Flux Loop

Discussion of rare things not covered in other fora.
Forum rules
Only taste in good post.
Post Reply
User avatar
Fonotec
Site Admin
Posts: 105
Joined: 03 Apr 2013, 03:13

Flux Loop

Post by Fonotec »

A good way to test the electronics of a tape machine without needing the tape transport's motors, guides, and switching to be 'dialed in' or the physical alignment of heads to be ideal, thereby isolating some of the other important factors of a tape machine's operation, such as head response and audio circuit performance, from those related to moving tape across the heads, is to use a (driven) flux loop. It's made of a coiled segment of magnet wire and is placed in electrical series with the output and return legs of an audio oscillator (i.e., 'test-tone generator'), after exposing the copper on each end of the insulated wire by gently scraping off the polymer coating with sand paper and placing a small load resistance (...not great enough to limit the necessary current from the oscillator that energizes the coil with magnetic radiation...i.e., 'flux'), also, in series, since the magnet wire has so little electrical resistance that, on its own, it would 'look like' a 'dead short' to the oscillator. Basically, if the signal generator has enough current-sourcing ability to make alternating flux lines emit from the loop of wire, the loop can be used in a fashion similar to placing a record head right against the repro head, doing away with the band of tape, and letting the alternating flux around the loop be what generates alternating electricity in the coil within the repro head. The driven flux loop and repro head coil are acting as audio transformers.

If the test signal is asymmetrical, having a different voltage-excursion on one half-cycle of its alternating current from the other (half-cycle), one can use the flux loop to test the absolute polarity (fidelity) of the (audio transferred through the) tape machine's repro circuit, since an oscilloscope can be used to view the signal going into the flux loop as well as the one exiting the machine, and if the asymmetrical signal exits the tape machine in the same orientation about the zero-crossing of the 'scope, then there is no net polarity-inversion taking place (even if numerous ones occur within the circuitry). If, on the other hand, it has an inverted orientation on output from the machine, one should swap the wires of the XLR pins, 2 and 3, on the output connectors to correct this.

For tape machines having recording functionality as well as playback, the process is more complicated, but not difficult to carry out. One signals to the inputs of the recorder with the asymmetrical test tone while recording onto tape and monitors the output from the machine, after having first ascertained that the repro-circuit (head, electronics, and output connector) polarity is non-inverting (and that the tape travels correctly across the heads - the physical alignment of the heads and the proper functioning of the motors, guides, and switching _is_ a necessary prerequisite for this procedure, of course!) . If the output signal from this tape recording's reproduction is inverted, swap the wires of the XLR pins, 2 and 3, on the input connector.

Of course sound is the result of a longitudinal pulse-train (with varying duty cycle) of atmospheric molecules which make repeating patterns of compression-, and rarefaction-, half-cycles of alternating sound-pressure. The pitch of a sound is described as its number of (whole-)cycles per second, also known as Hertz. The absolute polarity of the analog chain is supposed to maintain the correlation between an atmospheric 'compression' causing a deflection of the diaphragm of a microphone that generates a positive-swinging voltage from the capsule, and this positive-swinging voltage being transmitted through the stages of preamplification and, ultimately, amplification, during reproduction, along electrical cables having standardized connector pins for the two legs of the alternating current, so that it causes an outward displacement (towards the listener) of the correctly-wired loudspeaker's diaphragm(s), once again, producing an atmospheric 'compression'... Negative voltages created in the microphone from its diaphragm's deflection in response to atmospheric 'rarefactions', similarly, would propagate through the sound system and cabling used, resulting in the correctly-wired loudspeaker's diaphragm to move inwards (away from the listener), producing, once again, an atmospheric 'rarefaction'. Of course this would be happening tens to many thousands of times per second for the sound pressure waves to be audible.

Through listening experiments in 1957, with a single, amplified loudspeaker and an oscillator, Charles L. Wood, of the U. S. Defense Research Laboratory, proved that the human ear can distinguish when an asymmetrical signal has its polarity suddenly reversed, due to the slight, relative inefficiency of rarefactions of the same magnitude as compressions to produce the same degree of mechanical transduction into nerve-impulse current within the auditory system. (We might as well be 'wired for sound', innit?)


The absolute polarity of tape machines and other audio recording and playback equipment, including disk-cutting lathes and pickups, is standardized by the Audio Engineering Society. meaning that the standard recommends the internationally-observed practices for interconnecting audio signal leads so that the electron-holes of a positive-swinging voltage go to a specific contact on the chassis of a device (such as the red terminal on the back of a loudspeaker cabinet). Standard, AES26 (shows how to test the polarity of a reproduce head:
AES26_§6_B.jpg
AES26_§6_B.jpg (301.23 KiB) Viewed 3282 times
Notice that this conforms to Fleming's 'right-hand rule' as we view the front of the loop from the perspective of the repro head:
right_hand_rule(tape_head_view).jpg
right_hand_rule(tape_head_view).jpg (34.54 KiB) Viewed 3169 times

I first read about flux loops many years ago, on the website of Dale Manquen, here: http://www.manquen.net/audio/index.php?page=6
Wayback Machine has a 'hit' for his site from 2002, which was only a year after the AES publication, above, however, AES26-2001 §6 is merely a re-affirmation of AES26-1995 §6. Here's the rationale for the standard, as published in 1995:
Screen Shot 2021-10-22 at 9.00.33 PM.png
Screen Shot 2021-10-22 at 9.00.33 PM.png (63.83 KiB) Viewed 3247 times
Incidentally, I did attend the 111th AES Convention in New York in November and December of 2001 when AES26 was reaffirmed. I wonder if Mr. Manquen was there. I met mastering engineer and author, Bob Katz and his wife, Mary Kent, there then, and also toured Masterdisk and Sony Mastering, including the room of Vlado Meller, during the Tech. Tours on the 31st of November and 1st of December, respectively. I also visited Paul Gold, who was then working at Brooklynphono pressing plant, where he had his Neumann VMS-70 lathe and transfer console set up on the second floor. Tom Bernich showed me one of his Southern Machine & Tool presses in action and gave me a test pressing and a hardened-vinyl biscuit (with the labels on) as souvenirs.

At his website, Mr. Manquen explains the need of equalizing the frequency-response of the signals energizing a flux loop for more comprehensive analysis of the electronics and heads, since recorded flux on tape would have normally had such equalization applied (with RC networks). But 1 kHz falls withing the flat part of the NAB equalization standard, and for absolute polarity-discernment, gain is irrelevant to what is being looked for.

Although I could find no mention of it in the manual, I had been told that my MCI JH-110/M tape machine's XLR output connector was wired, pin-3-High, which made sense, since it was the customary polarity-convention for American audio equipment prior to the AES adopting pin-2-High as the global standard, and the circuit board in the machine has a silkscreened legend that actually reads 'Hi' pointing to pin 3 of the output XLR chassis connector (and, btw, the mastering machine has no input connector, since it's a reproducer, only, without any recording functionality). Futhermore, the input/output assembly of the Ampex ATR-100 is pin-3, High (and reads as much on the back of the penthouse), But, when I looked at the wires going into the backs of various tape heads of the JH-110/M, I noticed that there was no indication of their polarity, so, looking at the wires wouldn't confirm anything, and, also, there wasn't any guarantee that they could not ever have been incorrectly wired to a head assembly, either, at the factory during manufacture, or when heads are being polished or re-lapped at a tape-head specialist's workshop. So, I decided to bite the bullet and wind a flux loop and check... Here are the back sides of a couple of 2-track tape heads for reference. Note the absence of a + or - indication where the leads attach to the heads. ); The head-smith must have a flux loop!
tails.jpg
tails.jpg (102.33 KiB) Viewed 3282 times

Here's my sine wave:
1_kHz_sine.jpg
1_kHz_sine.jpg (151.45 KiB) Viewed 3282 times
Here's my sine wave with a diode in series that removes the negative half-cycle of the sinus, rendering the signal, asymmetrical, and, therefore, suitable for testing the playback head's absolute polarity with the electronics and chassis connector wiring:
asymmetrical_waveform.jpg
asymmetrical_waveform.jpg (250.45 KiB) Viewed 3282 times
('pin-2', in the legend of the photo, above, refers to the 'hot' pin of the XLR connector on the cable from the Tuchel connector under the PG70 oscillator (console cassette, plugged into the tray of the SP272), which is by Neumann, of former West Germany - so far, the only Germany I've been to, other than East...(DDR).
My custom harness redirects the 'hot'/'High' leads of its plugs to each device's jacks correctly, so, we're not getting inversions, even when a pin-2-High device is connected to a pin-3-High one.)

Here's my flux loop:
Flux_Loop.jpg
Flux_Loop.jpg (69.7 KiB) Viewed 3282 times
And here's the output of the playback head sent from the machine to the oscilloscope when the signal-energized flux loop is placed right against the repro head gap (note that the orientation of the displayed output signal is identical to that seen when the leads from the oscillator are attached directly to the oscilloscope, indicating no net polarity inversion taking place within the machine):
Flux_induction.jpg
Flux_induction.jpg (260.98 KiB) Viewed 3282 times


So, it's definitely a pin-3-High machine. Interestingly, my Studer A80 R, which was made in Switzerland where pin-2 was always the standard pin for the High signal, even before the AES standardized it (as a recommended practice, worldwide), was at some point in its history (probably when used at Jewel Studios) made pin-3 High so as to conform to the other American audio equipment being used there at the time without the operator having to remember always to use a pin-swapping cable or to flip polarity switches every time the Studer was used).


One can use a special test tape to test the absolute polarity of a tape machine, as well, but, since a reel of tape having a test tone can be flipped upside down and played backwards without sounding as if it's running backwards to the tape operator (since it's a test tone, for one, and since it's being looked at rather than listened to, for another...), resulting in an inversion of the voltage changes viewed from the output of the machine,the signal on such a test tape must be asymmetrical in, both, the voltage, and time, domains, in order to reveal the absolute physical orientation of the tape when its recorded signal's absolute polarity is being observed (on a 'scope). Whereas a 'living signal', emerging in real-time, rather than 'reel time', from an oscillator, presents a source-pattern that can't be flipped upside down before exiting the device's contacts, nor can it run in reverse from the outputs of the real-world (analog) oscillator. Note how additionally asymmetrical the recorded test signal must be on an absolute polarity test tape than the merely half-rectified sine wave that suffices when signaled from an oscillator in real time, where the orientation of the signaling leads is unambiguous:
Screen Shot 2021-01-18 at 8.04.45 AM.png
Screen Shot 2021-01-18 at 8.04.45 AM.png (57.12 KiB) Viewed 3282 times
Also of great value is the implementation and use of absolute polarity switching on the monitor and record paths of the console so that one can listen for the Wood Effect to discern the deflection of the switches that results in 'natural acoustic polarity' being transduced by one's loudspeakers.' Although AES26 is an international standard, standards aren't always followed, and a few recordings end up inverted, despite the intentions of the recordists. However, their playback can be made to sound a bit better if the sound system has such switching capability and also can resolve the subtlety audibly...

DAWs in personal computers depict the asymmetry of digitzed analog waveforms, and include an easy way to change a recording's polarity via the command, Invert (Phase). Below is a screen-grab of a recording of my voice, which uses two takes, edited together, that are of me singing the same phrases, back to back, with the same mic using two different K87 capsules (a decent-sounding, but not-as-clear, counterfeit capsule, which was identified by Klaus Heyne - when it didn't have the customary handwritten numbers on the bezel, and a NOS, echt Neumann capsule, that was sold in 2020, still in its factory case, by the Italian vendor, vintageneumann, on ebay, which does have the scribbled numbers written on the bezel) and the workstation correctly shows that I didn't vocalize a pure sine wave. (; That is to say, there is a difference in the distribution of positive (upper half of the waveform) and negative (lower half of the waveform) voltages in the recording that faithfully-enough captured the actual acoustic (sound pressure) asymmetry of the voice (in air)...
Neumann_capsule_DAW.jpg
Neumann_capsule_DAW.jpg (284.2 KiB) Viewed 3251 times
Here's an mp4 of that recording.
Ab,Pol.Aud_GB/Old&NewK87capsules.mp4

One can check to see how a recording's channels will sound in the opposite polarity from that which was recorded by highlighting the segments and commanding an Inversion.
Invert.jpg
Invert.jpg (867.32 KiB) Viewed 3251 times
However, the best way is to use a stereo switch which flips, both, the Left and Right channel polarity at the same time and can be switched back and forth quickly a number of times so that you can listen for the change in the sound of the frequency transients (such as the 'halo' of a high hat ' or snare 'hit', or the 's' sound (known as, 'sibilance') of a human voice) to determine the deflection of the switch corresponding to the playback where those transients sound somewhat higher in pitch. Interestingly, the low bass notes may also seem a little lower in resonance when the switch is in the position that makes the high frequency transients seem a little higher in pitch. This is the position of the switch that causes the recording to be reproduced with 'natural acoustic polarity'.
Post Reply