UA25 Input sensitivity too high on dynamic mic - clipping at zero

[kahl__hahgt]

I'm tyring to record my guitar amp (50 watt head into 2x12 cab) via an SM57.

I'm using the Edirol UA-25 interface as I don't have anther preamp for the mic at the moment. The problem is when I crank my amp to a nice level the UA-25 is clipping even with the input sensitivity set to zero. The clipping light on the front just flashes red, when I turn down the amp I can have the sensitivity higher and there is no clipping, but of course sounds not so good. Is the Edirol broken?

The internal limiter is turned off, phantom power is also turned off, advanced mode turned on like in the picture.

Before I go out and buy a proper mic preamp, could it be that I'm doing something wrong?

I am new to recording and have tried to follow the instructions, but maybe I am just an idiot and set it up wrong.

[Hugh Robjohns]

I can;'t find any useful specs on the UA25 sensitivity, but it sounds like the amp is so loud, and the interface so sensitive, that you're overloading the preamp.

If that's the case, you have two simple options:

1. Use an XLR-TRS adapter or XLR(f) to TRS cable and plug the mic in to the line input instead. The higher impedance will affect the mic's tonality slightly, but I doubt it will be enough to concern you.

2. Buy (or make) an in-line 20dB pad like this (other brands are available) to plug between the mic and the interface XLR input.



https://www.studiocare.com/neutrik-in-line-xlr-attenuator-20db-pad-200-ohm-3-pin.html

Instructions for making your own XLR pad are here: http://www.uneeda-audio.com/pads/

[Wonks]

I can;'t find any useful specs on the UA25 sensitivity,

From the user manual:

Input Jack L, R (XLR type): -60 to -20 dBu
Input Jack L, R (1/4 inch TRS phone type): -36 to +4 dBu

[Wonks]

Regardless of the level light, are you getting audible clipping of the waveform when recording? Is it clearly hitting 0dBFS and levelling out?

[kahl__hahgt]

Yes, in Reaper it is clipping on the recording, and meters hitting red. I suppose it must be that the guitar amp is too loud for the ua25 to handle, so I will need to find a proper preamp for my sm57.

[James Perrett]

Just go for an attenuator like the one that Hugh showed in his post.

If you are in the UK try
https://cpc.farnell.com/pro-signal/psg0 ... dp/AV17578

or elsewhere try
https://www.amazon.com/Whirlwind-IMP-Pa ... 45&sr=8-17

[Hugh Robjohns]

I can;'t find any useful specs on the UA25 sensitivity,

From the user manual:

Input Jack L, R (XLR type): -60 to -20 dBu
Input Jack L, R (1/4 inch TRS phone type): -36 to +4 dBu

Yes, I found those, but they don't count as 'useful' in this context, sadly, because there's nothing stating the reference. Is that -20 dBu the alignment level, the peak level, clipping, what? And if it is the nominal level does that equate to -20dBFS, -18dBFS, -12dBFS, or what?

What we actually need to know is at what analogue mic input level the interface clips. From that -- and knowing the sensitivity of the SM57 -- we can work out the maximum SPL that can be tolerated at the microphone from the guitar amp/speaker.

However, since it appears the minimum mic gain is around 20dB, is seems reasonable to assume that the interface is overloading. The usual solution in that situation is an in-line pad... 20 or 30dB attenuation should sort it out.

The alternative bodge, since the OP is using a dynamic mic and so it doesn't need phantom power, is to connect the mic to the balanced line input via an XLR(f)-TRS cable or adaptor since that has much lower sensitivity.

H

[Martin Walker]

Or as a dalek would say "Att-en-u-ate!"

Martin

[ef37a]

How I agree with Hugh that interface input speccs' are rarely useful! Since we mostly all know a dynamic puts out 1 to 2mV for a close "rock" voice, all we need is the input number that gives -18dB fs or 0dB fs or just SOME **** figure for fs!

But OP, a 50W amp at full chat into a decent 2x12 is going to hit 110dB SPL and if you keep that up for long you will go deaf.
You might consider a power soak or DI and software?

Dave.

[Wonks]

But OP, a 50W amp at full chat into a decent 2x12 is going to hit 110dB SPL and if you keep that up for long you will go deaf.

Whilst true, the OP may well not have it at 'full chat', just loud enough for it to sound good to the OP. But unless it is at 'full chat' with some added output valve distortion, the recorded amp sound will probably sound very similar if recorded with the amp at a much lower volume. You need to listen to the recorded sound, not the amp sound in the room. If necessary, add a pedal or two for more drive or compression.

[Hugh Robjohns]

But OP, a 50W amp at full chat into a decent 2x12 is going to hit 110dB SPL...

Just to throw some numbers around, according to my spreadsheet... if an SM57 is receiving 110dB -- which is easily possible with the mic placed right next to the speaker even when playing quite moderately -- it would be producing about -40dBu.

Since the minimum gain setting on the interface appears to provide 20dB but it's overloading, I suspect the actual level hitting the mic is substantially more in the OP's case...

H

[ef37a]

But OP, a 50W amp at full chat into a decent 2x12 is going to hit 110dB SPL and if you keep that up for long you will go deaf.

Whilst true, the OP may well not have it at 'full chat', just loud enough for it to sound good to the OP. But unless it is at 'full chat' with some added output valve distortion, the recorded amp sound will probably sound very similar if recorded with the amp at a much lower volume. You need to listen to the recorded sound, not the amp sound in the room. If necessary, add a pedal or two for more drive or compression.

True too Wonks but if it is a standard, EL34/6L6 valve amp they really don't start OP stage distortion until close to full power, up to say 10W (which is stll piggin' loud) they are pretty much "class A hi-fi") .

My experience some years ago miccing up a 15W Dommy clone was even with 90dB or so in the room and the 57 on the speaker fret we were still not getting a massive signal. I think therefore OP must be giving it very large?

Dave.

[Wonks]

The SPL a few inches from the speaker cone is going to be a lot louder than the quoted speaker efficiency SPL figure (at 1 metre at 1W) plus the amp gains. No idea how much more though.

[ef37a]

The SPL a few inches from the speaker cone is going to be a lot louder than the quoted speaker efficiency SPL figure (at 1 metre at 1W) plus the amp gains. No idea how much more though.

Hugh will know but I suspect it's a bit complicated? The speaker is a huge source area and the mic only intercepts a tiny fraction of that.

Then, just defining what a watt is into a speaker load with a guitar's waveform/spectral content is tricky? Not white or pink noise is you?

Hugh's estimate of around 10mV from the mic should not overload any decent mic pre so I suspect we are talking defenestration here?

Dave.

[James Perrett]

That 20dB minimum gain is even higher than the 13dB minimum gain on my Focusrite interface that often gives me problems with loud sources. While I've not had overloading with guitars on the Focusrite, they've come pretty close and I certainly wouldn't want another 7dB of gain on those channels.

[Hugh Robjohns]

The SPL a few inches from the speaker cone is going to be a lot louder than the quoted speaker efficiency SPL figure (at 1 metre at 1W) plus the amp gains. No idea how much more though.

If it was a perfect point source it would be 6dB louder for each halving of distance. In reality, though, because the speaker size is a significant portion of the signal wavelength over part of the usable,frequency range, it's more likely to be 3-4dB louder for each leaving of distance.

So if it was rated at 90dB/W/m, it would theoretically be 96dB at 50cm, 102dB at 25cm, 108dB at 12cm, and 114dB at 6cm, and 120dB at 3cm, which is probably about as close as the mic is likely to be -- so nominally 30dB louder!

In practice, then, be exposed to an SPL roughly 15-25dB louder than the speaker's rated efficiency... Per Watt of input power.

[Exalted Wombat]

...so I will need to find a proper preamp for my sm57.

Your Edirol UA-25 has a perfectly 'proper' microphone preamp. A different one might have more attenuation range, it might not.

[kahl__hahgt]

Thanks for all the input and reading material. The amp is loud, it’s a cornford mk50, I have it in another room because it’s too loud for me to be next to it, but yes I was cranking it when trying to record - to at least 6-7 on the head master channel.

I will probably end up getting an attenuator for the head so I can push the tubes at lower volumes, I will also order and try out the in-line attenuation suggested by Hugh to compare with.

I also tried going directly into my RME card from the SM57, but of course the signal was not suitable without a preamp.

[ef37a]

"I also tried going directly into my RME card from the SM57, but of course the signal was not suitable without a preamp."

Ah now, that is a possiblity. You can buy low Z to high in line transformers very cheaply (not the Shure one! Good I am sure but pricey)
These give about a 15-20dB level lift to a mic and with the high level you are getting would I am sure give enough signal into the RME card. Remember, you only need -20dB fs as an average and at 24 bits even -30dB fs would be perfectly usable. Do still get the XLR attenuators but a transformer could give another tonal option.

The transformers also act as a "balun". I shall seek out a link...

I know it was I that suggested a load/power soak and they are useful but few sound quite like the amp on full chat into speakers.

https://cpc.farnell.com/unbranded/35-59 ... 2sQAvD_BwE

Dave.