DIY Tape Delay

[midgetfidget]

I've been wanting a tape delay for ages, but I find most of the ones that come up on ebay are usuall rather expensive.

So naturally I came up with the idea of building one myself.

Anyone here have any experience with this, or can point me to some useful info?

[bigtex]

Here's one I've been planning on making since I saw it. The cool part is it's made with a cheapo consumer tape deck, and it appears to be rather easy to make. Now this is a flanger, which is a short modulated delay, but it could probably be adjusted for longer delay times as well.

http://homepage.ntlworld.com/henry01/flanger/flanger.htm

Check out the rest of the stuff on that Henry's Exciting Electronics page. Some really neat ideas on there (and some cool tunes under "Switched on Tijuana").

Henry's also on this board as Kerr_Avon so maybe you could ask him some questions about it.

[midgetfidget]

Looks interesting,

A little confused as to how it works, even after reading everything.

My idea was to get a tape deck/reel to reel and modify the erase / record heads so I could get it to delay. Then control the motor speed.

Would it be that simple?

[zipzap]

It is quite simple and it sounds great! I was experimenting with something like that till i killed some resistor in my tapedeck. Took some time to figure out, but now i´m with this project again.
I use a three head tapedeck. You´ll need one of those, as they can record and playback (with some delay) at the same time.
I found a little trimpot in series with a resistor labled tapespeed. Replaced it with a pot. I´m not sure what it´s doing, larger resistance means faster play. There is now a huge usable range. At very fast speed there comes a point where i get really nasty highpitched noise. It starts emediately when i get across a certain point. When it gets too slow there is a point where the tape just stops. Inbetween those points the range is huge. The sound changes very much, getting duller at slow speeds, of course. To get a longer delay you can chain the left and right chanel.
Now i want to design some controll circuit. It should have: Feedback, VC-Tapespeed, controll of paralell/serial mode. This gets interesting, since you could combine parallel and serial mode. Like chaining the two chanels, using both outputs, crossfeedback etc. Have to check out a few things there. At faster speed and in parallel mode this could work like a stereo tapeflanger as well, since the delay is quite little. Fun stuff.
About the vc speed i´m not sure yet. I would like to use a vactrol, but to get really slow the resistance has to be very small, maybe some vactrols in parallel will work.

[midgetfidget]

Did you have to modify the heads at all?

[zipzap]

No. It´s a 3 head deck. That means you have one head for playback, one for ereasing, one for recording. This is done so that you can monitor what´s actually on the tape while recording, not just what´s going to the tape. Since the tape first passes the record head, than the playback head you get a delay.
Reel to reel, are those the ones with the tape on big wheels?
If you use one of those you also need a 3 head recorder. What´s cool with those machines is that it´s possible to have the tape going through another loop between the write and the read head. Need some little wheels. I got a picture somewhere, see if i can find.

[Yorky]

The easiest way to defeat the erase head is to put a bit of cardboard on it

The easiest way to increase delay time is to loop the tape round a nearby chair leg (for Eno-esque delays) and you can change delay length by moving the chair in and out

Who needs electronics !!

[zipzap]

the chair leg, that´s what i mean. But why defeat the record head? The signal has to get on the tape somehow...

[bigtex]

no, you'd be defeating the erase head so the signal would stay on the tape... you definitely need the record and play heads

some erase heads could probably be attenuated, so you could control the ammount of erasure, in essence the opposite of regen or feedback of the delay

[zipzap]

But isn´t that normaly done by feedback of the signal from the tape, back to the tape? When the signal stays on tape, of course you could have 1hour of delaytime, but i´m not sure how that would work out in praxis.

[bigtex]

Well, okay, I guess you're right. The signal on the tape would stay the same, rather than going through generations or play and record, decreasing in quality. More like infinite echo, less like full feedback.

[blue hell]

The record head will partly erase and modify the "old signal" that's already on the tape (even without using the erase head).

[midgetfidget]

And I guess you would yous a looped tape?

[bigtex]

Well, you might work with a looped tape.

If using reel-to-reel tape deck, you could:
-use a loop of tape, which would have a fixed length but perhaps you could vary the tape motor speed
-use some mechanism to pull the tape away from the heads between the record and playback heads, thread the tape around something, and make it moveable so you can vary the length between those two heads
-use two reel-to-reel decks with either a loop or from the supply reel of one, through the heads, to the second deck, through it's heads, and to it's take-up reel

If using three-head cassette tape decks you could:
-use continuous loop tapes (like those made for answering machines, or make your own)
-use a regular tape, but still use the distance between the record and playback heads as your delay, and just vary the speed of the motors
-break open the tape decks and make custom cassettes that thread between two shells and run tape between two cassette decks (this is similar to the approach with two reel-to-reels, but with cheaper, more abundant tape decks) .. this could also be done with a dual cassette deck, just run the tape through two cassettes and create holes between the two tape bays so you can use both simultaneously, and put the deck into dub mode (record from one to the other) and then monitor both .. this would depend a lot on the tape deck you use, though

[zipzap]

Maybe it´s even possible to modify a cassette so that the tape forms another loop between the rec and play heads. Propably tricky with a small casette.
But look what i found.
The White plate covers the heads. You see something like a T. The bar going from top to bottom holds another wheel. The tape goes around that wheel. By lowering or rising the wheel on the bar you get longer or shorter delay. If you speak deutsch you can have a look in this Bandmaschinen Forum. Bandmaschine would be a good name for a detroit minimal techno project
http://mb.abovenet.de/forum2/showtopic.php?threadid=3041&highlight=echo

---

Koerting_5.jpg
Description:
Filesize:	63.31 KB
Viewed:	18303 Time(s)

---

[midgetfidget]

I don't speak german, but i'll have a look anyway.

Thanks for your suggestions everyone!

[deknow]

...i have lots of reel to reel tape in my basement (the widow i bought it from made me take all of it). i'm keeping the maxell mastering tapes with the metal reels, but there is lots (like 300 reels) of the cheaper stuff with classical recordings from the radio and tv (from the 50's to the 80's). i'd be happy to trade some of this for anything mildly useful.

deknow

[Kerr_Avon]

1950s? Think before you tape over those. Even crappy off-air home tapes can turn out to be the only remaining recording of some special performance, event or even old TV advert or jingle! I guess the tricky/impossible bit is spotting the gold amoungst the guff.

As for cassettes loops, Oooo eck! I do admire the use of tape and the use of things which are ready at hand: I delight in using two 3-head casstte decks for tape delay, feedback and flanging as noted above. But I remember attempting to edit compact cassette tape as a kid, when a manky recorder frequently mangled my cassettes and indeed making loops like this. It was a major nightmare, very fiddly with the microscopically thin tape and I think I'd say impossible to make a splice which didn't make an audible thunk. You have an additional problem if you intend to record your cassette loop after you've physically spliced it, in that there will always be a period of silence or unwanted noise corresponding to the distance between the erase and replay heads. I'm wary of offering "advice" if you're already having fun trying it the hard way, but have you considered that domestic open reel 1/4 inch recorders are available almost for free nowadays, and semi-pro ones for quite small amounts of money? Then you can do loops of any length you like round the studio, limited only by the number of bottles you have available to pass the loop round. I did this on a BTR/3 some years ago to make some Russ Conway music loops for DJs to blab over, and you can loop 60 seconds easily even at 15ips in a reasonable space. Recording the loop before splicing the ends together does away with the inter-head problem, which is easily possible because of the ease of editing the thicker tape recorded at higher speed and the ability to find the exact right spot on the tape in an open-reel situation. How about two cheap/free recorders in the same loop with electrical or acoustic feedback between them? Hmmmm Quatermassy:) Now where did I leave that old Nagra?

[deknow]

well, on the one hand, i agree....there may well be good stuff there...and given the build quality of the dynakit (kit) amp that he used (the best wiring i've ever seen...it's simply beutiful), i would expect he did a fair job recording these.

on the other hand...there are several things on each tape, about 300 tapes...if someone was going to go through them looking for historical recordings, i'd give them away....but i can't imagine doing that job myself.

deknow

[kokoon]

hey guys. i just bought a cheap 3 head tape deck (Kenwood KX-7030) which looks quite fancy. and seems to work perfectly too. the problem is though that it doesn't have a trimpot labeled "tape speed" or anything like that. there are only ~6 trimpots inside, just for various dolby NR settings, bias and rec+playback amplification.

so... as far as i can tell i have 3 motors but one is just for the casette door, so that makes 2 motors... i guess i have to look up a bit how casette decks generally work, but one motor is probably for the reel and the other is for the cylinder that directly pulls the tape. so i'd have to slow down both right? any good links for modding tape decks?

[deknow]

there should be only one transport motor i think (otherwise calibrating the two would be a nightmare).

deknow

[kokoon]

so i guess i have to rip it apart and see what exactly does what. would anyone care to take a look if i posted some photos of the insides?

[kokoon]

okay i did it sort of.

at the moment i just put a resistor on the motor supply but when i load it too much it just stops. now what i'm wondering is - does it stop simply because it gets too low voltage? would putting an adjustable voltage regulator instead of just a pot change that? cause the speeds are still too high, if i want delay > ~500ms (more than my BBD delay) i have to chain the channels... it's still not long enough!

[bigtex]

That resistor is probably limiting the current to the motor. So when you load down the motor, it can't pull enough curent to turn. A better solution would be to have a variable voltage power supply without the current limiting, such as an adjustable voltage regulator. Then you could turn the voltage down but the motor would still be able to pull enough current to turn slowly. You'd still need to set a proper maximum and minimum voltage, though, because too much will burn it up and too little still won't turn.

[kokoon]

ah so there is hope

something like this http://www.casperelectronics.com/motor-speed-mod/
should probably work right?

so - am i getting this right - according to the ohm law increasing resistance lowers the current (not the voltage?)

don't laugh at me i wanna learn it all