Biasing Magnetic Tape Head

[MattHollands]

Hi!

I am building a Tape Delay pedal. This requires writing a signal to magnetic tape.

I've read everything I can find online about writing to magnetic tape and apparently you can either DC bias the signal or AC bias the signal. Supposedly, AC biasing the signal is best - you mix the signal with a ~100kHz sine wave at ~10x the magnitude of your signal.

What I am doing is using an OPA2134 op-amp to drive the tape head using an ~400mV signal. What I have found is that when I DC bias the signal by ~50mV, I get the best results (this is what the tape recorder I took the tape head from did). When I AC bias the signal, the resulting sound wave (when read back) is very distorted and noisey. I have tried AC biasing with amplitudes between 50mV and 2V with little effect of the sound.

However, everything I read online says that AC biasing sounds better so I cannot explain why it sounds so bad to me.

Does anyone have any experience with writing audio to magnetic tape? Do you need different tape heads for AC biasing?

[mzacharias]

Not an engineer - but I often see audio bias at 100 volts peak.

[Alex Eisenhut]

Look up the "hp magnetic recording handbook", I can't find a PDF for it now but it must exist.

You sure you got good heads, transport, tape, etc...

[MattHollands]

Assuming your're refferring to this book http://books.google.co.uk/books/about/Magnetic_tape_recording_handbook.html?id=OtlFAQAAIAAJ&redir_esc=y I can't get hold of it anywhere Can't find the pdf and I've searched my university library.

I'm pretty sure I've got a good tape head. I'm using the parts from this tape recorder http://www.amazon.com/Panasonic-RQ2102-Cassette-Recorder/dp/B00004T1XK and it plays back pretty nicely, I just can't get AC biasing to sound good.

Furthermore I am using all of the mechanics directly from the tape recorder so all the transport should be the same.

I assume the tape is fine because it plays back well, but could give a different tape a try.

Does it matter that the tape recorder uses a permanent magnet as the erase head?

[MattHollands]

If it helps, this is what AC biasing sounds like when I try it, https://soundcloud.com/mattahol/ac-biasing, it is just plucking a guitars open strings

[flynwill]

If you have the whole working deck, I'd suggest putting a scope probe on the tape head while it is in record mode and see what the bias frequency and level is.  Alternately why not just scavenge the record & playback circuits from the deck?

Also I'm curious, did you buy two of these to work with?  Since cassettes normally use the same head for both record and playback I would assume your unit only has the one head (plus the erase head which probably won't work very well for either record or playback).  And your tape delay is going to need simultaneous record and play.

My memory is that the bias level is one of those tuning adjustments all tape decks had.  Get the level too high and the bias starts to erase what has just been recorded, get it too low and the distortion goes up.

[MattHollands]

If you have the whole working deck, I'd suggest putting a scope probe on the tape head while it is in record mode and see what the bias frequency and level is.  Alternately why not just scavenge the record & playback circuits from the deck?

I have done this and the record signal was 50mV DC biased, and this is what works well-ish when I try. However I had read online that AC biasing works better and so thought I would try to use that instead, but perhaps this head just isn't designed for that. Also I'm using this as an analogue learning experience so I wanted to build it all from scratch. The record & playback circuits are all integrated into one purpose made chip which is a bit boring! 

Also I'm curious, did you buy two of these to work with?  Since cassettes normally use the same head for both record and playback I would assume your unit only has the one head (plus the erase head which probably won't work very well for either record or playback).  And your tape delay is going to need simultaneous record and play.

Well, I am working to a budget so I actually got this for free. I also bought a rubbish old walkman (that stank of battery acid) on ebay but the head sounded very poor (and then a pin broke off it) so I'm trying to get another head for playback.

The erase head is actually a permanent magnet so definitely no good for playback or recording 

Do you have any idea roughly what AC biasing levels to use? I'm a little scared to put too much current through a tape head.

[Alex Eisenhut]

Assuming your're refferring to this book http://books.google.co.uk/books/about/Magnetic_tape_recording_handbook.html?id=OtlFAQAAIAAJ&redir_esc=y I can't get hold of it anywhere Can't find the pdf and I've searched my university library.

I'm pretty sure I've got a good tape head. I'm using the parts from this tape recorder http://www.amazon.com/Panasonic-RQ2102-Cassette-Recorder/dp/B00004T1XK and it plays back pretty nicely, I just can't get AC biasing to sound good.

Furthermore I am using all of the mechanics directly from the tape recorder so all the transport should be the same.

I assume the tape is fine because it plays back well, but could give a different tape a try.

Does it matter that the tape recorder uses a permanent magnet as the erase head?

Sorry, I just automatically assumed a reel to reel monster with their usually soft heads and cranky mechanisms. But as flynwill noted, you need separate heads for delay, and either variable spaced heads (good luck), or the kind of closed-loop system with slack that lets you vary the tape speed manually, or be happy with one delay setting.

This tape deck is like a dictaphone, who knows how they work.

PDF here ->
http://www.hparchive.com/appnotes.htm

AN 89

[flynwill]

Afraid it's been 30+ years since I looked closely at tape-deck record/playback circuits, so I can't be much more help.

If the head was designed for DC bias then AC bias may well be impractical.  All of your bias power may just be heating up the head's core.  You might try significantly lowering your Bias frequency.  I expect you could probably take it down to 10-20kHz and still not have much of it actually recorded to the tape.
 

[MattHollands]

Sorry, I just automatically assumed a reel to reel monster with their usually soft heads and cranky mechanisms. But as flynwill noted, you need separate heads for delay, and either variable spaced heads (good luck), or the kind of closed-loop system with slack that lets you vary the tape speed manually, or be happy with one delay setting.

Thanks for the pdf! Looks like an interesting read  I'm hoping to be able to control the speed of the motor just by sticking a variable resistor in series... haven't actually tried yet but I reckon that shouldn't be a problem.

Afraid it's been 30+ years since I looked closely at tape-deck record/playback circuits, so I can't be much more help.

If the head was designed for DC bias then AC bias may well be impractical.  All of your bias power may just be heating up the head's core.  You might try significantly lowering your Bias frequency.  I expect you could probably take it down to 10-20kHz and still not have much of it actually recorded to the tape.
 

I gave it a shot with 20kHz bias and it actually sounded a lot better! (any lower and I could start to hear it through my amplifier). Not as good as DC biasing but better. Having tried a whole load of configurations, I think I'm going to stick with DC bias. The circuitry is simpler and it seems to sound ok.

Thanks a lot for all of the help. There isn't too much information about tape recording any more! Although the pdf looks like a good source. 

[Alex Eisenhut]

Sorry, I just automatically assumed a reel to reel monster with their usually soft heads and cranky mechanisms. But as flynwill noted, you need separate heads for delay, and either variable spaced heads (good luck), or the kind of closed-loop system with slack that lets you vary the tape speed manually, or be happy with one delay setting.

Thanks for the pdf! Looks like an interesting read  I'm hoping to be able to control the speed of the motor just by sticking a variable resistor in series... haven't actually tried yet but I reckon that shouldn't be a problem.

Afraid it's been 30+ years since I looked closely at tape-deck record/playback circuits, so I can't be much more help.

If the head was designed for DC bias then AC bias may well be impractical.  All of your bias power may just be heating up the head's core.  You might try significantly lowering your Bias frequency.  I expect you could probably take it down to 10-20kHz and still not have much of it actually recorded to the tape.
 

I gave it a shot with 20kHz bias and it actually sounded a lot better! (any lower and I could start to hear it through my amplifier). Not as good as DC biasing but better. Having tried a whole load of configurations, I think I'm going to stick with DC bias. The circuitry is simpler and it seems to sound ok.

Thanks a lot for all of the help. There isn't too much information about tape recording any more! Although the pdf looks like a good source. 

I don't know if you said what kind of tape you are using? This kind of dictaphone is probably meant to record to Ferric tapes, not the metal/Chrome high bias tapes.

[MattHollands]

I don't know if you said what kind of tape you are using? This kind of dictaphone is probably meant to record to Ferric tapes, not the metal/Chrome high bias tapes.

Erm, I'm not really sure how to tell the difference. I walked into Oxfam and bought a 50p Mozart tape. I can't find anything on the tape saying what type it is. Sorry to be useless. Is there any way I can tell the type?

[SeanB]

Light brown is a ferrite tape, chrome and metal film tapes are a very dark brown to nearly black.

[MattHollands]

Ah according to Wikipedia, you can tell by the number of notches  This tape has only 1 pair of square notches, so I THINK it is a type 1 tape ie ferrite.

Light brown is a ferrite tape, chrome and metal film tapes are a very dark brown to nearly black.

Yes, this is light brown, so ferrite!

[Alex Eisenhut]

Light brown is a ferrite tape, chrome and metal film tapes are a very dark brown to nearly black.

(pedant mode ON)
It's ferric oxide. Ferrite is iron.
(runs away)

[SeanB]

All aside from the metal film are metal oxides, just the base metal differs. Ferric oxide tapes are mostly iron (III) oxide in a fine powder with a binder, the Chrome tapes contain a higher proportion of Chrome (IV) oxide as a blend.

[MattHollands]

Does this mean that DC biasing is more suitable? Or is this a bad thing?