Need help understanding "D.I out", XLR out, etc.

[jokerjkny]

nothing more to add, but to say, great thread!

 

[tadawson]

Yes, I too experience inconsistencies with engineers - some want more, some less and I just give them what they ask for. The thing that really boggles me is how a signal can have a known impedance, like ~100 K ohms. Impedance as I understand it, is simply resistance to alternating current (the signal). So, impedance relative to what? I understand how a signal can have frequency(s) and amplitude, but the impedance part is confusing. Beyond the scope of this thread for sure.

Thanks - Skel

A signal does NOT have an impedance - the hardware that is either generating or receiving it does . . . . it's a hardware spec!

- Tim

 

[dave_bass5]

A primary reason for the "idea" of having a level control at the DI can be found in your post.

The part where you say "I thought the DI goes into the mixer at mic or instrument level...."

Well because there are so many variables (in addition to "line level") on any given mixer plus the myriad of ways cabling, patch bays, etc. can influence the issue you additionally must realize that it varies from brand to brand among mixers, consoles, PAs.

So someone thought that having a level control on your DI is a convenience to you because its "normal" level may be too hot for any number of inputs on any number of mixers, recorders, PAs, foldback, etc.

However, in practice you may discover that you need it wide open all the time with anything you connect it to. You may discover that you can't turn it up enough. Case-by-case basis.

While I'm not familiar with the DI in your specific amp I will add that I usually prefer a separate, dedicated DI box like a Radial JDI. I say that because I've never met an itegrated DI that I liked mainly due to level weakness or hum. But I haven't tried all that many either.

I can see the point in that i can control the level but when i say mic or instrument level that is switched at the desk (well, the ones i have used) and not on the amps DI
I would think wide open all the time is the best way to use it but that makes having the level control a bit pointless. i would also guess that wide open is the same as a DI box without a level control and i have used 4 different amps with DI out and all have been fine with the level they produce.
One thing has just struck me though. with ym SA450 i can have the DI pre or post EQ. i guess if i was using post Eq with a lot of low end boost then turning the DI level down a bit might give a cleaner signal at the desk. just a thought though.

 

[Vic]

I can see the point in that i can control the level but when i say mic or instrument level that is switched at the desk (well, the ones i have used) and not on the amps DI
I would think wide open all the time is the best way to use it but that makes having the level control a bit pointless. i would also guess that wide open is the same as a DI box without a level control and i have used 4 different amps with DI out and all have been fine with the level they produce.
One thing has just struck me though. with ym SA450 i can have the DI pre or post EQ. i guess if i was using post Eq with a lot of low end boost then turning the DI level down a bit might give a cleaner signal at the desk. just a thought though.

It depends. AFAIK, not all power amps and/or boards are created equal with input impedence, yet they behave best when fed a fairly specific input level.

So, if a single DI box is to mate with multiple devices optimally, and if its output impedence is fixed (never seen one that wasn't), the only way to compensate for a variable load to the DI and maintain a constant output level, is to have a variable drive for the signal coming out of the DI.

...IMHO...

 

[KingOfAmps]

I can see the point in that i can control the level but when i say mic or instrument level that is switched at the desk (well, the ones i have used) and not on the amps DI
I would think wide open all the time is the best way to use it but that makes having the level control a bit pointless.

Not necessarily. There are some mixer/recorder inputs out there that are very sensitive to much "level". So if the DI has no level control (like yours does have) then one must resort to some kind of inline pad/attenuator. Otherwise the input of the next device in the chain will be overloaded and that device won't be able to do anything about it.

The best example I can think of right now is the result you get when trying to run a +4 level into -10. Though it varies from piece to piece it's difficult to turn down the input of the -10 device enough to avoid distortion. And if you are, it usually chokes the tone.

i would also guess that wide open is the same as a DI box without a level control and i have used 4 different amps with DI out and all have been fine with the level they produce.

Depends on the DI box. They can vary. Most of them try to output a balanced, 600 ohm, +4 signal that optimally would go into a "mic" input.

 

[KingOfAmps]

OUTPUTS
EFFECT SEND
unbalanced, max. voltage 12 Vpp (pre-EQ)
TUNER OUT
unbalanced, max. voltage 2 Vpp
LINE OUT
balanced XLR, max. voltage 25 Vpp
SPEAKER OUT
speakon, 1/4"
---------------------------------


Is this what your ouputs are like? If so, and at the risk of confusion I'll add:

They're not calling it a "DI". That doesn't mean it's not though. To me, "line-out" implies line level which is different than mic level.

Do you have the specs. on what level this output operates? And what kind of load it wants to see?

 

[KJung]

Per the above discussion, I can't imagine being in a situation with a mixing board that does not have an input pad that allows for an almost infinite range of input levels. I've never quite understood the need for the output level controls on the DI's of amps. I assumed they were for the player to be able to adjust the 'direct' volume from stage (which is a sound person's nightmare!). This is the reason that Mesa Boogie states for putting the DI volume control on the front of some of their bass amps.

 

[Vic]

Per the above discussion, I can't imagine being in a situation with a mixing board that does not have an input pad that allows for an almost infinite range of input levels. I've never quite understood the need for the output level controls on the DI's of amps. I assumed they were for the player to be able to adjust the 'direct' volume from stage (which is a sound person's nightmare!). This is the reason that Mesa Boogie states for putting the DI volume control on the front of some of their bass amps.

I've heard soundmen say they'd rather have a lower level going in vs having a hotter signal they have to pad more. I think the idea is the tone is better?

A BIG +1 on the player affecting FOH levels!!!!!! HUGE no-no. It should TOTALLY be set/forget at sound check and let the production folks do what they're paid to do from then on.

In fact, that's why I HATED the Mesa Walkabout DI layout where you could NOT adjust EITHER the pre OR post gains withOUT affecting FOH. I know technically why they did it, to preserve tone based on their design, but it basically made the DI unusable to me. I REFUSE to mess with FOH sound. REFUSE. Funny you should bring them up.

 

[Bob Lee (QSC)]

I've heard soundmen say they'd rather have a lower level going in vs having a hotter signal they have to pad more. I think the idea is the tone is better?

The input pad on a board shouldn't be so poorly designed that it affects the tone.

As Tim mentioned, signals don't have impedance, so you can't have a "600Ω signal." The source could have an output impedance of 600Ω, though, so ideally it should feed an input impedance of 6 kΩ or higher for optimum signal voltage transfer.

"Balanced" means that both signal-carrying conductors have equal impedances to ground, so that any induced noise will be equal on both legs, and therefore will cancel out in the balanced/differential input.

 

[Passinwind]

I've heard soundmen say they'd rather have a lower level going in vs having a hotter signal they have to pad more. I think the idea is the tone is better?

The primary idea is to reduce crosstalk between channels when a long snake run is involved. Take a look at the Jensen website
for a nice overview on DI design considerations. The "standard" Jensen DIs actually attenuate rather than step up the input signal, by a substantial margin.

 

[Vic]

The input pad on a board shouldn't be so poorly designed that it affects the tone.

Agreed, but from what I understand, this isn't always the case, and it's always safe to just feed it its favorite level right off the bat.

As Tim mentioned, signals don't have impedance..

I assume this is in re-reply to the other poster, as it's not relevant to what I was saying or anything I was at all confused about.

AFAIK...

In a matched impedence circuit, the only reason to have an adjustable DI out to the board is if some boards (or whatever you're plugging into) are designed to operate best at different input levels from others, and don't have well designed pads or gain stages at the input compared to the DI.

In a MIS-matched impedence circuit, the level will be affected, and also be a good reason to compensate either by an adjustable DI and/or a well-designed pad or gain stage (depending on the mismatch) at the device receiving the signal.

That's what I was getting at when I brough impedence into the picture. I was hoping that wasn't going to obscure my post, but thought it might.

 

[Vic]

The primary idea is to reduce crosstalk between channels when a long snake run is involved. Take a look at the Jensen website
for a nice overview on DI design considerations. The "standard" Jensen DIs actually attenuate rather than step up the input signal, by a substantial margin.

This is interesting. I thought about this, but then thought the opposite based on transmission line theory. In other words, the power company cranks up the voltage for less loss on longer runs and then steps it back down for the short local runs, so I woulda' figured the same for audio, but apparently somehow the case is opposite for audio signalling. I'll read the web site. Thanks!

I understand about impedence and level matching (in a perfect case where the cord connecting them has no characterstic, or at least vastly insignificant impedence), but haven't dug into the transmission line part of it (which would be the series impedence in the circuit vs the parallel impedence at each end) where the cord does have significant impedence.

 

[Vic]

The primary idea is to reduce crosstalk between channels when a long snake run is involved. Take a look at the Jensen website
for a nice overview on DI design considerations. The "standard" Jensen DIs actually attenuate rather than step up the input signal, by a substantial margin.

Ok, I looked and I think we're talking about two different things.

The attenuation they're talking about is to go from unbalanced to balanced (which is internal to the DI), but I'm talking strictly of the balanced output side of the DI to the balanced input side at the board... basically, the snake wire.

AFAIK, we're talking about DI's with level adjusters, so the discussion is why you'd have a DI OUTPUT with adjustable levels, so strictly XLR to XLR "balanced" circuit.

Ref: an003.pdf from jensen-transformers.com:

"Jensen AN-003
INTERCONNECTION OF BALANCED AND UNBALANCED EQUIPMENT
by Bill Whitlock"​

 

[Bob Lee (QSC)]

Agreed, but from what I understand, this isn't always the case, and it's always safe to just feed it its favorite level right off the bat.

Maybe for low-quality boards.

In a matched impedence circuit, the only reason to have an adjustable DI out to the board is if some boards (or whatever you're plugging into) are designed to operate best at different input levels from others, and don't have well designed pads or gain stages at the input compared to the DI.

In a MIS-matched impedence circuit, the level will be affected, and also be a good reason to compensate either by an adjustable DI and/or a well-designed pad or gain stage (depending on the mismatch) at the device receiving the signal.

That's what I was getting at when I brough impedence into the picture. I was hoping that wasn't going to obscure my post, but thought it might.

There is no need to "match" impedances with modern gear, as practically everything is based on voltage transfer. Therefore, interconnections are based on the concept of a bridging interface, in which the load (input) impedance is, ideally, at least 10× the source (output) impedance.

 

[Vic]

Maybe for low-quality boards.

Understood, but is this case super rare these days? I honestly don't know.

I'm just making the (potentially blind) assumption there's at least a halfway reasonable case for having adjustable DI outs... no?

There is no need to "match" impedances with modern gear, as practically everything is based on voltage transfer. Therefore, interconnections are based on the concept of a bridging interface, in which the load (input) impedance is, ideally, at least 10× the source (output) impedance.

Ok, I can totally understand this. Didn't know this was the case. Makes good sense tho, of course.

So is the whole idea of impedence matching pretty much universally deprecated then?

I really appreciate contributions from industry experts. Very cool! Thanks Bob!

 

[Vic]

In a moment of clarity, I think Bob's comments also helped me realize some other things...

With these connections, we're not dealing with power transfer, but instead, only voltage transfer, which is why impedence matching is not relevant, and also why the transmission line issue is a non-issue (because the 10x or greater input impedence at the receiving end nullifies it).

Corrections welcome if I still missed something, tho.

 

[dave_bass5]

OUTPUTS
EFFECT SEND
unbalanced, max. voltage 12 Vpp (pre-EQ)
TUNER OUT
unbalanced, max. voltage 2 Vpp
LINE OUT
balanced XLR, max. voltage 25 Vpp
SPEAKER OUT
speakon, 1/4"
---------------------------------


Is this what your ouputs are like? If so, and at the risk of confusion I'll add:

They're not calling it a "DI". That doesn't mean it's not though. To me, "line-out" implies line level which is different than mic level.

Do you have the specs. on what level this output operates? And what kind of load it wants to see?

Actually your right. it doesnt say DI, just line out. but i dont think i have seen many amps with a balanced XLR out that worked at line level rather than mic level (but im not that experienced)

I would agree with Ken though. i have never had a problem with sending a signal to a desk. they have always been able to cope with my signals from my MAG, ABM, two aphex pedals (both have a balanced DI built in).

I guess i will use it at some point though.

 

[Passinwind]

Ok, I looked and I think we're talking about two different things.

The attenuation they're talking about is to go from unbalanced to balanced (which is internal to the DI), but I'm talking strictly of the balanced output side of the DI to the balanced input side at the board... basically, the snake wire.

AFAIK, we're talking about DI's with level adjusters, so the discussion is why you'd have a DI OUTPUT with adjustable levels, so strictly XLR to XLR "balanced" circuit.

The same case holds whether you start with a a balanced signal or unbalanced. There's at least one Jensen DI schematic with switchable attenuation on the XLR output side, as well as on the input end. A level control on an amp's DI would have the same function. But I'm guessing that the real reason that level control is there is often for "non-DI" applications like driving a power amp.

Another consideration is that not all small mixers have an inline pad. Some of the smaller Mackies don't, for instance. That's caused me plenty of headaches over the years (although building pads is quite simple), and a level control at the amp end is a perfectly workable solution.

A hot signal from the bass amp can be useful with those small boards if you run out of mike channels and need to use one of the line input channels too.

 

[Vic]

Another consideration is that not all small mixers have an inline pad. Some of the smaller Mackies don't, for instance. That's caused me plenty of headaches over the years (although building pads is quite simple), and a level control at the amp end is a perfectly workable solution.

A hot signal from the bass amp can be useful with those small boards if you run out of mike channels and need to use one of the line input channels too.

Ahhh, good stuff... I figured there had to be a reason... even if not the most typical case...

Thanks dude.

 

[Bob Lee (QSC)]

Input pads are just resitive attenuators, so there's little difficulty in making them flat in or out of the circuit.

One application that doesn't always fit the 10:1 load-to-source Z ratio is dynamic mics into mixing boards. They're often lower because of the tradeoffs involving reducing noise with small signals. Mic inputs are often as low as 2 kΩ, so if the mic's Z is ~400Ω, that would be only a 5:1 ratio.

IMHO, having the versatility of adjusting the DI out, from mic up to line level, could be very useful at times.

Anytime you can read something by Bill Whitlock, it is time well-spent. He's truly one of the wise men of audio.