|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
209.179.58.201
I recently gave up on my Dynaco PAS-3X preamp since no amount of optimizing the original circuit seems to me to cure its problem of slow, muddy bass. I think I've become a more critical listener to the point where I can't forgive its failings anymore.
As an experiment, just for the hell of it, I built a simple "passive preamp" with a 100k stereo volume control, an enclosure, and four gold-plated RCA jacks all from Radio Shack for under $18. (The RS pot tracks surprisingly well.)
It beats the pants off the PAS-3X. The output from my Musical Fidelity A3.24 DAC has no trouble driving my ST70 as loud as I need it to.
It makes me wonder: If you're willing to give up tone controls and don't need a phono preamp or have a dedicated phono preamp, why do we really need active line stages? Compared with the expense and trouble of building an active line stage, it seems like a quality passive is a very attractive option.
Follow Ups:
A simple potentiometer can work wonders when enough drive is available. I built a simple 6P5GT line stage that sounded much better than the Dynaco too.
Going back to the subject line of the OP...
Except for impedance matching of sources and amplifier, level matching and all that jazz, a linestage is supposed to provide (1) volume control, and (2) switching between sources - right?
If I'd need to build something to assist with my AI M2C preamp, is improved switching between sources. I'd be interested in building a box with much more and better selecting provisions than the contorted and impossibly inconvenient and unnecessarily complex the earlier AI preamps have (I may need to check myself for this and other things with a doctor, but I can never remember what combination of buttons selects some given source - except for the one I'm using the most - and I end up fiddling with the buttons every single time...).
Not sure what the traps of the project are either. It would obviously be entirely passive and based on some high quality rotary switches, good wiring inside (shielded?), good quality RCA jacks. Not sure how I'd approach grounding either (RCA jacks screwed to the metal chassis and that's it?....).
But it would definitely be a worth while project for me.
Radu.
I went through the same phase many years ago and it proved to be unfounded.
The truth is that every system needs a volume control and whether it's active or passive is not the point. In design, you have to look at the whole chain and ensure that every stage is as good as it should be and ensure that the volume control is "matched" into the chain.
These days I build the volume control into one of the components to reduce the count and the number of cables. But if you have to have a separate Pre then a properly designed Active will outperform a Passive every time UNLESS impedances are right, ie the passive does not load the OP stage (I like at least 10X the impedance, or more if I can get it) and it can drive the Miller of the following stage properly (I like a -3db point of at least 200K but that's me).
There is no doubt whatsoever in my mind that your Active Pre is simply not up to scratch. Design topology and PSU are the two greatest potential culprits and I'm not fond of 12AX7s unless you need the gain for stuff like tone controls. If you do have tone controls then there's another potential problem right there, especially since you'd have to be playing with very high resistances using a 12AX7. I'll bet the math doesn't work.
Naz
Can you say something about the "state of tune" of your PAS3X? I owned one several decades to go and found it unsatisfactory even by the standards of those days and compared to a few other "good" tube preamps that followed it. (I know the PAS3X is legendary, and I was supposed to love it. I just didn't.) In particular, I found the bass to be slow and muddy. But that unit was probably stock or near stock; I do understand that your unit may have upgrades making it far superior to the one I remember, which is why I ask the question. Your passive linestage may well be excellent, but I would not necessarily leap to the conclusion based on this comparison that all active linestages are hopelessly colored and all decent passive linestages are superior.
Funny:
I just finished rebuilding a PAS-3 with stock circuit,
1. had to revamp heater supplies,as is typical, as the unit used a voltage doubler circuit and you need 15.2 volts to get heaters of the 12 AX&'s up to snuff. Inital heater reading read only 19 volts. with UF diodes got the filament up to 24 volts.
2, changed coupling caps, using a 1.5 UF unit polypropylene.
3. Used diode separation to separate each B+ stage of the tubes in lone stage Owner was not going to use Phono so I left the phono section alone.
result was a remarkably airy and open soundstage with good control over the frequency extremes. enough so I am in the process of doing the same to a couple of PAs units I had and was going to use for parts.
You'll be impressed by what adding a large amount of capacitance to the filament supply of a PAS will do to improve the sound quality, if you haven't already done so.
I used Schottky 31DQ10 diodes (about 50¢ each) and two 22,000 µF Nichicon KW caps.
I've tried it with an otherwise stock PAS, and ones with modified circuits, and never failed to hear an improvement. I've added (via umbilical) much more capacitance than that, but the Nichicons will easily fit inside the PAS chassis.
One size does NOT fit all. Passive control centers are inherently high impedance O/P and can't drive loads. IME/IMO, the place for a passive setup is in the same enclosure as the power amp circuitry.
For the most part, line stage gain is unnecessary, when digital sources are exclusively employed. If cabling with anything resembling substantial capacitance is to be employed between the control center and the power amp circuitry, active buffering circuitry is definitely in order.
Eli D.
If you are using a CD player, computer audio server, Ipod, etc, then you really don't need a preamp at all. These devices are low impedance and typically put out around 2 volts of signal.
Now if you use an vintage tuner, reel-to-reel or cassette deck, you may or may not be OK. And with tube sources it depends if the output is plate driven or has a cathode follower.
A very loose rule of thumb is the newer the device, the lower it's output impedance will be. For example when the products transitioned from discrete transistor circuitry to OPAMP output drivers. But of course this is a very loose claim. There are plenty of exceptions.
I don't see the output impedance of the source as being the problem.
Almost anything can drive a 100k stereo volume control but the output impedance of the 100k stereo volume control is what can cause problems with cable capacitance plus Miller capacitance plus stray capacitance.
Even if the source has a output impedance of zero, the pots output impedance will be 25k ohms at the -6db position.
For instance, at 100pF and the -3db point of the low pass filter is 63kHz and the phase is disturbed all the way down to 6.3kHz.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Have you measured the frequency response of your PAS? Have you ever looked at internal signals with a scope to determine whether it might be clipping on higher-amplitude bass passages? Sometimes it's simple old age that degrades this gear; switches and pots don't last forever. I would personally never buy a 40-year-old preamp for exactly that reason. There's also the fact that the PAS wasn't a great design, even new. All things considered, it's not surprising that your passive replacement sounds better.
--------------------------
Buy Chinese. Bury freedom.
Triode,
I went down the rabbit hole of replacing virtually all the parts in my PAS years ago with Holco resistors, Cerafine electrolytics, Hovland Musicaps, high-quality volume pot, etc... It's got used/tested strong Telefunken <> base 12AX7s, cleaned selector switches, the whole OCD laundry list of component upgrades. I definitely improved it, but, as Norman Koren revealed a few years ago in his computer analysis of the circuit, the PAS has some problems at subsonic frequencies. You can see this on a scope if you feed it a strong bass signal and watch what happens when you switch the signal off. The operating points of the tubes will bounce a little before settling down.
If the PAS worked as well in the bass as it does in the mids and highs, I'd have no problems with it, and I'm hardly the first person to notice this :-)
It's really a shame to put so much effort into an upgrade, only to see it fall down in some way that wasn't anticipated. Perhaps it's not too late to salvage it. There are dozens of very transparent preamp designs around, and some are extraordinarily simple to build. It might be worthwhile to consider how you could adapt one of those to the PAS. It might even be possible to reuse some of the high-end components you bought for the original circuit. I know your passive design is working well, but maybe it would prove advantageous at some point to have gain, or at least a low output Z that can drive various loads. If it were me, I'd put it aside and take some time to mull things over. I would decide what features and specifications are really useful, then look for a way to build that functionality into the original chassis using the existing power supply. Even just a rudimentary buffered linestage with an input selector and volume control would be more useful than letting it sit in the garage. :)
--------------------------
Buy Chinese. Bury freedom.
You live and learn in audio as a hobby. If I wanted an active tube line stage at this point, I'm pretty sure I'd build some version of the Aikido circuit from scratch.
Maybe I'll run a thorough diagnostic of the PAS, but it's got to be something that affects both channels. The last time I checked the 12X4, it was strong.
I appreciate the advice on situations where a line stage is definitely needed. Very useful information. My passive preamp will probably end up installed inside my ST70.
My current push to get this sorted out started, btw, when I was testing a Pioneer SPEC-4 SS power amp I'd fixed for a customer in my main system. It had outstanding bass control when hooked to my KEF 104s, but when I hooked my PAS and ST70 back up, I realized that I was missing something essential to the music. Within its power limitations, my Dyna SCA-35 also has more articulate bass than the PAS/ST70.
OK, I ran some diagnostics on the PAS. Tubes all test > 100%. Voltages are where they should be. No high ESR in power supply caps.
Listening: the PAS sounds gorgeous, but that's really the problem. It's gloriously euphonic, but not honest. It adds just a bit of compression. It's like one of those TV shows like CSI Miami where we know that everything is run through filters to make everyone look good.
Switching back to the passive volume control, some recordings are less pleasant to listen to, but the instruments sound more like I know they do when heard live. Bass guitars and drums are much more defined.
I may be preaching to the choir, but you did not mention exactly what you did to your PAS3.
I've been repairing and modifying PAS3s for several decades. From what little I have been able to glean from your posts, it sounds like your line section is not working up to snuff.
I usually start with the filament supply, which should be rebuilt using at the very least a silicon rectifier instead of the stock selenium. I also replace the filter cap in the power supply; failure to do these steps will result in poor performance down the road. Its not a bad idea to replace the tube rectifier but if you do, do it with HEXFREDs; otherwise keep the tube rectifier.
If you are removing the tone controls, you don't also need the line stage circuit that is otherwise stock in the preamp. I usually use a 12AU7 instead of the 12AX7 and run it without feedback (there is a limit to the current you can draw from the power transformer)- this can provide similar gain but with a whole lot more bandwidth. I direct-couple the first half of the 12AU7 to the 2nd half, which I run as a cathode follower- if using a tube power amp the output coupling cap is then about 1 uf. This works a whole lot better than the original circuit- my circuit clips at about 15 volts output.
I have often built up the volume control out of a custom-built 23 position rotary switch. The existing volume control tends to have tracking problems. I usually retain the balance control, mono and separation switches. The volume control *can* be used as the grid resistor for the line stage, but if you do this make sure you install a grid stop resistor in the line stage or you can have problems.
If you do this right you will have a stable circuit that will play bass fine no worries- and be less colored than a passive control. There are a lot of things to mess up though, and if you don't know how to do this stuff there are a million and one ways to cause the circuit to fall short of the goal. Its important to keep this in mind; if you don't you wind up leaving performance on the table for no good reason.
To recap: I completely rebuilt my PAS-3X years ago from stem to stern with audiophile-approved parts, including the filament supply, volume pot (Alps Black Beauty), tube sockets, all resistors (Holcos before they stopped making them), and all film and electrolytic capacitors. It has increased B+ supply capacitance using Elna Cerafines following suggestions I got > 10 years ago from Gordon Rankin. Bass controls are out of the circuit, and treble controls are out of circuit at center. The filament supply did not escape notice.
I appreciate your recommendations, but it's quite a long way from "your line stage is not working up to snuff" to designing a new line stage. At the point where you're using half a 12AU7 gain stage direct-coupled to a 12AU7 cathode follower, isn't that a tacit admission that one might as well go ahead a build a different preamp? It's not really a PAS-3X anymore.
I was under the impression that the PAS power transformer didn't really have enough current capacity to run 12AU7s optimally. At least I think that that was one of the conclusions of Norman Koren's analysis of the amp.
Considering the reviews I've read of various versions of the Aikido preamp, I've been thinking that I might build one if I decided I wanted an active line stage. I even went so far as to buy an enclosure intended for a new preamp, but that was before the economic crash eliminated my job, sent me scrambling to keep my head above water, and forced me to turn my hobby into a livelihood. At this point, I'm a professional Hammond/Leslie and guitar amp tech. I have a shop, and I repair and rebuild amps every day. I've neglected my own HiFi system for a few years since I've been too busy fixing everyone else's stuff.
If I built a volume control into my ST70 so that there were no cables between the pot/attenuator and the input stage of the amp, would that make it any more acceptable? There is room. My ST70 has a custom base that's deeper to allow for PP power supply caps.
"I'm a professional Hammond/Leslie and guitar amp tech. I have a shop, and I repair and rebuild amps every day. I've neglected my own HiFi system for a few years..."
Man, isn't that always the way? I had my own company for about eight years, although that was in the era of the big SS receivers and amps. We did warranty service for companies like Yamaha, B&O, JVC, etc. etc. If I was still working on audio gear as my day job, my own system would look like something out of the Middle Ages. In fact, it's pretty much neglected even though I work in other fields now (RF/microwave). Hammond, you say? I still haven't quite finished the custom output/headphone circuit on my M103. Started it two years ago, so now it's at the point of chronic embarrassment...
--------------------------
Buy Chinese. Bury freedom.
These days, I'm fortunate that most of the gear I work on is out of warranty. It seems to be a major pain to be an authorized service center in terms of what companies are willing to pay for warranty repairs. I know of several techs who have jettisoned Behringer because they end up losing money vi an endless flow of low-dollar work. A lot of companies are making it fairly clear through their policies that they have little interest in support for products, especially those out of warranty. Parts become NLA shortly after the unit goes out of production, etc...
There are exceptions. Parasound was very helpful in helping me fix an amp and preamp that got damaged by lightning.
I had to deal with Musical Fidelity once about my own DAC, and they were awful. The fact that the thing was assembled with special security fasteners should have been my first clue to the attitude I was about to encounter.
isn't that a tacit admission that one might as well go ahead a build a different preamp? It's not really a PAS-3X anymore.
I would have thought it was obvious that with all your work, that was already the case :) If you want the preamp to be a PAS3, expect it to have problems in the bass at the very least. IOW, saying that 'passive is better' is not the same as 'active is worse' if the PAS3 is left stock. You can make it a lot better!
If you are going to use a passive volume control, putting it in the amp is the thing to do. You may run into issues though- the line stage does 4 things:
input switching, volume control, add any needed gain, and control the interconnect cable.
Switching and gain may be your issues in this case.
At any rate, its a simple fact that the stock line section of the PAS-3 is problematic. I've been running 12AU7s in there for many years and they work fine, despite Norman Koren's remonstrations. To deal with the current limitations impsed by the power supply, I set the plate resistor of the first stage and the cathode resistor of the second stage both to 100K. This is not optimal for a 12AU7 but it allows for way more performance than you will be able to get out of the stock line section or variations thereof using a 12AX7! I've put the resulting preamp up against a lot of preamps that cost serious cash and its had no troubles keeping up.
The one missing value that you don't have to complete the line section as I have described is the cathode resistor of the first stage. That should be about 4.75K. This circuit is not hard to set up on the existing circuit board; try it.
Thanks, Ralph. I may try your suggestion.
My rebuild of the PAS-3X was based on what I knew and what I'd been advised to do a decade ago. It suited me for a while, but I think it's time for a change. My quick and dirty passive preamp box was mainly a cheap experiment to see how it would work. It may not ultimately be the ideal solution, but it does reveal the original PAS's colorations.
In many ways, I suppose we can credit Hafler & company for being smart about designing an inexpensive preamp whose failings are mainly euphonic.
-But even back then its failings were understood. Its always been fun to try and see what you can do with one; I draw the line where anything has to be outside the chassis to make it work. At that point you may as well get a chassis designed for what you want to do.
But I have had good success with them. There are a host of changes for the phono section too.
It's been getting raves from owners.
It's a sort of active/passive line stage but with no gain, no impedance matching issues
and can drive long cables. Looks interesting.
I'm in the same camp as twystd...I have a PAS-2 that mates well with my own builds. Many variables...
I have built several low gain line stages with good results. A Preamp with a simple tone control and a little more gain is nice to have around especially if you're rotating amps and speakers.
What ever path you take, be safe, have fun...
Stuben
Lots of factors, the sensitivity of the amp, and the output V and impedance of the source are the most obvious. Then there is the amount of attenuation you use in normal listening, and how that output impedance interacts with the Miller Capacitance of the amp driver stage. The dynamics and the bandwidth of the music is a factor as well.
I'm sure you know that you can do much better, with a stepped attenuator, or better yet, a TVC or AVC in a passive. Also a Dynaco Pas-3 is on what I consider to be the low end of the preamp scale. However, no doubt, in many applications, no preamp is necessary for line level sources.
twystd
In this case, this ST70 still has its 7199 pentode input stage, so Miller capacitance shouldn't be a big issue.
And I know that there are better ways to do volume controls. I just thought that it was significant that something built with <$20 worth of parts is clearly superior, in this case, to something I probably foolishly tried to improve with higher quality parts, removing bass tone controls, etc...
I know that many modern power amps have lower impedance inputs that are harder to drive and that that may present problems for some sources.
I don't think you should "give up on active line stages", just because you can't get a complex cathode follower circuit with loop NFB to sound good. As the late Allen Wright was fond of pointing out, there are serious alternatives to both CF and NFB.
I am VERY pleased with my recently completed DIY version of Allen's Six-Valve Preamplifier, which uses his SLCF out stage with no NFB. No issues with loose bass, and it sounds very transparent and dynamic to me.
Alex
Sure do miss Allen's insightful input here. The circuit you posted is one of the first I intend to test when my workshop is back together.
--------------------------
Buy Chinese. Bury freedom.
Allen was a great guy who earned the respect he enjoyed from his peers. I can tell you that his SLCF works extremely well and is a mile in front of standard CF ... relatively complex but well worth building.
Naz
Agreed, my "volume knob" is merely a 500k pot between my tubedac and amp. Who needs all that extra gain and circuitry to muck up the signal IMHO. JH
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: