Hum in Aikido Cathode Follower (ACF) - diyAudio
Go Back   Home > Forums > Blogs > My DIY projects
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
I returned to DIY electronics in 2009 after a 20 year pause by building a few kits to get in shape. This blog is for me to keep track of my progress.
Rate this Entry

Hum in Aikido Cathode Follower (ACF)

Posted 7th March 2017 at 03:00 PM by alexcp
Updated 13th March 2017 at 09:20 PM by alexcp

I purchased an Aikido Cathode Follower 2 (ACF-2) All-in-One 9-Pin PCB from Glass-Ware.com. The PCB is designed by John Broskie, the editor of Tube Cad Journal (tubecad.com), and the flexibility that he designed in is quite remarkable. The PCB allows to test various configurations, which is often not a feature of a design committed to a PCB. The quality of the PCB is outstanding; it withstands multiple soldering and desoldering without complaint. It is an outstanding product IMHO.

The schematic seems to be simple, with a cathode follower loaded by a triode current source. JJ E88CC with 221ohm cathode resistors shows 0.003% THD with pure 2nd harmonic, and select 6N1P-VI achieve 0.005%. (My 6N1Ps are hit or miss; some of them have much higher 2nd harmonic, and some of them have a mix of 2nd and 3rd at Ia=6ma and Ua=140V).

As I tested my ACF, I noticed excessive hum at its outputs - audible buzz in the speakers and 0.15% THD+N, well above the THD.

Soldering in "electronic chokes" instead of R4/R7 increased the hum level. (An electronic choke seems to be a gyrator, a transistor circuit simulating an inductor and in particular exhibiting a high AC impedance.)

I quickly stumbled upon a solution of adding an extra capacitor (4.7uF film was the only high voltage axial that I had available) between the anode of each of the top triodes (V1a and V2a - see the schematic) and the star ground.

After some head scratching as to why this solves the hum problem, I realized that there is no low impedance return path for the AC output current. In an Aikido amplifier with a single B+ power supply, the capacitors C4 and C5 would normally provide such a path. In ACF-2, they are connected between B+ and B- without a ground connection. The only AC return path for the output current is via C18/C19 and R4/R7, and the voltage drops across each of these are effectively in series with the output.

The voltage drops across C18/C19 and R4/R7 are a mix of the signal and a 120Hz sawtooth (rich in harmonics - I can count beyond 25th on an FFT). These voltages are in series with the output of the ACF. In the ideal Aikido world, non-signal contribution from C18 would cancel that from C19, while the voltage drop across R4 would cancel that from R7. That, however, requires that B+ and B- be perfectly symmetrical. Most importantly, the hum component of B+ must match that of B-. (This is irrelevant for a single-supply Aikido circuit, which uses a different technique to improve PSRR.) Naturally, the top and bottom halves of the cathode follower must be identical, too.

The simulation shows that even a slight imbalance between B+ and B- makes ACF-2 hum. For example, a 1% mismatch between the halves of the secondary transformer winding, or a 10% mismatch between the actual capacitance of C18 and C19 (which normally come at 20% tolerance), creates the hum level similar to what I measured in reality.

One solution would be to split C4/C5 into two capacitors each (shown in red in the attached schematic), one connected between B+ and the ground, the other between B- and ground. That wouldn't eliminate imbalance but would make it less relevant, as there is now a lower impedance AC path to ground from the anode of V1A and the bottom end of R5. It is not an elegant solution, as it is not Aikido-esque, and the PCB is not designed for it.

On a different but related topic, the capacitor C17, which connects heaters to the ground, is a puzzle. The hum level depends on the capacitor, which is expected. However, the capacitor seems to have an optimal value of 0.22uF that produces lowest hum in my test setup. Moreover, different capacitors of the same value produce different results. The best was PHE426, which improved hum by a few dB compared to WIMA MKP-10; I am at a loss as of why.

Note: the attached schematic was designed by John Broskie and published in his Tube CAD Journal at Tube CAD Journal. It is attached here solely for convenience. John is the author; I just built his design. I made changes in red to the original schematic to illustrate the post above.
Attached Thumbnails
Click image for larger version

Name:	ACF-2 Schematic Lg.png
Views:	164
Size:	63.9 KB
ID:	2318  
Attached Files
File Type: asc acf.asc (5.2 KB, 109 views)
Posted in Preamps
Views 403 Comments 0
Total Comments 0

Comments

 
Hide this!Advertise here!

New To Site? Need Help?

All times are GMT. The time now is 12:51 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki