Resurrecting the Preamp

[Exeric]

Agreed.

 

[The Computer Audiophile]

"Amp makers, ESPECIALLY high end makers, know that they have customers with more money than sense."

 

Hi Exeric - Can you name some "high end makers" who you think fit your description? Barrows is correct in my opinion and experience. Manufacturers build what people want to purchase. I was talking to a very prestigious component manufacturer who said his company would build many more integrated amps because of the quality it can get out of them. But the purchasing public won't spend money on integrated amps so he builds stereo and monoblock amps.

 

[Weerstandje]

I find this a very interesting thread, even if I only untherstand half of the technical stuff.

If a lot of the synergy between a Dac directly into a amp has to do with impedance, how much difference can the interconnect make? I am thinking of for instance the MIT's with a switch to adjust the impedance of the cable between 10 to 100 or so?

Or are people who buy those cables customers with more money than sense?

 

[sandyk]

The analogue output of a well designed DAC should not be influenced by the type and length of a normal interconnect. However many cheaper DACs have output stages where the sound is affected by the capacitance of the cable used.

For example a Blue Jeans 2 metre low capacitance cable can markedly lift the top end in comparison with a more typical 2 metre long interconnect, which is likely to have almost double the capacitance.

 

SandyK

 

[Jim Smith]

Maybe I wasn't clear enough.

 

There was nothing intrinsically wrong with the dCS driving the amps.

 

If I did so, I didn't mean to imply that it was weak.

 

As I said, it was clean and transparent.

 

I simply thought that a neutral line stage did a better job of driving the amp than the dCS. And not from a technical standpoint. From a musical involvement standpoint.

 

While I have a number of industry friends who agree, it must be said that not all feel that way.

 

Of course, they are wrong. ?

 

[Exeric]

Several people now have said the amp manufacturers build what people want to purchase. I agree. But the probable majority demographic for people most wanting to buy separates are older people, like myself (age 58). But it is no longer technically appropriate to divide duties between an amp and preamplifier for non-active speaker systems. So when I say the amp makers know that their customer base for separates has more money than sense I feel i'm correct. It's silly to still be building separates when you can better control the quality of the audio with an integrated.

 

Today's culture seems to be politically correct about the dumbness of the public about some things and to mindlessly say that the corporacracy has no complicity in it. I'm here to say that these products wouldn't still be nearly as big if companies and interconnect companies did not make more money from the arbitrary division between amp and pre. I'm not going to call out individual companies. That is more of a tactic by Chris to enlist more defenders of the corporate faith to weigh in against me.

 

[Exeric]

It seems to be obvious by now that preamplifiers do much more than just add control features to the amplification process. That is the reason why many power amplifiers just do not work optimally without a dedicated preamplifier. The power amplifier in these cases are not a complete unit electronically without the pre. This is a whole separate issue from the control features in a preamplifier.

 

I think where the industry needs to go is to educate consumers on this subtlety. Once that is done then you can start building headless integrated amplifiers. This would be an amplifier that is just as complete electronically as an integrated amplifier is, but without the control features. This is a pretty subtle distinction from a pure power amplifier and right now consumers would not get it. To them it would be a distinction without a difference.

 

Once this concept of completeness in the electronic amplification process is fully understood by consumers you could change the demarkation from pre/power to control/amplification. The control could then be in the dac and you could use any wimpy output from it as long as you had volume control and digital input that are selectable within it.

 

[sandyk]

Far from being made obsolete, a very high quality preamplifier, whether a stand alone unit, or in an integrated amplifier is esential if you wish to have high quality stereo sound with your large screen DTV.

These days, even many of the soapies have very high quality audio, often accompanied with special LFE.

Many movies also have surround sound, which through a good preamplifier and power amplifier often presents an excellent surround sensation through only the 2 stereo speakers.

TV series such as the different Startrek versions usually have "engine room" LFE that can sound very convincing through a high quality stereo system.

I use Toslink from my Samsung "LED" TV into one of the 3 inputs of my DAC. Due to the sound level already being

[kana813]

"While I have a number of industry friends who agree, it must be said that not all feel that way.

Of course, they are wrong. ?"

 

LOL!!!

 

Jim- Nice to see you on this forum, what's the ETA on your DVD?

 

 

 

[audiozorro]

http://www.high-endaudio.com/RC-Linestages.html

 

FUNDAMENTAL PERSPECTIVE

 

Any system that sounds better with the addition of an "Active" line stage is an unequivocal indication that:

 

A mistake has been made in the choice and "the matching" of the components and/or cables.

 

To make my perspective and experience crystal clear:

 

No properly designed and optimized system requires an active line stage.

 

However, there is no need to worry or panic if you have, or require, one at present. The vast majority of audio systems require an active line stage. The elimination of the active line stage is a long-term goal that requires each and every component to be optimized and matched with regards to both sensitivity and impedance (and cables for low capacitance).

 

This means that most systems, at most times, will require an active line stage.

 

Finally, never forget that all active line stages are actually "passive" line stages along with (active) amplification. In short- you can't avoid using a passive line stage in any circumstances (unless you want no volume controls). The only issue/choice/decision, in this instance, is whether you want an active stage in addition to the passive.

 

The Active Test and The Rule

 

This test is very simple. If you are currently connecting a phono stage or a CD player directly to an amplifier, or through a passive device (and then to the amplifier), simply add any decent active line stage, or replace the passive device with any decent active device (it doesn't have to be "the best"). Once this is done, then listen to the results. The Rule...

 

If there is any noticeable and obvious sonic improvement with the active line stage, then you need an active line stage. It's that easy. All that's left is the most difficult part, choosing the model that you like the most.

 

Most sources do not have the required output. When they don't, it's extremely easy to expose their sonic weakness(es). In fact, virtually any active line stage (short of total "junk") will sound better in some noticeable manner (deep bass, dynamic intensity, more natural "body" etc). (It will also sound worse in some manner, but that is irrelevant at this point.)

 

Alternatively, when the source does have the "required output" (which is my present situation), then no active line stage, no matter how good it is, will prove to be superior in any noticeable manner. In fact, it will rarely even equal the sonics of the direct connection in any manner (because of all the extra cabling, connections and an imperfect active circuit). Even a theoretically "perfect" active line stage can only equal an equivalent passive line stage with the required output, because they both must share the same passive parts (volume control, selector switch, wiring etc).

 

The above "test and rule" is based on multiple experiences, not only in my system, but in many other systems I am/was familiar with. It is NOT some speculative "theory" I've put together for some irrational or egotistical reason, and I've never heard any exception to this "rule". So...

 

In short, if you need an active line stage because your source is not up to the task of driving the amplifier(s), then...

 

Any good active line stage, from any era, will improve the sonics in some obvious and clear manner.

 

Alternatively, if your source is up to the task of driving your amp(s), then...

 

No active line stage, no matter how good it is, will ever equal the sonics of your direct connection (or an equivalent passive).

 

[Exeric]

I don't really think anything that you said contradicts what I said in my post previous to yours. You already have three separate inputs into your dac. The manufacturer could have easily added 3 more analog inputs and route them through switching that is isolated from digital input to get low cross contamination. All the control features would then be in the dac.

 

I think you may not be getting what I said previously. ALL the pre amplification gain structure could be incorporated into a headless integrate amp. This is really the purist way of doing things because the manufacturer has complete control of how that amplification sounds going into the speaker. And you would avoid putting in an incompatible interconnect between pre and power. When a manufacturer can control all things in the amplification between control and speakers he can generally do a much better job. And get the control features out of amplification unit because it does not belong there.

 

I think it is not that difficult to build a supremely high quality headless integrated amp in a reasonable size. It definitely is much easier than when you also have to incorporate control features.

 

[Part-Time Audiophile]

In short, if you need an active line stage because your source is not up to the task of driving the amplifier(s), then...

 

Any good active line stage, from any era, will improve the sonics in some obvious and clear manner.

 

Alternatively, if your source is up to the task of driving your amp(s), then...

 

No active line stage, no matter how good it is, will ever equal the sonics of your direct connection (or an equivalent passive).

 

This is a really interesting set of statements. If I get a positive result (preamp sounds better), it's because my source is flawed. Whatever that might mean. If I get a negative result (preamp sounds worse), it's because my source is not flawed. Whatever that might mean.

 

Fascinating. I love it.

 

Of course, it's a circular argument. Why? Well, because "flawed" is defined as "not being better than having no preamp". Which is question begging.

 

Honestly, this is a to-MAY-to/to-MAH-to issue. If you're looking for minimal signal manipulation, then no-preamp. If you're looking for "the best sound", then go with whatever works best. But dictating a priori what must sound best sounds like an ideology (a fools game), not a theory. At least, it does to me.

 

[SteveChicago]

... and in my system, adding the old solid state preamp was a noticeable improvement and now having replaced that with a BAT VK51SE is a MUCH bigger improvement beyond that.

 

 

[sandyk]

Exeric

I am not disagreeing with you. I am trying to reinforce the message that a preamp is necessary due to very different output levels from DVB etc. There is a big difference between what a manufacturer may be able to achieve with a far better than average Integrated amplifier, and a DIYer like myself can do physically in a similar sized case.

I am using a DIY Class A preamp and a DIY 15W/Ch. Class A amplifier with large heatsinks that dissipate a total of 100W of heat

The transformers for both units are in a separate 2 unit rack case. If something of similar sound quality to what I am using was built into a single integrated amplifier case, it would not be very affordable for most people.Both preamp and Power amplifier are fully discrete, except for offset corrector I.C.s, as they are fully DC coupled. My new preamplifier does not need switching facilities, as the inputs are remote controlled via the DAC. So our views aren't really that much different , are they ?

 

SandyK

 

[ziggyzack]

but then, we should expect nothing less

 

Exeric, nothing at all against your call for power amps to be made to accept any reasonable input, but I am curious why you are focusing on the input stages of power amps, and not the output stages of sources?

 

[Exeric]

SandyK, I had a feeling we weren't that far separated. It was more a matter of which parts we were choosing to emphasize. Other than that we seem to agree.

 

Ziggyzack, line level sources seem to have a defacto standard of 2 Volts output give or take a volt or so. What they don't have is a standard drive capability. This is basically what the pre part of an amplification package provides for. It can be contained in the preamplifier or in the front end electronics of an integrated amp. There isn't any good way to enforce a minimal drive standard for the myriad line level equipment manufacturers. So why not leave the compensation for that lack where it is today - in the front end of amplification. I'm just arguing for a clear understanding why straight in to a power amp doesn't work and trying to disentangle that concept from the need for an amp that will universally work with Those line level sources, but does not have any control features. Actually volume control still might be good in an otherwise headless integrated design.

 

[Panelhead]

He has always seemed to be a well trained listener. Got to agree with him here. Less is more.

I used to have two high quality full preamps. A CAT reference and a VTL Ultimate. Plus a couple lower cost units. The phono in the CAT was better than the VTL with FMS mc stage. But the line stage in the VTL was better than the CAT. Since most listening was CD or radio used the VTL the most.

One day hooked a modified Fisher 200B tube tuner directly to the Manley 350's. The Fisher had a cheap carbon pot volume control. Had to swap back and forth several times because I did not like what I heard.

Needless to say the sound without either was much better. Even the quality of broadcast FM the degradation was very clear.

These were almost new preamps. Cost 4,000-5,000.00 each 20 years ago. Very highly rated by all. Yet they failed the bypass test.

Sold both a a large loss. Ended up using my AI Modulus that Ric Shultz had reworked. It still lost the bypass test. But no worse than the VTL. Later found the Croft Epoch. It held it's own.

Today I cannot imagine even trying against the dac now used. It has a PGA4311 volume controller. This scheme is used in the newer Metric Halo units. And has balanced outs. Sounds better balanced. All but one of my preamps is SE only. Not a fair fight.

Gotta believe adding a preamp between a good dac with volume control and amplifier can sound more pleasing. But the resulting sound is at best only as good as it is without the gainstage(s) and extra cabling.

 

[barrows]

While you make some good points, it appears that you may not have a full technical understanding about what you are talking about. For example:

 

"I don't really think anything that you said contradicts what I said in my post previous to yours. You already have three separate inputs into your dac. The manufacturer could have easily added 3 more analog inputs and route them through switching that is isolated from digital input to get low cross contamination. All the control features would then be in the dac."

 

When the PerfectWave DAC was in development at PS Audio, there were suggestions (from sales) that analog inputs should be added, as the unit was designed from the start to directly connect to an amplifier. The lead engineer on the project vehemently objected to adding analog inputs, because the required switches in the analog signal path degraded the performance too much to be acceptable. My point is that there are many seemingly small details like this which can contribute to great sound when each of them is addressed appropriately. I suspect the best approach to play, say vinyl, in an all digital system will be a digital phono stage, which converts the signal to 24/192, and then drops it into the computer for equalisation, then out to the DAC from there. With good 24/192 conversion, and RIAA equalisation in the digital domain, it is likely this sytem could outperform traditional analog phono stages, and we avoid the switches in the analog signal path.

 

 

 

[NickG]

Steve, I'm curious what the BAT VK51SE replaced?

 

 

 

[Jim Smith]

what's the ETA on your DVD?

 

Supposed to be shipped to us at the end of next week.

 

If so, shipping begins during the week of the 4th.

 

Now, back to your regularly scheduled thread, wherein perfectly logical people continue to make absolutist statements.

 

Instead of saying that folks who hear the need for a preamp have systems that are not optimized, I could suggest an opposite position...

 

If their systems were voiced to their rooms adequately, they'd have no problems hearing the less musically involving, more sterile presentation.

 

But that would be a simple use of their tactics, so I won't suggest it... ?

 

 

 

[Exeric]

You always seem to nitpick what the thrust of what I'm saying is. If you don't like the example I gave that is fine but I worked over 20 years working on avionics. I have a good understanding of electronics. Not that this is at all important to what I have been saying. I think your main problem is that what I'm saying regarding how to think about amplification is unconventional. For some reason it causes you to feel so much discomfort that it requires you to not only argue on the merits but to try to discredit the messenger. I think it is not necessary for either of us to wear our credentials on our sleeve. If you find it necessary, as you seem to, it indicates to me you have an ego problem. Bit it is alright with me if you want to blather on about PS audio this and PS audio that and then take a small "for instance" and say I don't know what I'm talking about.

 

I think what I said about customers with more money than sense was a little too true for your liking.

 

[PeterSt]

Don't fight guys. :-)

 

You can (nit)pick on the one of your choice, but the following is how things go when *I* don't want a single switch anywhere, while customers (luckily) go along with this. Or maybe a bit the other way around : I provocate that if possible nothing should be in the signal path, and thus the customer thinks along with me to avoid that switch.

 

So what really happens is that there's switchable output (following digitally switched input - source is PC only), but that switch switches On/Off two complete DAC boards, both connected in parallel to two sets of outputs (with zero influence from the one being Off).

 

A crazy ($$) solution, but still no switch.

 

One could call this nit-picking by itself, but I wouldn't do it otherwise (similar to the volume which is a no-go) and customers luckily understand (urging for such a crazy solution).

 

All I can say is that Barrows tends to understand this intrinsically by means of the start of his last post. This doesn't invalidate other options and opinions, but maybe it is good to see that this is all nit-picking in the first place, or otherwise you won't get anywhere with this. All adds up, really.

 

Peter

 

[The Computer Audiophile]

Hi Exeric - I've met Barrows many times. Your assessment of him couldn't be more incorrect.

 

[PeterSt]

Going direct without the preamp, and while there were some gains (slightly improved transparency) the losses outweighed the gains (lost body, weight, dynamics). I tried two DACs, one with ~150 ohms per phase output impedance, and the other with 100 ohms per phase. The amp is 11K in per phase, and my preamp is 55 ohms out per phase. I do not believe the problems of the no pre set up are due to using a digital volume control (my gains are matched such that 10 dB of digital cut is enough in most cases) but are caused by the DAC's output stge not driving the input stage of the amp as well as the preamp.

 

Let's say we are talking about your Sabre implementation;

(at the time you posted about your finished out of case project I created an unfinished response, and in the end never posted it)

 

With preamp, you will have a more smeared sound. This is just general, and wether we call it "filtering" or "flattening", both will have a smearing effect. This occurs in the highs (good for many because of poor highs to begin with) but this foremost happens with the bass. More so called punchy sound, but in the end no way the real vibes of a real (double) bass emerge. Bass though. Could even be perceived as good bass, not knowing any better anyway. But :

 

The DC filtering you apply (had to estimate the resistor value, but thought you would be filtering at 10Hz) causes this :

 

With a testsignal of 1000Hz this still creates a phase shift of 24 degrees (seen by measurement). I don't recall the phase shift at e.g. 50Hz, but it will be way more. Now just for fun, draw an imaginary sine with the length of 50 Hz (it could fit in your listening room), and draw a second line with only 24 degrees offset. Calculate (or measure) the length of the offset.

Yeah, well ...

 

Now imagine that you let this pass through a non-molesting chain. It will be exactly this you will be listening to. This can only be a disturbing bass. One which tries its best, is tight by itself, but which arrives as much later as the length of the phase shift and the speed of sound implies. It goes right against any together playing drummer and bassist.

 

Keep in mind that the 24 degrees happens at 1000Hz, and will vanish somewhat higherup. Also, it will be more and more towards the lower frequencies. Actually it is quite a mess, and your non-molesting chain (like no preamp) nicely makes this audible, or worse : emphasises this.

 

Now put in the preamp, or other means of volume control. Anyway, a means which we (or I) tend to call a "filter", but which really messes up like my before example, but now throughout the spectrum. For the "volume" means in general we could say that the result is random (per person different), but each specific situation will have a "pattern". A color.

 

The first what will happen is that your nicely put out phase shift will be subject to a couple of more of these "phase shifts" or anomalies otherwise (very different responses to different frequencies). It will be more messy, but the strict anomaly caused by the DC filtering will be gone as such. It will be smeared, and bass most probably will be perceived lower. Darker. The physical output (in SPL) will be higher, because a nice sine turned into a more squary wobbling thing, and as you know, squares carry more "voltage".

You preamp etc. made the sound better, because the real anomaly is masked.

 

But it got more wrong ...

 

If I'd refer briefly to the hardware vs. software thread, you have seen there that there is much (MUCH) more going on, and that generally nobody will be able to have this non-molesting chain, and this starts with "the OS/NOS debate". The ringing you all WILL (guaranteed !) perceive, will do even more to your layed out 50Hz sine wave, and all will be a big mess in the first place. Remove the preamp, and the mess will be pronounced.

 

The story is longer than I'll ever be able to write out.

 

It will be arbitrary whether a preamp will degrade sound more, or whether it will make it better. Chances to the latter are near zero, because things *are* wrong in the base.

 

Why do expensive interlinks make sound better ? because they may filter all the bad stuff. I said "may", because there's much more going on in this department.

 

But it is really the bad stuff which should be taken out.

 

Peter

 

[PeterSt]

I always appreciate your tempting questions, just because they are correct and to the point. On this matter, I hope you can also "appreciate" my being vague as usual, not always (or always not :-) wanting to unveil things, no matter how much I would like to do that as a person. Yes, I could have a second life on DIYAudio I guess.

 

While I agree with almost everything that you say above regarding "less is more" I feel it is important to point one detail out:

 

While you claim one resistor in the signal path (and for this part I assume you mean one series resistor, yes?, and that you are not including the power supply rails as being in the signal path), you then state that you use an IC for your gain stage. Well, IC anplifiers typically have quite complex circuits internally, including many resistors directly in the signal path, so by using an IC amplification chip, you actually have a quite complex circuit.

I understand that you tried to design a simple discrete circuit for the NOS 1, and found better performance from the IC, and am not making any comment on how the NOS 1 sounds (as I have not heard it), but to claim simplicity for an output stage which uses an IC amplifier is a little misleading.

 

Yea, well, uhm ...

That depends.

 

Let's first say that I *am* misleading. Leave a few words out and still speak the truth. Take the example of the DC Offset "prevention". You (in another thread) came up with three general methods, of course incouraging me to come up with my fourth, but I won't. Still I claim nothing to be in the signal path (for that, if you want), and this is the truth. More true would be my explanation of my fourth method which I actually want to shout to the world. But how can those with a "commercial" product, and neither can I ...

 

Similarly it goes with the OpAmp I use. So, Yes I use that, and no it is not in the signal path. Dive into the freaking world of the PCM1704 and you might find this just other method I have been working on for over a year.

I'm still dead because of that.

 

To at least say something useful in this post, let me add one more thing which may not be all that obvious to most :

 

As my development route (explained elsewhere a couple of times) implies, I started out with exactly nothing in the signal path (but of course the active gain in the power amps, which I don't measure). This gave me the "opportunity" to see how straight noise lines, and signal (frequency) responses can be. So, this also gave the "opportunity" (I wish I hadn't have it) to see all what molests. Or better : all what is applied molests. If it is not this, it is that. It is always something, and it always is audible within the second. Of course this doesn't take just one random track for listening, so I have many tracks with specialties to listen for.

 

The real message might be : while everybody says (or claims) that the differences we perceive can not be measured, I am 100% sure all can. However, start off with something like (and here we go again, sorry) a massively ringing (DAC) filter, and all is destroyed so much, that indeed no differences can be measured (for testing with a signal that is).

 

It may be important to add that this is all not a case of "when we go on the DIY route we can solve it, and no other (commercial) means exist". This is not true, because all *can* be solved in the DAC, although in my case it includes the playback software to do it really well. Strangely enough the lot indeed *can* be solved, by means of taking out the baddies. And this is the preamp and/or volume control. So, all problems solved (but take out your ever solving expensive interlinks as well, because by now they destroy).

 

That's all,

Peter

 

 

PS: Yes, that resistor is in series :-(

 

PPS: Rereading your quote, I now wonder whether you refer to the "amplifying IC" from my before post as that being in the DAC, while I mentioned it as being in the power amp.

But anyway, in the DAC a few are there too.