Building a tape delay

[Uusilehto]

I've always wanted a proper tape delay for use with the guitar. The digital units are fine but the ones I can afford are not even close to competing with an actual tape delay.

Basically, what I'm thinking is taking two of those hand-held tape recorders and installing them in a box with a tape going in a loop. The signal goes into the first unit that records it to the tape. The second would have multiple tape heads placed right after the writing head, distance between the reading heads being equal to the distance between the writing head and the first reading head.

There would obviously be a variable feedback going from the second tape recorder output to the input of the first one. The "extra" read heads would probably need off switches for feedback trickery. A variable speed shouldn't be too hard to achieve with a floppy drive motor. I've tried a few and there's more than enough torque in them.

Now, the actual question. Would it be possible to use a tape head from one of those car cassette adapters as a reading head? After some twenty minutes of reading about tape and how it works, I've come to the conclusion that it should work. I've ordered a few $1 units from eBay. Maybe they'll actually work.

I've got a few schematics for tape head amplifiers here. Slightly more complicated than what I'm used to but I'll give them a shot when I have time.

Any ideas?

 

[dnsey]

Most cheap cassette heads are universal rec / play, and I doubt that those used in the car adapters are specially made for the job - they may be sub-spec examples though.

I'd suggest, however, that you consider 1/4 inch tape for this project rather than the fragile and difficult to handle 1/8 in cassete tape. A reliable transport would then be a much easier proposition.

 

[Uusilehto]

I actually have some 1/4" tapes/reels and a Grundig TK-141 reel-to-reel player. I did think about basing the delay system on the Grundig but alas, it only has two heads. One for playing/recording and one for erasing. Even if I had a three-head unit, it would be very bulky and would likely have too many unnecessary parts.

Actually, now that I think about it. If I machined a new roller/alignment system, would it be possible to use 1/4" tape with 1/8" heads? Sure, there would be lots of wasted tape space but it would work, right? The mechanical side wouldn't be a problem, considering I have a fairly precise miniature lathe and a mill at my disposal.

 

[poptart]

what about starting with an old vcr? (or two) I remember a friend had an 8 track recorded that used vhs tape once. Big solid tape and still has the ease of a cassette maybe.

 

[leadbelly]

IMHO: STOP right now and buy yourself a PT2399 chip on eBay to make your delay. So much simpler, cheaper, and so much more likely to wind up with a reliable end product.

 

[dnsey]

Using1/4" tape with cassette heads is really a non-starter. Apart from the drawbacks - poor head contact, uneven stretch and weave, high noise etc., any saving will be offset by the time and troubel involved in engineering a reliable transport.

The VHS idea is a good one, though - machines (and tapes) are free these days, there's room to fit a number of audio heads - especially if the head drum is removed -, and the capstan servo could be modified for variable speed quite easily.

Solid state is certainly easier, but I can understand why you want to go with tape.

 

[jethdub]

Outside of the digital domain, does any design exist that emulates tape delay sound using solid state circuitry...what kind of analog processing would give the closest results sound wise, some kind of VC pitch shifting cicuit and some chorus or phaser type effect?

 

[poptart]

If I remember right there was an era of analog delay pedals that used a bucket brigade chip. I don't know if those are available anymore, it was so close to being a digital sampling circuit they've probably all gone that way.

 

[Uusilehto]

Right then. I now have the heads, parts for a capstan drive and tape (VHS).

Am I correct in thinking that the erase head is basically just an output head with a strong 40khz signal "blanking" the tape? Will square-wave do? Would a simple Little Gem (http://runoffgroove.com/littlegem.html) amplifier work for powering the record/play/erase heads?

Also, let's say I was to ignore biasing completely. How terrible would it sound?

 

[Limhes]

For a good explanation on tape:

http://hyperphysics.phy-astr.gsu.edu/hbase/audio/tapecon.html#c1

 

[Uusilehto]

Ok, I slapped together a crude tape transport and Set up a few tape heads on it. Both heads work but I'm not able to record anything on the tape. Basically, what I did was run a guitar through an amplifier on to the tape head. Is there some special trickery needed for this?

 

[Uusilehto]

Update. I cranked up the amps and I can make out the lyrics to funkytown now. There's just absolutely no bass.

I'll try again tomorrow with a phono preamp after the playback head.

 

[Limhes]

Have you been biasing the signal after all? As it seems not, and that's what you should. Too much hysteresis at the magnetization of the tape, if you'd read the article . Just make some 100kHz carrier and AM modulate your signal onto it. Write that on the tape, and you're set..

 

[Uusilehto]

I did read the bit about biasing but, for some reason, decided to try without it anyway.

Any ideas on a simple sine-wave generator? Would a Wien bridge oscillator work?

 

[Limhes]

Would be just fine. Then add a summing network after it to sum bias signal and audio signal, using for example an inverting opamp with two input nodes. You can match all levels using this opamp, changing the three resistors. Attach its output to the write head and you're ready, I'd think...

 

[hailteflon]

If I read you correctly, you are looking for something similar to what Sam Phillips did at Sun in the mid-50s. "Slapback" echo for Rockabilly was obtained by using two of the early Ampex tape machines. I caught this on a cable documentary. Perhaps some research into early Sun equipment will help.



Mark

 

[el`Ol]

I ask myself whether the edge of an old 5 1/4" hard disc would be long enough for the task.

BTW,
all the big names have delays for 100€, was there really nothing for you?

A bit more expensive, but much more than a delay (and analog filters in the feedback!):
http://www.jomox.de/product_details.php?lang=2&category=&product_id=3

Maybe you like the guitar demo sounds.

 

[bullpeters]

still going?

I did read the bit about biasing but, for some reason, decided to try without it anyway.

Any ideas on a simple sine-wave generator? Would a Wien bridge oscillator work?

any progress with this. for me the mechanical side seems easier than the electronic. i also have a good tool shop. i would love to hear about your project or if you gave up
mick