Resonant Peak Filters (just FWIW) might be of help...

[Erik Mandaville]

I have looked through a few texts I have on speaker/crossover/cabinet design by David Weems, Martin Colloms, and Vance Dickason and can include here a short section from Weems -- I quote:

"Tweeters and midrange drivers often present a problem at their fundamental resonance. This occurs at the lower limit of the frequency response curve."

"To reduce the possibility of a peak in response, you can add a series-tuned circuit that resonates at the same frequency as the impedance peak." "To design a resonant peak filter, find the frequency of the peak at resonance for your midrange or tweeter, than use this formula:

C = 1/(50f)

*note: This formula assumes an 8 ohm driver. For 4 ohms, substitute 25f in the denominator

To find the inductance that will resonate with C at the peak frequency, use:

L = 0.025/(f2C)

For R, choose a value equal to the rated impedance of the driver or slightly higher.

(I can't reproduce the schamatic shown, but it is a simple series CLR filter installed across the driver terminals.)

He goes on....

"Peak filters for midrange drivers can require rather large values for L." However, "...you can substitute lower values for L along with higher values for C and get good results."

[leok]

Erik,

thanks for the info .. I've filed in a log I'm keeping on the Chorus-II.

I have made some progress:

In trying to damp resonance using my old resistor trick I found I just completely screwed up low and high frequency response without much reduction of the honky, and sometimes screechy sounds. So I went to the parafeed output transformer input and put a 22 KOhm reststor there (input impedance of the transformer is 2.5 KOhms driving 8 Ohms). That stopped the honk and screech and smoothed out the entire response for all tube combinations. So, the good news is, the parafeed Moondogs sound better than ever through Chorus-IIs which also sound better than ever.

The Chorus-II crossover cap changes were contributing to the confusion. The Jensens on the mids had some serious settling to do, and their behavior improvement may have accounted for what I thought was the 0.15Ohm resistor I put in there. I also have Hovland Musicaps on the tweeter (and that was a no brainer .. cleaner highs and no problems).

Chris Robinson brought over his Horus 2A3s (by Jeff Lessard). We did a little gain matching with his preamp (also a Jeff Lessard design) and then listened on the Chorus-IIs. I thought they sounded great .. no honk or screech (and they don't have the 22k Ohm damper I put in the 'dogs). Something about Jeff's design is cleaner (and remember, Chris's amps have the same parafeed with "ultrapath" storage cap as my 'dogs). The amps were close, but Chris's seemed to have slightly better control.

The Tripath amp continues to sound phenominal through the Chorus-IIs, which is interesting because it was originally designed for the Forte-IIs.

From here, asside from an experimantal tweak or two, I'll have to get a 'scope involved to figure out what's happening in the parafeed Moondogs. With all the messing around I've done in there I'm not surprised something's become a little hyper.

leok

[Erik Mandaville]

Sounds like things are going in the right direction, Leo. If you have not read anything from the references (or at least the authors)I mentioned, it is interesting reading. Colloms, in particular, is very comprehensive. Weems is able to reduce some of the technical information to more common terminology and application. If you're interested, I can provide ISBN numbers, too. Weems is good for quick references, and I have built my own modified versions of a couple of the projects in the book.

I bet that that amp you built does sound great with the Chorus. It seems I'm sort of working on two systems in the same room, here...

I decided to stop torturing my Moondogs, and they now are virtually in their stock form. I guess I sort of felt that if I wanted something new, I would just make something completely new. My Moondogs have been through too much surgery. I'm still breaking into the sound of the Horus, but I think there are some interesting things going on with the grid choke on the 2A3 -- things just have a very cohesive sound -- especially with strings, which I like quite a bit. My Lowthers are incredible with classical guitar, but they start to sound congested a bit with very complex multilayered material. Orchestral stuff just doesn't happen well with the Lowthers -- at least in the rear loaded horns I have. So I turn to the La Scalas, which can divide things up much more clearly. But sheesh! a total of four speakers in one room is kind of much, but they both don't seem to get in each other's way sonically. The visual impact is kind of imposing, though!

Glad to hear things are working out better.

Erik

[leok]

I have run a few more a/b tests with the parafeed Moondogs and P6D .. modified Moondogs have an edge in the high end with RCA 2A3s and Sylvania 6SN7s, but the resistor I added killed the very low end. I have to find the root cause of the resonance. I'd say that what may be a 4 Ohm impedance on the Chorus is the remaining problem, but I don't think Chris's Horus 2A3s had any problem with the bass.

I will probably adapt the Moondogs to work with the Chorus-IIs .. I just don't have time to build a new design. I'm wondering if I shouldn't have gone with a 300B design and used a 4 Ohm output.

Also, I haven't tried the pp amps on the Chorus-IIs.

I'm still looking into the crossover circuits you mention, but P6D and Horus evidence tells me this is a modified Moondog problem and the Chorus-IIs are fine.

leok

[Erik Mandaville]

I think I asked about that impedance matching question, and what you describe may very well part of the problem -- especially the lower bass you mention. I don't know what resistor you put in, but if you've got a lower impedance load than what your amplifier is set up for, it's going to have to work harder to get what you want in terms of frequency response. And with a 2A3 SET, there isn't that much room to work with.

I was going to say reconfigure your OPT for 4 ohms to see if that helps, but you don't have that one in your Moondogs anymore! The Chorus is really 4 ohms? So, this is sort of a series/parallel arrangement of the drivers it sounds like. Are the midrange and tweeters 4 ohms wired in series for 8 ohms, and then wired in parallel with an 8 ohm woofer for a total of 4 ohms.?

Honestly, you could put a simple variable L-pad in line with the the midrange horn (after the crossover)and experiment with loudness output to see if that helps get rid of the peak. It's just a voltage divider, is all, and you could make a fixed L-pad later on, once you know what resistance you need/want.

Leo? You have the most important parts to build a version of the Horus. This is meant to be just a humble submission. I totally understand your work with the Moondogs.

[leok]

Erik,

I'll be off-line until Fri. I will have an opportunity to study these circuits. For now, the 2A3 amp is supurb for chamber music, which was the original intent. Getting the last few low end Hz is mostly a challenge. My idea of a good time. Thanks for your help & I'll be back.

leok

[Erik Mandaville]

That sounds fine.

Have a good rest of the week,

Erik

[Deang]

Leo, do you use a preamp? I've always noticed you don't have one listed in your signature line.

Erik, I know you don't use one, though I believe you had the AES AE-1 at some point - yes? I can't remember for sure, but this unit went to Ed Whitesell, and I believe there was a problem with it of some sort. At any rate...

The first AE-25 Superamp I had was outfitted with a volume control. I removed my preamp from the system several times trying to evaluate the difference. In comparison to having the preamp in the loop, it sounded somewhat cold and life-less, and the treble was peaky as hell. As long as I didn't turn it up, it sounded pretty good -- but anything extra and it was not very good at all.

I only bring this up because this "resonance" issue sounds very similiar to what I heard coming out of the RF-7's without the preamp in the system.

[Erik Mandaville]

I still have the AES AE-1 Preamp, Dean. Ed Whitesell does not have it now. He sent it to me for repairs, and I ended it up buying it from him. There is an enormous list of repairs that I did on it in archived threads. The person who worked on it or built it before did some of the messiest electronics work I have seen. I replaced both high and low voltage rectifiers (it uses 1 low and and 1 high for each channel -- a total of two bridge rectifiers and 2 full-wave bridge circuits). All the filter capacitors in those have also been replaced. A couple of the caps that came with the preamp when I got it from Ed were crushed in from the under side. The builder had squashed them into the preamp in order to make them fit.

The circuit was grounded all over the place, when the schematic (which I purchased from AES, specifies a SPG.

I also initially thought that whomever replaced the switch that Ed said was broken had done a good job -- and I changed my mind on that fast. I redid the entire input and output wiring, as well as the AC leads that lead from the switch-grounded center tap and over to the front panel -- which is not the best place to put an AC switch to begin with, since it is very, very near signal carrying connections.

I also replaced both volume controls, and installed grid stoppers on the input tubes to quiet some slight oscillation that was taking place. All coupling capacitors and cathode resistors were replaced; the entire preamp was very nearly completely rewired -- this time using a ground system (the one I have used for many years is very much like Jeff Lessards)with a single point chassis ground. I have also installed a transient spike filter on the IEC connector, as well as a low value, high voltage snubber capacitor across the power switch -- this completely kills any 'pop' one can get if the preamp is turned off before the amplifiers. The Standby switch also was not wired correctly and didn't work. Standby switches use a switch on the center tap of the power transformer -- with the switch on standby, only tube filaments will glow; making the contact of the center tap is what provided B+ voltage to the circuit. The pilot lamp stopped working, and so many wires were broken that I can't count them all.

I had the preamp working perfectly several times, and then something went out. Each and every one of those times, the problem was traced to former wiring -- mostly because the previous builder had cut into the the wire so deeply while stripping insulation off, that the lead broke. This was the case with the bridge rectifier circuit -- the leads of the diodes had beed stretched so hard that they snapped. I simply could not believe this hadn't been noticed by the repair people at AES.

The AE1 is now in top working condition, and it is the preamp for our much less sensitive downstairs system. The preamp has such considerable drive, that it has way too much gain to be used with high impedance amplifiers and insanely sensitive speakers.

I have not been using a preamp for several days, mainly during the testing of the Horus amplifiers I just finished building. Chris Robinson, who also has the Horus amps built by Jeff, also needed to have a series resistor installed on the input in order to reduce the gain of his active line stage (SRPP Pantheon built by Jeff)not overload the input stage and cause noise.

In my case, my wife made a gentle remark about how it's difficult to go and look out the windows to the outside with the amplifiers right between the speakers. The Horus monoblocks look and sound incredible in that position, but I have agreed to move them to either side of the speakers. This means using active preamplification again. I have a few preamps I built myself that I can use for this. I also began a little research this morning on 5687s in either an SRPP or DC (that means direct coupled)cathode follower design. It is a fine sounding tube that I like better than the 6SN7 (make dependent). I have another preamp that uses a single, sub-mini 6021 for both channels (designed by Eric Barbour) that is an incredibly good preamp. The chassis needs to be better shielded for our high RFI envrironment, though.

If you asked me if I preferred passive vs active preamplification, I would say a passive design, but only if interconnects are kept to a bare minimum, and the attenuator I was using had a very high input impedance and a vanishingly low output impedance. Any circuirty is going to color the original recording, Dean -- I guess people just need to find (as you continue to search...nothing wrong with that!)their favorite colors.

Erik

[Deang]

"The AE1...has such considerable drive, that it has way too much gain to be used with high impedance amplifiers and insanely sensitive speakers."

In spite of the several excellent expanations offered up by many here to help me understand this, I still struggle with it. Most preamps provide 18 to 22 db of "gain" -- so what constitutes too much? Wouldn't the only real world difference be that with low gain the volume control at 8 o'clock produces the desired SPL, say instead of 11 o'clock? What difference does it make if the volume control is at 8 o'clock or 11 o'clock to reach the desired SPL? Am I still misunderstanding the issue?

"Chris Robinson, who also has the Horus amps built by Jeff, also needed to have a series resistor installed on the input in order to reduce the gain of his active line stage (SRPP Pantheon built by Jeff) as to not overload the input stage and cause noise."

I want to zero in on the underlined part. How does one know, or how can one determine if the input stage of the amplifer(s) is being overloaded? Is there a formula, or rule of thumb for matching preamps with amplifiers?

If you asked me if I preferred passive vs active preamplification, I would say a passive design, but only if interconnects are kept to a bare minimum, and the attenuator I was using had a very high input impedance and a vanishingly low output impedance.

Is there then a relationship between the preamp input & output impedances, the amplifiers input impedance -- and "gain"?

[Erik Mandaville]

Chris, I can reply briefly, since I'm in the middle of several differnt things that are going to take up some time this afternoon.

1. Without getting into the math, lots of tube linestages have a far greater output capability than many tube amplifiers need or require to swing them (amplifiers) into full output. Lots of vintage gear, for example, (as well as some modern tube amps) are set up for maybe .5 volts in for full output. That's why many modern sources are very easily capable of providing the line level needed to more than edequately power an amp without the extra gain provided by active preamplification. Coupled with this, is a tube amplifier's intrinsically high input impedance -- often greater than that for solid state equipment (and the reason lots of people -- my brother is one example -- who really likes the synergy between a tube preamp and solid state power amplifier.

2. If you have your amplifiers say perhaps 20 feet of your sources, then the system might benefit from an active preamp with a little extra gain. Interconnects can play a big role in this, especially if the preamp has a high (10K ohm +) output impedance. That kind of OPI (abbreviated term)can have trouble driving or working against the increased capacitance of a long cable run. That's why cathode followers, which are often used as impedance matching devices, can be useful in such a situation: They have a nearly infinite imput impedance, and and output impedance of only several hundred ohms in comparison to the previously mentioned 10K output.

3. So, if an active preamp has an input impedance of maybe 100k or 250K (potentiometer)+ a high gain circuit + a low output impedance (as in the case of a cathode follower), one can have such great sensitivity that even the slightest nudge on the volume control can be blastingly loud. One way to solve the problem (if one must use a linestage)is to pad the input by lowering the input impedance. Going to a 50k or even 10k pot can assist, since more of the signal is shunted to ground. But, some preamps circuits like to 'see' a higher input impedance, which in my experience also sounds a bit more transparent. Another way to pad the input is to use series resistance, which is what Jeff did on the input of Chris's amplifiers -- I've done that too. The problem, is that if you use too much, like in the order of 100k or more, it's possible to maybe lose some high frequency response. He could have actually also put the 20k resistor in series with the output of the preamp (after the coupling capacitors) and achieved about the same thing.

4. One thing that can work, is a volume control on the amplifier, which basically serves as a voltage divider. When used in conjunction with an active linestage, it's possible to find the best input resistance based on the position of the pot -- which, again is very much like what Jeff did for Chris's Horus, only the resistance is fixed rather than variable.

5. Just in general terms, if an amplifier can achieve 100% ot its rated output with .5, 1, or 2 volts, which can easily be supplied by most modern sources other than phono cartridges, what is the point of a preamp that is rated for 30 volts output or more? All it does is add complexity to the circuit -- but provides one advantage I have enjoyed myself: It looks cool (I really think so!); provides more tubes to play with; is just another neat thing to have and take care of as far as one's prized hi-fi system. I admit that for me, this has been part of the fun. I used to use a high output tube linestage within 3 feet of my amplifiers, and had all kinds of resistance on the input and output in order to make the two (amps and preamp) work together without too much noise, distortion, and hypersensitivity. I just liked the way it made my system look -- to coin a phrase of a fellow form member's wife...it looked like a "...city of lights" My wife also thought all the tubes looked impressive and sophisticated, as would friends who would come over, afraid to touch anything because it all looked so...well...fragile, important, and unapproachable.

5. Ok...I've got to do some other things after this!!! One of the most important factors in this specific equation has to do with the speakers we are using. This kind of efficiency can simply become a pain in the azz sometimes. They make clearly known every small noise, buzz, hum, and distortion downstream; they simply need very clean electronics to sound good. After working with and building tube circuits on a regular basis for several years, now (I still don't know anything compared to many of the people I communicate with -- include the Horus designer! Hi, Jeff.)I am able to make amps and preamps that, except in certain very high gain cases, are quiet and sound good through my amplifiers and horn speakers. The first preamp I made was noisey as hell! I tried rebuilding it for fun a couple of years ago, and it is nearly dead silent. I just have gotten more comfortable and learned things along the way.

That AE-1 preamp we were talking about? It was a reather noisey device through my Lowthers (not so much with the La Scalas), and I have used the best grounding scheme I know, a couple of grid stoppers, shielded cable on parts that need it, etc.) The AE-1 is now in our downstairs system with that pair of 86 (or there-abouts) dB speakers I built. And despite the fact that the amplifier is a high input impedance tube amplifier, the AE-1 is almost totally silent...really dead silent. With my ear actually touching the 6.5" woofer, I can just detect a small. This is with the volume turned all the way up. It is fabulous in this second system.

Ok...Lastly: I wanted to sell the AE-1 because I spent an extraordinary amount of time and money beyond the price I bought it for. I took it off the 'market' though, in order to save the prospective buyer some of the headache that I went through getting it to function correctly. I kept discovering cold joints, broken wires, etc., and I would have hated for someone to pay me good money for it, only to have it stop working two weeks after they bought it. I am able to fix whatever goes wrong with it because I know how to troubleshoot (sometimes damn difficult and time consuming as heck!), but that may not be the case for whomever might have bought it. That wouldn't have been fair. Plus, my wife very much likes the way it sounds. And, with the amount of time a put into that thing, it is sort of hard to just see it go. The inside was like a junkyard (to me) when I bought it, but I sincerely respect the first builder's effort in trying to make something! I have just changed it a great deal since then.

Gotta run, take care Dean...we look forward to your first listen and report regarding the Quicksilvers. BTW, if there is a hole punched in the chassis that can accommodate the size of those big capacitors, it may be possible to install the brackets from the inside so you don't see them.

I wrote this post very, very quickly, and if therre are spelling mistakes or whatever, too bad!