'''Electronically'''
 
  
'''Passive DIs''' are basically a roughly-12:1{{verify}} step-down audio transformer (used balun-style), so have an impedance ratio around 144:1.
+
'''Active versus passive'''
So a ~170kOhm source (such as an instrument pickup) will now plug into a 1.2kOhm mic input.
+
  
Also active sources like keyboards are fine with a passive (and you may need the DI for isolation).
+
Passive DIs are basically a roughly-12:1{{verify}} step-down (audio) transformer.
 +
So will have an impedance ratio around 144:1, so that a ~170kOhm source (such as an instrument pickup) will now plug into a 1.2kOhm mic input. They also step down the signal, which is why they are only useful for sources with stronger (usually powered and lower-impedance) outputs.
  
 +
Active DIs do the impedance conversion as described, but are also are phantom(/battery/AC)-powered pre-amps, so you won't have any issues with weak signals.
 +
These are primarily targeted at guitars, and may have some guitar-specific features.
  
Passive sources may have weak signals, and will have their sound colored by said transformer,
+
 
in which case an '''active DI''' (often powered from a battery or phantom power) is preferable.
+
Note that passive DIs are fine to use with instruments that actively power their output, like keyboards (and these may use the DI primarily for isolation).
 +
 
 +
Whereas passive instruments (in particular unamplified pickups) often have weak signals and high impedance, and would have their sound colored by the just-mentioned transformer, in which case an '''active DI''' (often powered from a battery or phantom power) is preferable.
 +
 
 +
In other words, active are required DIs for passive components (and will generally work for all), while passive DIs will work mainly for active instruments.
  
  
Line 991: Line 1,150:
 
<!--
 
<!--
  
Impedance matching is a wider concept, but in stage audio there are relatiely few things to match, and DI is already doing one of them, so these things refer to devices that:
+
Impedance matching is a wider concept, but in stage audio there are relatively few things to match, and DI is already doing one of them.
  
: Take a (possibly balanced) low-impedance microphone output
 
  
 +
...so 'impedance-matching adapters' often refer to devices that
 +
''roughly'' do the opposite of a DI:
 +
: Take a (possibly balanced) low-impedance microphone output
 
: Yield a (possibly unbalanced) high impedance input
 
: Yield a (possibly unbalanced) high impedance input
 
  
 
Mostly useful to plug a modern mic into a guitar amp or effect pedal,
 
Mostly useful to plug a modern mic into a guitar amp or effect pedal,
 
and in fact these things often come as a XLR-to-TS plugs.
 
and in fact these things often come as a XLR-to-TS plugs.
  
 +
Since this is usually a transformer (1:10ish?{{verify}}), it steps up voltage in the process{{verify}})
  
And yes, this is roughly the reverse of a DI, and you ''could'' sort of use a passive DI in reverse,
+
Could you use a passive DI in reverse? Sort of. It gives you ground isolation, but the output voltage is usually increased too much and will distort.
which does give you ground isolation, but the output voltage is usually increased too much and will distort.
+
  
  
Since this is usually a transformer (1:10ish?{{verify}}), it steps up voltage in the process{{verify}})
 
  
 
-->
 
-->
Line 1,012: Line 1,171:
 
=Digital logic voltage levels=
 
=Digital logic voltage levels=
  
[http://images.google.com/images?q=voltage+levels An image search like this] may be the simplest answer.
+
In the context of '''logic levels''': {{comment|(sorted from higher to lower voltage spans)}}
 
+
 
+
When, power-supply-wise, a distinctin is made between e.g. Vcc and Vdd, it's BJT and FET:
+
* V<sub>CC</sub> - positive supply, BJT
+
* V<sub>EE</sub> - negative supply, BJT
+
 
+
* V<sub>SS</sub> - negative supply, FET
+
* V<sub>DD</sub> - positive supply, FET
+
 
+
Which note, is referring to collector, emitter, source, and gate
+
 
+
V+ and V- is not specific.
+
 
+
https://en.wikipedia.org/wiki/IC_power-supply_pin
+
 
+
+
 
+
 
+
 
+
In the context of logic levels: {{comment|(sorted from higher to lower voltage spans)}}
+
 
* {{comment|(V<sub>CC</sub>)}}
 
* {{comment|(V<sub>CC</sub>)}}
 
* V<sub>OH</sub> - maximum output high
 
* V<sub>OH</sub> - maximum output high
Line 1,040: Line 1,179:
 
* V<sub>OL</sub> - minimum output low
 
* V<sub>OL</sub> - minimum output low
 
* {{comment|(Gnd)}}
 
* {{comment|(Gnd)}}
 
+
See also a few notes on margins below the summary
Usually/ideally, boolean levels should be in V<sub>CC</sub>-V<sub>OH</sub> for high, and V<sub>OL</sub>-Gnd for low.
+
 
+
 
+
The V<sub>IH</sub>-to-V<sub>OH</sub> difference is the '''high noise margin''' {{comment|(sometimes ''NM<sub>H</sub>'')}},
+
the V<sub>IL</sub>-to-V<sub>OL</sub> difference is the '''low noise margin''' {{comment|(sometimes ''NM<sub>L</sub>'')}}.
+
These two mean that noise (or voltage drop) of about this magnitude won't disturb the boolean interpretation.
+
 
+
 
+
The V<sub>IL</sub>-V<sub>IH</sub> interval is usually not defined as either logic level (devices ''could'' choose one or the other).
+
+
 
+
  
  
 +
'''Some voltage level systems'''
  
 +
[http://images.google.com/images?q=voltage+levels An image search like this] may be the simplest answer.
  
Some voltage level systems:
 
 
* '''TTL''': O to 5V. Boolean levels: 0V~0.8V should be low, and 2V to V<sub>cc</sub> should be high {{comment|(where V<sub>cc</sub> is ideally between 4.75V and 5.25V)}}. More specifically:
 
* '''TTL''': O to 5V. Boolean levels: 0V~0.8V should be low, and 2V to V<sub>cc</sub> should be high {{comment|(where V<sub>cc</sub> is ideally between 4.75V and 5.25V)}}. More specifically:
 
** V<sub>OL</sub>: 0.4V
 
** V<sub>OL</sub>: 0.4V
Line 1,064: Line 1,193:
 
** V<sub>CC</sub>: 5V
 
** V<sub>CC</sub>: 5V
  
* '''LVTTL''': 0 to 3.3V. Threshold levels identical to 5V TTL but V<sub>CC</sub> is 3.3V {{comment|(so only the V<sub>OH</sub> - V<sub>CC</sub> interval is smaller)}}
+
* '''LVTTL''': 0 to 3.3V. Threshold levels identical to 5V TTL. {{comment|(so only the V<sub>OH</sub> - V<sub>CC</sub> interval is smaller which generally affects nothing)}}
 
** V<sub>OL</sub>: 0.4V
 
** V<sub>OL</sub>: 0.4V
 
** V<sub>IL</sub>: 0.8V
 
** V<sub>IL</sub>: 0.8V
Line 1,071: Line 1,200:
 
** V<sub>CC</sub>: 3.3V
 
** V<sub>CC</sub>: 3.3V
  
* CMOS: various levels, levels mostly relative to V<sub>CC</sub>. Variants include: {{verify}}
+
* CMOS defines levels as a percentag of V<sub>CC</sub> (which can itself be 5, 2.5, 1.8, 1.5, 1.2V).  
** CMOS with V<sub>CC</sub>=5V
+
** V<sub>OL</sub>: 10% ()
** CMOS with V<sub>CC</sub>=2.5V
+
** V<sub>IL</sub>: 30%
** CMOS with V<sub>CC</sub>=1.8V
+
** V<sub>IH</sub>: 50%
** CMOS with V<sub>CC</sub>=1.5V
+
** V<sub>OH</sub>: 70%
** CMOS with V<sub>CC</sub>=1.2V
+
 
  
 
* '''ETL'''
 
* '''ETL'''
Line 1,091: Line 1,220:
  
 
* RS485, RS422
 
* RS485, RS422
 +
 +
 +
 +
'''Notes:'''
 +
 +
A given IC may deviate a little, so when in doubt check the datasheet.
 +
 +
 +
Usually/ideally, boolean levels should be in V<sub>CC</sub>-V<sub>OH</sub> for high, and V<sub>OL</sub>-Gnd for low.
 +
 +
 +
The V<sub>IH</sub>-to-V<sub>OH</sub> difference is the '''high noise margin''' {{comment|(sometimes ''NM<sub>H</sub>'')}},
 +
the V<sub>IL</sub>-to-V<sub>OL</sub> difference is the '''low noise margin''' {{comment|(sometimes ''NM<sub>L</sub>'')}}.
 +
These two mean that noise (or voltage drop) of about this magnitude won't disturb the boolean interpretation.
 +
 +
 +
The V<sub>IL</sub>-V<sub>IH</sub> interval is usually not defined as either logic level (devices ''could'' choose one or the other).
 +
 +
 +
 +
When, power-supply-wise, a distinctin is made between e.g. Vcc and Vdd, it's BJT and FET:
 +
* V<sub>CC</sub> - positive supply, BJT
 +
* V<sub>EE</sub> - negative supply, BJT
 +
 +
* V<sub>SS</sub> - negative supply, FET
 +
* V<sub>DD</sub> - positive supply, FET
 +
 +
Which note, is referring to collector, emitter, source, and gate
 +
 +
V+ and V- is not specific.
 +
 +
https://en.wikipedia.org/wiki/IC_power-supply_pin
 +
  
  

Revision as of 22:04, 31 July 2019

This is for beginners and very much by a beginner. It's meant to try to cover hobbyist needs, and as a starting point to find out which may be the relevant details for you, not for definitive information.

Some basics and reference: Volts, amps, energy, power · Ground · batteries · resistors · changing voltage · transistors · fuses · diodes · varistors · capacitors · inductors · transformers · baluns · amplifier notes · frequency generation · skin effect


And some more applied stuff:

IO: wired local IO wired local-ish IO · · · · Shorter-range wireless (IR, ISM RF, RFID) · bluetooth · 802.15 (including zigbee) · 802.11 (WiFi) · cell phone


Sensors: General sensor notes, voltage and current sensing · Knobs and dials · Pressure sensing · Temperature sensing · humidity sensing · Light sensing · Movement sensing · Capacitive sensing · Touch screen notes

Actuators: General actuator notes, circuit protection · Motors and servos · Solenoids

Some stuff I've messed with: Avrusb500v2 · GPS · Hilo GPRS · JY-MCU · DMX · Thermal printer

Audio notes: basic audio hacks · microphones · amps and speakers · device voltage and impedance, audio and otherwise ·

Less sorted: Common terms, useful basics, soldering · Microcontroller and computer platforms · Arduino and AVR notes · ESP series notes · Electronics notes/Phase Locked Loop notes · mounts, chip carriers, packages, connectors · signal reflection · pulse modulation · electricity and humans · Unsorted stuff


See also Category:Electronics.


Theory: Impedance when connecting two things

Output impedance is larger than the load's input impedance

Impedance matching

Impedance bridging

Impedance mismatches

Audio

Analog audio voltage levels

...and impedances.

This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, fix, or tell me)



Most audio levels are not so well standardized, and some have changed over time, somewhat quietly. So assume each can be a factor of two off and can require a little knob twiddling at least.


  • phono input
on the order of a milliVolt, and seems to often be the not-yet-amplified output of phono cartridges
There are two common types:
Moving Magnet (MM) pickups give ~2.5mV,
Moving Coil (MC) give ~0.2mV
MC versus MM is one of those debates. Higher-end is usually MC, but quality also significantly depends on other factors.
phono pre-amps will amplify this to (typically) consumer line level (and impedance)
phono directly on line-level (or mic level) input as-is will be very too quiet (or if you manage to amplify it, very noisy)
avoid connecting non-phono and in particular line-level outputs to phono inputs, it may be possible to blow the preamp
impedance
cartridge output: varies, order of 500 Ohm or lower (verify)
phono amp imput: 47k Ohm


  • consumer microphone level
on the order of ~10mV (verify)
because that's the order of what ~1cm electret (FET output(verify))s give, which is what most of these are
computer sound cards have
a voltage bias for electret mics - as electrets are the typical mic you plug in
internal amplication to get it to the same level as line level. (Relevant mainly in that plugging line level into mic in will distort)
impedance
mic output - high-impedance microphones are typically cheaper, e.g. the common electret mic is often 1-2kOhm but some 10kOhm+ (verify)
PC mic in impedance: 1..10kohm (varied over time and with cards)


  • consumer line level
on the order of ~300mV (~310mV RMS, ~440mV peak, 0.9V peak-to-peak)
also sometimes known as -10dBu (mostly in situations that also do +4dBV pro levels)
but has varied somewhat over time.
I've seen amplifiers with a sensitivity of 250mV, older ones with 150mV
Some recent devices are moving to higher voltages - amps may choose to deal with up to 1V, or 2V in the case of DVD, Bluray(verify) (perhaps in imitation of pro line level?)
when the the other end is not aware, you may need to attenuate the output, and/or keep the amplification low, to avoid distortion.
impedance
line in impedance is often ~100 Ohm. Possibly higher, up to 1 kOhm
line out impedance is often ~10 kOhm. Possibly higher, up to 1 MOhm


  • professional microphone level
order of 10mV, but can vary:
Can be ~1mV, can be ~200mV (in theory more but this is atypical)
more varied designs, and possible amplification at the mic, means more variation with design and per use
(e.g. dynamic mics are somewhat intentionally low, to be able to deal with the louder things)
...so you will need that gain knob
impedance:
mic output: most are (higher) in the 50..200 Ohm range, with deviations (see more notes around here)
mic preamp/mixer input: order of 1..2kOhm


  • professional sound line level
the standard also known as +4dBV means 1.2V RMS (1.7V peak, 3.4V peak-to-peak) (verify)
with some variants a little higher and lower, so think 1V order of magnitude


  • instrument level has no standard, though is generally quite predictable:
voltage is often somewhere between mic and line level
output impedance
pickup impedance is often quite high (see also notes below on pickup impedance)
input to a mixer will typically need a direct box (a.k.a. DI) to convert most things to typical XLR
output to guitar amps is unchanged - they expect high impedance from a directly connected pickup (DI boxes tend to have a thru on the input side so that you could do both)


Less standard / more varied:

  • headphone level
roughly commercial line level, but less of a standard - can be higher.
headphone amps tend to aim to power a ~30-60Ohm headphone (with a few milliamps(verify)
there are ~4Ohm headphones, but you really woudn't plug those into everything (likely to distort)
there are 250Ohm-600Ohm headphones, but these need their own preamp (the idea is that you can design for slightly better THD with less load on the amp)
  • Car audio tends to be on the order of 2V, sometimes 4V (verify)
so a headphone amp can be a good cheat to connect consumer-level things to this
  • consumer speaker wires
The voltages are proportional to the amplifier's/speaker/s ability (and relate to be).
For ~100W speakers you'll see up to a few dozen volts
a tiny desktop speaker may be <1W (verify)
impedance
speaker load is often 8 or 4 Ohm (sometimes 2, sometimes 16)
amplifier output impedance is typically very low, say 0.1 Ohm (this is also why the whole 'match your speaker impedance exactly to your amp impedance' thing is nonsense in a literal sense -- but with lower-impedance speakers you should limit how much you turn up the volume, because the maximum sensible power output happens earlier - and above that you get both distortion (THD increases with load) and risk of damage)
  • pro speaker wires
not really a thing. Most speakers are connected by one of:
XLR: carrying typical XLR line signals to active speakers
Speakon-connected: already-amplified signal to a passive speaker
6.3mm TS: already-amplified signal to a passive speaker (Sometimes avoided to avoid smoky mixups)
Note these cables are different from TS instrument cables, basically in that instrument cables use a thinner core-and-shield and these are beefer and not shielded (just 2-lead stranded(verify))


See also:

dBV and dBu

Audio device differences

Balanced audio / pro audio

Balanced in electrical terms
Some other terms you see

Plugs's relation to balanced/unbalanced, voltage levels, etc.

Connectorwise:

  • XLR3 is pro mic level, always balanced/differential, always mono.
mono, because one signal requires a differential pair.
if you want to carry stereo over XLR, use two cables (in practice, stereo is often about inter-device, and may well be two balanced 6.35mm TRS instead).
  • 6.35 mm is (typically) pro line level (6.35mm but people are lazy typers)
6.35mm TRS is balanced mono, see notes on that above.
or, sometimes, unbalanced stereo. This is an exception and will be noted.
6.35mm TR is unbalanced mono, often instruments, which is also often lower voltage levels (but close enough(verify))
when a device instead uses 6.35mm these for mic in, aux, or controllers like pedals, or stereo, they will be marked as such (or switchable)
  • mixer outputs are often two 6.35mm jacks (balanced, TRS)
  • RCA on a mixer are typically only used for phono in, or consumer in (aux)
so common mode. Ideally it's isolated (to avoid conductive ground loops)


On 6.35 TS versus TRS:

  • 6.35mm (1/4") TRS is most typically balanced/differential mono. A mixer input will often mark this as "balanced"
Tip and Ring is the pair, Sleeve is shield.
No, it's not stereo. And using a 3.5mm-to-6.35mm converter to plug in consumer line level will easily do weird things (depending on the case).
  • 6.35mm (1/4") TS - instrument cable
is mono, and not differential
from actual instruments it won't
Tip is signal, Sleeve is shield,
  • mixers tend to accept both TRS balanced and TS unbalanced. If on the same socket they usually mark it (e.g. "bal/unbal")
Note that unbalanced inputs are not always isolated, so connecting unbalanced things (other than floating instuments) could create common mode issues.
contrast with...
  • 6.35mm (1/4") TS - speaker cable - that is, amplifier-to-passive-speaker
basically a pair of thicker wires than instrument cable would use, and no shielding (it doesn't have to care at all because it's higher voltage, low-impedance load so coupling falls into nothing)



Things that won't work / things to avoid

On pickup impedance

Things that are't pure bridging

Other notes

On microphone impedance
This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, fix, or tell me)
On 600 Ohms, and impedance matching

DI

This article/section is a stub — probably a pile of half-sorted notes, is not well-checked so may have incorrect bits. (Feel free to ignore, fix, or tell me)

Direct box, DI box, DI unit, DI. (people argue over whether it stands for direct input, direct injection, direct induction, or direct interface).


Functionally

Takes a high-impedance, possibly-unbalanced signal (often roughly pro line level), e.g. from a passive instrument - probably most frequently electric guitars and electric basses.

Outputs a low-impedance pro-mic-level balanced signal, usually on an XLR plug.


In other words, usually plugs a high-impedance instrument into a (line that ends up at the) mixer.


Impedance-matching adapter / impedance-matching transformer / line matching transformer

Digital logic voltage levels

In the context of logic levels: (sorted from higher to lower voltage spans)

  • (VCC)
  • VOH - maximum output high
  • VIH - minimum input high
  • VT - theshold level, used in a few definitions, and applies to devices that transition at a given level(verify)
  • VIL - maximum input low
  • VOL - minimum output low
  • (Gnd)

See also a few notes on margins below the summary


Some voltage level systems

An image search like this may be the simplest answer.

  • TTL: O to 5V. Boolean levels: 0V~0.8V should be low, and 2V to Vcc should be high (where Vcc is ideally between 4.75V and 5.25V). More specifically:
    • VOL: 0.4V
    • VIL: 0.8V
    • VIH: 2V
    • VOH: 2.4V
    • VCC: 5V
  • LVTTL: 0 to 3.3V. Threshold levels identical to 5V TTL. (so only the VOH - VCC interval is smaller which generally affects nothing)
    • VOL: 0.4V
    • VIL: 0.8V
    • VIH: 2V
    • VOH: 2.4V
    • VCC: 3.3V
  • CMOS defines levels as a percentag of VCC (which can itself be 5, 2.5, 1.8, 1.5, 1.2V).
    • VOL: 10% ()
    • VIL: 30%
    • VIH: 50%
    • VOH: 70%


  • ETL
  • BTL
  • LVDS
  • PECL
  • RS232
  • RS485, RS422


Notes:

A given IC may deviate a little, so when in doubt check the datasheet.


Usually/ideally, boolean levels should be in VCC-VOH for high, and VOL-Gnd for low.


The VIH-to-VOH difference is the high noise margin (sometimes NMH), the VIL-to-VOL difference is the low noise margin (sometimes NML). These two mean that noise (or voltage drop) of about this magnitude won't disturb the boolean interpretation.


The VIL-VIH interval is usually not defined as either logic level (devices could choose one or the other).


When, power-supply-wise, a distinctin is made between e.g. Vcc and Vdd, it's BJT and FET:

  • VCC - positive supply, BJT
  • VEE - negative supply, BJT
  • VSS - negative supply, FET
  • VDD - positive supply, FET

Which note, is referring to collector, emitter, source, and gate

V+ and V- is not specific.

https://en.wikipedia.org/wiki/IC_power-supply_pin



See also: