tube fanatic

Found this in a Radio Maintenance magazine from 1951. Maynard

The easiest solution is to order some JJ 12dw7s from Tube Depot for $12.95 each. Maynard

There are 2 ways to reduce distortion in SEP amps. One is to use it at a fraction of its maximum output power, and the other is to use nfb. I prefer the latter (as voltage fb) as it reduces the typically high output impedance as well as increasing the bandwidth. If applied within reason, it does not ruin the resultant sound in my experience. In fact, it can provide a very triode like experience in some systems. Maynard

Class AB refers to a specific set of operating conditions which are usually somewhere between class A and class B. So unless the amp has provision to change that, it is running in AB at idle through its maximum power output. It combines the advantages of both classes while minimizing their respective disadvantages. Maynard

Here is the perspective of an EE from the Steve Hoffman forum. I agree with him. I believe that using multiple caps in parallel, regardless of type or price, craps up the sound. My ENT doc friend showed me some frequency response plots of typical males in their mid to late 70s which have convinced me that some of what we believe we hear is wishful thinking. Also, the nonlinearity of the human auditory system is so great, and so varied from one person to the next, that universal recommendations about audio components are worthless. Maynard John DysonForum Resident Location: Fishers, Indiana If I was making the decision for myself, I wouldn't change the cap unless I had a concrete reason to do so. Some people like to change things -- and that is fine also, perhaps part of the enjoyment. Put my EE hat on (I am an EE RF/Audio/Systems, SW Operating Systems/DSP -- and generally pretty good at all of them ), so here is a quick (back of scrap pad of paper) analysis of the situation: Lets see -- cathode resistor/capacitor (I dont know the tube, so I'll guess a gm of about 2mmhos or 2mS.) That would imply an effective resistance of about 500ohms at the cathode (without bypass.) With an electrolytic with perhaps 0.1ohms of ESR, lets guess at 10% of that is nonlinearity. That would mean a delta of 0.01ohms with a 500ohm impedance at the node... The additional amount of distortion caused by the electrolytic changing by the 0.01ohms would be less (much less than) 0.002% additional distortion, and I am guessing VERY, VERY HIGH. In reality, the amount of distortion caused by the nonlinearity of the resistance would be pretty small. Now, we have an issue about the nonlinearity of the capacitance of the capacitor, and the best success there is to make sure that the capacitance has a reactance that is very small (relative to the lowest audio frequency) in relation to 500ohms. Also, unless the tube/valve is in a negative feedback circuit, the changing gm will totally swamp the nonlinearity of the capacitor (either the resistive or the capacitive reactance nonlinearity.) If I was making a design decision, I would simply try to choose a reasonably high quality capacitor with fairly low ESR -- the rest will take care of itself. The difference in distortion of a reasonably good electrolytic vs a perfect capacitor will be vanishingly small, and probably not even measurable. Of all of the things that might need to be improved (other than as a failed/degraded parts replacement) I don't think that changing the capacitor would be all that helpful. Audio source material and transducers (speakers, headphones, microphones, etc) are much more important. John Dyson, Apr 25, 2018

Here’s what Rozenblit had to say about OTL output capacitors on the Transcendent forum: Maynard Bruce Rozenblit03/29/04 There seems to be a lot of confusion about what an OTL is on other message boards. OTL stands for output transformerless. The idea is to eliminate the output transformer from the signal path. That's what I do. It is a very difficult task because of the impedance mismatch between tubes and speakers. There also seems to be a great deal of resistance to the use of an output capacitor. Many people for some unexplained reason, have gotten the idea that an output capacitor will greatly injure the signal. That is baloney. A transformer is about 1000 times more injurious to a signal than a capacitor. Most tube amps have output transformers but that is OK. Most speakers have internal coupling capacitors but that is OK. When an amp has an output capacitor, its the end of the world. I don't get it. Many people have spent a couple of hundred dollars to replace the stock caps in the SE OTL with black gates and all have reported back that there was no improvement. All have wasted their time and money. There is a big difference between conceptual idealistic notions and what works in the physical world. I concern my designs with what works in the physical world. The T16 is direct coupled and the SE OTL sounds even better but it has an output capacitor. Its the results that matter, not the concept. QUOTE

There’s no question in my mind that the OPT is the weakest link in a conventional tube amp design. Anyone who is willing to deal with a larger number of tubes should certainly consider OTL. If I weren’t so cheap, and did not spend my time with similarly minded audio folks, I would do a few OTL designs myself. Maynard

Richard, I am not the least surprised that you enjoyed an OTL amp as many audiophiles consider them the best. Having a capacitor in the circuit does not in any way degrade the sound. Remember, interconnect cables are capacitors as are the tubes themselves! I am sure that Ralph Karsten and Bruce Rozenblit, both of whom sell some excellent OTL designs, would have a great deal to say about this. All amps are flawed in some respect, and OTL designs are certainly no more flawed than any other. Maynard

His description of his early days in audio are a riot. I totally agree with his conclusions about mono listening. I went back to full time mono, with a single speaker in the corner, some time ago and will never go back to stereo. Even my collection of poorly mastered stereo cds sound terrific when listened to in mono. http://www.drowninginbrown.com/dib_sp.htm Maynard

Actually, I think Mike Stehr’s builds are nicer than mine! Maynard

I have always thought that PWK favored Brook push pull triode amps but could be mistaken. Maynard

This was done in 1956. The results were not stated in the article. I’m surprised that such powerful amps were used considering that PWK had once stated that far less power would be needed to fill a huge concert hall. Maynard

Thanks for posting the link Carl. One of the OPTs looks like it was replaced, and the other presumed original looks a little questionable from the picture. For what he is asking, I would pass. Maynard

Fifty bucks could cover any needed parts if both output xfmrs are in good shape. A tech’s labor would be another matter. 3 watts would power your speakers, but the unknown is whether it is enough to satisfy your listening levels. You need to be careful with stuff like this since many have been stored in damp basements or sheds for long periods of time. Can you post the link to the CL listing? I can look at it this afternoon. Maynard

If I recall, this one uses a single ended 6v6 per channel for about 3 wpc. Correct me if I am mistaken. Has it been restored professionally? If not, you should factor in the cost of that. Can you post pics of the amp chassis? Would the seller let you pull the chassis to photograph the underside? The gamble here is whether an output xfmr, or the power xfmr is blown. Maynard

Maynard

A comparison with the SYS isn’t valid since it has no gain. The gain of the GG is probably higher than that of the Adcom as a design choice. As I recall, its maximum output voltage is far greater than that of the Adcom. The position of the volume control does not correlate with output voltage. That can be measured if you have a signal generator or test cd. Maynard

Jeffrey, I am surprised you are discussing your amp in a thread about George Wright’s amp. I am confused! When evaluated at Dennis’ house you and the others concluded that your KT88 amp sounded better than any amp in existence (including Dennis’ SET). So how did it go from that to almost unusable to hear at its best? Also, why does a cable need to be operated at a “higher level of drive” to sound its best? Are you saying that a cable performs better at the 2 volt RMS maximum output from a CD player than it does during a quiet passage at, say, 0.2 volts RMS? Please explain the physics behind this to further my understanding. Maynard

The 12AU7 is a fine tube which will not detract from the sound quality of an amp. If you are pleased with the sound I recommend that you do not seek modifications which are based on the opinion of someone whose auditory system is totally different from yours. Maynard

This is where we definitely part ways! Resistors in parallel do not “fight” one another as the voltage across them is identical. It does not matter if you use 50 in parallel- the voltage across each is the same. As a plate load for a voltage amplifier, the tube will not know if you use a single 100k resistor, two 200k resistors in parallel, or 500k in parallel with 125k. The dc plate voltage and ac signal voltage will be identical. I, and every amp designer or builder I know, cannot hear the phenomenon you describe. And we have tried this many times to determine if what you are saying is real. Have you ever evaluated your concept blind? If not, have a trusted friend do some changes for you to determine if what you perceive is real, or wishful thinking. Further, what is the advantage of using resistor wattages vastly greater than what is needed? What is a “skew” in the music? Maynard

I can’t count the number of times I have experienced this. Maynard

The best thing is to call Craig and ask him about this as well as the best kt88 tubes to use in his amps. We designers get insulted if folks ask others for advice about our creations 😀😀😀 Maynard

Of course! We know this. I want Jeffrey to explain why he believes his assertion to be true. I am also confused by some of his recommendations which presumably are meant for hum reduction. Yet, he insists that some hum is needed for music to sound good. I look forward to reading his responses. Maynard

Nothing to add. Tubes and Klipsch speakers make the proverbial marriage made in heaven in my experience. Definitely post your impressions! Maynard

I share your frustration. Occasionally, Barnes & Noble or CD Universe has had better prices. It seems that those in charge are trying to force streaming as the only source of music. I prefer a physical medium. Maynard