captainbeefheart

I know which BAF conference you speak of. He was talking about "Schade" feedback being old tube technology, this is to connect the plate with the grid via a resistor which is negative voltage feedback. This takes the pentode curves and makes them into triode curves. He was discussing the static induction transistors or SIT's used in the Sony amps. Those transistors naturally have the triode like curves so Nelson wanted to make an amp that everyone could build with regular Mosfets and by using "Schade" feedback will give the same curves as the SIT transistors. He made the SE 50 watt Class A amp but instead of directly connecting the drain to gate with a resistor he used an input transformer to process the feedback increasing input impedance instead of lowering it by just connecting the drain to gate. I actually built this amplifier which is why I know so much about it and remember the discussion very well. There is nothing negative here and I am not "determined to be right", I just want to get all the facts straight for all our sake's as anyone in the world can read this and I feel it is important to have the facts of the matter in order.

I read nothing on page 4 about this coming from old tube amplifiers. I do know there were a handful of tube amps that did utilize this approach but it was extremely uncommon. Grommes and Bogen for example used current feedback from the speaker, in the case of the Bogen you could adjust the current feedback from positive to negative for variable damping control. The Grommes used a set amount of positive feedback. It is the opposite. No engineer would ever dream of having a summed amount to positive feedback. What I mean by that is like the F7, Nelson says he uses a moderate amount of negative feedback with a touch of positive feedback. So the amp doesn't sum to an oscillator, the positive feedback adds to the negative feedback making the total feedback still negative. For example moderate amount of negative feedback for Nelson is usually around -20db and if he uses like he does in the F7 +2db of positive feedback the sum is -18db and we still have an amplifier not an oscillator.

Positive current feedback is much different than what you are talking about. I am not sure if you are speaking about adding resistance to emitters or leaving the cathode resistor unbypassed which is current degeneration negative feedback. You also mentioned making pentodes behave like triodes, there are two ways of doing that, one way is to just tie the screen grid to the plate and the tube actually is a triode at that point. Or, there is what has become known as "Schade" feedback by Otto Schade who didn't really discover this where you bring plate/collector/drain signal through a resistor back to the grid/base/gate of the device and you end up with triode like curves and distortion. Schade feedback is negative voltage feedback. Positive current feedback is rare to see in amps and needs to be done carefully in conjunction with negative feedback. All the other feedback methods I mentioned do not take error information from the speaker, the current feedback I am talking about senses the current through the speaker and processes it somewhere back into the amp at an earlier stage.

Not all SET amps lack in bass performance. Sadly many of the companies that make them market them towards people that have a fear of negative feedback, ironically the reason why triodes are so linear and have low impedance is because of internal negative feedback intrinsic to them. Many very nice push pull triode amps use negative feedback but for the reason I stated many SET amps on the market do not. It's funny because push pull cancels even harmonics so they naturally will have less distortion vs single ended so you would think they would use negative feedback with SE amps more than push pull amps. The best part about triode output stages is you do not need much extra negative feedback to get excellent results. Single ended outputs need very high quality output transformers because of the standing DC current. The gap needed so the core doesn't saturate reduced inductance and so you end up needing larger core and more copper to get the inductance up to what is needed. Add to this that fact of having more copper you need very ingenious winding methods to keep good high frequency performance. Negative feedback will help increase bandwidth so even if some high frequency bandwidth is reduced from the gapped output transformer the negative feedback will bring it back to well above what is needed for audio. Another HUGE benefit of adding negative feedback is you can now start to also add some positive current feedback from the speaker to actually create a negative output resistance and go from a damping factor of 8 all the way up to 100!!! Seriously just +2db of positive feedback can have huge advantage into lowering the output impedance to control speakers. Nelson Pass uses this trick with some of his amps like the F7. The first generation people complained of boomy poor bass control so he added just +2db of positive feedback and drastically improved performance. Sadly far too many tube amp manufacturers just do not desire to improve their designs that are basically just recycled old designs with fancy capacitors that do nothing. I have listened to 150 watts per channel SET amplifier that had extremely tight bass performance. From Outkast to Opera this amp was amazing, lush midrange you would expect from a SET amp but clean and tight enough to play difficult complex music like a SS amp. It's all in the engineering, it took many years to design that amp and the output transformers were $10k a pair!!

Before you get ahead of yourself I would take the "broken" power supply and run it on a resistive load. Take the 24vdc section and load it with a 220 ohm resistor. This is about 2.5 watts of power. See if the voltage stays at 24v with this load. Next test the 5v section. Load it down with a 10 ohm resistor, again this is around 2.5 watts of power. If the power supply cannot regulate the voltage under these loads then I would surmise it is officially toast.

I am going to have to steal that fishing lure quote from you!! So true. We have been using Panasonic ECQE non-inductive metallized polyester capacitors for a while now. Very tight tolerances, low esr, low DF, no shock or audible noise or piezoelectric effects. Just fantastic capacitors really.

Wow very interesting amplifier!! These must be handmade built to order small company? Any more info on them? Since you are asking about what are the best tubes for the amp I would find out the exact primary impedance of the output transformer and what the B+ is. Then use the power tube that fits best. EL34 would want a higher load, 6k p2p vs 4k p2p for the KT88. In triode mode both tube types are only 200 ohms difference for internal impedance so they would be ok with similar plate loads. If you use triode mode the most then no worries. I also find myself listening to tubes nowadays more than solid state. I do love some of my Pass type amps but they are not the same as a lot of other solid state amps. I do enjoy any great amp but tubes tend to be powered up here for more often lately.

When I purchase veneer the choice of thickness is usually between .9mm-3mm. I am not exactly sure how thick Klipsch veneer is, I know it is thin so I am giving a guess of 1mm. Furniture grade Birch plywood's outer ply (face veneer) is on average 1.5mm. Both are technically "veneer".

Very true, the furniture grade raw birch ply used on the La Scala has 2 outer very fine grade ply's which similar to veneered plywood you need to be careful not to sand too much to go through these ply's. I am not sure the exact thickness difference between the veneer and outer ply's, both are thin.

This is how it came out.

I have done exactly what you are inquiring about. Raw birch which was painted black, wanting to revert back to wood finish. Sorry to say but yes first you MUST remove all drivers. I used a paint stripper instead of sanding. Choose one that is safe for wood. I used New Generation Super Remover with great results. Once you have the paint off all you need to do is sand paper by hand. Start with 100 grit and end with a finish grit like 400 or above. I dislike polyurethane based stains and finishes so I used Tung-Oil finish which is both protective and looks great. Only problem is you want to clean and re-coat every couple of years to keep protection and appearance up. Depending on how deep the paint got into the grain you will get a slight exotic look as long as you do not have to aggressively sand to get a nice finish. After applying the oil finish coats (two coats was enough) I got a better appearance vs just plain raw birch. The extent of this is how much the wood has absorbed the black paint. Good luck!

You overlooked the part about the increased noise floor differences. The boutique capacitors especially the Jupiter ones mentioned are much larger in size compared to capacitors actual engineers chose to use and the physical size is what picks up more noise. It is not really just a economical issue, noise is certainly considered part of performance. The capacitors named by decware showed the higher noise floor when tested. If it does only come down to price and performance is exactly the same, I completely agree with you. If you just want or like fancy expensive "boutique" things then that is totally cool also and I have zero problems with that. People strive for originality and want to be different from the pack. I guess my only gripe is that many of these companies that install these boutique components claim they have performance enhancing quality which in a coupling capacitor application is going to be BS 99% of the time, or worse they add more noise which I would think nobody would want.

That method used in that video about how much power is needed is grossly inaccurate. He is using a digital multimeter that is not fast enough to display any reasonable information, with actual program material like he played it will just bounce around readings. The best way is to use a digital source while listening to music and make note of where the volume control is at. This should be the loudest you will ever listen to music. Next record a 240Hz sine wave at -20db and play it through the system with the same digital source used for the listening test. Take your measurement at speaker terminal. We know the loudest a digital source can get is 0db, since you are measuring a constant sine wave at 240Hz the digital meter will be accurate. We also know that -20db is a gain of 10x. So say you measured 700mV, you know that the largest transient in the recording you were listening to while doing the listening test will be .7*10=7v. That is 6.125 watts into 8 ohms. This method is going to be the closest way to measure your power needs and have it be accurate. Jim I agree, a Fisher 400 would be great!!

I stay as far away from any company that states using cryo treated beeswax caps as a performance part. Coupling capacitors are of such low impedance compared to the circuit you don't encounter phase shift unless they are undersized and suffer low frequency roll-off and at those frequencies you will encounter phase shift. Coupling caps low impedance means there really is no signal ACROSS them, which means they really are not doing anything to the signal at all. Since there is no signal across them they do not add any distortion. I have tested so many coupling capacitors and overlaid the results and there is no difference between electrolytic and any boutique capacitor you can find me. In fact, the only different with these extremely large beeswax capacitors is they will actually increase noise which was the only thing I saw different. Our local audio group did blind listening tests and nobody yet has been able to reliably tell the difference between any coupling capacitors. I know this is a heavily debated topic but there are other capacitors in the signal path that actually make a difference. People believe because the coupling capacitor is in series that the signal has to pass through it like a water through a filter which just is not correct. It is when a capacitor has the whole signal across it's terminals, like a capacitive load, the capacitor qualities then come into play. I do not expect non-engineers to understand exactly how capacitors function as it is very complex but cryo treated snake oil capacitors are just not needed, sorry. The decware amp you are looking at is 6 watts. The forte is a 98db speaker at 1 watt. To have enough dynamic headroom you need 20db, will this be enough for you? Depends on average listening levels. The 6 watt amp will roughly bring you to 106db at 1 meter away, this puts your average listening requirements down to 86db. If you average 86db or less then it should be fine for you. Unless you are running a class A2 amplifier you never want to run it near max output so to be safe maybe say 84db average. You can download an app that will tell you the db level of your average listening volumes, I suggest you figure this out before looking at amplifiers so you can at least find out how much power you will need to have adequate headroom. Decware does not state any distortion specifications. It does state that it uses zero negative feedback. I will go ahead and say that this may not be the amp for you if you want to listen to complex music at even moderate sound levels. It will just have too much distortion for complex music and without any negative feedback the damping factor will not be great, you will get what they call "one note" boomy bass most likely with this setup so bass heavy and complex music tones will not be great. This particular amp is marketed to people that think feedback is bad, there is a myth that "no feedback' amps are superior but in this case I do not get the warm fuzzy feeling that no feedback is a good idea. Well, unless you listen to very simple music like vocals, chamber, simple acoustic, piano, etc.... Start cranking Metallica after a few beers and it just will not hold up to well. I say figure out your average listening levels and report back. Download the app and let us know, then we can roughly give you a power rating you want to shop for and go from there.

I agree to a degree Yes tubes tech peaked a long time ago and almost everything has been done. But because we have access to better parts and all sorts of silicon to help we can improve on designs. Yes they used Pentodes for current sources as loads but now we can cascode two cheap transistors and drastically increase AC impedance that far out performs a pentode. We can build far better active regulators now with sand. Although we are not reinventing the wheel we can continue to improve upon performance. Actually one thing that was not done back in the 50's or 60's was active impedance bridges like the David Berning ZOTL amps. Yes they had OTL amps but that was paralleling ridiculous amounts of tubes and a total PITA. Now we can cutout the transformer and change impedance all with transistors.

Some older circuits were using pentodes as constant current sources but today we have the means to get extremely cheap transistors to cascode them into a much higher AC impedance for plate loads. Using LM317's etc... for cathode bias I don't like, it creates more problems than helps. Long tailed pair phase inverters LOVE CCS as the tail load but requires a negative rail which is not hard to do. For directly heated tubes like 300b, 2A3, 45 etc..... a current source is MUCH better than AC and DC voltage sources. The Current source on filaments avoids nasty in rush currents when filaments are cold. 90% of DHT failures I see are the filament burning out, using a current source they will last A LOT longer until their emissions run out. I even used Mosfets as source followers for screen drive on sweep tubes, easily got 100 watts from just two tubes push pull

I personally prefer single ended amps. I built a Brook 12a a long time ago and it did sound great. I was just saying Klipsch should partner up with an amplifier manufacturer to re-make the Brook amp with some upgrades to the circuit because PWK himself was very fond of that amplifier pairing with his speakers. I can see it now, "heritage series amplifier" to be paired with Klipsch "heritage series" speakers.

The Grommes amp shows both negative voltage feedback and positive current feedback to increase damping factor. The Bogen amp shows the same but instead of fixed positive feedback it shows a simple variable damping control. I think it would be really cool for Klipsch to make a "heritage amplifier" specifically for their heritage speakers based on the Brook amp PWK himself said was the greatest amp for the K-Horns. Wink Wink to Vox to hire me for the design

Nelson Pass uses positive feedback for damping control in his amps like the F7. There are many vintage tube amps with variable damping. From memory PWK said you want the highest damping factor in 1956 and setting anywhere else made the sound worse. He was referencing the type that has both positive and negative current feedback and so negative current feedback would reduce damping so he preferred the Klipshorn with the most damping so he would only need positive current feedback to increase damping factor. Remember this was the 50's and since they really only had tube amps with output transformers you couldn't really use more than -30db negative voltage feedback before stability starts to become an issue, so output impedance was somewhat limited unless you were also using positive feedback to increase damping. I like to have the option to have both +/- current feedback but the point of all this is discussing damping of tube amps and that people believe we are limited to a minimum DF. This is untrue because with positive feedback we can achieve better damping control vs solid state, but modern tube amp builders are not using this technology to improve performance as I feel it is very misunderstood.

For us Klipsch users with sensitive heritage series speakers wires don't matter as much because we are not using 100's of watts of power and the speaker cable does not see large currents. With very low currents we do not see the losses across the wire. Same for the "Skin Effect", you will not notice anything different between stranded or solid wire because at audible frequencies the skin depth is still deep enough to not need anything special, especially since we push very little signal current from the amp to the speaker because of their efficiency. So for people probably thinking I am crazy about being able to increase damping performance and actually overall low frequency performance with variable damping control here is a link about the subject. The cool thing is they are referencing the Klipschorn for the reading material. The article actually references "Variable Damping" written by Paul Klipsch October of 1956. I am still looking for the actual article about variable damping that Paul himself wrote so stay tuned as I will post it as soon as I can find it. http://www.rfcafe.com/references/radio-news/positive-curent-feedback-november-1957-radio-tv-news.htm

Skin effect is real and we pay close attention to it with things like radio frequency equipment and microwaves but for audio frequencies it really isn't anything to worry about. I do see some exotic speaker wires bringing attention to skin effect and why their wire is superior causing many audio enthusiasts to lose sleep over it. Let's all pretend we are 14 years old again and can hear up to 20kHz, yes at the higher frequency the skin depth has decreased (I won't get into the math) but we need to keep it all into perspective. Easiest way to think of the skin effect is similar to using smaller gauge wire with higher currents. The resistance of the wire at smaller gauges is increased and depending on the length and the current you will get a loss across the wire. With skin effect the skin depth decreases as frequency increases which means the signal uses a smaller portion of the total conductor. What this does is increases the impedance effectively giving a loss across the wire. Why this effect does not worry me? First of all the increase in resistance at 20kHz is not enough to notice especially by ear, the losses are very small and inconsequential. Why inconsequential? Have a look at the size of the tweeter on our speakers. The driver is tiny compared to the woofer and the power rating of the tweeter is much much smaller because in music there is really not much content up there in frequency. The tweeter does not have to dissipate 100 watts like the woofer. For example the output of your amp may be at 100mW for the tweeter while the tweeter sees 99mW from the loss of the conductor. Worrying about losses at bass frequencies will trump losses from skin effect at high frequencies every time. But that doesn't stop some companies from making you worry about the detrimental skin effect, yes it is real and yes it is measurable, barely but when put into perspective isn't nearly as troublesome as some companies lead you to believe. I tend to focus on things that are a magnitude greater importance to improve my sound quality. So for the grand scheme of things solid wire seems to perform better than stranded wire so I go with that as it has more pros vs cons.

I do not own a specific First Watt model but after Nelson made his static induction transistor amps and talked about them at the burning amp festival I had to try one. Very simple design with only 4 giant Mosfets for both channels and an input transformer. Two Mosfets per channel, one acting as a constant current source load for the other it is a single ended arrangement with "Schade" feedback. This drain to gate feedback lowers output impedance and gives a triode like output characteristic. The 1:2 input transformer processes the feedback in a boot strap fashion in order to increase input impedance instead of lowering it making it much easier to drive with normal sources and preamps. It can produce up to 50 watts before clipping but distortion is high at that power level but at 20 watts it is around 1% THD. It may not be as magical as some real SET amps I have had but it is darn close and much more power vs 2A3 and 45 amps which is attractive to many. Big heat sinks are needed as the Mosfets are biased around 3 amps ~85 watts dissipation. Not your average SS amplifier by any means I found they were a great pair with Klipsch speakers, currently running on Reference Bookshelf's but I really liked them on my Cornwalls. Also great on my La Scala's but they hardly ever go above 1 watt and only for dynamic transients so I run actual triodes with them. I am surprised more Klipsch owners do not use Pass amps. I am very new here so maybe I could be wrong.

Current is what moves a speaker not voltage. As you can see with a voltage source the voltage does not change but since the resonance has a very high impedance the current has decreased substantially through that region. With a current source amplifier with a high z out the voltage actually changes along with impedance, so less impedance lower voltage, higher impedance higher voltage to keep the CURRENT the same. How does this add up to sound? A current source or transconductance amplifier will track the impedance curve and distribute the power evenly across all frequencies no matter the frequency/impedance. The thought is 1 amp of current for example will move the motor the same no matter the impedance. Since it is the current and magnetic field that is moving the motor you would want to have a current source. Unfortunately most speakers today are designing their speakers for a voltage source and not a current source things do not work out how they should shown by not only looking at the impedance vs frequency plot but the SPL vs frequency plot. The sound pressure or sensitivity changes so with that high impedance bump in the bass region does not need the same current to reach the output level as other frequencies closer to nominal impedance. So with a current source amp you may have very high increased "one note" bass where the resonance is.

Solid single conductor like Romex.

I agree my La Scala's seem fine with any good amp in both SS or tube camps. Possibly because I keep my 14awg solid copper speaker wire as short as possible. I found in the old days doing pro audio when we didn't have self powered sub-woofers I had some Peavey power amps that had variable damping control to make up for variables in speaker wire impedance on long runs and different speakers. It actually worked pretty darn good, I did notice that with long runs the bass would sound off if I didn't adjust the variable damping. Other power amps that didn't have the feature seemed to be hit or miss with long cable runs and bass performance.