0

If a step signal (sudden signal change, square waves etc.) is recorded onto a vinyl record, what would be the resulting waveform actually captured on the final disk?

  • 1
    It really depends on a lot of things. If you send a digital music file or even an analog tape to a place that presses vinyl records, they are going to process what you give them a bit before they master the record, and a lot of that processing limits what will be engraved in the master. On playback it would probably be more like a thump, as like a dry bass drum hit. If you engrave it yourself, there are limits to what you can engrave, and probably the best you could do is make the needle jump off the record. – Todd Wilcox Oct 12 '15 at 18:25
  • Well, I guess a theoretically vinyl cut "by hand" may contain a groove suddenly displaced, that make the needle jump out. However the record master cutting machine would not be able to cut such a groove I think, I don't know how it works exactly, but the cutting tool will obviously have a quite limited motion and a size itself. So using this machine fed with a step signal would be the closest experiment to answer my question. – dronus Oct 17 '15 at 18:06
  • 1
    @dronus - why do you want to do this? What precisely are you trying to discover or achieve? – Andy aka Oct 17 '15 at 18:36
  • I was just listening to electronic music LPs where some tracks including recordings of several classic audio signal faults like broken cable, plugged and pulled plugs, dusted vinyl etc. Also analog synthesizers where used to produce step-like signals I guess. So I wondered about how exactly extreme signals like that may be transfered by the vinyl disk. – dronus Oct 27 '15 at 17:56
2

From my experience as a vinyl cutting engineer, the phase response makes it look more triangular in shape, but it sounds very similar.

1

Thing is that such a signal only exist in theory - practically this signal will be made up of a number of sine waves with varying frequency and amplitude.

The generator or the sample rate format determines the upper frequency limit (i.e. how "sharp" that seemingly discontinuous signal edge will be a the corner/peaks). This alone will transform your step signal in a similar way to how it will react to later filtering at the vinyl factory (notice the smoothing of the corners below in the red line).

Furthermore the response of the sudden step change will create ringing (capacitances and inductances in the circuits that causes resonances.). With square waves this is known as the Gibbs Phenomenon.

square wave reconstruction

It is thinkable that you have an analog square wave generator that operates in true DC offsets, but even so at some point you have to record that thing, and then you start to hit the "real world" of preamp frequency ranges, AD converter sample rates and/or analog tape frequency range limitations. Long DC spans are likely to get filtered out by DC coupling capacitors (DC removal)

At the factory they will probably filter it even further using high and low passes. If the stereo image is too wide for fast changing low frequency material, they will also reduce the width in lower region.

  • It might bear clarification from the asker, but a step signal and a square wave are not necessarily the same thing. I assumed the asker was talking about a non-periodic, one-time change in level, as close to instantaneous as possible. Of course it would be "rounded" and have hysteresis, so the square wave discussion is not entirely off. – Todd Wilcox Oct 14 '15 at 13:14
  • Well, will that change much? I mean the filters will react the same way (and furthermore DC removal circuits or filters will eat that DC offset quite fast). Isn't a square wave under the category of step signals? I can't seem to find an exact definition.. – Michael Hansen Buur Oct 14 '15 at 13:21
  • @ToddWilcox could you clarify what you mean when you say it would have hysteresis? – Andy aka Oct 14 '15 at 16:51
  • @Andyaka Looks like the word I was looking for was "ringing", that might be the result of system hysteresis. en.wikipedia.org/wiki/Ringing_(signal) – Todd Wilcox Oct 14 '15 at 17:19
  • 1
    Ringing as in successive over and undershoot of a signal? Not sure that I've ever seen hysteresis cause this. – Andy aka Oct 14 '15 at 17:21
1

Guys who master vinyl will tell you something like this: -

A pure step signal has an infinite spectrum and, considering you can't hear above 20kHz (a lot less as you get older) there's no point doing what you suggest.

You may then say "oh please do it" and the vinyl mastering guy might say "OK we'll give it a go" and then completely ignore your request because you won't be able to tell the difference i.e. the mastering guy will apply the normal low pass filtering to a "tape" to prevent too much high frequency rubbish being put on the vinyl.

The basic problem is that a step change cannot be put onto vinyl or any other type of media (including CD, MP3) because the basic media cannot support an infinite bandwidth, so it is limited to whatever bandwidth the media supports. Common sense really.

  • @ToddWilcox - I cleaned up the comment thread here. This question and answer have nothing to do with mastering of a record. Andy mentioned the low pass only because it is the aspect that makes what the OP is asking about irrelevant. The question is about the impact of putting a pure step on the record and what the waveform looks like. This isn't a question (or an answer) about mastering and going in to general detail about mastering is irrelevant to the question asked. – AJ Henderson Oct 14 '15 at 23:10
  • I'm not sure how the answer has nothing to do with mastering a record when the answer includes the words "master" or "mastering" three times. A low pass filter is not the only thing that will change the step signal. Compression and other equalization will also change it. My first comment and downvote are predicated on the fact that the answer is incomplete since it only discusses the low pass filter and not anything else that will change the waveform before it reaches the vinyl. – Todd Wilcox Oct 15 '15 at 11:12
  • 1
    @ToddWilcox if several things in a row can convert an instantaneous change into a non-instantaneous change then why bother going into all the details - low pass filtering changes an instantaneous event into a non-instantaneous event and no amount of other techniques is going to alter that basic fact - it seems like you want to modify the question to suit answers that go into the full vinyl mastering process - it isn't about that. – Andy aka Oct 15 '15 at 11:39
  • I think I've explained my down vote (although my original explanation has been removed). And the fact that there are several things affecting the signal, with which you seem to agree, is central to my reason. I favor more detail over less, which is why one might "bother" going into all the details. Details are important. I don't think it's clear what the question is about, which is why i never answered it. It is possible to carve a step into a circular piece of vinyl. If that's what the question is about, then this answer is kind of wrong, actually. – Todd Wilcox Oct 17 '15 at 18:33
  • @ToddWilcox I hear what you say so maybe if the guy asking isn't able to come clean on what he wants this question be closed down. – Andy aka Oct 17 '15 at 18:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.