Jeffrey D. Medwin

Hi Les, Two Type 45 structures in a single bottle ( parallel connected internally ) is a " bi plate 2A3 " tube. It is my understanding, early 2A3 bi plates, ( introduced after the original 2A3 - which was made with a HARP filament inside a single plate ), were developed to cut-costs by tube Manufacturers. Biplates are considered to not be the sonic equal of the original mono plate with HARP filament. Tube Manufacturers simply put two Type 45 plate and grid structures, connected in parallel, into a single ST shaped bottle. Les, can you spot the EIGHT STRING symmetrical filament structure, fine wires, ( and their easy-to-see supports ) in this 1990's AVVT mesh plate 2A3 close-up I just photographed, as an example of a " HARP" filament structure? Today, moderately priced, readily available JJ 2A3-40s have a similar harp / symmetrical filament . As offered by one or two others. Some people go "GA GA" over this AVVT meshplate tube. Used or NOS pairs are rare, cost 1,000s of dollars, as they are no longer made. It IS pretty, and impressive to see !!! Not to change the subject of Type 45 tubes, but for output tubes, these days, I satisfactorily prefer this : I am very pleased in particular, with how this tetrode effortlessly plays the transition from the midrange, to the lower and middle highs on cymbals, energy-delivery-wise, to my compression driver . ( As the output tube - in a DIY Loftin-White SE amp design ). Demanding drum kit playback, has never sounded more realistic - to me. This tetrode's extra grid, G2, as configured in the new circuit, does some good things ( that my previous triode tubes.......eg : the deluxe 300B-XLS, ST Type 45s , JJ 2A3-40s can't do ) ! Also of note, .....these E.H. KT88s have a nice cathode structure - " ample ". Your mileage, with a similar tube however, is going to vary, due to our circuit differences. Performance differences in my DIY creations - are primarily power-supply related. Also, parts choices and energy-transfer-efficiency ( wiring, internal ), play a role. Everything matters. Although this thread is about the Type 45 low mu triode, it is not just the Finals tube choice, which is important to everyone's amplifier's results !! As a matter of FACT, with a low mu triode output tube ( having a gain typically of about 4 Xs ), about 85% or more of what you always hear through the entire triode amp, is the front end stage(s) implementation. Why ? The front end may have a total signal gain of 35 to 100 Xs, VS: 4 Xs for the Finals tube. How carefully, and how thoughtfully, do we implement the amplifier's front end tube ????? If we skew the music signal in any way, or loose ANY information at all in the Input stage, the signal is changed forever. It never ever gets to be " made up ", by the low mu Finals tube, ............does it ? Extraordinary thought and care is required with a tube amp, and most often, something gets overlooked .

Having designed and built a number of these amps, ( such as described - in the 1995 Sound Practices article ), I thought I would offer some practical written suggestions, to any persons reading this thread, who may possibly be contemplating building such a mono amp, or a pair, today. This should save a DIYer money and improve either reliability, performance and / or both. We have learned many things in the past 25 years, about this specific type of direct coupled circuit. Today, we enjoy different / better parts choices, which often are more performance - effective. Here we go : First, build it simply.......... as ½ a 12AX7, DCed into a JJ 2A3-40. Mono amps only !! Order, for the output transformer, either a Softone RW-20 ( from Softone ) or a Hammond 1627 SEA ( from Hawk ) . Change the V1 rectifier tube, from a mercury vapor Type 83 ( no mercury tubes in your home !!! ) to a E.H. 5U4GB. It is very important to eliminate R1, a 330 Ohm power resistor, in series with the B+ to the finals. Use a Hammond 274BX power transformer. ( Hawk Electronics ) . By using the Hammond 274BX, a 375-0-375 VAC rated power transformer, ( instead of the Italians' 410-0-410 VAC custom-made tapped unit ), we can nicely eliminate unwanted DCR ( D.C. Resistance ) feeding the 2A3 Finals tube. 330 Ohms that was present in series resistances, is GONE..........superb. ( No R1 ) CAPACITORS : The men from Italy in the 1990s used very expensive CSI metalized Teflon, and also Polyester film capacitors, as their “ main or base cap ”, in every “ C” location in the circuit. They also ingeniously bypassed these with a 0.22 uF Paper in Oil, and a 0.022 uF Silver Mica. FYI : The 0.22 uF bypass capacitor value enhances the amplifier's all-important middle midrange. The smaller 0.022 uF bypass capacitor value enhances the middle of the high end ( cymbals, etc ). In 2020, I would NOT look to use their 1980s-1990s chosen expensive Teflon or Polyester caps, but instead substitute modern WIMA DC LINK film caps, ( the ones with FOUR pins for their output connections, not two pins. ) These are industrially priced, employ modern cap-design technologies from Germany, and may likely outperform MUCH more expensive choices. I use such WIMAs often, as a go-to “ main or base cap ”. The WIMA DC LINK caps with four PINS exhibit no detectable / audible "storage-time" anomalies to us. Very nice. In my limited experience, a great sounding 0.22 uF bypass capacitor available today would be a Mundorf Supreme Silver and Oil. It is VERY special-sounding in the midrange, and it is costly. C7, ( the new 50uF cap ) feeding the plate of the 12AX7, in particular, certainly needs that MSSO 0.22 uF cap as a bypass. This specific capacitor, in that critical circuit position, will really perform !!! And - have the 0.022 uF ! For the Italians' specified 0.022 uF value, I suppose Silver Micas will do. Be sure it is high-voltage rated. Microfarad VALUES : In general, keep the main or base caps, in any one circuit position, at 50 uF maximum, film caps ONLY, and you will be fine. Here are some newly re-computed capacitor values, with important design changes applied : C1, eliminate it completely. C2, change from 100 uF to 35 uF. C3, change from 140 uF to 50 uF. C4, 15 uF retain this value, order specifically the WIMA 4 PIN 15 uF at 900 VDC rated cap, a lovely performer. C5, change value from 15 uF to 6 uF, a quality film. C6, change from 47 uF to 15 uF ( 900 VDC 4 Pin DC LINK again ). C7, change from 15 uF to 50 uF. ( 4 Pin WIMA, 50 uF at 800 VDC in particular, is appropriate ). RESISTORS : R4, the 12AX7's plate resistor, denoted as Ra, and R3, this tube's cathode resistor, denoted as Rk, specifically should be either Roderstein Resista MK8 two Watt, or Mills MR-200-8 two Watt resistors, available from Michael Percy. There exists no other choices for this circuit location, I could recommend. ( No carbon resistors, in any circuit position, please. ) CHOKES : By eliminating R1 and C1, we have neatly converted the topology of their power supply into a "double pi choke input filter." It can be expressed in E.E. symbols as L1/C1/L2/C2. This filter topology, after the rectifier tube, is the most desirable topology possible in E.E. to apply. ( It is typically done with L-Critical value chokes, of high DCR .) Sadly, ( and common even today ), the Italians were unaware of low DCR power supplies. Their 10 HY chokes, are each about 155 Ohms in DCR. Two such chokes would be 310 Ohms total of inductor - associated DC resistances. *** Inductors in 2A3 amps ( actually, ANY good tube amp ) need to be 20 Ohms or less, IF you want the amp to play a grand piano, or reproduce a drum kit properly !!!!! *** Today, I would estimate over 99 percent of tube amps in existence, fail to meet this 20 Ohms or less parameter. A shame !!! Also, what ever choke you use as L1, MUST be duplicated for L2, or .... audible skewing in a Single Ended amplifier's power supply, is always heard, and this is unwanted. Make L2 = L1. In such a mono amplifier, the very best inductor choice I can suggest, in 2020, is a Hammond 159ZA, for L1 and L2. It has a 6 Ohm DCR . Two in series, will be 12 Ohms of series inductor resistances. The 159ZA needs 75 hours minimum, to break in, to sound proper. A 159ZA inductor does best with m22759/11 Mil Spec wire-augmented leads, in and out. Do you recall what wire the two Italians used, to build their amps ??? ( the same ! ) ie: " We used cables of aeronautical military specifications in silver plated OFC. They cost less than "signature" cables intended for the hi-fi market and offer high quality musical performance. " Next, ...let us compare these two discussed power supply filter circuits, for the output tube, on the basis of DCR : Stock Marzio-Jelasi amp had a 330 Ohm power resistor, R1, ( which gets immediately eliminated ), and two 155 Ohm 10 HY chokes. Add their resistances together please ................ 330 + 155 + 155 = 640 Ohms. Suggested : is 6 + 6 Ohms = 12 Ohms. The 2020 suggestion is 53.3 times lower in series filtering resistances, as it feeds the 2A3 output stage. Do you think the power transformer and tube rectifier will have an easier time, powering the 2A3 playing a grand piano, or a drum kit, pulsed music information, through 640 Ohms of filter resistance,.... or .....12 Ohms of resistance ????????? :-) If you have ANY questions about what is suggested above, regarding chokes, simply read ( the past - Dean of E.E. at University of Alabama ) Dr. Charles A. Halijak's “ Figure of Merit of a Power Supply “. It's been posted twice - on this Forum. Look it up. Allow it to sink in. The twenty five cent word is inculcate. Practical considerations, and a WARNING : Please note this, *** I have NOT built the Marzio and Jelasi circuit as they suggested, NOR..... as suggested herein. *** Whoever makes these changes, ( of which I am confident ), will likely want to co-ordinate with me privately. A final schematic and a listening report of the completed amplifier, can certainly be posted, on this Forum for those who can DIY. The simple amp suggested herein, once built, will need to be carefully voltage-measured, and analyzed, to determine if the Direct Couple, with the values suggested, requires a resistor value or two, of change. Before a person starts on this, please contact me, Klipsch private mail, so we can coordinate communication with each other. I will help others for gratis, if someone is serious, and if I am able to do so. I have the right to choose a person I would feel comfortable working with !! I have from-scratch designed at least two dozen direct coupled amps of this type, and constructed nine such amp types, each slightly different, in the last 15 years. I urge folks, to NOT casually attempt to replicate the original amp, strictly from the Italian's 1995 published schematic and parts list. Hence, .... this post !! Kudos, to the Italians' 1995 written descriptions, superb. I loved it ! http://www.ispra.net/audio/DCSingleEndedAmplifier/index.php EDIT 9-19-20 : Upon further studying the Italian's circuit, there exists additional parts that can and should be beneficially eliminated !! KISS rules !! ELIMINATE totally R6 and leave the previous path UNconnected . Eliminate the Milliampere Meter ( but NOT the connection ). C4 should be eliminated totally, it is unwanted - an unnecessary feedback loop, cathode to B+. Now, upon removing these parts, and thus even further simplifying this simple amplifier circuit, it becomes a slightly better amplifier in my opinion. Jeff

Thank you.

Nelson Pass's concluding statement, his feedback article : ( Suggest you read it all, very good. ) Another point of Nelson's I love, keeping in mind that triode vacuum tubes can be fairly linear , VS other amplifying devices : I always like to say, " All the King's horses, and all the King's men, could never put Humptey together again. " You may deduce easily today, who I learned that from !!!!!

Yes Mike. More often in commercial amps we see an electrolytic across the Final's tube cathode resistor. This could be improved upon, in my own opinion. How do you like to usually bypass a 2A3's Rk ?? Jeff

For your information, I hope some enjoy this person, a perspective, and that some get something out of it . This was taken from Sound Practices Magazine, Joe Roberts editor, in 1992, from Joe's very first issue: cent genius,gave us the vibrations on foil. ty. The minute we did that we were re- Note please : Robert Fulton was a musician. He played piano, coronet and trumpet, and designed trumpet mouthpieces, by ear. Doc Severinsen used Robert's mouthpieces on his instruments. https://en.wikipedia.org/wiki/Doc_Severinsen

Coytee, I know you're not technical, and I don't think you realize this, but likely every single tube amp you have ever owned, and heard at your home, has had negative feedback ( NFB ) in it . I believe, ( am not 100% sure ) that all the OTLs you have asked me about, need to use NFB. The Rosenblit SE-OTL , for sure, uses NFB. The George Wright 2A3 amp didn't need it, but it has NFB, as part of the topology. A Loftin - White type amp, typically, has zero NFB. ***** With zero NFB, we hear the music, in the proper timing of the music, not in the amp's timing. ***** A zero NFB tube amp can be like a breath of fresh, natural, air ! The first dictate, from my musician-audio mentor in 1982, was " Be sure to design the amp so it does not need to use negative feedback ". The second dictate was, " Be sure the chokes, in a 2A3 amp, are 20 Ohms or less, IF you can find them. " Since 1983 Coytee, all my tube amps have had zero NFB, and the chokes were all 24 Ohms or less. The G.W. 3.5 2A3 amp, in addition to having NFB added to it's circuit, employs a 195 Ohm Hammond ( 158M ) choke. That is approximately ten times too high in DCR - from ideal !!! I have brought up this discussion of NFB and DCR, ( in relation to tube amp design ) to point out to you, and all others, that there exists a big gap, between commercial tube amps typically available and used in homes, and ..........what can be achieved. A Hammond 159ZA choke, an available part for about two decades, is 6 Ohms in DCR. Employed in a double pi series B+ filter ( L1/C1/L2/C2 ), there exists a total of only 12 Ohms of inductor resistances - to power the output stage. After about 75 hours, ( to break them in ), it does a wonderful job, particularly in delineating the bottom-end of a well recorded grand piano. Jeff

What caused you to sell your four SE-OTLS ??

Yes to the above, thanks for highlighting it Travis . .......... And this is precisely why this ( above ) circuit is ALL I ever DIY build, for my personal-use amps, ever since 2005. ( started building this, right after reading the first on-line " 6 - Moons " RMAF 2005 review by Stephaen Harrell. I was already well-acquainted with the above 1995 Sound Practices magazine write up ). 15 years, ....... working on the SAME basic circuit, ( often called Loftin-White ) , learning ......how to design it, build, and upgrade it. Here is an actual build progression, of those " Loftin-White " designs and builds in my own home : from 2005 to 2020 : First - a super-deluxe 300B-XLS amp, next - two Type 45 amps, a Type 46 amp, two Type 2A3 amps, two Type 6005 amps, a 2019 KT88 stereo amp, and now, coming up, late 2020, a rebuild into KT88 and / or KT150 monaural amplifiers. Ten Single Ended amps I've designed, over the past 15 years, ..... ............all eleven amp builds share the same circuit topology, as the amp in the post highlighted up-top, by Messers Marzio and Jelasi : Two-stages,....... directly-coupled, ........zero negative feedback.

Hello David, Nice to see someone experimenting with this !! I am not 100% sure, but it looks to me like R6 is associated with R5. They are part of the amp's power supply, in that they are a voltage divider, off of the amp's highest B+, the B+ going to the output transformer. I believe the ratio between these two Rs is establishing the B+ voltage for the front end tube's plate resistor, the Ra. Notice, R6 and R5 are 3 and 5 Watt power resistors. This would be easier to define if they would have posted VDCs and VACs on their schematic. !! - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - An appropriate high-performance output transformer for a 2A3 tube is easy in 2020. My overall favorite, a Magnequest DS-025, is no longer being made. My 2020 choices, are in this order: 1) A Softone RW-20 on its 2.7K primary tap, and 2) a 2.5K Hammond 1627 SEA. We have heard world class 2A3 amps play in Montana 14 months ago, using all three of these specific SE output XFRs. I am particularly fond of the latest Softone, because it sounds to be LOW in distortion, and was very pleasing for us to hear, ( being an R-Core XFR, the primary and secondary windings are on opposite sides of the core, not wrapped over each other as in an E-I , and the measurement graphs show desireable - very smooth rolloffs at its extremes ). See this David : http://softone.a.la9.jp/english/RW20.htm Also, do download its Manual, and read all about it. It is reasonable - about $110 per transformer, very reasonable, cost VS performance. Takes 400 hours, to FULLY break in ( bass extension ). A second good choice, is the 2.5K Hammond 1627 SEA, and HAWK has the best prices, at $129.80 each, two at a time. See this David : http://www.hammondmfg.com/1627.htm One of my audio buddies prefers his 2.5K Hammond SEA, to the "Level 3 " ( highest level ) O-Netics outputs he has owned. The O-Netics he paid $600 each for. $129.80 VS. $600. He even bought a second pair, after eBaying the O-Netics Level 3s. I would do the Softones for their sound, but either will be fine !!! Look at the specs for each, as they both differ in primary and secondary configurations DocC. Flip a coin !! As for updating the Italian amp, either contact me through Klipsch P.M., or at my usual email address. We should arrange a time to TALK on the phone about the Marzio and Jelasi amp in 2020, to be able to communicate most effectively. Typing is tiresome !! On the phone, one can spontaneously exchange information, the easiest. Jeff Medwin

My take away from Rosenblit's SALES blog : 1) Bruce certainly seems to tell us above, that his SE OTL amps have some sort of Output Coupling CAP, unlike what has been posted earlier, above. These were words of the Manufacturer, who should know his own circuit topology. Any OTL that is direct coupled to the speaker, ( has no cap couple, ie: his T16 ) is a push-pull design. This requires a phase inversion, which is typically, an extra stage, and always ADDED/ extra circuit manipulation, ...which the SE amp totally avoids !!! 2) Of course Black Gate WKZs didn't help, if the overall amp itself is built to a popular price point, and likely has enough price-sub-optimizations throughout, to be beyond help. 3) A OTL amplifier begins its mediocrity with the basic OTL topology - pointed out in 1) above, as being imperfect, too complex, ( an output cap or a phase inversion added , and the use of negative feedback loops ) ................this is problem Numero Uno !!! 4) No discussion of the OTL Transcendent amp means anything ( to me ) when my two stage DIY 12AX7-6005 Direct-Coupled Single Ended zero negative feedback amp far far outplays the Transcendent OTL, (except for power - 1/4 Watt VS. 2 Watts), in actual listener testing. The proof of it all, is how it sounds. Due to it's OTL topology, as mentioned in 1) and 3) an OTL amplifier can become a " Grade B " amplifier, at best. Nothing can ever compete with " Grade A " amplifier topology. ( two tube stages, directly coupled to each other ) . " Grade A " topology, is a take -off of the 1929 Loftin-White topology. It is the simpler. An example, from 1995, is nicely written up in different / current thread, entitled " A Simple Triode Amplifier". 4) Bruce ends his sales pitch........ " Its the results that matter, not the concept". ...........True !! For results of the Transcendent OTL : read how the amp Bruce designed and sells compared, to a simpler two stage DC amp . Simply read the Feb. 9th, 2019 email report, embedded in this post : https://community.klipsch.com/index.php?/topic/194394-jeffrey-d-medwin-otl-question/&do=findComment&comment=2534656 That email says it all !!! WHY Coytee should I - on this Forum, or anywhere, endorse " Grade B " solutions? No sir. Realize this : KNOW that " Grade A " solutions exist ( as I claim, and as this ( now "ex" ) Transcendent OTL amp owner reports to us ). A simpler circuit, when well executed, can EASILY be markedly superior . The output transformer is not the problem. Eliminating it is a wrong solution, as it serves a good function in matching impedances. Results matter !!

Are you talking about a bypass capacitor across the cathode resistor??? If so, most SET amps I see, maybe, just guessing 98%, regularly use electrolytic caps across the Rk, as a bypass. IT is IN the circuit !!!! Would these same people use a high quality electrolytic as a coupling cap. Why...NO !! But they unconsciously use electrolytics across the Rk, of a 2A3, in its cathode return to ground. That seems thoughtless to me !!! When I bypass the Rk of a DHT FInals tube, I use films, with as "S", because it takes about five high-quality film caps, of different types and uF values, in parallel, to get the Rk bypassing musically correct. Also, it takes a lot of money, to do it right.

Appropriate immensely to this thread and discussion, a Klipsch Forum post I wrote in April 2019 : Most people have little or zero experience, direct listening wise, hearing a well executed amp with LSES topology Erik, is Transcendent Sound the same company run by Bruce Rosenblit?? If so, we are in luck. From 7-3-18 to 12-31-18, I prototyped a SE triode-connected 6005 DC amp, with a 159ZA LSES supply.. Only seven people have ever heard it. One of them, in February 2019, was a DJ and astute audiophile that had a low powered ... name....?? Transcendent Sound ?? amp on his reference Lowther system, it was a maybe 2 Watt OTL amp. Let me go to my email directory, and try to find his words to me, upon hearing the 2018 DC 6005 SE prototype. I'll post this entry now, and add his email comments below in this post, later. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Here we are, yes, he did use and A-B my 6005 SE 2018 prototype amp to his existing Transcendent Audio OTL: This 2018 SE DC triode-connected 6005 amp is only 0.25 Watts clean, 0.33 Watts maximum, and he was using a T.A. 2 Watt amp. He still wanted to hear mine, and he is reporting back to me, and to my audio friend / mentor Dennis. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - The person who wrote that email to me (above) is NO FOOL. He at that time, was the President of an Energy Company, owned by Bill Gates, and was using special Lowther speakers. Since my amp, at only 1/4 of a Watt ( 6005 triode connected ) was not sufficient to power his speakers, he found a best solution. He contacted Dennis Fraker, my Mentor, and bought reasonably, a used pair of monoblocks, customer-trade in of 2A3 two-stage DC amps, which Dennis had refurbished. This person has been delighted, listening-wise, since then. When you move, as he did, from only hearing a Class " B" scoring amps ( Transcendent SE OTL ) , to Class " A " scoring amps, there is NO going back !! Jeff

Coytee, As I said, I understand what you heard, and think its good that you trust your ears. You asked me, and I gave you my opinion, from me having experiences, and hearing amps, the likes of which, you have never heard. If all you have been exposed to, and owned, is tube amps that score a Grade " C ", ........any Grade " B " amp, (in this case, hearing your SE-OTL ), will be the best you heard. But a Grade " B " score, is NOT the best score possible, is it ???? I have a different set of listening experiences. I have DIY built for my own use, ( and have heard manufactured ), Grade " A ", amps. World-class amps. This is all that me, and many of my closest audio buddies, ever consider using . I find the world's BEST possible sounding amps always have the same topology, ( and...... it is never OTL ). The output transformer is not any big problem, IF......the basic topology has STARTED OUT right, ( the proven topology ......... two-stages, directly-coupled, zero negative feedback. ) A three stage amp will never do what a two stage amp will do. An amp using a coupling cap, or an interstage transformer, will never do what a directly-coupled amp will do. Simple and UNBEATABLE, when done well : 1) 2 inch Silver wire, RCA Input jack to GRID of DriverTube 2) Silver wire, PLATE of Driver Tube to GRID of Output Tube 3) Output transformer, from Plate of Output Tube, to power the Loudspeaker 4) No negative feedback There is one other aspect that I failed to mention. OTLs often require paralleling of tubes. As soon as you parallel tubes, you have a degrade in the best possible performance, a " choir" effect. This is easily noticeable ( IF you are used to using Grade "A" scored amps with well-set-up, highest efficiency speakers. ) So, add to our list " Any amp that is an OTL, will not be able to do what a two-stage direct-coupled amp zero negative feedback amp can do. " Why, ....... because of the OTL's typical use of paralleled tubes. If all of a person's experience and lifetime has been mired listening to Grade "C" score amps, ( because you have insufficient funds, or don't know any better ), you will never ever know about the sonic differences when using one tube, versus paralleled tubes, and, hearing a zero negative feedback amplifier play music. Nor will anyone know about the differences in using two tightly-matched highest-quality plate resistors, VS. using one. No, one will not know this, if one has only been exclusively experiencing Grade "C" and "B" audio amplification. So Coytee, you asked me,........... and here you have it, from me, ... my friend . Jeff Medwin Thank you, Moderators, for allowing me to answer !!!!!

I understand what your listening experience was, and given the list of amps you mentioned owning, precisely WHY you wrote and asked me such a question. I am fully understanding, Coytee, of what you heard. I do NOT like OTLs at all, because the coupling device between the amp and the speaker is a capacitor. A capacitor is possibly the worse sounding part we have in audio, to couple through. Sure, you got rid of an output transformer, but NOW, you are listening through something far worse, a CAP !!!! The OTL's basic topology is flawed, not the best possible, when you have to couple to the loudspeaker with a capacitor !!! Also, they have to use negative feedback , a huge NO NO in quality amplifier design. OTL amps ( eg , the Ro****lit 2 Watt SE-OTL ) are EASY - to thoroughly and totally " trounce" - in performance. I did this in 2018, and can post the Listener's evaluation for anyone to read, ( his email back to me ), if, there is any sincere interest. Jeff