Active Speakers, Cable Runs & Passive Pre-Amps

[NicS]

I have a dilemma...

I'd like to use a passive Pre-Amp to send an attenuated signal from my phono preamp to my active speakers. The intention is bypass my Emotiva AV processor altogether with the cleanest fully balanced signal path from my phono cartridge to the speakers. My dilemma has to do with impedance. I'd like to properly impedance match the Passive Pre to all the other components in the system. Hopefully getting the least "colored" transfer of sound. This is the intended chain...

Musical Fidelity MX VYNL - 8V RMS output via XLR >
Mogami W2534 2' with Neutrik XLR plugs >
Goldpoint SA2X-I or Khozmo >
Mogami W2534 25' with Neutrik XLR plugs >
ATC SCM100ASL active speakers (these have an input sensitivity of 1v and an input impedance of >10k Ohm via balanced XLR)

With the Khozmo I have the option of 10K, 20K, 50K, 100K & 200K versions.

First is any of this a good idea, given that my signal run to the active speakers is 25'. And if it is, what impedance value for the passive pre-amp should I opt for? Is there any information missing that I need to get to figure this out.

Help very much appreciated. I've hit the end stops of my current knowledge.

 

[DVDdoug]

10K is probably best. I assume you can ask Khozmo what they recommend. It's generally better it the source (preamp) has a lower impedance than the load (speaker input) but you wouldn't want a lower impedance on the preamp input "loading" whatever is plugged into it. 10K is sort-of the traditional minimum for a line-level input.

There's really nothing wrong with using an active preamp. Especially for line-level signals, it's easy to make it transparent. That solves the impedance problem by having high-impedance inputs and low-impedance outputs driving another high-impedance input. The (active) preamp is more for impedance than actual amplification... Like the passive preamp most of the time it's being used as an attenuator rather than an amplifier.

The low-impedance output means any electrical noise getting into the cables "sees" a low impedance making the connection more immune to noise. But that's usually not a problem either, especially with shielded cable and balanced connections.


...My biggest concern would be the surface noise on the records and other potential or occasional vinyl limitations.

 

[Speedskater]

How would a passive phono pre-amp work?

 

[NicS]

How would a passive phono pre-amp work?

I don't think you could make a passive phono pre-amp viable. You need to increase gain on such a weak original signal.

 

[LTig]

I have a dilemma...

I'd like to use a passive Pre-Amp to send an attenuated signal from my phono preamp to my active speakers. The intention is bypass my Emotiva AV processor altogether with the cleanest fully balanced signal path from my phono cartridge to the speakers. My dilemma has to do with impedance. I'd like to properly impedance match the Passive Pre to all the other components in the system. Hopefully getting the least "colored" transfer of sound. This is the intended chain...

Musical Fidelity MX VYNL - 8V RMS output via XLR >
Mogami W2534 2' with Neutrik XLR plugs >
Goldpoint SA2X-I or Khozmo >
Mogami W2534 25' with Neutrik XLR plugs >
ATC SCM100ASL active speakers (these have an input sensitivity of 1v and an input impedance of >10k Ohm via balanced XLR)

With the Khozmo I have the option of 10K, 20K, 50K, 100K & 200K versions.

First is any of this a good idea, given that my signal run to the active speakers is 25'. And if it is, what impedance value for the passive pre-amp should I opt for? Is there any information missing that I need to get to figure this out.

Such a long connection could have too high capacitance. Together with the rather high output impedance of the passive preamp (the volume pot) forms a low pass. Deoending on the position of the volume pot a loss of treble could be audible. Hence a lower impedance for the passive preamp might be required as long as the phono pre can handle it.

 

[NicS]

Such a long connection could have too high capacitance. Together with the rather high output impedance of the passive preamp (the volume pot) forms a low pass. Deoending on the position of the volume pot a loss of treble could be audible. Hence a lower impedance for the passive preamp might be required as long as the phono pre can handle it.

This is precisely my concern.

The ATC speakers are designed to be run with XLR cables of lengths around the 20' mark. The speakers themselves recommend input impedance of no less than 10K (shown as >10K). My gut reaction is to buy a passive pre-amp and try it out.

How do I measure the cumulative impedance of the cables and the Passive-Preamp if I do?

 

[LTig]

This is precisely my concern.

The ATC speakers are designed to be run with XLR cables of lengths around the 20' mark. The speakers themselves recommend input impedance of no less than 10K (shown as >10K). My gut reaction is to buy a passive pre-amp and try it out.

It's not the input impedance of the speaker which matters, but the output impedance of the preamp. With a passive one it depends on the position of the volume pot but the worst case is half of the nominal resistance of the pot, that is 5k for a 10k pot.

How do I measure the cumulative impedance of the cables and the Passive-Preamp if I do?

With an LCR meter or a dedicated L meter. You can also ask the maker of the cable. Usually they state something like xx pF per length and you multiplicate it with the length to get the resulting capacitance.

 

[Speedskater]

I don't think you could make a passive phono pre-amp viable. You need to increase gain on such a weak original signal.

That was my point.

 

[NicS]

That was my point.

My phono stage has an output of 8v and increases gain adequately to the listening volume I need. What I'm trying to figure out is how to impedance match a passive pre-amp if used to attenuate the phono stage connected directly to my active speakers.

 

[MRC01]

It's not the input impedance of the speaker which matters, but the output impedance of the preamp. With a passive one it depends on the position of the volume pot but the worst case is half of the nominal resistance of the pot, that is 5k for a 10k pot.
...

The output impedance of a passive is actually 1/4 the nominal impedance - 2500 for 10k. This worst case highest output impedance is at -6 dB where both resistors are equal. With 20' of normal RCA cable, the RC bandwidth should be well above 100 kHz, perfectly flat in the audio spectrum. Unless the cables are some kind of wacky high capacitance type.
More on the subject here.

 

[NicS]

The output impedance of a passive is actually 1/4 the nominal impedance - 2500 for 10k. This worst case highest output impedance is at -6 dB where both resistors are equal. With 20' of normal RCA cable, the RC bandwidth should be well above 100 kHz, perfectly flat in the audio spectrum. Unless the cables are some kind of wacky high capacitance type.
More on the subject here.

Very interesting...

So considering this an all balanced setup, with Mogami W2534 cable terminated in Neutrik plugs, with a choice of 10K, 20K, 50K, 100K & 200K impedance values on the passive pre-amp, which would you recommend?

 

[NicS]

Very interesting...

So considering this an all balanced setup, with Mogami W2534 cable terminated in Neutrik plugs, with a choice of 10K, 20K, 50K, 100K & 200K impedance values on the passive pre-amp, which would you recommend?

I have plenty enough gain from the phono stage BTW. I just need to attenuate the volume it by around 50% to prevent the wife from killing me.

 

[MRC01]

Very interesting...

So considering this an all balanced setup, with Mogami W2534 cable terminated in Neutrik plugs, with a choice of 10K, 20K, 50K, 100K & 200K impedance values on the passive pre-amp, which would you recommend?

With a passive, use the lowest impedance that gives you at least 10:1 impedance ratio on both sides. For example, if all your sources have 500 ohm or less output impedance, use a 5k attenuator. That 5k attenuator has a worst-case highest output impedance of 1250 ohms, so your target device should have at least 12.5 kOhm input impedance.

With a passive, 5k or 10k are most common and it's unusual to need anything higher. But check the specs of your equipment to be sure.

I'm a fan of passives, having designed, built and used them. But to be fair I must give the pragmatic caveat: of course, the alternative is to go with a standard active preamp. They buffer your devices from each other, obviating the need to match impedances. And good ones are almost as clean (low noise & distortion) as passives. Even if an active preamp isn't quite as clean as a passive, it is usually cleaner than anything else in the signal chain, in which case the marginal improvement of a passive won't matter.

 

[LTig]

The output impedance of a passive is actually 1/4 the nominal impedance - 2500 for 10k. This worst case highest output impedance is at -6 dB where both resistors are equal.

Yep, you're right. Thanks!

 

[NicS]

With a passive, use the lowest impedance that gives you at least 10:1 impedance ratio on both sides. For example, if all your sources have 500 ohm or less output impedance, use a 5k attenuator. That 5k attenuator has a worst-case highest output impedance of 1250 ohms, so your target device should have at least 12.5 kOhm input impedance.

With a passive, 5k or 10k are most common and it's unusual to need anything higher. But check the specs of your equipment to be sure.

I'm a fan of passives, having designed, built and used them. But to be fair I must give the pragmatic caveat: of course, the alternative is to go with a standard active preamp. They buffer your devices from each other, obviating the need to match impedances. And good ones are almost as clean (low noise & distortion) as passives. Even if an active preamp isn't quite as clean as a passive, it is usually cleaner than anything else in the signal chain, in which case the marginal improvement of a passive won't matter.

Great advice. Thanks. I'm trying to bypass my AV processor which I don't think does my system justing in stereo mode.

 

[Waxx]

I think you need a buffer preamp, not a passive.

In my experience, long cable runs from a passive preamp are causing a lowpass filter on the signal. I tried passive preamps with 2 systems, an diy one with a 10K pot and a (passive) transformer balanced xlr out, that was ok, and a Tisbury Mini Passive. Both work well when the cable to the amplifier is very short, but both have problems with cables over 2m I used QED Profile RCA cables and Neutrik/Mogami Gold based diy cables.

A buffer preamp has (mostly) an opamp in unity gain (no gain), but provides a static lower impendance to the amp, and so has no lowpass filter effect. A transformator can also do that, but it's way more expensive (for a good transfo with low enough output impendance, what was my fault with the diy one) with no real quality gain. The same can be done with a transistor or a tube btw, it's just that a low noise opamp is perfect for that and much cheaper/easier to do. And even the very oldskool and cheap NE5534 opamp is already good enough for that actually, no need for relative expensive newer opamps (it does not make a difference in reality when done right). But even the expensive opamps is only a few €/$ a piece, so...

 

[MRC01]

In my experience, long cable runs from a passive preamp are causing a lowpass filter on the signal. ... A buffer preamp has (mostly) an opamp in unity gain (no gain), but provides a static lower impendance to the amp, and so has no lowpass filter effect. ...

This is true, but the devil's in the details. That lowpass filter is an RC circuit whose bandwidth (-3 dB point) is 1 / (2*pi*R*C). If you do the math you find it is unlikely to be anywhere near audio frequencies in most normal systems. With a 10k passive attenuator, the worst-case R is 2500 so you need 1.6e-9 Farads of capacitance for the bandwidth (-3 dB point) to be 40 kHz, and even that is an octave above 20k. Typical Some RCA cables have around 15-20 pF per foot, so that would require about 100 feet of cable. And that wouldn't affect audio frequencies - the cable would have to be even longer.

PS: if you had a really bad high capacitance RCA cable at 100 pF per foot, then your max safe cable length would be 16 feet, putting -3 dB @ 40 kHz.
By comparison, Blue Jeans LC-1 cable is 12.2 pF per foot.

 

[MRC01]

... it's just that a low noise opamp is perfect for that and much cheaper/easier to do. And even the very oldskool and cheap NE5534 opamp is already good enough for that actually, no need for relative expensive newer opamps (it does not make a difference in reality when done right). But even the expensive opamps is only a few €/$ a piece, so...

The opamps are an insignificant cost of a good preamp. The real cost is getting the design & layout right to minimize power supply noise, shielding, the volume/gain design, etc. Good preamps are not cheap for these reasons. The same can be said of passive attenuators. You can buy 1% metal film resistors for what, 10 cents each? But a good well designed passive can cost hundreds of dollars to get the proper grounding, high quality switches, etc. Either way, the device can only be as good as its design and implementation.

 

[NicS]

I think you need a buffer preamp, not a passive.

In my experience, long cable runs from a passive preamp are causing a lowpass filter on the signal. I tried passive preamps with 2 systems, an diy one with a 10K pot and a (passive) transformer balanced xlr out, that was ok, and a Tisbury Mini Passive. Both work well when the cable to the amplifier is very short, but both have problems with cables over 2m I used QED Profile RCA cables and Neutrik/Mogami Gold based diy cables.

A buffer preamp has (mostly) an opamp in unity gain (no gain), but provides a static lower impendance to the amp, and so has no lowpass filter effect. A transformator can also do that, but it's way more expensive (for a good transfo with low enough output impendance, what was my fault with the diy one) with no real quality gain. The same can be done with a transistor or a tube btw, it's just that a low noise opamp is perfect for that and much cheaper/easier to do. And even the very oldskool and cheap NE5534 opamp is already good enough for that actually, no need for relative expensive newer opamps (it does not make a difference in reality when done right). But even the expensive opamps is only a few €/$ a piece, so...

Define "long".

According to ATC, the manufactuere of my active speakers, 25' isn't considered long,

I think you need a buffer preamp, not a passive.

In my experience, long cable runs from a passive preamp are causing a lowpass filter on the signal. I tried passive preamps with 2 systems, an diy one with a 10K pot and a (passive) transformer balanced xlr out, that was ok, and a Tisbury Mini Passive. Both work well when the cable to the amplifier is very short, but both have problems with cables over 2m I used QED Profile RCA cables and Neutrik/Mogami Gold based diy cables.

A buffer preamp has (mostly) an opamp in unity gain (no gain), but provides a static lower impendance to the amp, and so has no lowpass filter effect. A transformator can also do that, but it's way more expensive (for a good transfo with low enough output impendance, what was my fault with the diy one) with no real quality gain. The same can be done with a transistor or a tube btw, it's just that a low noise opamp is perfect for that and much cheaper/easier to do. And even the very oldskool and cheap NE5534 opamp is already good enough for that actually, no need for relative expensive newer opamps (it does not make a difference in reality when done right). But even the expensive opamps is only a few €/$ a piece, so...

I'm fully balanced through the entire signal path, from the cartridge to the speakers. We are't talk about the inherent issues with RCA; my issue is with the specifications of a passive pre-amp when used with active studio mid-field monitors which expect a load on the end of balanced signal cables that are around 20' long,

 

[NicS]

With a passive, use the lowest impedance that gives you at least 10:1 impedance ratio on both sides. For example, if all your sources have 500 ohm or less output impedance, use a 5k attenuator. That 5k attenuator has a worst-case highest output impedance of 1250 ohms, so your target device should have at least 12.5 kOhm input impedance.

With a passive, 5k or 10k are most common and it's unusual to need anything higher. But check the specs of your equipment to be sure.

I'm a fan of passives, having designed, built and used them. But to be fair I must give the pragmatic caveat: of course, the alternative is to go with a standard active preamp. They buffer your devices from each other, obviating the need to match impedances. And good ones are almost as clean (low noise & distortion) as passives. Even if an active preamp isn't quite as clean as a passive, it is usually cleaner than anything else in the signal chain, in which case the marginal improvement of a passive won't matter.

This where my knowledge hits a wall.

The phono stage is a Musical Fidelity MX VNYL:

Musical Fidelity | MX-VYNL Phono Stage

MX-VYNL | Phono Stage The MX-VYNL is a Fully balanced, configurable, high performance phono stage. Find a...

www.musicalfidelity.com

The cables are 25' Mogami W2534, which I assume add to the impedance of the MX VNYL:

MOGAMI® - Neglex Quad Cables

Mogami Neglex Quad Cables are perfect for home studios suffering from wiring and grounding problems.

mogamicable.com

The speakers are ATC monitors which are expecting an impedance of >10K.

What would you say was the ideal impedance of the passive pre-amp sitting between the speakers and phono stage?