MODIFY THE UL40-S2 STEP BY STEP
In my earlier articles on this website I explained that not every modification is a good one. Many modifications cost lots of money and they even might not work. I prefer cheap modifications with lots of results. Last year I did many experiments to discover and evaluate such useful modifications. The UL40-S2 in its original form, as delivered by me or Amplimo, is excellent. But when one uses an amp for many years, you will find small issues that can be improved. That is what I am writing about in this publication. I asked some of my UL40-S2 clients to help me and to check my findings by repeating my experiments. They were my control group. I sent them some components with the explanation where and how to use them. I did not tell the effect on sound reproduction. I asked them to tell me in their own words about their findings. Surprisingly (really surprisingly) all my findings were confirmed. So, after a year of experimenting, I am absolutely sure that the modifications following are no rubbish. They are good and impressive improvements of the UL40-S2.
My advise is: try these modifications step by step, they are inexpensive, money is not the issue. It is very important to take much time to listen to each modification. Learn the soundstage and its character by heart. By doing so, each next modification will easily be recognized and appreciated. This is the way to enjoy the Ul40-S2 the most, by finding out that this fine amp really improves step by step.
At the end of this article I explain where and how to order the components.
The UL40-S2 in its original DIY kit form, can be switched in three different
modes by rearranging the connections of the screen grids of the power
tubes. See the manual for more details how to do so. The three modes are
push-pull pentode, push-pull ultra linear and push-pull triode. Please
take your time to listen to each of these modes and to recognize their
goodies. This modification costs no money, it is standard available in
each UL40-S2 amp.
The sound reproductions as found in these three modes can easily be explained by measurements. For instance by looking at distortions or damping factors. In all modes the -3dB frequency range largely exceeds the human hearing range of 20 kHz and is close to 80 kHz. So, it is certainly not the frequency range which explains the differences in sound character.
The UL40-S2 can handle many power tubes without any further changes. It is just a matter of plug and play. Many types were tested and the most applied are EL34, 6550C and KT88, all of Svetlana. Their behavior and sound character are as follows:
EL34: very dynamic and a warm round sound character with many details. The low end is a little uncontrolled. The amplifier is extreme stable (no oscillation or any kind of electrical instability). Originally I designed this amp for EL34 tubes. So, it is not surprisingly that this tubes feels stable and at home in this application.
6550C: the sound character is strong and well controlled at the same time. The bass reproduction is fast and not weak at all. The amp sounds clean, still with a warm tuby touch, but certainly not too warm. Electrically this setting is very stable as well, no oscillations or any other instability. The 6550C compares well with the EL34, but has the advantage of a larger maximum anode dissipation, which explains its controlled character.
KT88: the superior choice? Most people think that the more expensive, the better, but that is not the case in this situation. The KT88 shows a little more details, a little more control, a little much more expensive compared to the 6550C. But, the UL40-S2 with KT88 might become unstable, might behave like an oscillator. Sometimes the amp is stable, the next day it might behave as a transmitter. In this amp the KT88 is at the edge of stability. I did not design this amp for the KT88, and I am not intending to adapt the design for it. Next week another tube can appear on the market and then I should redesign again? No way, I have more to do. Earlier I gave some advise how to stabilize the KT88, but this was not always successful.
Therefore: with my long experience I really advise as the optimum choice the 6550C for this amplifier. This tube shows the right balance and beauty and control, combined with excellent electrical stability. I do not advise to use the KT88. When you are a guitar player, then the EL34 has the right tonal balance and dynamics. See the end of this article for ordering and prices of the 6550C.
Sometimes a modification gives excellent results and you even don't know why. This modification is such a one. In the original kit the wiring from input section to the Alps volume control and to the main PCB is with normal stranded wire. When you replace this wire by 1 mm diameter high purity five niner gold doted Siltech wire with Kapton isolation, the sound quality tremendously improves. It sounds as if a curtain is removed and much more details easily can be recognized. I never would have believed that such small length (about 30 cm) of wire would have such an impact. I tried many brands and qualities of wires, also less expensive silver wires, but again and again the superior quality of the Siltech wire was confirmed. Especially the input circuit of the UL40-S2 and its wiring seems to be very sensitive for the quality of materials applied (also see modification 5). When you decide to change to Siltech wire, I deliver 2 meter length. This will enable you to put the wires from input section to volume control in the corners of the metal case. It looks nicer because the input wiring is not running on top of the main PCB. It also prevents distortions (from the mains transformer) and amplified signals (from output transformer) to enter the sensitive input section.
In earlier publications on my web site I already explained how to convert the UL40-S2 into a Super-Triode amplifier. You need to replace the screen grid resistors by 1kOhm resistors. These should be connected to their UL-taps of the OPT. Between the control grid and the anode of each power tube a series combination of a 2M2 resistor plus 10 nF capacitor is added. From driver section to control grid a series resistor of 82 or 100 kOhm is added. I also explained that you can trim the three most important parameters of the power tubes by changing this 82 kOhm resistor into a trim pot P5 plus 47 kOhm series resistor. I really advise everyone to use this trim pot solution, because then you can get the most advantage out of this superior circuit. Harmonic distortions and low hum and linearity all can be trimmed to an optimum. The final result is an extreme open and clean sounding amplifier. The behavior of the amp is complete triode. However, the output power is close to 30 Watt. The price of this modification is almost nothing. In spite of that this is one of the most powerful modifications.
Again the KT88 might become unstable with this modification. To prevent such, I advised to add small 100 pF Styroflex capacitors in parallel with the series resistors from driver section to control grids. When you order this modification kit, you will see that I replaced the Styroflex by better sounding Silver-Mica. Besides suppressing of oscillation, the function of these capacitors also is to widen the -3 dB bandwidth from 25 kHz to the original 80 kHz. So, you should think that these capacitors do an excellent job. They do, but in spite of their good behavior, I advise NOT to use them. Without these caps the amp sounds superior, more balanced and quiet and open. The 25 kHz bandwidth limitation, caused by the extra series resistors between driver section and control grids, followed by a first order low pass filter behavior up to 80 kHz, where the second order low pass of the OPT starts, can not be noticed by our ears. I can explain this with a deep going research into the complex domain and psycho acoustics. But for now I propose to leave theory where it belongs (in books) and to rely on my ears. They told me that without caps the sound quality is better. However, when the amps oscillates, you should use them. See the other article in this section of my site to find all the details about this important modification.
This modification is the next logical step after modification 3. There it became clear that the input section of the UL40-S2 is very sensitive for the quality of the applied materials and components.
I decided to test in detail the resistors in the input section. In the kit I apply Beyschlag metal film resistors and they sound good. But when I replaced these by carbon, I encountered huge improvements. I tried many brands and finally I selected the optimal sounding carbon resistors. The impact of changing metal film into carbon is of the same level as changing from stranded wire to high quality silver wire. It is as if again a curtain is removed. The soundstage gets wider, more open, more to the front and more to the back. The sound starts to surround you. The acoustics of the listening room disappears and the acoustic field of the recording environment gets dominant. I gave this very impressive and important acoustic effect the name "enveloping".
Later I started to research the famous work of H.J. van der Heide (former employee of the Dutch Radio sound laboratory, also a researcher in alternative healing). In his article (*) he describes in an excellent way the changes in experiencing sound when the materials and components of the amplifier are replaced by human friendly materials (like carbon). Following you find a translated listing of his findings, as mentioned in the article:
1. The remarkable experience as if one is objective and subjective
connected with reality.
(*) H.J. van der Heide:Waarneming en Ervaring van Geluid; Natuurlijk
of Electro-akoestisch"; Prana 113, juni/juli 1999.
My findings coincide completely with what Van der Heide wrote, although I did not check 8 and 10. Earlier I introduced the term "enveloping" to describe my experiences. However, out of respect for Van der Heide, from now on I will use his wording. He introduced "structure" or "regulation" information (in the meaning of the terminology of alternative healing). This regulation information has two faces. Firstly, it describes the arrangement of the source of sound in the position of this source in its environment. Secondly, it describes the time-space manifestation of the actual sound.
I know that I can not completely quote the work of Van der Heide in my small abstract. I really advise a closer study of his research. I now return to my issue. My experiments with the human friendly carbon resistors support his findings. It is a huge change, of a multi dimensional nature. My control group reported exactly the same results, independent of me, because I did not tell them anything in advance about what they might experience. I also tried to change the other resistors in the amp into carbon, but the further improvements was negligible. Therefore the modification kit only contains the most sensitive resistors of the input section.
Suppose you have a lot of money and don't know where to spend it. Let me give you a good advise. Change the good Vanderveen toroidal output transformer into a Silver wire version. In my transformers I use high quality copper magnet wire, but now we have two new versions with silver wire. The VDV-6040SILpp has a silver wire secondary, while the VDV-6040SIL2pp is completely Siltech silver wire. At www.amplimo.nl in its high-end section, I give a complete description of the impact of changing the copper wire OPT into a silver wire version. But, first let me tell you that not all silver wire is good. I tried several brands, from inexpensive to very expensive. What I say following is only true for the Siltech silver wire that I applied.
You do not see it at the outside of an OPT, but don't be surprised by the huge length of wire in it. More than a kilometer is not unusual. Optimizing the quality of the wire should have a large impact, and so it is. We use high purity five niner silver with gold doting, with Kapton isolation (which functions as a kind of glue at the same time). This wire largely improves the sound quality in such a manner that the influence of the core material becomes negligible. It even is useless to start discussions about push-pull or SE with these transformers. Their impact is the removal of a thick curtain in front of the speakers. In my listening room I have standard two versions of the UL40-S2 available, one with silver, the other with standard copper OPT's. Both amps are excellent, but the silver version shows much more details. I call that reproduction of "micro information".
MODIFY STEP BY STEP
I would like to advise everyone to start with modification 1, next number 2 and so on. It is very important to take enough time to listen to and to learn and to remember each sound character. It always takes a lot of time when you modify an amp, so the previous sound should be well remembered, else you might not recognize the subtle improvements of the next mod. Take your time and it almost costs nothing. Only modification 6 is expensive and I certainly will not blame you when you decide not to go for the silver OPT's. But all the earlier modifications each are very valuable and substantial. They will change your excellent sounding amp into a superior amplifier. It took me more than a year to test and evaluate all changes and fortunately I had my control group. They checked me, independent and unaware of my personal opinion. They described their activity as a very nice and challenging adventure. I hope you will experience the same joy with your modifications.
HOW TO ORDER THE MOD-KITS
I now describe the content of each modification kit.
MOD-1-P/UL/T: you don't need to order anything. All is available in the original UL40-S2 kit. Only rewiring of the screen grids is needed for and all the information already is in the manual.
MOD-2-6550C: Four matched Svetlana 6550C tubes: € 140,- including VAT. To be ordered at Vanderveen or Amplimo.
If you plan to buy the UL40-S2 kit, I advise to order directly the 6550C version. The price of the kit then equals € 1350,- including VAT. To be ordered at Vanderveen or Amplimo.
Vanderveen does not sell the KT88 tubes. You can order them at Amplimo. I only get troubles (extra work) when my amp starts to oscillate.
MOD-3-silver: 2 m 1 mm diameter Siltech silver wire: € 70,- VAT included. Order at Vanderveen or Amplimo.
MOD-4-ST: all extra components for € 25,- VAT included. (In the past the price was a little lower, but now I use silver-mica caps). You only can order at Vanderveen, because I wish to closely follow all results and feedback about this very important modification.
MOD-5-carbon: Selected carbon resistors for the input section of the UL40-S2. The price is € 15,- VAT included (mainly costs for administration, the carbon resistors are not that expensive). You only can order at Vanderveen, because I wish to closely follow all results and feedback about this very important modification.
MOD-6-OPT-silver: See the website of Amplimo for current prices. You can order at Vanderveen or Amplimo.
How to order at Vanderveen: send an email to email@example.com with your order. I will send you an invoice (including extra costs of shipment) by email. When the monies involved are received at my bank account, shipment will immediately follow.
How to order at Amplimo: through their web site www.amplimo.nl or by phone 0031-545-283456 or by email firstname.lastname@example.org