New single-ended power amp Valvolone Sbrillu-UL – the magic of the triode, the grit of the pentode

When pure listening pleasure meets thoroughly “handcrafted” ingenuity, Valvolone Sbrillu-UL is born—the single-ended tube amplifier in pure class A that promises to turn every note into emotion. Thanks to the exclusive ultralinear connection—that “mystical” balance between silky triode and muscular pentode—Sbrillu-UL delivers up to a fifteen—well, actually 4 to at most 5—sparkling watts per channel, ready to fill the room with sonic warmth.

Favorite tubes? EL34, KT88, 6550… you choose! With a simple gesture you can switch from velvety triode to roaring ultralinear, no screwdrivers or complications. All housed in an elegant gourmet matte black chassis with VU meters winking at you, because looks matter too.

Why Sbrillu-UL

• Single-Ended Pure Class A: the classic recipe for engagement.
• Ultralinear “sweet spot”: triode detail, pentode drive.
• No-stress rolling: compatible with the great iconic tubes.
• Immersive soundstage: presence, depth, air between instruments.

Hit play and let yourself be won over by a sound that marketing would call “purer than pure”, while your friends will only ask: “Where’s the stage?”

Lately I’ve been getting more and more questions and requests for opinions about circuits found online that someone would like to replicate.

First, though, a playful preface: the slightly tongue-in-cheek opening pokes gentle fun at a certain kind of audio marketing that loves to talk about “musical watts,” often twice the real ones— a unit of measure invented more to fire the imagination than to describe electrical reality.

Having clarified that the opening joke is just an icebreaker, let’s get to the point. The article that follows is not meant to be partisan or polemical; it aims to demonstrate—with verifiable data and explanations—a well-defined technical reality.

I had long intended to tackle this topic because I keep seeing the same issue: in several projects circulating online, the famous “ultralinear connection” appears applied to single-ended stages. My answer, inevitably, is always the same: forget it. In a single-ended amp, ultralinear does not work; it only makes sense to choose either a triode or a pentode output.

When I say that the ultralinear connection in a single-ended “doesn’t work,” I don’t mean the amplifier will explode, emit no sound, or sound disastrously distorted. I mean that a conceptually unsuitable approach has been chosen: yes, the unit will play, but sub-optimally, with no real benefits and with the risk of introducing further issues. That’s why I invite you to read the following observations carefully: they clarify the technical reasons for my position and prevent misunderstandings in the comments.

Ultralinear in single-ended is a glaring mistake, yet it has been copied and spread throughout the audiophile world—even by prestigious brands—without anyone bothering to truly verify how it works.

The DIY and audiophile community keeps treating it as a valid connection for both push-pull and single-ended, as if the name itself were a guarantee: “Ultra-Linear”! Not just linear… ultra. An effective psychological and commercial hook, but utterly misleading from a technical standpoint.

What the Ultralinear Connection Is

The ultralinear connection is a configuration possible only with pentode tubes, in which the screen grid is connected to a tap on the output transformer. In this way, the screen-grid voltage is not fixed but partially follows the variations of the plate voltage.

The result combines the high efficiency typical of a pentode with the linearity of a triode, yielding behavior halfway between the two modes. This solution, devised by Alan Blumlein, was created precisely to exploit the advantages of both tube types and finds its natural application in push-pull output stages, where it allows overall performance to be maximized.

Advantages of the Ultralinear Connection

By precisely setting the percentage of the screen-grid tap, it’s possible to obtain an ideal balance between triode and pentode characteristics. Within a fairly narrow range, distortion can drop to unusually low levels—sometimes lower than those achievable in pure triode or pure pentode—while energy efficiency remains only slightly below that of full pentode operation.

The optimal tap percentage depends mainly on the tube type. For KT88s, for example, the value commonly considered optimal is 43% of the primary turns on the anode side; for 6V6GTs, 20% has often been recommended. Mullard circuits made wide use of 20% distributed load, while some LEAK amplifiers pushed up to 50%.

The features that make distributed loading particularly suitable for audio power amplifiers—compared with solutions based on triode, beam tetrode, or pure pentode—can be summarized as follows:

1. The output impedance is reduced to about half that obtained with a pentode.
2. Distortion is reduced to approach that of a triode, and can even be lower in ultralinear operation.
3. Output power is higher than that of a triode, approaching that of a pentode.
4. Output power is more constant, since distributed loading combines a transconductance amplifier with a voltage amplifier.

Alan Blumlein conceived and applied the ultralinear connection exclusively in push-pull circuits, and the same choice was followed by all major manufacturers of the time. There are no historical examples of single-ended amplifiers with an ultralinear connection: it’s an idea that appeared only recently. Let’s see why! Below is the schematic of a Single-Ended Ultralinear amplifier built by an SB-LAB customer featured in this article (click). As you can see, the screen grid is connected to an intermediate tap on the output transformer.

I’d like to draw attention to the value of the resistor under the KT88 cathode in the schematic, marked as 360 ohm. Here are the words of a customer who built this circuit:

“If you could add a note not to be influenced by the 360 ohm resistor: I replaced it with a measured value of about 190 ohm, after various attempts, because the correct bias current wouldn’t line up.”

This is a crucial detail: remember it. The very fact that you need to halve the resistor value highlights an underlying oversight. It’s a strong clue that the published schematic was not experimentally verified under the stated conditions: otherwise the discrepancy in the output tube’s bias would have emerged.

Yet this schematic has circulated on the Internet for decades, replicated and praised, with almost no one investigating the reasons for the bias discrepancy. Many probably didn’t even notice, happily listening to distortion they believed was the “right sound.”

Anyone with basic design skills who can read a tube’s characteristic curves—triode or pentode—knows the drill: pick an operating point (voltage/current) within dissipation limits, then draw the load line based on the output transformer’s impedance.

But with a tube wired in ultralinear mode, you cannot freely select the operating point: behavior is constrained by the screen-grid characteristic. Any change in screen bias significantly alters the entire family of characteristic curves. To clarify this, we can refer to the curves in the Genalex KT88 datasheet (click to enlarge):

First, note that in the KT88 datasheet, ultralinear operation is described only for push-pull configurations. At the time, the idea of applying it to a single-ended amp simply held no interest. Still, it’s easy to imagine the reasoning of a less careful designer:

“The typical load for a KT88 in single-ended is 2500 ohm. Roughly, I can set 250 V with 120 mA of bias and a grid voltage around ?32 V…”

Such a hasty calculation leads to drawing a load line that at first glance can seem correct, as shown in the following graph:

To avoid lab trials, we can use LTSpice to simulate the bias of a KT88 in this configuration (2.5 kohm primary, 8 ohm secondary, 50% ultralinear tap). The KT88 model used is by Norman Koren, renowned for high accuracy: simulation results overlap with those of a real tube. In theory, we’d expect to measure a current of about 120 mA at the cathode…

Here’s the simulation result: the bias current is only 24 mA! At this point, anyone with a minimum of experience (hobbyist DIYers can be forgiven, but anyone calling themselves a designer should notice) should ask a key question:

“Why do the characteristic curves indicate a current of about 120 mA, while in practice I get barely 24?”

A small discrepancy due to tube tolerances is normal and can be compensated by slightly adjusting the bias. But this isn’t a few milliamps: dropping from 120 mA to 24 mA is a huge gap, which should seriously cast doubt on the correctness of the theory used to determine the bias.

Yet in most cases this red flag is ignored. The bias is “pulled” to force the tube current without questioning the cause. Let’s try a clearer signal…

The circuit now seems to work, but the oscilloscope tells another story: the waveform is heavily distorted (the blue trace is the input signal, the green trace is the output). A natural question arises:

Why, in a single-ended amplifier in ultralinear mode, do the bias current and transformer impedance not match expectations?

To find the answer, we must examine the datasheet characteristic curves more closely…

Did you notice the dashed line marked Va,g2(0) = 425 V? Before proceeding, it’s worth a quick review of how tubes work—both triodes and pentodes—focusing on their internal structure. Let’s start with the triode: it has a single grid and an extremely thin plate (anode), almost “toast-like,” placed very close to the cathode.

Now let’s look at a tetrode or a pentode, which contain two or three grids. In a beam tetrode, the third “grid” is actually two thin metal plates, but we won’t dwell on that here. What really matters is that, unlike the triode, the plate (anode) is located much farther from the cathode.

In triodes, the electric field generated by the plate (anode) directly attracts electrons, while the control grid (G1), kept at a negative potential, slows and regulates their flow. In tetrodes or pentodes, however, the plate is too far from the cathode to attract electrons on its own (or would do so only weakly). Here the screen grid (G2), placed right after G1 and positively biased, accelerates the electrons toward the anode.

Since G2 is made of very thin wires, most electrons don’t settle on it; thanks to their acquired speed—in a sort of “slingshot effect”—they continue on to the anode’s electric field, which captures them. It’s therefore clear that, in a pentode, the anode current depends not only on G1’s negative voltage but also on the positive voltage applied to G2.

In the ultralinear connection, at idle the voltage applied to the screen grid (G2) is practically identical to the plate voltage, because the transformer winding’s internal resistance is negligible. Consequently, any change in plate voltage causes an equally significant change in tube current: the G2 voltage inevitably follows the plate. For this reason, in ultralinear configuration we can speak of “dynamic” curves, whereas in triodes and pentodes wired as pentodes the curves remain substantially “static.”

The dashed lines in the previously cited Genalex datasheet essentially indicate that the operating point may be placed at any current, but it must remain above that line, i.e., at 425 volts! If you change the operating voltage, the curves shown in the datasheet are no longer valid—they change completely!

Let’s analyze this using uTracer, which can be configured to acquire curves in ultralinear mode. However, for the reasons already mentioned (and due to a missing software implementation), uTracer acquires dynamic curves only below the specified voltage (the Genalex dashed-line value).

To better illustrate the behavior of these dynamic curves, I’ve highlighted with a black dot an intermediate point corresponding to 300 volts, with the control grid G1 biased at –25 volts.

With a “stop” voltage at 400 volts we have 80 mA at 200 volts with G1 at ?25…

If we bring the “stop” voltage to 300 volts, the current measured—still at 200 volts with ?25 on G1—drops to a bit under 40 mA

If we then further lower the “stop” voltage to 250 volts, we end up with a current below 20 mA

You can also observe that as the “stop” voltage decreases, the tube’s current-delivery capability drops noticeably, while its internal resistance increases, as shown by the curves’ reduced slope. This means the tube is far less capable of supplying current—and thus power—than under optimal conditions. For example, with a stop voltage of 400 V the KT88 can reach a peak of about 170 mA at 50 V; with the stop voltage reduced to 250 V, the peak drops to about 60 mA.

As if that weren’t enough, the change in curve slope also requires a change in the transformer’s impedance to avoid heavy distortion. The power actually delivered to the loudspeaker ends up almost identical—or only slightly higher—than what you get with a pure triode connection. However, in triode mode the tube operates far more linearly. In short, if you don’t want to use the tube in pure pentode mode, the sensible choice is to use it as a triode, without even considering the ultralinear option: in a single-ended amp, it almost comes off as a gimmick to add distortion.

It’s essential to note that these considerations apply to class A operation (both single-ended and push-pull), where the idle voltage is not high and the ultralinear curves at various voltages are not known. The ultralinear connection was instead conceived for use in push-pull class AB, where the idle voltage is higher. In this context the tube works correctly and offers real benefits in terms of reduced distortion and sometimes greater output power.

For example, a pair of KT88 can safely deliver about 50 W in class AB in pentode; beyond that, the screen grid (G2) tends to glow, since the plate voltage can drop below the screen’s, generating screen-current spikes. When KT88s are wired in ultralinear mode, the current is better controlled and 70–75 W can be obtained without G2 glowing.

Let’s now examine the load line set with the following operating conditions: 425 V plate voltage, 75 mA current, –50 V on the control grid, and a 6k impedance transformer.

In this simulation we obtain a current of 66 mA, very close to the expected value (small discrepancies stem from the mathematical model). The bias current therefore returns to expected levels because the operating point was chosen on the datasheet’s dashed line at 425 V. This result fully confirms the points made so far. Now let’s see how the circuit behaves when driven with a sine wave:

Once again, the output signal shows strong distortion, far from expectations. A clearly flattened half-cycle is visible—but what causes it? The asymmetry of ultralinear curves is plain to see: just open the images and look closely. On the left side of the graph, the spacing between curves is much greater than on the right. This means that, for any hypothetical operating point X, one half-cycle will inevitably be longer, the other shorter. This phenomenon is intrinsic to the ultralinear connection and explains why the circuit was designed for push-pull use, where the twin tube—working in the opposite phase—cancels it out.

You can see a real-world circuit’s behavior in this article, where I first examine an amplifier that used a KT88 in single-ended UL on a 6k load, shown below with its captured waveform (yellow is the generator signal and blue is the circuit output).

As you can see, the real behavior fully confirms what emerged in the simulations. I’d also like to highlight a few other aspects my experiments brought to light:

1. A KT88 in single-ended ultralinear can deliver about 6–6.5 watts in practice, but with marked distortion. To obtain clean sound, negative feedback is needed—unless you truly enjoy distortion.
2. The KT88 in pentode reaches 12 watts with feedback.
3. The KT88 wired as a triode reaches about 5 / 5.5 watts with low distortion.

My conclusion is clear: if you’re working single-ended with a pentode, it makes sense to use it as a pentode to prioritize power, or wire it as a triode to prioritize linearity. In this context, the ultralinear connection offers no real advantage: the power gain over triode is negligible, while distortion remains high and forces you to resort to feedback. At that point it’s far more logical to choose a pure pentode, which at least delivers more watts, or a pure triode if maximum fidelity is the goal. Personally, I consider ultralinear in single-ended (especially without feedback) little more than a trick to introduce distortion.

Naturally, this judgment applies only to single-ended: in push-pull configuration, by contrast, the ultralinear connection brings real and significant benefits.

On this note, it’s worth pointing out a far more effective alternative to ultralinear: the Schadeode connection. This configuration, suitable for single-ended stages as well, combines the full power of a pentode with the triode-like linearity. It also offers major advantages, such as a high damping factor and reduced phase rotation—exactly where the single-ended ultralinear connection fails spectacularly.