Line Level question?

[ROBSCIX]

Hello, I am not sure this is in the right place but, I know Line level is a very small voltage. Isn't it true though that the line level specification has a safety built in? I mean the line level inputs can take input voltages X5 or more of normal line level voltage? does this make sense? A freind told me this info and I am trying to get it confirmed or not...thx

[Leolabs]

Not "safety build in" but rather than "overload margin" and depends on the design.Some specified on 200mV normal input level but able to handle input level up to 1V.

[youyoung21147]

Are you talking about the voltage at the input of a power opamp (like LMxxxx) ?

If so, there is no particular limitation, as far as the amp doesn't clip (try yourself)

If talking about soundcards, preamps etc... the line level is limited by the PSU rails of the preamp. Again, as far as you don't make anything clip, there is nothing particular that could deter you from using the highest possible line voltage (hence giving the lowest noise)

[AndrewT]

Hi Rob,
it's all down to overload margin.

In the decades when grooved records were the dominant medium, it was found that a clipped transient pulse type noise was more audible than an unclipped pulse that went through the amplifiying chain.

On that basis, plops and clicks coming off the records (due to dust in the groove and mains switching interference) were very significant voltage peaks that had very high gain applied (+20db to +60db) before the main amplifier (+26db to +36db).
The amplifying chain had to let these peaks through without clipping until some HF filter could reduce them to being audibly tolerable.
A riaa peamp often strived for 30db overhead margin above a high average modulation level. The following pre-amp/volume control usually tried to match this or better (infinite overload margin on the resistive volume control only limited by power dissipation in the pot track) and our usual 10db to 20db margin in the power amp and speakers between average level and individual peak signals.

A similar but less demanding situation still exists. Noise coming from switching can be higher than the average signal and in worst case higher than the peak signal. Similar overload margins will assist in making these noises less audible when the filtered signal finally exits the speaker.

Power amps often work at maximum output power when a signal of between 200mV and 2000mV is fed in to the input. Average listening levels are usually 20db below this (20mV to 200mV).
The input to the preceeding stages will be similar since additional gain is nowadays less likely to be needed.
The result of this is that to play a 20mV signal through a power amp all preceeding gain stages must be able to cope with 2000mV to be deemed universal, and do this with an acceptably low distortion. This low distortion condition may only be achievable by designing the stage with a potential maximum output level approaching 20V (rare, but some advocates follow this route, mostly valve affionados).

Here we have an extreme case of adequate signal to noise ratio at 20mV signal level and potential to pass 20,000mV (=20V) unclipped to the power amp.

[ROBSCIX]

This is what I mean. What would be the maximum signal voltage that I could send a amp(recevier) -not a opamp without doing damage to the input.? WHen I say amp I mean a set top unit such as a preamp or receiver etc...thx for the help guys.

[AndrewT]

Hi,
what is the voltage you normally pass to the input?
What is the input impedance of that receiver?

Use P=V*V/R and get the power delivered to the input.
If your answer is below 1mW then damage is unlikely to occur.

eg. 1000mV into 10k is just 0.1mW.
Most inputs will only see a few tens of uW.

[ROBSCIX]

FOr instance, if say I had a source that had a line level signal of ~4-5 volts would this damage a nromal line level input. I was told long ago that there is a bit of a saftey factor on line levels meaning the say they can handle 1 volt but they can take more to prevent damage to the inputs. Does this make more sense?

[youyoung21147]

I see no way to damage any line input with normal overvoltage (well, not 3kv) since an input has an almost infinite impedance. In most cases, the 10kohms specified come from a resistor placed between input and ground, that can handle up to 0,25W (typically)


I often apply overvoltage up to 18V p-p on my line or microphone input on the soundcard, and never got any problem.

The indirect risk is to damage the speakers at the amplifier's output

[ROBSCIX]

OK, so connecting a soundcard that has a output of 5 volts with the volume at 100% will not damage the inputs then?

[CarlosT]

I found that most components like CD players put out like 2V so most preamps/amps seem geared towards that kinda "line" level input and I guess most GCs are setup with a gain of like 20:1. Do the math...is that 40 volts at full output assuming you got that much voltage at the power supply? Yeah...that could be loud