OT: Vpp - eh?

[DominicPerry]

ESI have a new USB interface. I want to know whether it operates at -10 or +4 and what the maximum input level is. However, the only spec they give is:
Input Level(0dB): Max 6.0Vpp

Can I draw any conclusions? How do I translate Vpp to anything?

Thanks

Dominic

[Cary B. Cornett]

Vpp means "Volts peak-to-peak", which for a sine wave (or any other symmetrical waveform) is exactly twice the "peak" voltage, which for a sine wave is 1.414 times the RMS voltage.

6.0 / 2.828 gives about 2.12 volts RMS, or about +8.74 dBu. For the standard +4 dBu reference level this would only give you about 4.7 dB of headroom, which is way too low by any of the usual standards. With a -10 dBv (10 dB below 1 volt) reference level, OTOH, you would have about 16.9 dB of headroom above "0 VU", which is acceptable.

If you feed this interface from any "pro" mic preamp, just figure you have added 12 dB of gain to what the mic preamp has and make sure you set levels according to the SAW record meters rather than any meters in the mic preamps.

[DominicPerry]

Cary,

Thanks for your explanation. I don't get the derivation or the calculations, but I appreciate your clarity in pointing out the headroom available, which is what I now realise I needed to know.
I have a mic pre with a hot output, although not unusually so for 'pro' gear, +22dBu on XLR and +11dBu on a 3.5 mini jack. So it blows all the consumer ADCs I've tried out the water if I turn it up. I was hoping that the ESI would take it. Not so. Thanks again.

Dominic

[Cary B. Cornett]

Cary,

Thanks for your explanation. I don't get the derivation or the calculations, but I appreciate your clarity in pointing out the headroom available, which is what I now realise I needed to know.
I have a mic pre with a hot output, although not unusually so for 'pro' gear, +22dBu on XLR and +11dBu on a 3.5 mini jack. So it blows all the consumer ADCs I've tried out the water if I turn it up. I was hoping that the ESI would take it. Not so. Thanks again.

Dominic The real question about any ADC is not "how hot a level will it take?" so much as "how far above the noise floor of the converter is the hottest level it will take?".

Here is an experiment you can try:

First, fire up your DAW and open up either the sound card's own mix/metering application or SAW in "Live" mode with a really wide-range level meter. You want a meter that will read down to at least -90 dBfs. With nothing hooked up to the input of the ADC, you will see a very low level reading on your level meter, and that level is the noise floor of your converter. Now hook up your mic preamp to the ADC input, but don't have anything plugged into the mic preamp. Odds are the reading on your meter will go up, and if it does what you are now seeing is the noise floor of the mic preamp (try this at both high and low preamp gain settings). If the noise floor does NOT change when the mic pre is connected, your ADC is what limits your dynamic range.

Now, if the mic preamp has a higher peak output level capability than your ADC can "take", and you saw the noise floor increase when you connected the mic pre to the ADC, you can use a "pad" (attenuator) between the mic pre and the ADC to drop its output. If you drop the preamp output just enough for its noise floor to match that of the ADC, you are now set up to get the best dynamic range you can.

There is one more sneaky bit that often gets ignored here: sometimes the noise level in the room you are recording in is well above the noise floor of both the ADC and the preamp. If you are using a condenser mic, the odds of this can be pretty good, in fact. When I do choir concert recordings, for example, there are all sorts of noises like HVAC rumble, page turning, audience noise, etc. that the noise floor of my recording system is simply not an issue, and this is with the gain set so that the highest level peak in the entire program is maybe around -6 dBfs.

My setup isn't all that fancy, either: Behringer ADA8000's for preamps/converters fed by a pair of AKG C451/CK2 condenser mics.

It is just possible that you can work with less "fancy" converters and get along just fine. Don't give up until you have done some tests.

[DominicPerry]

Thanks Cary, I'll try what you suggest. The problem I've had with more than one interface is that the SAW meters don't clip at all, or they clip way after the ADC is saturated. The M-Audio Transit I'm using is a case in point. I can't get the SAW meters to clip, but even at lower levels, the waveform doesn't get anywhere near the boundary, but its clearly flat-topped and the distortion is easy to hear. This makes setting levels a nightmare - I can't trust the mic pres 7 level LED, because I know clipping is somewhere in the middle, I can't trust SAWs meters and the Transit has no meters of its own. Anyway, more experimenting to go...........

Dominic

[Bob L]

The SS meters will display clipping at exactly the last bit of resolution... therefore they are exactly accurate in the digital domain... if the signal is clipping before that, then your analog section is clipping before the signal is digitized and passed on to SS.

Bob L

[DominicPerry]

Thanks Bob. I guess I'm just overloading the analog stage then. Must be a rubbish design!

Dominic

[DominicPerry]

I used Sonalkis FreeG meter.

Transit on its own :-89dB RMS
Transit with FP24 plugged in, no mic, phantom on,
zero gain: -91dB RMS
about 40dB gain: -84dB RMS
full gain (66dB): -71dB RMS

Dominic

[Cary B. Cornett]

The problem I've had with more than one interface is that the SAW meters don't clip at all, Meaning that the converter analog section is clipping at a level somewhat below 0 dBfs, which, as you say, means a bad design.

or they clip way after the ADC is saturated.
To me, "saturation" and "clipping" of the ADC would be the same thing, which would mean that once the ADC hits saturation/clipping, it simply cannot produce numbers that are any higher. Now, it is possible that a badly-designed analog section could start producing appreciable distortion before it hits clipping. That would be the only way the second "case" you describe could make sense.

The M-Audio Transit I'm using is a case in point. I can't get the SAW meters to clip, but even at lower levels, the waveform doesn't get anywhere near the boundary, but its clearly flat-topped and the distortion is easy to hear. This makes setting levels a nightmare - I can't trust the mic pres 7 level LED, because I know clipping is somewhere in the middle, I can't trust SAWs meters and the Transit has no meters of its own. Anyway, more experimenting to go........... First of all, you need to make sure that you are not attenuating the signal digitally (either in an input channel or with the record level adjust) before it hits the meter, because doing so would cause exactly the problem you describe.

If you are NOT doing that, all you need to do is find out what level on the SS record meter corresponds to clipping in the converter. Pretty simple, really. The highest reading the SS record meter will show is the converter's real clipping level. If, for example, the highest reading you can get on the SS record meter is -6 dBfs then you just have to remember not to let any signal reach that level on the meter and you should be OK. If the maximum level you can get on the meter is not always the same, you are changing something somewhere without realizing it.

[Cary B. Cornett]

I used Sonalkis FreeG meter.

Transit on its own :-89dB RMS
Transit with FP24 plugged in, no mic, phantom on,
zero gain: -91dB RMS
about 40dB gain: -84dB RMS
full gain (66dB): -71dB RMS

Dominic I haven't tried the FreeG meter, so I don't even know if it does RMS readings. Other than that, offhand those numbers don't seem that far out of line. What would be interesting to know at this point is how far above the Transit's max input level the FP24's maximum output level is. If the difference is more than 10 dB, you might want to consider using a pad between the FP24 and the Transit.

OK, I just looked up the specs of the two devices, and if I read correctly, the output clip level of the FP24 is at least 12 dB above the input clip level of the Transit. Unless, that is, you are using the Tape output instead of the Line output, in which case the clip level difference is less than 4 dB, which IMO is not worth worrying about. Based on what I see in the FP24 user guide, you should use the "Tape" output of the FP24 to feed the line in of the Transit for the best level match between the two.

Gee, I sure learn stuff when I take the time to read :eek::):cool:

[DominicPerry]

Cary,

FreeG has RMS and peak readings. I chose RMS just because you mentioned RMS in your earlier explanation.

I was using the tape output of the FP24. It has a -18dB (0dBu) ref tone (1KHz) and that reads -14dB RMS / -11 peak in SAW. Which is irritating, frankly.
So the difference is about 7dB. Or is it 4dB?

I can't find any official input level max for the Transit - did you find a figure or calculate one?

I have a 10dB mini-jack pad, I just worry about whether I lose much sound quality going through it - I don't know how they work.

Dominic

[DominicPerry]

The peak figures are

Transit on its own :-80dB
Transit with FP24 plugged in, no mic, phantom on,
zero gain: -84dB
about 40dB gain: -70dB
full gain (66dB): -56dB

Dominic

[DominicPerry]

Good news - I tried the main XLR outs to a mini jack with a 10dB pad, and the -18dB ref tone now gives -17.98 dB peak on the FreeG meter (well, what's 0.02 dB between friends, except perhaps to BrainCrasher).

Thanks for your patient help Cary.

Dominic

[Ian Alexander]

Thanks Cary, I'll try what you suggest. The problem I've had with more than one interface is that the SAW meters don't clip at all, or they clip way after the ADC is saturated. The M-Audio Transit I'm using is a case in point. I can't get the SAW meters to clip, but even at lower levels, the waveform doesn't get anywhere near the boundary, but its clearly flat-topped and the distortion is easy to hear. This makes setting levels a nightmare - I can't trust the mic pres 7 level LED, because I know clipping is somewhere in the middle, I can't trust SAWs meters and the Transit has no meters of its own. Anyway, more experimenting to go...........

Dominic
There's a female VO talent I often get tracks from. She uses an M-Audio FW410. She has to be very careful with levels because even when the levels look safe on SoundForge, the signal is clipping somewhere in the chain at about -6 dBFS. As you describe, she gets the flattops and obvious distortion. We tried everthing we could think of, playing with settings in the M-Audio mixer and SF. AFAIK, she is simply aiming for levels around -12 to be safe. I don't think that is too low, but it sure is frustrating to get clipped at -6. If you figure this out, please let me know how. Thanks.

[Cary B. Cornett]

Good news - I tried the main XLR outs to a mini jack with a 10dB pad, and the -18dB ref tone now gives -17.98 dB peak on the FreeG meter (well, what's 0.02 dB between friends, except perhaps to BrainCrasher).

Thanks for your patient help Cary.

Dominic There is a very good chance that your 10 dB pad is mono, and if so I don't think you can make proper stereo recordings with it. Sorry.

[DominicPerry]

There is a very good chance that your 10 dB pad is mono, and if so I don't think you can make proper stereo recordings with it. Sorry.

I'll check. If it is, I'll just get a couple of XLR-XLR barrel pads.

Dominic

[Cary B. Cornett]

I was using the tape output of the FP24. It has a -18dB (0dBu) ref tone (1KHz) Are you sure they didn't say it was "0VU"? A lot of people get confused about reference levels. Depending on the equipment, 0VU can mean any of several reference levels. The most common 0VU levels are +4 dBu, +8 dBu, and -10 dBv. Again, looking up the FP24 user guide, it says the tone is at +4 dBu, but elsewhere it says that the tape output is 10 dB lower in level than the XLR outputs, so the level of the tone from that output will actually be -6 dBu.

and that reads -14dB RMS / -11 peak in SAW. Which is irritating, frankly.
So the difference is about 7dB. Or is it 4dB? Based on my calculations, with the FP24 tape out connected to the Transit, the reference tone from the FP24 should read about -15 dB peak in SS.


I can't find any official input level max for the Transit - did you find a figure or calculate one? Looked it up.

[DominicPerry]

Well, now the whole thing has gone nuts. I have no idea how I got a -18 dBu reading but it's not doing it anymore, with or without pad, XLR or tape mini jack. Still works fine with the Mytek ADC, and I'm assuming SAW hasn't changed,

I give up with cheap stuff.

Dominic