teach me about magnetic tape

[Dimitree]

hi everyone
as experiment, I'm trying to make a simple and cheap tape echo.
of coursem I'm not expecting quality out of this thing. I have enough delay units so I just want to experiment with DIY.

I'm using 3 cassette magnetic heads (erase, read, play) that I took out of an old Aiwa cassette system.
I'm using a loop of magnetic tape around a motor (like Roland space echo and similars), so I'm not using the whole cassette as is, or reels.

At this stage my doubts are:

- can I use 1/4 inch tape instead of cassette tape (given that I properly align the tape to the heads)? I believe that 1/4 inch tape is stronger than cassette tape, so it should last longer. I think cassette tape would be almost dead after some hours of looping.
Do different tape need different electronic/heads or they just differ in width and durability?

- how do I interface line levels and guitar levels to the heads?
the Aiwa used a BD3881FV chip ("audio sound controller") and surrounding circuitry to interface AUX/tuner inputs to heads and heads to outputs.
Is there a simple way to do with discrete parts? How should I handle noise reduction?

[PRR]

Tape is tape. Cassette tape will give good service. Use what you got.

Tape playback is a simple preamp, like a phono preamp with different EQ. I know we discussed this here in the last month.

Recording.... tape was around for 40 years before good recording technique was developed. It is NOT simple. The build but especially the calibration. Strongly advise you to pull the whole electronics package along with the tape heads and just use that.

Hiss reduction: running the tape fast (bigger loops) is far simpler than any electronic scheme.

[PRR]

http://tubebooks.org/technical_books_online.htm#Other%C2%A0

Start with Burstein & Pollack.

Continue with Tuthill.

Quartermaine may be of interest. It is short, and tells how to cobble-up a tape deck out of scrap metal and old radio tubes. Today you would not do it this way, but the process is educational.

[Mark Hammer]

Well, I suppose tape IS tape, but know the following:

• 1/4" tape was developed for open-reel use, and open reels could be anywhere from 3" to 10" across, and of no particular length/duration.  As such 1/4" tape came in a variety of thicknesses, with thinner tape permitting more time on a single reel, but thicker tape being hardier and allowing less print-through.  Cassette tape (1/8") was intended for a fixed-size format, and pre-defined lengths (C-20, C-45, C-60, C-90, C-120).  The longer the time, the thinner the tape needed to be to fit in there.  If you can still find them, what you probably want is the C-20 size that was used for storing programs to tape in the early days of personal computing (VIC-20, ZX81, etc.).  Because less time was expected, the tape could be thicker and still fit in the case.  As well, where music can still be heard even if the tape crinkles or stretches in places, digital data HAS to be precise, so thicker tape assures that.  You can also pack a little more particles on thicker tape for a more robust S/N ratio.  I may have some old computer tapes kicking around, but it will be tricky finding some of those C-20s all these years later.  Your best bet may be yard sales.
• All tape really is, is ferrous oxide particles (fancy rust) adhered to an acetate strip.  If you have used press-and-peel blue, you will know that there is only so many times you can flex and bend that sheet before the blue particles fall off.  You also likely know there are only so many times you can whip the wiper across the surface of a pot before scraping off the resistive strip to the point where contact becomes intermittent.  So, the smoothness and tension-minimized aspects of thetape transport, coupled with the sturdiness of the tape itself, determine the longevity and fidelity of the tape.  And naturally, the longer the tape loop, the less abrasion any single point in the loop has to endure, and the longer it waill last

T'wer I, I'd lean towards 1/4" tape, but then fundamentally one is constrained by whatever tape transport system and accompanying electronics one is using.

[Dimitree]

thanks for the replies.
I have a lot of 1/4" tape, so I'd prefer to use it instead of cassette tape..and since (in the same conditions) should be more durable, I'd be happy with it.
So from what I understood, since tape is tape I can use my cassette head with whatever tape, even my 1/4" tape.

Luckly I'm not constrained by the tape transport system, since I'm studing my own system (after all I just need to move the tape around the heads and keep a constant tension)

[Mark Hammer]

You can't just stick the one tape on the other type of tape transport.  The transport system has to line up the portion of the tape holding the signal with all the various heads.  Gotta pick the one system or the other.

[Brisance]

You can't just stick the one tape on the other type of tape transport.  The transport system has to line up the portion of the tape holding the signal with all the various heads.  Gotta pick the one system or the other.

That's a shame since I recently aquired a soviet tape echo for very cheap which works, but the tape on it is so degraded, that any sound put on it gets overpowered by noise and gets mangled very fast, I was hoping I could just get some 1/4" and loop it to replace.

[wavley]

You can't just stick the one tape on the other type of tape transport.  The transport system has to line up the portion of the tape holding the signal with all the various heads.  Gotta pick the one system or the other.

That's a shame since I recently aquired a soviet tape echo for very cheap which works, but the tape on it is so degraded, that any sound put on it gets overpowered by noise and gets mangled very fast, I was hoping I could just get some 1/4" and loop it to replace.

Mark is talking about 1/8" cassette taper vs. 1/4" open reel tape (and by extension, 1/2", 1", and 2" tape)

If your machine is made for 1/4" tape then you will be fine.

[PRR]

A big question is: do the cassette heads have tape-guides? If so, 1/4" tape won't fit; you may be able to bust the guides off.

A less critical point: signal/hiss ratio is related to track width. Cassette 4-track (stereo 2 ways) has much narrower tracks than say 1/4" Full Track Mono (great stuff!), 1/4" half-track, or even 1/4" 4-track. You used to be able to buy the head(s) of your choice, but today it's all scrap-finds. OTOH, signal/hiss ratio is also about tape speed, and for a tape-loop you are not concerned with minutes/reel, so you can run quite high speed (at the cost of a bigger loop-board to handle the increased loop-length for a useful delay).

Hmmmm.... a 3600RPM fan-motor, with a 1/4" shaft, makes a crude "capstan" giving 47 inches/second. This is faster even than pro-pro 30ips and could give fine S/N even with narrow tracks. However for say 7 second delay (broadcast swear-word buffer) you need about 28 feet of tape, 14 feet out and back, or a lot of little rollers (and friction) on a foot-square board. However for a 1-second delay (fast rondo) the loop fits a 2-foot stick.

Loops do not have to be horizontal. I often let them run vertical. With a TEAC in a rack, 7 feet doesn't touch the floor. Longer loops can be spilled into a clean cardboard box and pulled out again, but maybe only for Electronic Music Lab work-- you have to keep a constant watch for tape-tangle, which a solo guitarist could not do well.

It is possible to build a box 0.26" thick, a foot or more the other ways, with a roller in a bottom corner. Tape spills into S-curves. The narrow clearance does not allow tape to foul itself. The bottom of the pile is pulled-out under the roller and back up to the drive. Once working *smooth*, such contraptions were fairly reliable in tape-duplication work (where the master was an endless loop 20 minutes long).

[Dimitree]

my question was indeed not related to any particular mechanical system.. my biggest doubt was: assuming a theorical perfect mechanical system (and perfect alignment of the heads with respect to the tape), a cassette head can record on 1/4" tape?
looks like yes,
I thought there would be some differences, for example a different eq or different bias

another question I have is..since cassette heads are usually stereo, and I'm happy with mono..how could I exploit the 2nd channel?
can I improve S/N in some way, maybe using the 2 channels in a differential amp fashion?
or maybe doubling the delay time putting them in series?

[PRR]

EQ/bias *will* be the major problem.

(Which is why I suggested stealing the *whole* electronics package.)

For mono, just strap the record heads in parallel, playback in series. (This will give slightly more headroom on record, a slightly higher output level on playback.)

[Dimitree]

EQ/bias *will* be the major problem.

ok, but at least now I know that this problem should be indipendent of the kind of tape I will use

[Digital Larry]

Another practical matter, which you may or may not care about, is that tape is abrasive and will eventually wear down the heads.  Nobody much cares about that because it takes a pretty long time.  At first this will result in a loss of high frequencies.  Main point would be to turn it off when you're not using it.

[Dimitree]

Another practical matter, which you may or may not care about, is that tape is abrasive and will eventually wear down the heads.  Nobody much cares about that because it takes a pretty long time.  At first this will result in a loss of high frequencies.  Main point would be to turn it off when you're not using it.

thank you, this is really helpfull.. I could stop the motor when the effect is in bypass (but it would probably give a huge pitch shift effect once activated again) or I could mechanically remove the contact between tape and heads in some way I should think of..

[Mark Hammer]

S/N will improve with faster speed.  You can use any of a variety of electronic approaches to noise reduction, including companding and/or pre-empasis/de-emphasis, both of which are regularly used in BBD-based effects.

[wavley]

For mono, just strap the record heads in parallel, playback in series. (This will give slightly more headroom on record, a slightly higher output level on playback.)

This is what the space echo does, stereo heads wired mono.

[wavley]

Another practical matter, which you may or may not care about, is that tape is abrasive and will eventually wear down the heads.  Nobody much cares about that because it takes a pretty long time.  At first this will result in a loss of high frequencies.  Main point would be to turn it off when you're not using it.

thank you, this is really helpfull.. I could stop the motor when the effect is in bypass (but it would probably give a huge pitch shift effect once activated again) or I could mechanically remove the contact between tape and heads in some way I should think of..

The earlier Space Echoes stop and start the tape, it gives fantastic effects stopping and starting.  Not as much pitch shift as you would think stopping and starting if you don't touch the rate control, BUT whatever echoes you had when you stopped the tape are still there when you restart giving a nice "womp womp womp" when you restart, barely noticeable when you're playing but kind of awesome if you do it during a rest.  The other fun thing is to ramp up the speed and intensity, stop it at the top of a crescendo, turn the speed and intensity down, and turn it on... now that's some fun pitch shifting.  This is why I consider my space echo as part of my instrument and less as an effect.

The later Space Echoes free run the tape always, no fun stop/start effects, but there are tails when you kill the effect send, you can also set it to self oscillate but kill the input so you can play over it without adding to it.

I've settled into using my 101 for guitar and 501 for vocals, the two style compliment the sources in my opinion.

S/N will improve with faster speed.  You can use any of a variety of electronic approaches to noise reduction, including companding and/or pre-empasis/de-emphasis, both of which are regularly used in BBD-based effects.

The 501/555 space echoes do this, they are much quieter than the earlier ones... to the point of me missing the tape hiss when making whooshy self oscillation noises. 

[Dimitree]

that bad thing is that I can't find anymore the associated electronic circuit of my cassette head..
I only got the 3 heads (record/playback, playback, erase) and their shielded cables.
am I lost, or can I use them even if I don't have any datasheet of them and I don't know their specs?

[Blitz Krieg]

EQ/bias *will* be the major problem.

ok, but at least now I know that this problem should be indipendent of the kind of tape I will use

the deck is usually calibrated toward a specific tape formula.  have you looked up the technical reasons for bias? 

I know you're just trying to get this 'off the ground', but there are so many considerations.  And that is why you don't see anything like this out there.

Continuing down this avenue, your best approach would be to find a functional open reel tape deck.  These can be found cheap, but shipping is prohibitively expensive.
In that case you've essentially acquired an antique, and now you're going to hack into it.  For less than stellar results. 

The old stereo open reel decks are sometimes described as 4 track.  These are not 4 tracks like a tascam portastudio.  They are 4 track because they are stereo in both directions of the tape.  You can get delayed signal off of these tape machines.  But even then, mixing the original signal with the delayed signal is difficult.  And might even require two amplifiers depending on your equipment.

We all encourage you on this pursuit, and will be super excited if you get this working.  But it's good to have achievable sonic aspirations, rather than merely using components that are laying about.

[PRR]

> indipendent of the kind of tape I will use

Pretty-much. Tape is tape.

Black-oxide will take a little more signal than red-oxide. Chrome tape even a bit more. But the differences are small for your purpose.

As for the EQ- this is always a compromise. The basic "bend" in the EQ curve can be proportional to tape speed: 3KHz for 30ips, 800Hz for 1.875ips, though 1.6KHz(?) for Chrome at 1.875ips to counter the severe losses of low speeds.

The major obstacle to DIY recording is BIAS. You need many dozen Volts of very pure 80KHz tone to linearize the tape. You can't do this with common audio parts. You really want a specific coil. The amount of bias for erase is substantial. The amount of bias for record is less, but very critical, and you really have no clue where to start.

As Blitz and others are saying: scrounge a WHOLE tape recorder. Get it working as reel/cassette. Then start hacking it into a loop player. You get weeks worth of development and calibration already-done.