John Broskie's Guide to Tube Circuit Analysis & Design

07 February 2014


In my last post, I revealed the arrival of the Balance Trim PCB and rotary switch.

The Balance Trim is a simple ladder attenuator, with six positions, which can be used with a volume potentiometer (or fancy stepped attenuator). The switch is a 2-pol, 6-pos, shorting design that never places more than two resistors in the signal path.

The volume control's resistance must be factored into the ladder attenuator's resistor calculations. The following tables show -1dB decrements for both metal-film and carbon-film resistors.

E96 1% Values Potentiometer Resistance
10k 20k 50k 100k
4.42k 8.66k 22.1k 44.2k
3.65k 7.32k 18.7k 36.5k
2.94k 5.9k 14.7k 29.4k
2.05k 4.12k 10.2k 20.5k
1.07k 2.15k 5.49k 10.7k
82.5k 165k 412k 825k
38.3k 80.6k 191k 383k
24.3k 48.7k 121k 243k
16.9k 34k 84.5k 169k
12.7k 24.9k 64.9k 127k
E24 5% Values Potentiometer Resistance
10k 20k 50k 100k
4.3k 8.2k 22k 43k
3.6k 7.5k 18k 36k
3k 5.6k 15k 30k
2k 4.3k 10k 20k
1k 2.2k 5.6k 10k
82k 180k 390k 820k
39k 82k 200k 390k
24k 47k 120k 240k
18k 33k 82k 180k
13k 24k 62k 130k

Each channel gets its own Balance Trim control, which will undoubtedly bother some old-timers, who fondly remember the single balance control knob of their youth. But as Jeanne Moreau so aptly put it, "Nostalgia is when you want things to stay the same. I know so many people staying in the same place." Not that for a second am I endorsing the idea that all change is good, as it certainly isn't. No, what I want is small, reversible, incremental changes for the better.

For some insane reason, many believe that they must turn both Balance Trim controls at once. Not so. The whole point is to establish balance. If the right speaker is playing louder than the left speaker, just turn down the right speaker.

By the way, the Balance Trim can be used a volume control, particularly when cascaded into another Balance Trim. Why? Two Balance Trims in cascade equal 36 steps of attenuation, which is 12 more decrements of attenuation than the usual 24-position stepped attenuator from GoldPoint or DACT; more importantly, the two Balance Trims in cascade never present more than four resistors in the signal path due to their being ladder types, not series attenuators. This two-knob, 36-position, ladder stepped attenuator would require only 20 resistors per channel, not the 36 resistors that a series attenuator would need.

By the way, three Balance Trims in cascade would yield 6 * 6 * 6 steps of attenuation, or 216 steps. Why would anyone need so many? Perhaps you are doing test measurements and you require a precise amount of attenuation, for example -90dB. Well, the following stepped-attenuator box would be one easy and shockingly inexpensive solution. (If you have ever priced lab-grade attenuators, you will know how expensive such a stepped-attenuator box costs.)

Alternatively, you might wish to make precise adjustments to the balance, which such a stepped attenuator would allow. In this example, from 0db down to -97.5dB.

But what if you own a piece of old audio gear that already holds a balance control, albeit a not very good one, as it uses a dual potentiometer that doesn't center or makes scratchy noises? Well, you could use just one Balance Trim as a dual-channel balance control. How? Just use one Balance Trim on the left channel, parking the knob on the -2dB position; then, use two resistors on the right channel to drop its signal by -2dB. Thus, with the Balance Trim set at -2dB, both channels get the same signal level. But as the single Balance Trim knob is rotated, the left channel will get either more or less signal, at 1dB increments or decrements, allowing a small and subtle changes in balance. Of course, a seven-position rotary switch would be better, as it would allow equal number of decrements and increments, but such beasts are not made, alas.

Or, depending on the circuits used, you could use a Trim-1 series attenuator as a balance control replacement.

The Trim-1 offers 11 positions, which is perfect, as an odd number was required to center the switch. All that is required is to replace R11 with a jumper wire and use the same value resistors throughout. (You could get fancy and put in place some fancy attenuation curve, but I would stick to 10 equal-valued resistors. Sometimes it's better not to be too clever. Just do not quote me back to me in the future.)

Of course, the amount of possible balance adjustment is up to you. The best thing about DIY audio is the freedom—but, then, it is also the worst thing about it. Hell, I can imagine two balance controls per channel, one coarse and one fine, much as a good microscope offers two focus control knobs.

I once modified a tube-based guitar amplifier that held a single hum-null potentiometer by adding a second fine null potentiometer. The procedure was simple enough, center both knobs, then adjust the coarse hum-null knob until the deepest null was attained, then adjust the fine control for an even deeper null in hum. Nostalgia dictates one knob, intelligence says otherwise.

The Balance Trim is available now at the Yahoo-GlassWare store and is sold with or without the resistors.



More Sonic Control
I received a few e-mails that asked how to implement the sonic width control on existing tube designs, such as the CCDA and Aikido. Since I already covered this topic with regard to the Aikido, back in blog number 33, I will move on to the CCDA. "CCDA" stands for Constant-Current-Draw Amplifier. Adding a sonic width control is easy enough, as single Balance Trim PCB, switch, and four resistors can be used.

When parked at the third position, the control leaves the signal unchanged. Turn the knob all the way clockwise and CCDA puts out a mono signal. Turn in the opposite direction and the CCDA puts out wider sound-stage, with less center-fill.


Adding a Tilt Control to the CCDA
Ideally, the Tilt Control will see a low-impedance signal source and a high-impedance input from the following audio circuit. The CCDA's second stage is a cathode follower, which will achieve the first requirement, but what about the second? We could use a second cathode follower, but we might end up using at least two more internal coupling caps and the cathode follower would do nothing to improve the CCDA's somewhat poor PSSR of only -6dB. The following circuit uses a CCDA and an Aikido cathode follower (ACF) to both meet the second requirement and to strip the B+ noise from the output, while providing a low-impedance output. Not bad.

The CCDA's first stage, a grounded-cathode amplifier, splits the B+ voltage and the B+ noise at its output. The CCDA's second stage, a cathode follower, then relays this halved B+ noise to the Tilt Control, which in turn passes it on to the ACF, which then applies a countering current flow through the output to null the B+ noise at its output. Magic. Well, it's like magic, if you cannot figure out how it works. The following schematic shows how it works.

The two 10µF capacitors in series define a 50% AC voltage divider, which attaches to the Tilt Control's ground pad. Note how the above B+ noise null only occurs if both the CCDA and ACF stages share the same B+ connection. In other words, adding an RC filter to CCDA's B+ connection will undo the magic.

Of course, there is no reason why the sonic-width control could not be added to this circuit.


Adding a Tilt Control to the Aikido
The Aikido line-stage amplifier, known for its stellar PSRR and low-output impedance, makes a good circuit to drive the Tilt Control. But we could use the same approach as was used in the CCDA example by undoing the Aikido's power-supply-noise null. What!

I often get requests for an Aikido with balanced outputs capable of driving 600-ohm loads. My recommendation is the following circuit.

The un-Aikido gain stage now spills 50% of the power-supply noise at its output; but then, so, too, does the two 30µF power-supply capacitors, so the output transformer's primary never sees a power-supply-noise delta, so no power-supply noise couples to the secondary. No delta, no signal. Now, if the two big capacitors didn't also help the rest of the circuit out by working as power-supply-filter capacitors, I would be less inclined to recommend this topology. But since we do need power-supply capacitors, this approach makes a lot of sense. Think about this: even if the two 30µF capacitors are flaky in some way, perhaps due to excessive inductance or ESR, the un-Aikido stage will halve this flakiness and the output transformer will ignore it, dropping it from its output. Remember, no delta, no signal.

So, we can add a Tilt Control to an existing Aikido stage thus:

Note the two 10µF capacitors, which are needed to "ground" the Tilt Control and the elimination of the Aikido's voltage-divider resistors.

Of course, we could use a normal Aikido to drive the Tilt Control.

In this example, the Aikido presents a clean output to the Tilt Control, which then relays the signal to the ACF, which then undoes the power-supply-noise leakage that a simple cathode follower would spill out. Yes, simple cathode followers are not blameless in terms of PSRR. Note the single 10µF capacitor that attaches to the Tilt Control's ground pad, effectively "grounding" the control. (A much smaller capacitor could be used, say 1µF, but the larger value hammers home the point that  the Tilt Control is being grounded through this capacitor.) Note how the Aikido, the Tilt Control, and the ACF are all DC coupled.

Just like the previous CCDA circuit, there is no reason why a sonic-width control could not be added to this circuit.



Robo Speaker Stands
I caught some grief from someone who thought my idea of an active speaker stand ludicrous in the extreme. "Who would need such a thing?" he asked. Well, I would. About 30 years ago, I had two strong buddies over to help me figure out how to place my new speakers. I had just finished building two-way speakers that held 6-inch woofers and 1-inch dome tweeters. The internal volume wasn't all that large, as I had used 1.5-inch thick panels and much internal bracing. In other words, although they were not all that big, but they were certainly heavy.

Each fellow held his speaker, as I orchestrated their position, much as an airport plane-marshaller directs the pilot's path. "Up, tilt back, a bit further apart, no, closer, move away from the back wall, that's it, yes, perfect, stop fidgeting..." In less than a few minutes, I had found the perfect placement; well, at least for that LP.

What I can never forget is that during one brief moment, the two guys had fallen into perfect harmony, each placing his speaker at the optimal position, resulting in the most amazing sound stage, truly holographic. Then, alas, it was gone. Indeed, I could never replicate the placement, as my stands were not tall enough and didn't allow for tilting the speakers back. Nonetheless, the memory lingers—even after 30 years.


Ripping LPs
Apparently this is a hot topic. Christian Rintelen, in Zurich, sent me a good e-mail, which explained some of the subtleties involved. I particularly liked his recommendation of not to apply the RIAA EQ on the signal before it hits the ADC; instead to apply it, after the fact, in the digital domain. Interesting. If nothing else, this would greatly simplify the design of the phono preamp. Still, I would only take half of his advice and only apply the 2122Hz (75µS) low-pass equalization to the signal, so as not to overload the ADC or feed it frequencies that it cannot handle, which creates very unnatural distortion. In nature, we get overtones, higher harmonics, such as 2kHz being the 2nd harmonic of a 1kHz tone. But if an ADC is presented with some frequency higher than half of its sampling rate, the result is not higher harmonics, but sub-harmonics—very unnatural. Thus, analog signal presented to the ADC must be scrubbed free of high frequencies beyond the ADC's input frequency limit.


Adolph Hitler and Speaker Cables
If you have already seen this, forgive me. I just saw it and I loved it.

Hitler finds out cables don't make a difference to HiFi sound quality


Next Time
New PCB designs and, of course, new circuits.





I know that some readers wish to avoid Patreon, so here is a PayPal button instead. Thanks.

                                John Broskie


Kit User Guide PDFs
Click image to download

Download 5687 Aikido USer Guide PDF

BCF User Guide

Download PS-3 User Guide

Janus regulator user guide
Aikido Phon Preamp user guide

E-mail from GlassWare Customers

Hi John,

I received the Aikido PCB today - thank you for the first rate shipping

Wanted to let you know that this is simply the best PCB I have had in my hands, bar none. The quality is fabulous, and your documentation is superb. I know you do this because you love audio, but I think your price of $39 is a bit of a giveaway! I'm sure you could charge double and still have happy customers.

Looking forward to building the Aikido, will send some comments when I'm done!

Thank you, regards,


Mr Broskie,

I bought an Aikido stereo linestage kit from you some days ago, and I received it just this Monday. I have a few things to say about it. Firstly, I'm extremely impressed at the quality of what I've been sent. In fact, this is the highest quality kit I've seen anywhere, of anything. I have no idea how you managed to fit all this stuff in under what I paid for it. Second, your shipping was lightning-quick. Just more satisfaction in the bag, there. I wish everyone did business like you.

Sean H.

High-quality, double-sided, extra thick, 2-oz traces, plated-through holes, dual sets of resistor pads and pads for two coupling capacitors. Stereo and mono, octal and 9-pin printed circuit boards available.

   Designed by John Broskie & Made in USA

Aikido PCBs for as little as $24


Only $9.95
to start designing
tube-based crossovers
and much more...

TCJ Filter Design

The Tube CAD Journal's first companion program, TCJ Filter Design lets you design a filter or crossover (passive, OpAmp or tube) without having to check out thick textbooks from the library and without having to breakout the scientific calculator. This program's goal is to provide a quick and easy display not only of the frequency response, but also of the resistor and capacitor values for a passive and active filters and crossovers.

TCJ Filter Design is easy to use, but not lightweight, holding over 60 different filter topologies and up to four filter alignments:


While the program's main concern is active filters, solid-state and tube, it also does passive filters. In fact, it can be used to calculate passive crossovers for use with speakers by entering 8 ohms as the terminating resistance. Click on the image below to see the full screen capture.

Tube crossovers are a major part of this program; both buffered and un-buffered tube based filters along with mono-polar and bipolar power supply topologies are covered. Available on a CD-ROM and a downloadable version (4 Megabytes).

Download or CD ROM
Windows 95/98/Me/NT/2000/XP           Copyright © 1999-2014 GlassWare           All Rights Reserved