MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Is it possible to burn audio in with a hot Serge signal?
MUFF WIGGLER Forum Index -> Buchla, EMS & Serge  
Author Is it possible to burn audio in with a hot Serge signal?
vibralux
Some of my friends claim that they destroyed some equipment due to a very hot signal coming out of their Euro rigs - one of them busted an audio interface audio in and the other burned mixer audio in. So my question is - is it really possible?? I always thought You rather bust Your speakers before doing any harm to an audio interface. Should I always put a limiter on my Serge? So far I am doing OK going straight to my Lynx Aurora but those stories scared the shit out of me woah
Ranxerox
Of course you can damage any audio equipment designed for line-level (1.2volts) with modular signal levels (up to 15volts). You should always attenuate the modular output with a VCA or a dedicated output attenuator.
vibralux
Can I measure it somehow? I mean - when I am recording my Serge I am always using a VCA in the UAP and the meters usually show the expected -20 to -10 peak levels. Should I still worry? I have a spare A-Design ATTY passive attenuator, so I can always use that...
Ranxerox
If the output on your serge VCA is reporting the output voltage relative to 0 dBV, then you should be ok. If you were to bypass this and, say, feed the output from an oscillator directly into a line-level input on your sound interface, that's when you would start burning stuff.
vibralux
Thats what I thought - I mean - if its in the ballpark, its in the ballpark Mr. Green Thanks for a clarification!
hiawog
yes, it's possible to damage gear with with audio signals, but it's also very easy to avoid. suggestions:

1. definitely (as you mentioned) keep an eye on your input meter. no clippy.

2. use a dedicated output attenuator. you mentioned you've got one. keep it low. you can always turn it up.

3. generally, always start with your sounds at -infinite and work up. i find this is better and safer that starting at zero db and working down.

4. if you still feel the need, use some kind of limiter on the input. i know there are products that can do this.
vibralux
Thanks - I always do that. I do mixing for living so I am familiar with gain staging concepts, I just freaked out cause I am no expert as it goes for modular stuff, and when I hear those stories I got really scared. I am always recording withing the -20 to -12 peak levels and I have a really good limiter inserted prior the audio in (MC 77) so I would be really surprised if I messed something up. Although You can never be sure with such devices as WAD MY ASS IS BLEEDING
rydan
In my experience, most (or all) pro or prosumer stuff, running at +4dBU, will handle modular levels (+/- 10 or so volts) just fine. With too much level in, they will distort, but not burn. Cheaper consumer amps and stuff, consumer level 3,5mm soundcard inputs and such, well, that's another story...
michaelvertex
good to consider, thanks for the info!
rydan
Just checked, my SSL preamp has a max output of +24dBu, meaning about 12V (24V peak-to-peak). So, modular levels aren't THAT hot in pro audio systems. (It's also nice to be able to plug the preamp straight into the modular, and get a good signal level =) )
Ranxerox
rydan wrote:
In my experience, most (or all) pro or prosumer stuff, running at +4dBU, will handle modular levels (+/- 10 or so volts) just fine. With too much level in, they will distort, but not burn. Cheaper consumer amps and stuff, consumer level 3,5mm soundcard inputs and such, well, that's another story...


I think this is a little misleading. Line level at +4dBu is equivalent to 1.23 volts RMS, or 3.47 volts peak-peak. That's a lot less than a 10 volts RMS modular signal (20 volts peak-peak) - definitely possible to do damage with these voltages. If you haven't it's because you've been lucky and your 'pro' gear is idiot proof, but I wouldn't go regarding it as a safe policy.

Bear in mind that a lot of 'pro' gear (particularly computer interfaces, effects and digital mixers) has ICP (integrated circuit protector) fuses hard-wired to the input circuit board in order to protect the A/D converters from over-voltage. When these self-destruct they disable the input, requiring a repair job - a pain in the ass, and not cheap.

rydan wrote:
Just checked, my SSL preamp has a max output of +24dBu, meaning about 12V (24V peak-to-peak). So, modular levels aren't THAT hot in pro audio systems. (It's also nice to be able to plug the preamp straight into the modular, and get a good signal level =) )


It's actually more like 34 volts peak-peak. Here's a handy link to a calculator for converting between dBu, dBV and voltage:

http://www.sengpielaudio.com/calculator-db-volt.htm

Mic preamps are designed to deliver that sort of gain into professional component A/D converters calibrated to accept +24 dBu = 0 dBFS. But under no circumstances would you run your whole system gain at +24 dBu. At the very least you could expect your signals to distort, and most likely cause damage to your line-level gear.
rydan
Yes. 34V p2p. 12V RMS. Typo...

Correct, of course it's not a good idea to run +24dBu as system gain, that's not really what I'm saying either. All I'm saying is that, equipment made for +4dBu, is made to use that as the calibrated gain, but with a headroom, meaning they are generally designed to handle a bit more than that (usually at least 18 or so dB above). Also, they are generally designed not to break directly at the clipping point, so, in general, that would mean it would be hard to break them using a modular, even at full level. Might not always be the case, but if not, I would say the equipment is rather badly designed if specified for +4dBu, but breaking at +18.

But, of course, there's no guarantees...
rydan
Also, euro levels will be, at the most, 24V p2p (and that's rail voltage, so, it's more likely to be 20 or so V p2p), meaning 8.5V RMS.
sempervirent
Since a lot of people in this thread seem to know what they're talking about, I wanted to ask, why are modular levels so hot anyway? So that they don't lose as much signal as they're passed through various modules on their way to the final output?
Ranxerox
That's an interesting question. I don't know for sure, but at a guess I would say it's because of the 1 volt per octave standard for pitch CVs - you need quite a few volts to cover the audio spectrum, and if you're going to use your audio signals interchangeably as modulators then they will need to have at least the same amount of voltage swing.

Edit: to expand on this, if 0 to 10 volts (or e.g. -5 to +5 volts) gives you your pitch CV, -10 to +10 would be required to give you full modulation at either extreme using e.g an oscillator for FM.
sempervirent
Yeah, that's probably the most likely reason – seems obvious now.
Graham Hinton
sempervirent wrote:
I wanted to ask, why are modular levels so hot anyway?


To maximise the signal to noise ratio and because analogue computers, which early synthesizers were based on, were normalised for 10V signals. All analogue equipment has a noise floor. One decent quality op amp inverter stage would be around -95dBu, ones commonly used in synthesizers would be higher at around -80dBu so being able to handle +10dBu gives at least a 90dB dynamic range.

Most synthesizer modules are unity gain devices, only some filters have gain at their resonant frequency. In a mixer there are lots of sources of gain, faders often have +10dB and EQs have +/-15dB per band, so the normal operating level is lower, but most line inputs handle +28dBu and have a trim stage to bring it down.

Channel line inputs of low end mixers are really using the microphone preamps with a padding resistor and will clip at a lower level. Stereo channels and aux inputs usually accept higher levels. Essentially synthesizers and pro-audio are both built from the same technology running on similar power supplies and will be mostly compatible. You only really damage something by applying voltages higher than its power rails and there should be protection against that. A valve mic pre can kick out over a hundred of volts on a transient so pro audio gear has to handle that.

Also be careful of connecting to mic/line inputs with global phantom power because most synthesizers don't have protection from +48V on their outputs.

Consumer toys often run on lower power supplies like +/-5V or +/-2.5V so will need some attenuation. Not a lot in dB terms because 5:15 is a factor or 3 or about 10dB. If you can damage something driving it from a synthesizer output with a 1k current limiting resistor you are better off without it.

Beware of dB calculations and specs because they are different for different waveforms. A dB maximum input rating is for a sinewave, so use that to find the peak voltage and then that applies to any other waveform. Different waveforms at the same peak voltage will give different readings on an audio dBu meter.

+4dBu is not the working signal level, it is only the reference tone level used to line up 0VU on meters. Digital systems set a 0VU at a level less than 0dBFS and there are several confusing "standards" between -11dBFS and -20dBFS. The best way is to match analogue and digital clipping so that there is no reduction going one way or another.
Ranxerox
@Graham, that reads like a nice bit of reverse engineering, and it sounds credible from an elecctronic standpoint. Only, in reponse to the question, I'm not sure in the beginning that Moog or Buchla had much reason to be influenced by early analog computers, nor that they were all that concerned about maximising headroom in the audio path. In fact Buchla went so far as to run his audio path completely distinct from his CV, on different connectors and at much lower voltage. Whereas Moog always intended his audio and CV to be interoperable, thereby mandating use of the same voltage range.
Graham Hinton
Ranxerox wrote:
@Graham, that reads like a nice bit of reverse engineering, and it sounds credible from an elecctronic standpoint. Only, in reponse to the question, I'm not sure in the beginning that Moog or Buchla had much reason to be influenced by early analog computers, nor that they were all that concerned about maximising headroom in the audio path.


Anybody working with operational amplifiers was influenced by analogue computers, whether they knew it or not, because that is where they come from. Even when built with three transistors like Moog did.

Nobody was working in complete isolation, they were exposed to semiconductor manufacturers' data and application notes and engineering magazine articles and all of these at that time were heavily influenced by analogue computer technology.

Also check out Alan R. Pearlman's career before Tonus Inc. and Dennis Colin's ARP 2500 filter.
Synthesizers are just specialised analogue computers, a fact acknowledged by Analog Devices in their Non-Linear Circuits Handbook.

Quote:

In fact Buchla went so far as to run his audio path completely distinct from his CV, on different connectors and at much lower voltage.


That's a good example of getting it wrong and being stuck with it. I've never believed the post justification.

Quote:
I don't know for sure, but at a guess I would say it's because of the 1 volt per octave standard for pitch CVs - you need quite a few volts to cover the audio spectrum, and if you're going to use your audio signals interchangeably as modulators then they will need to have at least the same amount of voltage swing.


1V/octave is arbitrary, but comes from having a 10V range in the first place. The actual voltage on the transistor is around 18mV/octave so the external voltage can be scaled to anything. EMS used 0.32V/octave.
syncretism
Hey hey, necrobump.
I have an old, v1 DACS Freque that I'd like to use with my Serge matrix mixer. It doesn't have any sort of metering, and I'd like to avoid adding more hardware (other panels, effects units, etc.) to the mix, if I can.

Per the processor's manual, input signals should be between +2dBu and +12dBu. Naturally, I'll be conscious of gain and mixing my signals, starting from zero on the mixer's output, etc. I'm just wondering, given a mixer module, crossfader or any other output that's not a direct out from an oscillator, do I really have much to worry about, here? I was under the impression that most Serge signals are on the high side of line level, but rarely likely to damage external equipment.

I know we've talked about this in general terms in this thread; I'm just wondering, given the specific numbers I've provided, if we're in the clear.
CosmicFlight
And what do you think using a DI like a Radial JDI and then to a preamp ?
Demi Jon
I'm thinking about getting a Zoom LiveTrak L12, and wondering if using the 26 dB Pad would drop the level enough for modular input? I've never used a mixer with Pads before.
flashheart
26dB is more than enough, that'll drop a 10Vp-p modular signal to 0.5V p-p.
MUFF WIGGLER Forum Index -> Buchla, EMS & Serge  
Page 1 of 1
Powered by phpBB © phpBB Group