Gain Curves on Interfaces Preamps

[TxManx]

Hi,

 

This seems to be the bast forum for this question. IF it should go elsewhere, lemme know.

 

Any insight into what a reasonable gain curve would be for a preamp in a digital interface? I am asking because I recently bought a Clarett 4Pre USB interface and the gain does basically nothing until you turn the knob well above 85%. I am guessing that all gain curves are asymptotic in nature - not linear. They start flat from 0% and then beyond 50% they start climbing steeply to 100%. But, come on, starting to climb after 85% is tedious and annoying.... you basically have to only 80-95% to play with...

 

So, I conducted a couple of experiments with three mics (Rode M5, Aston Origin, and a DIY MicParts S47) and three preamps (Clarett 4Pre USB, Scarlett 18i8, and a FMR Audio RNP Preamp into the Clarett #5 line in). I basically sang a flat tone, turned the gain up, tapped the mics at 20%, 50%, 80%, and 100%, and then back down. Then I captured the wave forms and labeled the taps. While my testing technique could be improved, I doubt that more precision or accuracy will change my conclusions....

 

See the attached graphs. I am not an recording engineer and my practical experience in this area is next to nil. That said, the FMR and Scarlett strike me as having reasonable gain curves. The Clarett does nothing until you hit 80%+. Literally, I am thinking, "The Clarett is broken". Why would any design engineer build a device where the useful gain is always >80%?!? I mean, the return rate must be extraordinarily high on this device.

 

So.... my question... Does my concern about the Clarett gain curve have any basis in reality? Or, if I tested 50 digital interfaces would I find that 50% of them would operate similar to the Clarett? Should I get a different interface?

 

Your insight appreciated. TIA...

 

P.S. My only desire is to record my piano and voice, but have found microphony to be an interesting topic resulting in me building a MicParts DIY kit, swapping out the capsules on my Rode NT1-A's, etc. I just want to get to a straight forward game plan... Thx

 

.: eric

[Mike Rivers]

Your observations are correct. Preamp gain controls typically have a relatively small amount of change for the first half or 2/3 of their rotation and a fairly large change near full gain. Part of this is in the design of the gain control, and part is related to the way we hear changes in volume. This control action can either be helpful or annoying, depending on how you're using the preamp. In short, there are two ways of using a mic/preamp combination. When the source is very close to the mic, you don't need a lot of gain, but it's helpful to have good control of the amount of gain. This is why the control action tends to be spread out over about the first half of its range. If you need a lot of gain, it's either because the source is at a good distance from the mic or you're recording something very quiet. There's usually no need to make a fine adjustment near full gain - either it's all the way up or it's, up as far as it will go before the preamp's noise becomes objectionable.

 

When I first started reviewing this sort of thing, a lot of users were bugged about how gain controls worked, so I started measuring it in a more controlled way than you have. I used an oscillator as a signal source, connected directly to the mic or line input, rather than singing into a mic, and plotted the output level versus the rotation of the control. Try that if you want to make some more quantitative measurements. There are several phone apps that will turn your phone into a signal generator - I like Keuwlsoft's Function Generator - though you'll have to cobble up a cable to connect it to an input on your preamp or interface.

 

[matthew mcglynn]

Mike's answer neatly captures the design philosophy behind non-linear gain controls.

 

The question of why the Clarett has unusably low gain until you're at 80% of the knob's travel range remains open. I would be inclined to share your concern that the preamp is broken. I've used any number of inexpensive preamps/interfaces from M-Audio, Focusrite, Presonus, and Mackie, and for vocal use with mics such as you list, I rarely have to go past 50% gain to get a strong signal.

 

I'm not familiar with the software audio routing config for this device. You might wish to review that to see if there is anything there that could affect gain staging.

 

I scanned the reviews at Amazon and didn't see any mention of the situation you describe. If this unit was supposed to behave that way, I sort of think someone would have mentioned it. If it were mine, I'd look into a return or exchange, or at least get a second unit for comparison.

 

Did you try calling the company's tech support?

[TxManx]

Mike, Matt..... thank you for responding 3 months ago. Apologies for the delayed follow up.... (new role at work, more travel, yada yada yada). Anyway, after more testing, I returned the Clarett because I could *not* tell the difference between that and the Scarlet. Root cause is either (1) my lack of experience in recording and/or (2) my many summers decades ago of working in a quarry crushing rock, mixing blacktop, and running heavy machinery all without hearing protection : / Now I am burdened with tinnitus in my left ear, etc.

 

.: eric

 

[KuruPrionz]

Some thoughts on this. I had a Mackie interface some time ago. It used pots to adjust the gain for the preamps.

Pots can get scratchy, these did. It drove me nuts.

 

I went to a MOTU interface, no pots anywhere. All knobs were used to adjust digital controllers so there was no audio signal running through them. Gain settings had a number on the screen.

You could use the interface without the free software or you could control the unit from your computer. I liked it.

 

But, my computer (2008 Mac Pro) was rapidly getting left behind. I got a 2014 MacBook Pro with Thunderbolt 2 and a Presonus Quantum interface. This is hybrid in that the preamps all use digital controllers but the output fot the monitors and 2 headphone amps use pots. This means there will never be pot scratchy noises on the recordings. Maybe someday there will be on playback, right now it's all good.

They also have software to control the gain stages or you can press a couple of buttons to select the input channel and turn a knob that displays a number on the screen of the interface. It remembers your last settings so it would be the same as not touching your pots after finding your happy gain place, it will be there when you return.

 

I am happy with the Quantum and don't worry about gain curves anymore. If it takes 52 with a particular mic at a specific distance to a known sound source I can type that in to a page of settings saved on the Desktop and just set the gain at 52. Something to consider if/when you ever upgrade.

 

Even though it is mostly just me, I've found having 8 inputs allows me to set things up and leave them. Makes it very fast to plug in my bass or do a vocal, etc. Cheers, Kuru