Hardware Insert - Why Am I Losing Low End?

[ITHertz]

Hi Folks,

I generally work "in the box" however I also have a small collection of outboard gear that I use primarily for tracking. For something different I thought I'd try integrating some of this outboard gear into my mixbus processing in Studio One using its Pipeline plug-in.

As a starting point I thought I'd add my Warm Audio WA-273-EQ, which I'm connecting to via the Line In/Line Outs of a Scarlett 18i20 (2nd Gen.) interface.

Initially, and to check the routing, I ran the Outputs of the Scarlett to its Inputs. Apart from a small loss of signal that was corrected by using the Scarlett's Input Gain knobs, this worked as expected.

However, when I ran the Scarlett's output into the Line Input of the WA-273, my low end disappeared! This loss seems to be linear across the frequency spectrum and a bit like a tilt filter - what I'm getting is a very "transistor radio-like" sound. I used SPAN to check this and found that a noise source produces a linear sloping output with the low end removed.

Investigating this further, I connected the Scarlett's output to the Insert Return of the WA-273. The insert is designed so that a compressor can be added before the Warm's EQ section. In this case, the signal appears at the Line Out of the Warm in tact with no loss of low end.

Just wondering why I lose the low end when going from the Scarlett's Line Outputs to the Warm's Line Inputs? Is this some sort of impedence problem? Am I missing something obvious?

Cheers,

Chris

[The Elf]

Sounds like a classic case of balanced/unbalanced cable confusion. What connections are you using between these pieces of gear?

[Wonks]

No block schematic for the unit in the manual but presumably you are using the line input selector switch to disconnect the mic preamp circuitry? And you are using TRS to TRS cables (and have swapped cables in case one is faulty)?

Otherwise, have you got a mixer handy that you can test the WA with? It may be a faulty unit.

[ITHertz]

I'm using TRS cables and the I/O is via a Neutrik patchbay.

Yes, I'm switching the unit to Line Input (don't think it works otherwise).

I did some tracking through the WA-273 a while ago and all seemed fine (some bass actually so it would have been noticeable).

Could this be a TS/TRS problem? Could the Line In be expecting a TS?

[ITHertz]

Here's what the WA273-EQ's manual says:

6. Line Input
This TRS input is padded and is designed for line level instruments or sources that put out line level signals such as keyboards, electric instruments or audio interface outputs. This input travels through the entire preamp circuit including both Carnhill Transformers.

[Aled Hughes]

I’m sure you’ve checked this already, but you haven’t got the EQ set up to roll off the bass end have you?!

[ITHertz]

I’m sure you’ve checked this already, but you haven’t got the EQ set up to roll off the bass end have you?!

No EQ is flat. Actually set to Off.

[Hugh Robjohns]

What kind of signal levels are you sending?

If you're sending digital levels peaking anywhere close to 0dBFS, that will be around +20dBu analogue levels from your interface and that will undoubtedly be saturating the transformers in the 273, with a consequent loss of bass.

[ITHertz]

What kind of signal levels are you sending?

If you're sending digital levels peaking anywhere close to 0dBFS, that will be around +20dBu analogue levels from your interface and that will undoubtedly be saturating the transformers in the 273, with a consequent loss of bass.

The test signals were Pink/White noise around -12dBFS.

[Hugh Robjohns]

It looks like the line outputs of the Scarlett 18i20 interface are around +15dBu for 0dBFS, so a signal around -12dBFS will be about +3dBu which is where it should be. So it doesn't look like transformer saturation should be a problem... although it would be well worthwhile trying with a lower level just to check.

Rather than using noise, try and find a sweep-frequency sine tone and watch the levels on the meters (keep the speakers turned down!)

The Scarlett's output impedance is quoted as 430 Ohms which is unusually high (most are 100 Ohms or so), but that shouldn't be anywhere near high enough to be a problem.

I'm wondering if the Warm Audio needs to feed into a 600 Ohm terminated input to tame the transformer's resonance peak. The 60K input impedance of the Scarlett's line input might be too high to load the output transformer properly....

[ITHertz]

I tried the setup with a sine sweep and it's as I described it above. As the frequency of the tone increases, the volume goes up accordingly. It more-or-less goes from silence at the lowest frequency to the full signal strength at the highest frequency although there's something strange happening around 18kHz - the sine tone seems to become modulated.

[ITHertz]

The problem is there regardless of whether the EQ section is switched in or out.

[ITHertz]

Just plugged my bass into the WA273's Instrument input and monitored via a split input going into a Yamaha MG10 mixer - all seems fine, plenty of low end.

[Mike Stranks]

Just plugged my bass into the WA273's Instrument input and monitored via a split input going into a Yamaha MG10 mixer - all seems fine, plenty of low end.

That would seem to confirm the impedance mismatch using the Scarlett... I can only think of 'clunky' solutions. I'm sure Mr R will have a more elegant solution.

[ITHertz]

Just tried the original setup but running into a DBX162SL - all good. I guess this means the problem is with the WA273.

[Sam Spoons]

Can't see how it would be in this case but it sounds a bit like a polarity inversion of one channel would affect a stereo signal. Obviously if you're running in mono that can't be the case but.......

[Mike Stranks]

My money's still on an impedance mismatch... Hugh's facts 'n' figures are also pointing in that direction.

In simple terms what's needed is something to put a heavier load in the path so the respective outputs and inputs are at peace with each other.

[Hugh Robjohns]

What happens if you feed via the instrument input (you might need to keep the signal level down on the Scarlett output). That might help identify whether it's the line input at fault (which seems likely) or the line output.

Egg sucking time... but you have checked your cables are all good haven't you? A lack of bass and a rising top end is often indicitive of a broken signal wire...

H

[ITHertz]

Hi Everyone,

I contacted Warm Audio and they said it sounds like a faulty balanced cable. I'll check that first but it means I have to get to the back of my rack which requires moving some other stuff. I'll report back when I've checked the cables on the rear of the patchbay.

Cheers,

Chris

[ITHertz]

I swapped the leads at the back of my rack and I think it's fixed the problem. The original patch leads were Hosas and I noticed that they were a tiny bit loose in the sockets. I replaced them with some leads with different plugs that were a tighter fit and that seemed to fix the problem.

Cheers,

Chris

[Mike Stranks]

Good work

I stand corrected...