Become a Site Supporter and Never see Ads again!

Author Topic: What input stage clipping looks like  (Read 3461 times)

0 Members and 1 Guest are viewing this topic.

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
What input stage clipping looks like
« on: August 24, 2020, 09:31:21 PM »
Hi. This won't be anything new for most people here, but some people might benefit from seeing it, or it may just be mildly amusing, I dunno.

I needed to make a quick test recording of a piano today, and happened to use a recorder (Marantz PMD-661 Mk III) that has balanced XLR microphone inputs and phantom powering, but its input stage evidently needs to be "padded down" considerably to accept the signals from professional condenser microphones even at modest sound pressure levels. The screen shot below shows the result; I wasn't playing loudly at all, and the microphones weren't super close to the piano.

I'm surprised that the clipping is so clean, though. That's not usually the case with analog clipping. But it definitely is the input stage (or a stage before the record level controls, if there's more than one), because when I adjusted the record level controls upward a little (before I became aware of the clipping), the clipped signals got recorded at somewhat higher levels.

--best regards
music > microphones > a recorder of some sort

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: What input stage clipping looks like
« Reply #1 on: August 24, 2020, 09:41:25 PM »
yes it seems all the ‘bricks’ like the marantz pmd and tascam dr100 series need the 18-20 dB pads by default. they’re pretty sensitive
« Last Edit: August 25, 2020, 06:36:31 AM by jerryfreak »
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline justink

  • Trade Count: (4)
  • Taperssection All-Star
  • ****
  • Posts: 1973
  • Gender: Male
Re: What input stage clipping looks like
« Reply #2 on: August 24, 2020, 10:50:15 PM »
happened to me once using the dpa4021 > Oade mod R-44.  might have been smashing pumpkins, iirc.
Mics:
DPA 4023 (Cardioid)
DPA 4028 (Subcardioid)
DPA 4018V (Supercardioid)
Earthworks TC25 (Omni) 

Pres and A/D's:
Grace Design Lunatec V3 (Oade ACM)
Edirol UA-5 (bm2p+ Mod)

Recorders:
Sound Devices MixPre10 II
Edirol R-44 (Oade CM)
Sony PCM‑M10

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: What input stage clipping looks like
« Reply #3 on: August 25, 2020, 06:21:33 AM »
Marantz doesn't specify the max input level for this recorder, but the relatively SNR and dynamic range give some clues that it isn't going to take a very high level.
https://www.manualslib.com/manual/98972/Marantz-Professional-Pmd661.html?page=73#manual

DSatz, what mics were you using?
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: What input stage clipping looks like
« Reply #4 on: August 26, 2020, 12:30:45 PM »
I guess they'd be called CMC 122s; they were MK 22 "open cardioid" capsules on CMC 1 amplifiers (= 15 mV/Pa). I was playing at a soft-to-at-most-moderate level of loudness, with the mikes about two feet in front of the piano. The lid was closed. That should be no challenge to any preamp.

--Some testing with a tone generator seems to indicate a clipping point around -53 dBV when no pad is applied, which is only about 2.2 mV. The gain increase when the XLR inputs are switched from line to mike is apparently about 65 dB, which is a surprisingly large interval. I would expect something more like 40 dB, or maybe 50 at most.

Attenuation at the mike inputs can be switched on from the menu in 6 dB steps up to 18 dB, which puts the overload point at a little less than 18 mV. But a Schoeps mike with this type of capsule puts out 18 mV at just 95 or 96 dB SPL. Thus the problem.

So I'm glad I made this little experiment. The only way I will ever use this recorder with professional condenser microphones is to use 20 dB outboard pads on its inputs, unless I'm recording people's unspoken thoughts, or a clavichord that no one is playing, or an electronic instrument that's not plugged in. Further attenuation from the menu settings would probably still be needed, but then at least there would be a way to make it work.

Just in case I made some kind of stupid mistake, I hope that someone else here will measure this independently and confirm or dispute these numbers. I don't want to trash an otherwise pretty-good-seeming recorder.

--best regards

P.S.: While feeding in steady tones and fiddling with the record level controls, I heard a small but definite and repeatable "stepping" in the action at one point at least, about 2/3 of the way up. It was only a fraction of a dB, but that isn't a welcome feature. Again I'd be very much interested to know whether I'm alone in hearing that or whether others hear it, too.
« Last Edit: August 26, 2020, 01:33:48 PM by DSatz »
music > microphones > a recorder of some sort

Online yug du nord

  • ...til things never seen seem familiar…
  • Trade Count: (56)
  • Needs to get out more...
  • *****
  • Posts: 5533
  • made with natural flavor
Re: What input stage clipping looks like
« Reply #5 on: August 26, 2020, 02:16:42 PM »
The PMD661 and PMD661mkII mike inputs NEED to be set to -18 in the menu.

I can't say if that is the same for the PMD661mkIII..  but maybe??.
.....got a blank space where my mind should be.....

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: What input stage clipping looks like
« Reply #6 on: August 26, 2020, 04:28:33 PM »
...unless I'm recording people's unspoken thoughts, or a clavichord that no one is playing, or an electronic instrument that's not plugged in.

Thank you; I needed that laugh today!  :yack:
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

Offline MakersMarc

  • Trade Count: (17)
  • Needs to get out more...
  • *****
  • Posts: 8681
  • Gender: Male
  • 😈
Re: What input stage clipping looks like
« Reply #7 on: August 26, 2020, 07:07:01 PM »
I dumbly ran an SD MP2 without pads for some time. I didn’t do computer editing back then, never really saw the wave forms until a few years back.  :facepalm: With MBHOs, didn’t really show up much in the audio. Mostly embarrassed.

Luckily Doug was explicit about setting the Marantz pads at -18.
😈 Mk4v/41v>nbob actives>Baby nbox>Oade warm mod Marantz 620.

Open: 4v/41v>nbobs>Nicky mod Naiant PFA>Oade warm mod 661.

Home: the Stereo Hospital budget refurb rig: Lappie>DragonFly Cobalt/Red with Jitterbug>Nikko NR520/Sansui 221>B&W V202 speakers.

Online yug du nord

  • ...til things never seen seem familiar…
  • Trade Count: (56)
  • Needs to get out more...
  • *****
  • Posts: 5533
  • made with natural flavor
Re: What input stage clipping looks like
« Reply #8 on: August 26, 2020, 08:49:18 PM »
The only way I will ever use this recorder with professional condenser microphones is to use 20 dB outboard pads on its inputs,

Get an Oade mod.
I think that I know that you don't prefer to "flavor" the sonics..   so maybe check out the "Concert" Mod.  Nice and clean.
Well worth it over stock preamp IMO.

https://www.oade.com/digital_recorders/hard_disc_recorders/PMD-661MODS.html

....although the Warm mods go great with Schoeps!  8)

.....got a blank space where my mind should be.....

Offline jerryfreak

  • No PZ
  • Trade Count: (31)
  • Needs to get out more...
  • *
  • Posts: 6205
  • The plural of anecdote is not data
Re: What input stage clipping looks like
« Reply #9 on: August 26, 2020, 08:56:48 PM »
The only way I will ever use this recorder with professional condenser microphones is to use 20 dB outboard pads on its inputs,

Get an Oade mod.
https://www.oade.com/digital_recorders/hard_disc_recorders/PMD-661MODS.html

AFAIK none of their mods change the sensitivity, and the pad is still indicated
Unable to post or PM due to arbitrary censorship of people the mod doesn't like. Please email me using the link in my profile if you need to connect

Offline datbrad

  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 2295
  • Gender: Male
Re: What input stage clipping looks like
« Reply #10 on: August 26, 2020, 09:09:52 PM »
D, I can also confirm that -18 menu setting is where you have to run, even more so with the newer pmd-561/661mkIII units. If you recall the thread from a while back discussing the digital input gain you will see the connection. The level knob on the original 661 & 661mkII is a digitally controlled analog gain, and the internal menu settings control stepped gain of the preamp, so if you have the menu setting at -6 or zero, you can get overloading along with a clipped waveform. The current version of that deck has a fixed gain preamp stage, and the menu
adjusts the digitally controlled analog gain. The "record level" knob only changes the size of the recorded wave form, nothing more. The digitally controlled analog gain "clips" before it can overload, with results like you described. Bottom line, always run -18 on the menu setting, and don't worry if you have to crank the gain knob to 9 or 10 and only see peaks to -12 on the display. You can easily bump that up in post with no audible defects. Hope this helps
AKG C460B w/CK61/CK63>Luminous Monarch XLRs>SD MP-1(x2)>Luminous Monarch XLRs>PMD661(Oade WMOD)

Beyer M201>Luminous Monarch XLRs>PMD561 (Oade CMOD)

Online yug du nord

  • ...til things never seen seem familiar…
  • Trade Count: (56)
  • Needs to get out more...
  • *****
  • Posts: 5533
  • made with natural flavor
Re: What input stage clipping looks like
« Reply #11 on: August 26, 2020, 09:19:48 PM »
The only way I will ever use this recorder with professional condenser microphones is to use 20 dB outboard pads on its inputs,

Get an Oade mod.
https://www.oade.com/digital_recorders/hard_disc_recorders/PMD-661MODS.html

AFAIK none of their mods change the sensitivity, and the pad is still indicated

/\
True.
But the Oade Mods lower the noise floor and clean up or slightly flavor any stock preamp.  You know that..   ;)

.....got a blank space where my mind should be.....

Offline datbrad

  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 2295
  • Gender: Male
Re: What input stage clipping looks like
« Reply #12 on: August 26, 2020, 11:04:34 PM »
Doug told me he drops the preamp gain on the chip he uses for his concert mod -6db below the level of the stock chip, increasing overall headroom because of the lower noise specs. I like the transparent sound of the cmod with dynamic mics, but I still prefer the sound of the warm mod with bright mics like my old AKG 460s.
AKG C460B w/CK61/CK63>Luminous Monarch XLRs>SD MP-1(x2)>Luminous Monarch XLRs>PMD661(Oade WMOD)

Beyer M201>Luminous Monarch XLRs>PMD561 (Oade CMOD)

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: What input stage clipping looks like
« Reply #13 on: August 27, 2020, 05:01:49 PM »
Umm, if I may--sensitivity is one thing; overload limit is another. The relationship between them is a matter of the circuit design. A preamp or recorder can offer high sensitivity without necessarily being easy to overload.

Does anyone here remember the Advent 201 cassette recorder by any chance? It had a large record level knob which controlled both channels at once, plus smaller ones for the left and right channels individually. It wasn't perfect, since the large knob was a dual potentiometer and the tracking between channels could vary by a dB or more as you turned the knob up or down (especially down). And the source impedance "seen" by the next amplifier stage in the circuit would vary according to the user's settings, which requires a high input impedance for that next stage, and a purely resistive impedance at that, or else the level control would also be a tone control (a problem which some otherwise very nice Rane preamps have had). That requires some extra homework on the part of the circuit designer.

But the neat thing about the main control was that it was right at the inputs; the center pin of the deck's RCA sockets was connected to the top of that control, with some RF protection as I recall, but no transistors or ICs in between. (Actually, back then ICs weren't a thing yet in consumer audio equipment.) So you could feed in any signal levels that you had available--50 Volts peak-to-peak, even--and the input circuit would never overload once your levels were set, because that level control functioned as a continuously variable pad that could be set as high as you would ever need.

I'm not saying that other manufacturers should all adopt that design. The Advent 201 was unusual in that it had aux inputs but no microphone inputs. For live recording, Advent sold a little outboard preamp that had transformer-balanced inputs and was powered by a +18 VDC socket on the side of the recorder. They figured that most people would use the decks in their home hi-fi systems, so those people shouldn't have to pay for microphone input circuitry--nor should the designers of the deck be forced to minimize the cost of that circuitry as tightly as they would have to do for a feature that wasn't a selling point for most potential customers, who only wanted a cassette deck so that they could copy LPs or FM broadcasts. I'd be surprised if as many as 10% of the people who bought Advent 201s also bought MPR-1s (the little outboard preamps).

But the point is that "sensitivity" = the minimum input voltage that can drive the recorder to 0 dB levels, while the overload ceiling can be rather independent of that parameter. Plenty of good recorders and preamps have high input sensitivity but also higher overload thresholds than this Marantz recorder seems to have. -- A third (mostly) independent parameter is the noise floor of a recorder or preamp; that isn't directly a function of its sensitivity, either.

In summary, it's fair to look for equipment with a low noise floor, enough sensitivity for your typical applications plus a little extra for special applications, AND a high enough overload ceiling to let us record without messy pads and belts (didn't the old Tampax ads teach us anything)? Also, the internal pad(s) on a recorder or preamp shouldn't require stopping a recording or diving into a menu to set them. When you need them, you generally need them right away.

As a less antique example than the Advent cassette deck, I know that I can connect any Schoeps microphone to my Grace Lunatec V3 preamp; its inputs can handle the maximum voltage that those microphones can put out. Period. And at any given setting of the gain control knob, the preamp can deliver output levels all the way up to 0 dBFS on the meters. Period again. So all I have to do is set levels so that I don't quite reach 0 dBFS on the meters, and that guarantees that the preamp won't overload--neither at its input, its output, or anywhere in between. It will also contribute only a minimum level of noise--generally less than the inherent noise of the microphones, let alone the venue, at any normal or near-normal gain setting.

So it's perfectly possible to meet all three requirements at the same time. But the designers have to understand that simply adding balanced XLR inputs and phantom powering don't make a piece of equipment "professional" unless it is also capable of handling the signal levels that professional microphones put out.
« Last Edit: August 29, 2020, 07:36:41 AM by DSatz »
music > microphones > a recorder of some sort

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: What input stage clipping looks like
« Reply #14 on: August 27, 2020, 10:00:38 PM »
I repeated the test, this time using visual observation of the waveforms rather than listening through the headphone output. I changed my generator's units in the meantime from dBV to dBu but the result in terms of the actual input voltage was the same; rounded off to the nearest 0.1 mV, it's 2.2 mV. A waveform for an input signal that's ~1 dB higher is shown below.

With 18 dB of attenuation switched in, the clipping point would be only 17.3 mV; a Schoeps mike (with the MK 22 capsules that I was using for my piano test) puts out that much at ~95 dB SPL. Thus it would take about 40 dB of external padding (in addition to the 18 dB internal pad) to bring the mike inputs to where their clipping point was any greater than the clipping point of the microphones. There's no way that that's reasonable. These preamps don't have nearly enough input headroom in my opinion.

--best regards
« Last Edit: August 28, 2020, 10:44:37 PM by DSatz »
music > microphones > a recorder of some sort

 

RSS | Mobile
Page created in 0.082 seconds with 43 queries.
© 2002-2024 Taperssection.com
Powered by SMF