The real TAPE EMULATION circuit

[chrissugar]

Consul,

In a tape recorder the magnetic field is closed by the tape in that small gap not on the entire surface of the head. So there is no need to use anything between the two heads.
The gap of the record head should align with the gap of the playback head, and then the magnetic field will close the magnetic circuit.

chrissugar

 

[Consul]

Okay, I get it. Thanks for putting up with me. Like I said, I like mental puzzles.

 

[chrissugar]

Ok now i should go to sleep, it's 3:45 AM and at 9 in the morning I have to work.

chrissugar

 

[pucho812]

Pucho, please stop the jokes.

sorry :roll:

 

[fazeka]

[quote author="chrissugar"]
So my questions would be adressed to the tape recorder Gurus. What impedances/inductances are common for some of the high quality tape recorders. I remenber that I found on the net a table with lots of data for various heads, but now I don't know where it is.
Also at what level heads start to saturate?
What electronics do you recomend for this project. I was thinking about some old tube Studer electronics. Any sugestions? Any schematics?
Also, do you think it is possible to emulate the two head assembly with a transformer with the right magnetic material and the right impedances/inductances?
It would be necessary to have a small gap to emulate it as close as possible?

Lets do some brainstorming.

chrissugar[/quote]

I think the best place to start would be on the Ampex list. They have some people there that know their stuff.

 

[pstamler]

Chris, at the risk of sounding like a party pooper...

First, I think an awful lot of the "sound" of magnetic tape comes from what it isn't -- digital. In other words, it's not that it's adding some kind of analog magic, but rather it's not adding various kinds of digital crud. Which means that running digitally-recorded signals (or signals about to be digitally recorded) through your simulator setup will not make them sound anything like well-recorded tape. Might be a fun way to make signals sound weird, but it's not gonna sound like an Ampex.

Second, what remains of the "sound" of magnetic tape comes, for the most part, not from the electronics or the heads, but from the characteristics of the tape itself, notably its saturation behavior at various frequencies. Which you can't simulate just by using heads and electronics.

My $.02, and others will no doubt disagree, but that's what it looks like to me.

Peace,
Paul

 

[chriss]

you can get pretty good tape-machines at very low price these days. so the easiest way to check things aut would be to buy one, take out the relevant electronics an heads and experiment. worn out heads shouldn't matter as well...

maybe the idea with a transformer instead of heads face to face is the better choice. but i think the heads face to face work like a transformer, don't they? and if you drive the transformer hard enough it will saturate as well. would be interesting to compare the sound to tape-saturation. so the transformer should have not to much headroom, right?

that would give a new meaning to the word 'interstage-transformer'...

chris

 

[chrissugar]

[quote author="pstamler"]
First, I think an awful lot of the "sound" of magnetic tape comes from what it isn't -- digital. In other words, it's not that it's adding some kind of analog magic, but rather it's not adding various kinds of digital crud.
...
...
Second, what remains of the "sound" of magnetic tape comes, for the most part, not from the electronics or the heads, but from the characteristics of the tape itself, notably its saturation behavior at various frequencies. Which you can't simulate just by using heads and electronics.
[/quote]

Paul,

Thank you very much for your thoughts.
From what I know many mastering engineers use high quality tape recorders as a processing tool and there is a significant difference between the signal that pas the recording unit even after digitization.
So I think with quality conversion the recorder sound is well preserved even in digital.

Unfortunately many people use the tape only for the saturation effect but that is not all it does. I think the glueing effect is from the fact that the recording/playbeck equalisation produce different level and phase response at different frequencies and the magnetical circuit react in a nonlinear way to the whole audio spectrum. It is clear that the saturation effect will not be present but it is not what I want.

The guy that tested the "Masterpiece" Analog Mastering System say it works:
"It sounds like the real thing - fattens the bass, punches up the kick and glues the mids and highs. It doesn't attempt to give the limiting distortion of tape overload, but it sure gives a realistic ooomph for days."

So I suppose it can be done.

chrissugar

 

[DrFrankencopter]

I've thought about these tape emulation circuits too, and I think that if I were going to do it I would use a small (easiliy saturated) gapped transformer as a simulated tape head. I would send a DC current (variable) through the primary to saturate the core and get some magnetic hysteresis happening (I think this is a huge part of the tape sound). From there you could tailor EQ circuits and add bias level (if desired). Basically, you'd have to tailor it all to the sound of the transformer, but that would be the fun part.

Bias and EQ networks are just means to get around the hysteresis problem, tehy may not even be necessary if you can directly control the level of hysteresis.

Cheers,
Kris

 

[gyraf]

[quote author="DrFrankencopter"]
Bias and EQ networks are just means to get around the hysteresis problem, tehy may not even be necessary if you can directly control the level of hysteresis.[/quote]

I'm not sure we want to skip the bias/eq. After all, that is an important part of how tapes saturate distributed over frequency - and how they behave at low levels.

What could be interesting to try is having a complete recording/playback setup driving/picking up from two magnetically isolated coils. And between these two coils, you add a small magnetic "connection" made from some form of ferromagnetic substance. (a frontpanel drawer named "insert magnetic sample here" comes to mind) This would be your "tape media", and would respond different from type to type, specially when it comes to overload/hysteresis/optimum bias etc..

Jakob E.

 

[DrFrankencopter]

'm not sure we want to skip the bias/eq. After all, that is an important part of how tapes saturate distributed over frequency - and how they behave at low levels.

Yeah, I just spent about 30 seconds and thought about it.... since the saturation is a non-linear element, any filtering done before it will affect the output even if an inverse filter is placed after the saturation.

Then I thought of this idea " Instead of using fixed NAB or AES EQ curves why not use an active variable crossover circuit. That way you could adjust the break point of the filter and choose the balance between low frequency saturation, and high frequency saturation (i.e. the less LF information that gets into the transformer, the less LF saturation).

Bias may not be necessary if the hysteresis is low enough....I see bias as a potential headache since it will need to be filtered out (though maybe the crossover would have enough attenuation at the bias freq to make it not a problem).

So, my approach would look like:

Input -> Variable crossover (High pass side) -> gapped transformer with bias current applied -> Low pass side of crossover.

I think that would be interesting, yet simple enough to build....anyone have a suitable transformer in mind for something like this. Maybe an LO1166???

Cheers,

Kris

 

[chrissugar]

Jakob,
I like your idea with the "tape sample" probably it would be an interesting idea to include in a unit some some pairs of record/playback head assemblys with different tape samples for different sound.
That is good, we have a starting point.

DrFrankencopter,
I was also thinking about DC biasing the transformer to emulate some kind of saturation, but that would be useful only if a transformer can emulate the head pair assembly.

chrissugar

 

[chrissugar]

I have an idea :idea:

What if we simulate the two heads, with a gapped transformer with similar impedance/inductance like the equivalent heads, and put in the gap a piece of magnetic tape. :?:

chrissugar

 

[chrissugar]

I have even a better idea :!:

I know that some heads have two coil sections that can be connected in series. If one section is conected to the recording, the other to the playback electronics and attach a piece of magnetic tape to the gap with some adhesive tape, there is no need for transformer or two heads.
I think this would be the best tape emulation system.

chrissugar

 

[gyraf]

Chris,

We deliberately don't want direct-coupling - that is far to easy (read:clean).

The artefacts you hear from tape is maybe 99% from the tape media - very little from the head's magnetics unless they're very badly designed..

Jakob E.

 

[Bear]

It occurs to me that if you do it without tape, you completely loose the aspects that are specific to the tape that relate to oxides and such. Depends what the goal is - if you wan't a tape saturation effect this is probably essential. If you just want some of the effect of what a pristine, ideal tape deck would sound like, the tape peculiarities may be less important - pre-emphasis and de-emphasis would seem to be the better part or the equation then.

Bear

 

[chrissugar]

[quote author="gyraf"]Chris,

We deliberately don't want direct-coupling - that is far to easy (read:clean).
[/quote]

But if we use one head with two coils the coupling is done through the piece of tape that closes the magnetic circuit in the gap.

I thought about the two head configuration with a piece of tape in the middle. I dont think it will work well. Have you ever tried to play a recorded tape by reversing it. I mean you put the non oxide covered side of the tape on the head. Because of the not so perfect contact it behaves like a low pass filter.
So in the "dual head plus tape between them" setup wil not work so well because it will suffer at the recording or playback side depending where is the oxide.

[quote author="Bear"]
Depends what the goal is - if you wan't a tape saturation effect this is probably essential. If you just want some of the effect of what a pristine, ideal tape deck would sound like, the tape peculiarities may be less important - pre-emphasis and de-emphasis would seem to be the better part or the equation then.
Bear[/quote]

It is the second part (more clean) more interesting for me but if I can have all the aspects of the tape sound that would be even better.

chrissugar

 

[aronaut]

i'm a newbie here and don't have much to offer technically in this discussion, only that i've been thinking about this same idea for years. please keep it going... i'm very interested.

A

 

[Consul]

http://recforums.prosoundweb.com/index.php/t/4767/737/

Apparently, Chris and Rupert Neve think alike. :wink:

Well, at least now we know this is a worthwhile endeavor.

 

[TomWaterman]

Yeah I spent a few mins looking for heads today.....looks like it could get quite pricy for anything above 2track work.

I had a stupid hairbrain idea for a 16track emulator feeding a 16-2 mixer.....might have made a nice mixdown machine but I expelled the thought when I saw 16track heads were a $1000+!!

I'm interested in the Neve Portico....but if there's a viable DIY alternative I'm there.

-Tom