passive v. active preamps, could use some input

mgalusha · « **Reply #40 on:** 26 Mar 2004, 12:51 am »

Quote from: maxwalrath

That's not the worst of it though. Sometimes when I start the Technics up there are bad pops or thumps coming from the speakers each time there is an action by the player. When the player first reads a disc or when I hit play, there's a pop/thump. It's when there's just about any action. After that there won't be any music coming from the system exept the fuzz sound, except after the pops the fuzz is a little worse than before. The weird thing is it seems to be completely random as to when the Technics decides to play discs without problems, and when it makes me think I'm blowing out my speakers.

Max,

What is likely happening is the Technics is controlling the volume digitally before it gets passed to the analog output section. The analog section is probably running at full gain and when you initiate an action on the player it generally mutes the output for a second. Normally this would be attenuated through a preamp but since it's driving your amps directly, any noise in the analog stage will be at line level.

One thing to keep in mind when using the CDP's volume control, some (many) digital volume controls in CD players change the volume by dropping bits. In other words, at low volumes your 16 bit CD might only be 12 bits (or something less than 16). This is one reason some folks don't like the sound of a CDP connected directly to the amps.

On the original topic, there is one other type of preamp and that's a buffered passive. It's not technically passive since there is a buffer (tube or solid state) but they are not really active as because there is no gain. These can eliminate the impedance matching problems and provide an easy load for the source. The downside is the fact that if the source doesn't provide enough voltage to drive the amp to full power this won't help. A solid state example here and a tube based unit here. I think the IRD purist can run this way as well.

just another nickel's worth.

mike g

Oxia · « **Reply #41 on:** 26 Mar 2004, 04:09 pm »

Quote from: mgalusha

there is one other type of preamp and that's a buffered passive

Ah, the active linestage with unity gain. In practice, reducing the gain on an active preamp to would yeild an increase in S/N ratio simply because you're not amplifying noise. So if you reduce the gain on a preamp from, say 6db to 0db, your S/N ratio would increase by 6db. This would bring you closer to the ideal (by doing the least harm to the signal), while retaining the key advantage of an active pre (having a non-fussy and stable output impedance).

From a personal point of view, I use an active preamp with selectable gain. At its default gain setting of 6db it has an S/N ratio of 120db, which is pretty darn quiet -- certainly better than most sources, so I know it won't be the limiting factor in my system. While I could set it to unity gain, the improvement in S/N (given its already exceedingly low noise floor) would be a largely academic excercise. Yes, a S/N ratio of 126 db would look better on paper, but my ears tell me that the difference is indistinguishable. From a pure useability standpoint, I also fear that it would leave me with too little gain for some recordings, so I'm content to leave things as they are. In my mind, a well designed active pre gives you the best of all worlds: low noise, transparency, neutrality, and no electrical compatability issues to worry about.

_scotty_ · « **Reply #42 on:** 26 Mar 2004, 07:52 pm »

Oxia,I use a DACT pot ahead of a discrete single ended class A mosfet buffer. Better than a
30Mhz bandwidth,.003 THD,100dbSN broadband,25ohm output impedance,one active part in the signal path,and three passive parts,two resistors and a coupling cap. Zero gain and no negative feedback involved. It works for me. It helps that my speakers are 96db efficient however. My phonostage has just enough gain to work with the rest of the system.

Oxia · « **Reply #43 on:** 26 Mar 2004, 08:38 pm »

_scotty_,
Wow, 30MHz?! That's some incredible bandwidth. Even a Spectral tops out around 3MHz. I also love the description of its design -- simple, elegant. If you don't mind divulging, how much did Stan Warren charge for that example of his handiwork?

As an aside, regarding gain, most of the time 6db is more than enough for me. However, on certain recordings (mostly classical works) I have to turn up the volume knob to 12 o'clock in order to get sufficient volume. If I used 0db gain, I would have to max out the volume.

_scotty_ · « **Reply #44 on:** 27 Mar 2004, 04:54 am »

Oxia,I constructed this buffer from a schematic supplied to me by Stan.
The total cost was about $400 in parts. The MOSFETs, diodes,and resistors
were the cheap parts. The Blackgates,Jensen,DACT pot,and 225VA toroid transformer plus chassis, is where I got nickeled and dimed to death. There are plans afoot to perhaps bring this thing to market in an improved form hopefully in the next year or so. If this happens it should be a usefull product for two channel fans. The circuits sound quality is mostly a function of the passive parts used in construction. You can hear the impact on the sound reproduction accuracy due to RCA jacks,internal wiring,coupling caps,power supply design,and how the board is positioned
in chassis and its proximity to the metal walls and floor.

Paul L · « **Reply #45 on:** 27 Mar 2004, 07:19 am »

Scotty, anything contain active component even if only buffering cannot be classified as passive controller. It is unity gain active preamp.

_scotty_ · « **Reply #46 on:** 27 Mar 2004, 01:46 pm »

Paul,you are correct,this is an active impedance matching circuit.It has zero gain,no negative feedback with a cascoded active current source and is singled ended class A in operation. The circuit has no gain to be set to unity via negative feedback.
To describe it as a passive buffer is inaccurate. There really is no such animal that I am aware of. Good clarification. Scotty

Paul L · « **Reply #47 on:** 27 Mar 2004, 04:16 pm »

There are quite a lot of posting in audioasylum regarding passive vs active. Here are a few threads by me.

http://db.audioasylum.com/cgi/m.mpl?forum=amp&n=21315&highlight=passive+Paul+L&r=&session=

http://db.audioasylum.com/cgi/m.mpl?forum=amp&n=20829&highlight=passive+Paul+L&r=&session=

http://db.audioasylum.com/cgi/m.mpl?forum=amp&n=29613&highlight=passive+Paul+L&r=&session=

Psychicanimal · « **Reply #48 on:** 27 Mar 2004, 08:44 pm »

I have solved this dilemma by using a Dan Wright modded Marchand X-9 Deluxe electronic crossover in tandem with my Channel Islands passive pre. All interconnects are short runs of Ridge Street Audio MSE (30" to 1 m).

I had the x-over hooked up for a couple of days in the summer and then stored it, since not all my cabling was RSA and I was concerned about the subs bothering my neighbors. Now the x-over's been hooked up again for the past week and a half and is breaking in. Since the x-over boards were cryo'ed and the power supply modded, the unit is opening up and the results are nothing short of fantastic. To me there is no turning back. A passive pre w/ an electronic x-over following is the ticket to the best of both worlds: neutral_transparency and dynamics. The Channel Islands passive varies impedance as output volume changes, but is feeding the x-over, which in turn it is the one actually driving the power amps with just a little bit of gain. Take this from a hardcore salsa fan.

Any questions, children?

_scotty_ · « **Reply #49 on:** 27 Mar 2004, 09:59 pm »

I am not familiar with the Marchand Xover is it tube,ss or ss opamp and what did Dan do to it mod wise.

Psychicanimal · « **Reply #50 on:** 28 Mar 2004, 12:41 am »

Quote from: _scotty_

I am not familiar with the Marchand Xover is it tube,ss or ss opamp and what did Dan do to it mod wise.

The X9 is SS. The deluxe has upgraded Burr Brown op amps and mine has Cardas RCA jacks and soldered w/ Cardas Eutectic solder. Dan upgraded the rectifiers, filtering capacity and installed a smaller capacitor in parallel to give better speed to the power supply. The assembled x-over boards were sent to cryo treatment. It was a relatively cheap & simple--yet effective mod. As the unit breaks in the soundstage, noise floor and extension keep improving. My system is 100% SS and the sound coming out is really good and non fatiguing, w/ a high degree of "foot tapping factor". I am using a Forté 4 class A amp to drive a pair of Modwright Swans M1 mini monitors and two Kenwood L07-MKII monoblocks to drive my subs. I'm a happy camper.

If I was to use long IC runs I'd use a Marachand tubed x-over modded w/ higher output. Dejan Veselinovic (DVV) has been right all along. The pre and the x-over should be one unit.

DVV · « **Reply #51 on:** 28 Mar 2004, 08:43 am »

Quote from: Psychicanimal

...
If I was to use long IC runs I'd use a Marachand tubed x-over modded w/ higher output. Dejan Veselinovic (DVV) has been right all along. The pre and the x-over should be one unit.

Well, that's what I use multiple outputs for on my own gear. Maybe I should go commercial ...

When you incorporate electronic XOs into the preamp, you pay by having more output sockets (and these do cost money, even if you, like me, decide to buy the same socket as a VERY famous one, from the same factory, but minus the famous name), more space is needed and you do have to make the small PC boards for each crossover. Also, actual physical incormporation begs for some solutions, mine being to use pins on the board, which are inserted and soldered onto to the main output board.

This may sound inflexible, and anything involving soldering generally is inflexible, but in return you do get the best possible contacts, very modest overall size, and quality up to any level you choose to go to.

Most important, you are sure to have less noise, mostly because people generally bust their behinds over preamp power supplies, but not so with add-ons (at least in general terms), certainly because you have less cabling involved and consequently less contact points, and lastly, you save some space as well.

The usual objections are that this is inflexible regarding XO points; I suppose one could add some controls, but my feeling is that XO boards and points are not toys and shpuld not be messed around with unless there is real need. Thus, I have no problems with fixed 18 dB/oct (3rd order) XO points; the next time I will need to change them will be when I buy new speakers, and that happens once every 10 years or so. Since I also always leave a straight pre out, I can always use the unit as a straight pre.

Another valid option is to use the power amp to the same effect, meaning to have the XO board installable at the power amp's input. This is even more convenient, because it requires even less space.

If you are a designer, and are making specific power amps for active speakers, you can actually use the power amp input stage as an active filter, thus simplifying the whole lot even more. This is an idea I am toying with.

Cheers,
DVV

maxwalrath · « **Reply #52 on:** 29 Mar 2004, 03:27 pm »

To follow up on my problems, connecting the Technics to the Stratos caused a fuse in the amp to blow. I never really got to A/B the two, I was always focusing on the problems the direct hook-up. I'm back to using the modwright beta, and I'm happy.