# AC-bias in tape heads: could someone explain?

I understand that the purpose of AC-bias is to improve the quality of a tape recording by countering some otherwise adverse physical properties of practical magnetic materials.

I've had this explained to me in a number of different ways, often contradictory, always vague. Online sources like Wikipedia don't help because they're more muddled and vague than most!

Could someone who is eloquent and knowledgeable give the definitive answer as to the physical processes which are involved and how AC bias helps? Actual equations preferred to hokey analogies.

• There is a non-linear region around zero inherent in recording tape. The purpose of the AC bias is to raise the minimum signal level out of this region. Sep 12, 2015 at 10:29

It's to overcome the hysteresis of the magnetic tape.

The record head is a small electromagnet pressed against the tape. If a current is applied, it magnetises the oxide particles on the tape. the stronger the current, the more magnetic domains are re-aligned and the stronger the signal recorded.

However, a very tiny current will have no effect on the tape at all. This would mean that weak signals would end up distorted (as only the peaks would be recorded). Even weaker signals would not be recorded at all.

Adding a high-frequency bias ensures that there is always enough magnetic field to magnetise the tape. However, the frequency is higher than the tape is capable of recording, so it doesn't affect the result.

• Note that the record bias does affect the recording - it tends to erase some of the signal. The higher the bias, the more signal gets erased. This is compensated with record equalization adjustments. Sep 12, 2015 at 19:32
• Ah, ok, so this is starting to make sense. The much mentioned non-linearity is down the bottom, near zero, where it takes more current than expected to change magnetisation? Is the bias "added" (in voltage terms) to the signal like DC bias (which would still be a bit confusing), as most sources seem to imply, or is it multiplied, like AM (which would seem to make more sense)?
– Dan
Sep 12, 2015 at 21:59
• The bias is added. If it were multiplied, AM style, it would not work. I'm going to avoid hokey analogies at your request, but the high frequency signal "stirs up" the magnetic domains, while the recording signal "biases" where the domains will settle. So, in a strange sense, we named things backwards. Sep 13, 2015 at 7:27

I can't give you equations, but I can tell you that the main purpose of AC Bias applied to the record head of an analog tape recorder is to overcome the inherent hysteresis of the magnetic oxide contained in the binder on the recording tape.

I'm no longer competent to give you an extremely detailed explanation - I last worked with analog tape recorders more than 20 years ago and this kind of memory is one of those "Use it or Lose it" type of things.

In general, we used to supply more bias than necessary to improve the distortion figure that we could get from analog tape - the number that is still burned into my brain is 3dB of over-bias.

I do remember playing the amount of over-bias for the various tape brands that we used - Ampex 406 and 456 tape was the most common tape in use by the recording studios and broadcasters that I was working with at the time. We spent days messing with bias and equalization to find the best compromise between distortion and S/N ratio. This was massively fun work way back then but I wouldn't care to repeat it now.

I went looking for the calibration tapes that I used the most recently and found their home page: the company is Magnetic Reference Laboratories and much documentation can can be found at their home page. I haven't reviewed this documentation in several decades and can't tell you which file contains the information that you are looking for but it is a great place to start looking.

These people are what I would consider to be the modern-day experts in regards to analog magnetic tape recording.

Look further down the page for technical documentation.

MRL

[Edit 2]

One of the links on the above page is to a PDF that does a great job of describing what AC Bias is and why it works. The link is: AC Bias

The non-linearity in the way the tape is magnetised usually exists at both low and high levels of magnetisation with a region in between which is more linear. Before High frequency bias was discoverd, heads were biassed with a DC current to try to shift the magnetisiation into this 'linear' region. It was not all that successful and the advent of HF bias transformed magnetic recording. The exact frequency is not critical but should be 3 to 5 times greater than the highest audio frequancy to be recorded. The bias current is critical and depends on the type of tape you are using. The optimum adjustment is a compromise between achieving the minimum distortion without erasing too much of the high frequency audio; an unfortunate side- effect. Techniques such as feeding the bias to a separate head or modulating the bias current dependent upon how much HF there is in the audio signals (Dolby HX Pro) were techniques developed to try to overcome tis.