Need help designing an audio circuit - Tape delay unit

Hi, this is my first post here. For the past few days I've been building a tape delay unit. Mechanically, it works well, but I still need to design a circuit to link all the parts together. This is, however, beyond my current level of ability, and I was wondering If I could get help.

Basically, the circuit must do the following jobs;

1. Split the signal coming from an instrument or microphone into two separate signals.
2. Send one of these signals directly to an output mixer section, which mixes the two signals back together.
3. Amplify the second signal to the point where it will drive a tape write head.
4. Take the output from the tape read head and feed this back into the write head, with a controllable amplitude. Also, put this back into the mixer.
5. Output a signal from the mixer at a reasonable level.

The input signal level will be on the order of 100 mV, and the instrument will generally have an impedance of around 10kΩ.

The tape heads are cassette tape heads, and have a resistance of around 300 ohms, and an inductance of 80mH. I'd like to be able to drive the write head to around 1V, and the read head will output something on the order of 10mV. It should also be possible to amplify the feedback signal to a level about twice the amplitude of the input signal.

The mixer should mix the signal from the tape read head with a dry signal from the input, and it should be possible to control the amplitude of these signals allowing the result to be fully wet or dry. The output should be approximately line level, with an output impedance of around 300Ω with a peak voltage around 0.5V.

I understand what I am asking is quite complex, but is this project impossible for a medium level hobbyist with plenty of free time? Is anyone an expert in this area? Does anyone have any tips on how to design and build a circuit like this? Thanks.

Welcome DHughes!

Is this an echo producing device/circuit you're looking at?

Yup, tape echo unit.

Look for a schematic for an "Echoplex", model EP-3.

A to link jbeng's suggestion [irrelevant advertising URL removed - moderator], is spot on.

As you can see, no odd or unusual components. Somewhat complex but with the features you list.

Your level of expertise will dictate any attempt at construction.

cowboybob said:

A to link jbeng's suggestion ([irrelevant advertising URL removed - moderator]), is spot on.

As you can see, no odd or unusual components. Somewhat complex but with the features you list.

Your level of expertise will dictate any attempt at construction.

Click to expand...

**broken link removed**

Is this the link you mean?

This is excellent, but I have a few questions. Firstly, is there any way I can modify this to run from a 12v source? I don't need the motor section, the SOS (?) head, erase head or footswitch section. Also, what do the circled numbers correspond to? Are they just sections of the circuit?

The classic echo unit was the Watkins Copicat, here's a link to a 'later' version of a schematic for one.

http://www.vintagehofner.co.uk/britamps/watkins/schematics/ic300.html

However, in any such circuit the main problem is likely to be the bias oscillator transformer.

Surely you'll need an erase head? Otherwise old audio will be mixing with new.

Nigel Goodwin said:

However, in any such circuit the main problem is likely to be the bias oscillator transformer.

Click to expand...

Why not just use a simple phase shift oscillator set to like 100khz?

Surely you'll need an erase head? Otherwise old audio will be mixing with new.

Click to expand...

I was just going to use a strong permanent magnet, but on further reading I think I may use the bias oscillator with a dedicated head.

DHughes said:

Why not just use a simple phase shift oscillator set to like 100khz?

Click to expand...

Because it doesn't provide the high voltage required (which is what the transformer is for).

I was just going to use a strong permanent magnet, but on further reading I think I may use the bias oscillator with a dedicated head.

Click to expand...

AC erase is FAR, FAR superior to a magnet, and essential in this application where you are continually erased and recording over a short piece of tape.

And as you need to have AC bias, it's no more work to have AC erase as well.

If you want fairly high fidelity then you must tweak the high frequency boosts in the circuit to match the heads and the type of tape.
I remember that cassette tape was in 4 different types and each type had its own equalization and amount of high frequency bias.

DHughes said:

**broken link removed**

**broken link removed**

Is this the link you mean?

Click to expand...

Yes. (Wonder what the heck I posted??) Anyway, your link was the same thing.

DHughes said:

This is excellent, but I have a few questions. Firstly, is there any way I can modify this to run from a 12v source? I don't need the motor section, the SOS (?) head, erase head or footswitch section. Also, what do the circled numbers correspond to? Are they just sections of the circuit?

Click to expand...

Since T1 is a custom unit, you're going to have to come up with a PS. 12VDC alone won't cut it.

Don't know what the circled #s are for: maybe for PCB locations...

Oh my, reinventing the wheel! Go digital, or for a low cost/quality approach use the PT2399 echo processor IC by Princeton Tech! E

Oh my, reinventing the wheel! Go digital, or, for a low cost/quality approach use the PT2399 echo processor IC by Princeton Tech! E

Thanks for all the advice so far everyone, this has been extremely helpful and enlightening.

canadaelk said:

Oh my, reinventing the wheel! Go digital, or for a low cost/quality approach use the PT2399 echo processor IC by Princeton Tech! E

Click to expand...

While I'd definitely follow this path if I was short on time or ambition, at the moment I really just want to make this work. It's already been such an education so far and it's only been a few days.

While looking through some more schematics, I found one that seemed like it would be a lot easier to produce and understand. It does, however, use tubes rather than solid state components. Is there any way I could change the circuit to work with transistors? Here is the schematic;
http://www.vintagehofner.co.uk/britamps/watkins/schematics/copicat.jpeg

Also, I'd probably want to lose one of the inputs, and 2 of the 3 play heads. Can this be easily done?

DHughes said:

Thanks for all the advice so far everyone, this has been extremely helpful and enlightening.


While I'd definitely follow this path if I was short on time or ambition, at the moment I really just want to make this work. It's already been such an education so far and it's only been a few days.

While looking through some more schematics, I found one that seemed like it would be a lot easier to produce and understand. It does, however, use tubes rather than solid state components. Is there any way I could change the circuit to work with transistors? Here is the schematic;
http://www.vintagehofner.co.uk/britamps/watkins/schematics/copicat.jpeg

Click to expand...

That's the site I posted the link to above already (didn't you try clicking on it?), I specifically posted the link to a simple IC circuit, but there are transistor ones and valve ones there as well.

Also, I'd probably want to lose one of the inputs, and 2 of the 3 play heads. Can this be easily done?

Click to expand...

Incredibly easily - they even include switching for the different heads.

How far apart are you mounting your heads?, the delay is (obviously) dependent on the distance between the heads.

Nigel Goodwin said:

That's the site I posted the link to above already (didn't you try clicking on it?), I specifically posted the link to a simple IC circuit, but there are transistor ones and valve ones there as well.

How far apart are you mounting your heads?, the delay is (obviously) dependent on the distance between the heads.

Click to expand...

I did look at that circuit in detail, but it seemed like the schematic I posted might be easier for me to build, if transistors could be used instead of valves.

I'm mounting the heads 5cm apart, and I have already got a variable speed motor system working.

DHughes said:

I did look at that circuit in detail, but it seemed like the schematic I posted might be easier for me to build, if transistors could be used instead of valves.

Click to expand...

Look at the other circuits on the same site, there are at least a couple of transistors ones (older versions than the IC ones).

However, the IC versions are probably the simplest? (opamps make things easy).

Here's one transistor example from that site:

http://www.vintagehofner.co.uk/britamps/watkins/schematics/copicatmk4.jpeg

I have already got a variable speed motor system working

Click to expand...

To maintain a given frequency response you will need to vary the amplifier equalisation in sympathy with the speed.

Nigel Goodwin said:

Look at the other circuits on the same site, there are at least a couple of transistors ones (older versions than the IC ones).

However, the IC versions are probably the simplest? (opamps make things easy).

Here's one transistor example from that site:

http://www.vintagehofner.co.uk/britamps/watkins/schematics/copicatmk4.jpeg

Click to expand...

Ahh, this is excellent. I think I understand this one. I have a couple of questions. What voltage is the rectifier outputting? In other words, if were to use some kind of DC power supply, what would I use? Also, how would I go about getting/making the bias transformer? Thanks btw, this is so helpful.