erik2A3

In complete agreement with wdecho. Call A, despite its inherent inefficiency and heat is, to me, the best of the best. I must qualify that it just happens to be my personal preference. If you can just enjoy the sound you have, if you do enjoy it, might it be worth perhaps allowing the numbers to be what they are? But by all means, experiment with the many options available. Maybe you will find something that satisfies all priorities. I have a single-ended OTL amplifier that runs blisteringly hot -- quite literally, yet is entirely within proper working voltage and current parameters as far as its component parts are concerned -- and designed by a highly regarded audio EE, whose work I have followed since before he even developed his kit offerings. I have many amplifiers, both which I've built and not, and this hot-tempered little thing, sonically, is an absolute gem. edit: "Call A" (second sentence)??? Not sure what that is erik; I think you probably meant to write "Class A"

Yes, you're right...without a doubt. I'm kidding myself and '...it adds noise and distortion.' I will absolutely keep that in mind. edit: wink

Right, I'll say it again: In the right circumstances, all that is needed for attenuation can be obtained from a variable resistor (aka potentiometer). The point remains, however, that, despite the fact that such a 'passive' device might suffice, "the right circumstances" to which I just referred above include more than electrical compatibility; what remains, which may be an even more important consideration, is whether or not the end-user prefers one end result over the other. In my experience, 'passive' and 'active' do not necessarily sound the same, despite both being satisfactory in terms of suitable impedance matching, cable length, cable capacitance, surrounding electrical environment (EMI, RFI, etc.). I have more than one line stage and active phono stage (MC), and, with speakers capable of 104 dB and 106 dB, background noise is truthfully zero, except for residual noise associated with the power amplifier (a great assortment of single-ended and PP OTL, push-pull pentode, single-ended triode (yes, directly heated). My vehement disagreement had more to do with broadly sweeping statements about active preamplifiers being essentially nothing more than sources of noise and distortion. That is quite simply NOT a universal truth. I have both the building experience and understanding about PSU design to know that is very possible to build and/or buy an active preamplifier with virtually zero residual noise artifacts. And again: EVERY single element in the signal path, active or passive, is going to impart something of itself onto the signal passing through it. When it comes to passive parts, such as your metal film resistor example above, the net result of the effect on the tympanic membranes in the ears of the listener is going to be both a matter of degree and highly variable from one individual to another. If active preamplifiers are nothing more than noise and distortion boxes, I suspect someone should take it upon themselves to inform the great many companies devoted to the development and production of high quality audio equipment that, to push their active preamplifiers into the marketplace is a practice devoid of worth and just plain wrong. Summing up: The advice I am comfortable giving those who ask about what preamplifier (or any component, for that matter) might be a good candidate to audition, would be to share that there are a number of different possibilities in terms active and passive devices, and to try a few of both to determine what might be most suitable for both the electrical compatibility between components in their system, as well as what they, as distinct individuals with their own priorities, tastes, and preferences prefer. If asked for my personal opinion, I would offer it, but want to make very sure that I qualified what I said by explaining that it is simply my own opinion based on what I like; and, as such, can't be taken as what would necessarily be best for someone else. Were I to say something along the lines of "I know which is best because I've heard them all, so you should buy the same thing." would be a compromise of my own integrity. Moreover, it would deprive the person to whom I was giving that advice of having the opportunity to choose for him or herself -- something about which I feel strongly.

ANY component will add something. I also already stated that passive attenuation can work very well under the right circumstances. "fly past" : meaning speed, unencumbered by undesirable effects of long runs of IC (my situation). Do you own and listen to an amplifier of any kind? An amplifier can only add noise and distortion. Do you listen to loudspeakers? Passive or active crossover? You like the colorations contributed by passive devices called capacitors, resistors, and inductors. We ALL have our own preferred selection of colorations. We are listening to electronically and mechanically reproduced music -- by its very nature a coloration. Listeners to music in the distant past were thus in a sense far more advanced than their present modern counterparts:. They didn't rely on machinery to vibrate the air for them; they listened to musicians playing instruments. So, certainly. I enjoy my chosen compromises and colorations tremendously. I hope you do as well, because that's what you are listening to. We all are.

Not quite sure what you're talking about in some cases. I was on the Peach roadshow when Mark D. first introduced it. My wife adored it. It was, and still is, an absolutely outstanding preamp. And we in fact used the Peach with our Klipschorns. Mark is a most respectable designer who has influenced others on this forum. I wish I bought one at the time. Coytee, I joined this forum in 2002, and have been through quite a bit of its development. I left for a time to focus on my own building, but hopped on again several years ago. So......nice. You're back using a preamp. swell

Coytee, It may have to you under those circumstances. I think it good you stumbled on something you like -- really great things can be discovered by accident. That said, it is not an automatic truth that a preamp degrades sound quality. Perhaps in your case, but it depends greatly on the preamp you were using, and all preamps are arguably not the same in terms of sonic quality in various systems. I will say I have heard the same as you, and quite honestly stopped using the preamp I had been using. At the present time I have preamps that I would say absolutely fly past ANY passive I have ever used. The one I am using now is based on the 300B triode, and has outstanding current delivery into long runs of IC. It is like polished glass, with really great frequency extension. But, by all means! use the CD without the preamp if it sounds better to you! I would guess, then, that it has variable output capability, because if you ran its full (fixed) output straight into a power amp, the walls would have come down.

Schu, Absolutely NOT doing it wrong. We have discussed here and I on other sites where an active line stage often bring much more to the proceedings than a passive. Again, a matter of preference. Simple carbon pot can be fun, I've made quite a few of them -- if appropriate impedance matching precautions are met. Get that messed up, and things will not be good. Yours is the only case where I have heard a single negative comment about transformer-based gain control (by those who like them). Maybe you just don't care for the way they sound, and if that's the case, you, as I and everyone else here, will choose another signal coloration to his or her liking. Use what YOU like and what sounds good to YOU, not what someone else tells you you SHOULD like because they've been at it a million years longer. Suggestions, fine; final choice is for the individual, IMO. edit: My reference to the transformer VC in the third paragraph above was directed toward Mike, not Schu. My fault for not making that clear.

And to sort of finish my thought on this, we shouldn't neglect to mention that, if one were in favor of an in-active (aka 'passive') means of controlling the output gain of a source, one approach which is thought to be greatly better than a basic audio-taper pot involves the use of transformers instead of potentiometers. They provide a very high input impedance for the incoming signal to work against on the primary side of the winding, while offering considerably lower output impedances on the secondary (output) side. In fact, there are those examples which can actually provide a certain degree of what we might call 'passive gain.' Cost of entry, though, is exponentially higher for dual mono transformers in comparison to either a two-section stereo or dual-mono potentiometer arrangement. Maynard's well-made point, however, remains: That modern source components usually have more than enough signal strength (gain) on their own to drive a large majority of amplifiers to full output.............and beyond that into clipping. And so he is simply saying that adding more circuitry to an already very strong and pure signal may (subjectively speaking) ultimately do more harm than good. Whether or not that is the case, though, depends on an assortment of different factors. Which brings me to what I think to be an oddly ironic thing about active preamplifiers: That they are in actuality more often used to attenuate a signal by turning it down than the reverse of that. How? What will be the result of taking the full, unattenuated output of a CD player, DAC, etc., and connecting directly to a power amplifier? In other words, a preamplifier is not too unlike the oxygen regulator on a scuba diving air tank or the control knob on a gas stovetop. They are used as a means to control, harness, or attenuate an otherwise extraordinarily strong potential.

Maynard, You observed, "...a preamp can only degrade the sound." I'm in agreement regarding the benefit of a simply in-line attenuator between a high output source component, and have built many such "passive" devices in the past. They can work well if certain prerequisites have been met. Appropriate impedance matching is critical as you know. Where a preamp can arguably be beneficial has to do with its potential impedance buffering ability -- a quick example of course being a simple cathode follower (which not all people care for). There can also be considerable increases in capacitance associated with longer runs of interconnect, and that in turn can be aggravated by higher source output impedances -- both of which make it difficult for a signal passing through a passive attenuator to get much traction. We of course know of active line stages with sonic characteristics that well out-weigh and compensate for the fact that, being either of a hollow state or solid state design, they provide signal gain, impedance matching/buffering (or both) and have a power supply.

There was (perhaps still is) a Lexicon DC-1 being offered in Garage Sale. I have owned two of them, and will never part with one that was very professionally modified by another forum member. I'll leave the specifics of that out here, but the stock version is a very competent multi-channel performer. There are various listening modes available, including three-channel (RLC). The Lexicon was recommended to me on the day I mentioned my having built PK's mini-box. That was under a different handle, which was my full name, but the post is still out there, including responses from a number of veteran forum members. Those were interesting times! The Lexicon is a supremely transparent sounding line stage that also has a stereo bypass and subwoofer out with selectable crossover, as well as the same for main, side, and rear surround speakers. Once one gets used to the menu tress, the amount of flexibility it provides is really pretty amazing; and the one for sale in GS is priced very fairly. They were a few thousand dollars when they first came out in the mid 90s (if memory serves me right here with regard to their introduction).

Build one, taking the output off the speaker level outputs of your amplifier, and dropping them to a summed line-level input for a blended mono signal, plus attenuator, for your dedicated 3rd amp and center channel loudspeaker. If your amp has line-level outputs, you can also take advantage of those. 27k ohms comes to mind for the necessary series resistance, but, as I said, it was a very long time ago. We discussed PK's mini-box many times (an equal number -- many -- years ago). You might check the archives. I tried it one Saturday morning for a quick weekend morning project, and absolutely loved the result.

And a Happy Birthday anniversary to you Bruce! I'm hoping you and your family will get to enjoy some amazing Philippino cuisine -- lumpia is absolutely amazing! I'm sure you know that though, with dipping sauce!

Alzinski, Your thoughts on this are appreciated and....yet... definitely your own. I am able to respond virtually sentence by sentence to your last response, but am simply too tired, etc., to bother. The Moondogs are beloved by those who own them, and to me that speaks for itself. They also have been very well received by experienced reviewers. I have many years with this particular design, as well as many other single-ended approaches - direct coupling, interstage coupling, parallel-feed, and so on. As Bruce stated, your responses represent your personal opinion, which, like my own, makes neither of us universally correct. Regarding boutique parts, meaning passive components such as resistor and capacitor types and brands, is very much a choice of personal 'seasoning to taste,' as it were. And the Moondogs are, in my opinion, well-worthy of such modification if the end result is to a particular owner's/user's liking. I joined this forum in 2002, and we spent much too much time trying to prove who was right and who was wrong -- or what or was not a 'worthy' design. Right, so the designer used two octal slots when he could have gotten by with one. The audio biz is also full of companies adapting suitable parts from one design to another. The only difference in this case is what amounts to slightly longer lead length between sections, as well as an extra filament; which is well within the operating parameters of the power supply. Jensen interstage coupling capacitors are a little more expensive than others, but one can fork out much more $$$ than what these reasonably-prioced PIO caps cost. You were kind to not sound harsh, and you don't.....really. But I have never really supported criticism of what someone else likes, audio or anything else. I submit we can offer suggestions for getting this amp up and running again without our potentially sharp and/or overt expressions regarding the efficacy of its design or abilities of its designer.

All good Bruce... Series connection is conceivable, and I may have seen that all those years ago -- where the time has gone, I just don't know. If it is an error in wiring, I would be the first to admit I've made more than my share in the past. In fact, making mistakes and learning from much trial and error has hopefully helped me avoid fewer of those as I also continue to learn. That cap might also have been used as a bypass elsewhere, perhaps. And agreed. Those little grid chokes are gems in both performance and price. I don't have any right now, either, having sold my only pair within the Horus 2a3 amps I built (or, more accurately, my own version of the design, which was most definitely and strongly derived from Lessard's outstanding original circuit). Without his work, the pair I built wouldn't have been possible.

Well Bruce, Seems there are a few too many letter o's in your opening statement there. That was a long, long time ago. I also know who built them to begin with, and know he wouldn't have made a mistake like that. You indicated you thought you were the 4th owner, correct? That means these poor fellows have had a surprising number of owners before you. Who knows what was done in the name of 'improvement'. How about a very close image of the actual connection? Is there a chance they aren't actually in parallel? The way Ron set them up was for those not accustomed to building to be able to refer to positions on the terminal boards, many of which I by-passed for shorter signal paths on my own amps. It can be difficult, sometimes, to tell exactly what is going where. If you have been using them for years without problems -- then good. I'm glad to hear it. Best luck in tracking the problem down. My effort in this was to try to help you. erik

Ah, I see. That may have been installed in series to lower the filament voltage (as you mentioned) if, by nature of the variability of mains voltages, to achieve the necessary 5 volts. And you are definitely correct about the use of two dual triodes! That's perceptive, and you're correct too, about the use of two 6sn7s not being necessary. The amp will work exactly the same if one were to use both sections of just one. There is a reason, however, though not a technical one: The Moondog was built on the same platform of another amp by the same designer called the Laurel 300B. It was also a single-ended amp which was in fact designed to use an additional necessary tube. In other words, the two amps use the same chassis plate, but the Moondog made use of what would have been an empty chassis punch by installing the extra octal driver, but only using one side of both tubes.

If those two Solens are wired in parallel, they are wired incorrectly. That would give you a net capacitance of 20uf (even worse for the GZ 37), AND the voltage rating for the paralleled caps (which essentially form one capacitor) would only be that of the 400V capacitor.

I would want to measure all voltages against the schematic, as well as test, in particular, the rectifier. The GZ37 is a nice looking tube, but I have purchased (apparently) NOS that measured very disappointingly for the price paid. I mentioned the value of input Solen cap for a reason -- it's not an aspect of power supply design that should just be taken as a 'whatever.....no big deal' sort of thing. 10uf is more than twice that listed in tube data for that valve, and, as such, is not insignificant. You might also want to check the state or quality of the epoxy that was originally used to attach the nylon stand-offs beneath the terminal strips. Mine, as well as those of a couple of other Moondog owners I have known, was either old or incorrectly mixed -- turning into something akin to chewing gum. A definite, greatly time-consuming pain to remove and renew with fresh, carefully mixed epoxy. Ron abandoned this approach with the DRDs he developed with Electra-Print (who designed the DRD circuit), using actual hardware for the standoffs instead the decidedly more messy and less permanent glue. There are a couple of other Moondog owners among us here, so hopefully some will benefit from at least a bit of this info. It's a good, fairly traditional single-ended design, and definitely worthy of the maintenance and technical checkup as mentioned above. Someone wrote me a couple of weeks ago wanting to buy my pair. Never.

alzinski, Where do you see a series resistor in the rectifier filament supply?

Looking at the pictures you posted, have a closer look at the main filter capacitors and filament connections on the 2a3. I know you know what I'm about to mention, Bruce, and I say this as a matter of habit: MAKE SURE RESERVOIR CAPS ARE NOT HOLDING HIGH VOLTAGE BEFORE DOING ANY WORK. The Moondog does not use a bleeder on the output of the power supply, so this becomes doubly important.

As others have said, the Solen 10Uf used as the input to filter capacitor is well within the limits of the B+ coming off the rectifier. What you are seeing is simply a mild deformity of the outer insulation. However, there is in fact another issue related to BOTH the 10uf cap and the rectifier Ron W. chose for the Moondog circuit (and as you older forum fellows might remember, I'm also a Moondog owner/user, and have been so for nearly 20 years. Here's the thing (and I know the amp circuit so well I can see it in my minds eye): I have run into a number of more modern designs, both single-ended (as with this amp) and PP, where the 'designer' did not take into consideration the maximum allowable capacitance for a given valve rectifier when using a cap-input-to-filter design. So, if you are still using the stock rectifier, Bruce, the max value given for it is 4uf. There seem to be varying degrees of attention to this detail, as well as how important it ultimately is; but it seems tube data would not just attribute random values of reflected capacitance for the many rectifiers using the same general filter design. I'm NOT saying that this is your issue with the POP you heard, but just mention it as something I think to be very worth considering for modification since you have the amps on the operating table anyway. I'm in agreement with John and Craig on this. in my experience, popping is often associated with PSU connections, such as broken/intermittent connections on either the primary or secondary sides of the power transformer, connections to filter capacitors, etc. Don't only suspect high voltage secondaries -- check also filament winding connections for the rectifier, input/driver stage, and output tube. Confirm the grounded high voltage center tap connection too; dropping resistors, plate load resistors, etc. The power switch is also of course in the recessed area of the wooden chassis, so you might also inspect there, as well as fuse connections. Even something as simple as dirty or loose tube socket pins can cause such a problem, and pins can be easily tightened up. And while you're at this, put one of those MQ grid chokes I mentioned years ago on the grid to ground in place of the grid leak resistor on the 2A3. I must acknowledge Dr. J. Lessard for sharing this with me when I built his marvelous parallel-feed Horus amps. I still kick myself for ever selling those things. With the K-horns and Lowthers we were using at the time, that amp sounded easily like a crystal-clear 25 watts.

What brought all of you to that conclusion? I'm assuming this comparison was based on listening to each with all things being equal except line-level active vs passive frequency division. Is that correct? If so, were the differences immediately obvious? edit: Please disregard the bottom portion of the quote. I'm still used to the old days, and it's the first time since then I have even tried to use the quote function when responding to another member's post.

Wdecho - It seems others have referred to you by name as 'William.' I have enjoyed your concise descriptions of experiments done with this, particulalrly since you made comparisons to the AL network. The PIO capacitors you described (and I refer to Jensens in this case) did in fact seem to pull down HF response a bit. I have to qualify this by saying that was the case when used for very low voltage crossover use -- which is completely different from interstage coupling capacitor applications in amplifiers. In fact, I have seen cases where people are using not only the most expensive film caps available, but also those with the highest available voltage ratings; which for crossoever use in loudspeakers is simply not necessary. It is a fact that even a 50 volt cap is good for the vast majority of amplifiers with which a speaker is intended to be used, particulalry in the case of very high efficiency horns. On the other hand, the GE motor runs I used to use were much more to my liking in that respect.

Mike, I have some history under my belt with this, and was building speakers and their attendant crossovers exactly twelve years befor I even ventured to this forum in 2002. Of course measurement can be an extremely effective tool. To assume otherwise is a misinterpretation of what I wrote. Someone above made what I would describe as a completely accurate statement about the simplicity of the majority of Heritage networks -- particularly the type A. There is just nothing that remarkable about it. The majority who have worked with these networks or some of their own variations of them, do so with a general understanding of the characteristics of the drivers in question. We are also able to go through the simple calculations for the values of capacitance and inductance for reactance at specific frequencies in relation to the nominal impedance of the of the drivers with which they are associated. How complicated is it to make a first order, 6kHz high pass filter for a tweeter!? If one is doing the basic math for appropriate reactance at specific frequencies for specific impedances, and particulalrly for a simplistic, very low order design, one is already in the ball park. I also mentioned that I wound (and very carefully MEASURED the inductance) of chokes I wound myself -- which is not a big deal. There are an enormous number of audio DIYers around the world who do the same and much more. So you bet! With something as basic as a first order crossover, used with drivers whose response behaviors have been generally pretty well established here over the past 15 years or so, I am entirely comfortable with refining the final sound, as do many designer does, by ear. Even the second order tweeter network at 3.kHz was just not a big deal. One thing I didn't care for about the A was that its filter is derived from the band-pass portion of the network rather than being independently terminated. That's one of the first changes I made to it, and preferred the sound to the original. Also, I think it was mentioned above somewhere that an autorformer should not be used into an L-pad. If that is really true and accurate, than one should pass (cough) on McIntosh amplifiers, which use an autoformer on the output. Nowhere did I put down the importance of measurement -- I use it myself. What I did say is that I have a thorough understanding of the behavior of drivers I have used for nearly 20 years, their wild impedance swings in particular. I also know of the measured (and decidedly more visibly -- ON PAPER -- linear response) of correction filters that were designed using measurment tools. And I know how much life they sucked out of what was (for me) formerly an incredibly life-like, immediate, clear, and engaging presentation. So, since the sound of reproduced music is as highly personal and subjective a thing that it is, shouldn't I do what sounds best to me, and allow others to do as they see fit without criticising their personal choices? It's getting late, time to go!

Dean, We all choose our own sets of design compromises - and you're correct that an L-pad does not permit the same power transfer function as that of an autoformer. I'm aware of that. However, were I to base all my choices on plots and graphs, I would probably toss the majority of my equipment out the window (or at least sell it to get something in return). The Lowther drivers I use in our rear loaded horns measure terribly, particulaly their impedance characteristics, which are all over the place. It's the same speaker that Art Dudley of Stereophile used to have; he and I got into Lowthers at about the same time. Yet, they are very, very special, despite their flaws - a single, extremely efficient driver, nearly full-range, and one of the reasons I like them so much: they don't need a crossover. All manner of filters have been made for them to tidy up their response, and I've done most of them, only to find they just sucked the life out of the music. So, with the Lpad type A: If I like what I hear, that will win every time. I don't need a graph to confirm what I should or should not be hearing. It's really as simple as that. The other part of this equation is that the tweeter filter was on it's own, and so probably got some immunity from the ESR effects associated with the usual input cap into the autoformer - which is also in series with the tweeter cqpacitor in the A network. Using Bob's tweeter, I also crossed to the tweeter as low as 3.8khz, but used a second order filter for a sharper cutoff at that lower frequency. Quite good sounding, too. That was a completely different animal from the type A though -- but a network for which I wound and tested all the chokes it needed myself. I am sorry to hear of your back pain. I've had multiple surgeries and fusion after a crash on my mountain bike reaulted in bilateral fractures at L4/L5. Back pain can be so very exhausting....I'm sorry and I hope you are able to get some relief from it.