What a simple and elegant idea - use the bias signal as a clock; 
wow+flutter=jitter; heterodyne - same as recovering audio modulation 
from an RF signal.  Congratulations to all involved.

One criticism I heard when I mentioned this technique to someone in the 
restoration community - is the bias signal sufficiently stable to ensure 
that the derived signal is accurate?

Just my 2 cents,

Mike Gray

David Glasser wrote:
> I can answer some of the questions raised about Plangent, the need for 
> specialized replay electronics and heads, and the machine at my 
> company, Airshow:
> 1) The bias signal was never intended to be recoverable. That this is 
> possible is a happy accident, but stock replay electronics and heads - 
> by design - will not reproduce the bias signal.
> 2) The Plangent replay electronics and playback head have several 
> features designed to make bias signal recovery possible - the head 
> itself is a wide-bandwidth design and the first amplification stage is 
> built into the headblock itself. The Plangent electronics are designed 
> for wide-bandwidth, low noise reproduction, and they do not have a 
> bias trap filter. The bias signal of most tape recorders is too high 
> in frequency to be captured at 192kHz, but the Plangent electronics 
> contain a very clever heterodyne mixer that will downconvert the bias 
> signal, after its nominal frequency has been identified, to 24 or 
> 48kHz which is easily captured by 96k or 192k conversion (and recorded 
> to a separate digital track). A dedicated bias output feeds a PC-based 
> spectrum analyzer used to identify the bias signal. An earlier 
> incarnation of Plangent used a separate head to recover the bias. A 
> separate head is no longer used. The electronics were designed by the 
> late David Smith, and refined and hand-built by John Chester. We will 
> post photos of the electronics, etc on the Airshow website soon.
> 3) Bias from a tape recorded on an ATR-100 is indeed not recoverable. 
> ATR-100 tapes, however have other tell-tale signals that can be used 
> as a reference. The ATR's logic circuits emit rf noise at several 
> frequencies and tend to bleed into the record circuitry. For example, 
> simply turning off the ATR's LED time display eliminates several noise 
> spikes observable on our spectrum analyzer. (Consequently, we have 
> installed an on/off switch on the display.)
> 4) While the transport of an ATR is not as gentle as a Studer A820, it 
> can be safely used for all but the most fragile tapes. If we see 
> enough of a demand for this service, we will add Plangent heads to our 
> A820.
> 5) The Plangent processing is done at Jamie Howarth's lab in 
> Massachusetts. We are only offering optimized transfers, and a conduit 
> to Plangent's services.
> 6) As tape speed drifts, so to will the frequency of the recovered 
> bias. For this reason, Plangent recommends that a reel be transferred 
> song-by-song, and the bias frequency identified for each song. This 
> obviously can increase the time and cost of the transfer.
> 7) All of this would be besides the point if it didn't sound good. The 
> Plangent replay electronics are on a par with other high-end tape 
> replay electronics like the Aria. I am naturally skeptical of digital 
> signal processing, especially processing on the scale of the Plangent 
> Process, but after hearing some of Jeffrey Norman's Plangent-processed 
> Grateful Dead  masters, I am convinced that this is a major 
> breakthrough in audio restoration technology.
> I've been involved in audio restoration for many years, but am new to 
> the ARSCLIST, so I apologize if this post is inappropriate. We are 
> very excited to be associated with Plangent, and I invite anyone who 
> is interested in this technology to contact me directly, or contact 
> Jamie at Plangent.
> David Glasser
> Airshow Mastering
> Boulder, CO
> 303-247-9035
> On Apr 29, 2010, at 7:31 AM, Shai Drori wrote:
>> Hi Andrew
>> I meant  that the recorded signal is so high that it is not present on
>> tape, I think that is the magic of it. It still biases the heads but
>> it's just too high for the tape itself. Don't forget that at 15ips
>> you're describing a theoretical 56kHz at 1 7/8 ips. It's just not there.
>> I love it. I like the way it sounds much better than the reocridngs I
>> made on otari or Studer, dare I mention tascam. I can see the benefits
>> of this system for film, but for nice audio tape in good condition, an
>> ATR recording will be hard to lock to. Did they transfer any command
>> recordings? I would love to hear the Borelo!!! What a recording (I know
>> I will now ge tons of emails about my taste in music. In my defense, my
>> wife would be on your side and add her thoghts about my clothes as 
>> well).
>> Shai
>> On 4/29/2010 1:52 PM, Andrew Hamilton wrote:
>>> Dear Shai,
>>>      As I wrote in your quoted response, below, there are other
>>> signals besides bias that the PP can lock to.   They mention "logic
>>> control" signals.   I suspect there are other ghosts they could bust.
>>>      The ATR is only bias-free while reproducing.   But when
>>> recording, it should be present, at 432 kHz - almost as high as the
>>> Dave Hill Aria bias.  Even if you recorded at 30 ips and then, on
>>> playback, select 3.75 ips and then also  vari-speed the oscillator to
>>> 50% of that, the 432 kHz bias signal would still be at 27 kHz.  Did
>>> you remember to digitize at 2x F/s?  Otherwise, it would have been
>>> filtered out by the ADC.
>>> Please audition the samples on the PP website.  They made a believer
>>> out of me.   There's a Waves plugin to inject wow and flutter into a
>>> digital recording.   The PP would not be able to undo this since it is
>>> simulated and does not contain a veiled clock.
>>> Andrew
>>> On Apr 29, 2010, at 5:53 AM, Shai Drori wrote:
>>>> I can see that working with 160 or even 200 (you can hear it if you
>>>> slow down the tape enough), but the ATR is bias free, at least on
>>>> tapes I made so far. The system is interesting though. How does it
>>>> monitor to see that the original bias did not drift while recording?
>>>> This would make you track the wrong frequency..
>>>> Shai
>>>> On 4/29/2010 10:24 AM, Andrew Hamilton wrote:
>>>>> There are other forensic time stamp signals besides bias which would
>>>>> allow for an ATR-100-recorded tape to be de-fluttered by the PP
>>>>> DSP.   However, even though a tape may be played back by the Airshow
>>>>> ATR-100, it's entirely possible that the tape was recorded elsewhere
>>>>> by a different machine (having a much lower bias f).  I believe that
>>>>> Airshow are offering this service with PP for already-existing
>>>>> analog tapes, rather than for creative layback transfers.
>>>>> David Glasser is chief engineer at Airshow and he has mastered a
>>>>> huge amount of audiophile CDs, DVDs, and SACDs.   Great ear; great
>>>>> rooms; great gear.
>>>>>  From the PP website:
>>>>> "software algorithm, developed with researchers at Cambridge
>>>>> University in England, which identifies a steady-state ultrasonic
>>>>> reference tone (such as tape bias or logic control) embedded within
>>>>> the original analog signal and then performs continuous
>>>>> high-resolution pitch correction in order to keep the reference tone
>>>>> at a fixed frequency..."
>>>>> Andrew
>>>>> On Apr 29, 2010, at 4:15 AM, Shai Drori wrote:
>>>>>> My experience with the ATR is just the opposite. I have tested
>>>>>> various transport and some tapes were handled only by the ATR. Does
>>>>>> the system figure out bias frequency automatically. What does it do
>>>>>> with tapes recorded on the atr where the frequency is so high it
>>>>>> doesn't show up on playback (400kHz+)?
>>>>>> Shai
>>>>>> On 4/29/2010 8:54 AM, Paul G Turney wrote:
>>>>>>> Well they only use ATR 102 machines which are notoriously rough
>>>>>>> tape handlers....
>>>>>>> They use software to track and maintain a bias frequency so that
>>>>>>> any speed anomolies are and wow and flutter are reduced by
>>>>>>> maintaining perfect pitch with this tone.
>>>>>>> Not worked with Airshow mastering.
>>>>>>> It appears to be a monopoly on the software so Airshow would be
>>>>>>> subbing the work out to PP.
>>>>>>> PT
>>>>>>> -----Original Message-----
>>>>>>> From: Chandra Lynn [mailto:[log in to unmask]]
>>>>>>> Sent: Thursday, April 29, 2010 02:12 AM
>>>>>>> To: [log in to unmask]
>>>>>>> Subject: [ARSCLIST] Airshow Mastering&  Plangent Processes
>>>>>>> I noticed some earlier postings about Plangent Processes. It
>>>>>>> eliminates wow,flutter and speed aberrations from analog masters.
>>>>>>> They are now working withAirshow Mastering to offer optimized tape
>>>>>>> transfers. The announcement is onAirshow¹s site at
>>>>>>> any of you
>>>>>>> worked with Airshow or Plangent? If so, what has been 
>>>>>>> yourexperience?