AMS Neve 1073 Sweeps

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
For giggles...Here's the mod without a zobel or a terminating resistor.


(graph removed due to inconsistent measurement)
 
Thanks James,

do this for me, read me the percentage values for THD at 20hz and 20khz. you have to move the mouse to that freq for it to display its percentage.
 
Stock
20hz  5.93%
20K  0.021%

Mod unterm
20hz  5.74%
20K  0.016%

Mod term
20hz  5.83%
20K    0.017%

(Numbers not truly indicative of "normal" performance of preamp due to input levels too high)
 
The high 20hz numbers could be from hitting the mic input transformer with line level.  Keep in mind this isn't exactly a 1073 clone with a dedicated line input.

I should probably do a sweep at a lower output level from REW.
 
Here's a graph with the input level down...

20Hz  0.090%
20K    0.027%
 

Attachments

  • nv73_mod_low-level.png
    nv73_mod_low-level.png
    55.6 KB · Views: 36
Actually nope you are measuring correctly,

I send line out into the mic in set to 20dB for all my measurments.

What's weird is you're getting way too much distortion in the low end, yet less than all my modules in the high end.

Is your input the VTB9045?
 
Thanks so much for your efforts running all of these tests.  I know it takes a lot of time to set them up and record them.

For those of us following along at home, it would be very helpful if you guys doing the tests could define some more test parameters.

What level are you feeding the input transformer? 
dBu, dBm, dBv? 
Balanced, unbalanced, ground reference? 
What is the source impedance?
What position is the trim or fader?
What is the output level?

Best,
Bruno2000



 
TheJames said:
For giggles...Here's the mod without a zobel or a terminating resistor.

Are you sure you zobel resistor is the correct value? It seems that with out the zobel your high end is extended appropriately but with it on it barely makes it past 12k.
 
Without the zobel on a 1166 the high end should increase. Try increasing your pf on your input transformer, that will bring the high end up. Before I had my EZ 1073s wired the way I have them now, it took a massive 680pf to get the high end right.
 
I don't have much time to write at the moment...

But the zobel values on the output are 1.5k and 0.010uf.

The zobel on the input transformer is 180pf.

I think I should pull down the captures from last night and focus on some new ones with the same parameters that Aaron was using.  This is based on the reasons that Bruno listed above.  We need to make sure we're doing apples-to-apples comparisons, and my previous graphs, (which I made an effort to match up pretty closely), are all at levels that don't correspond to what Aaron was using.  Therefore, these graphs are not really valid in the grand scheme of things.

There are quite a few things to look at, I might have some cycles tonight to provide some better, more consistent output.

James
 
Pay attention to the DB scale (1db window).  The top end isn't that radically out.  I'm down about .5db at 20k.  The thing that I was frustrated with hunting down was the response out at 30-40k.

Anyway...I'll put together some new graphs and get back to the group.
 
dont forget to turn down the output of the interface by 50db. going from line to mic level you need a 40db-50db pad depending on if you are coming from -10 or +4…

Line level signal to microphone input adapter
Sometimes there is need to convert line level signals to such signal that it can be connected to microphone input. Because the line level signals are typically in range of 0.5..2V and the microphone signals are in millivolt range, quite much attenuation is needed to match the signal levels. This means that typically you will need 40-50 dB of attenuation.

40 dB PAD
Consumer audio line (-10dBu) level to microphone matching can be done using a "40 dB L-pad attenuator". This circuit is indeed quite simple:

+Line level input -----R1----+-- +Mic level output
                                    |
                        +---R2---+
                        |
Ground (input)----+--------------- Ground (output)

R1 = 10 kohm
R2 = 100 ohm

Circuit technical data:
Attenuation: 40 dB
Input impedance: 10 kohm
Output impedance: 100 ohm
Input signal: Unbalanced line level input
Output: Unbalanced microphone output
50 dB PAD
Professional audio audio line level (+4dBu) to microphone matching can be done using a "50 dB L-pad attenuator". This circuit is indeed quite simple:

+Line level input -----R1----+-- +Mic level output
                                      |
                        +---R2----+
                        |
Ground (input)----+--------------- Ground (output)

R1 = 33 kohm
R2 = 100 ohm

Circuit technical data:
Attenuation: 50 dB
Input impedance: 33 kohm
Output impedance: 100 ohm
Input signal: Unbalanced line level input
Output: Unbalanced microphone output

Balanced version
Balanced Taper Pad: 50dB (for 600 ohm impedance, XLR connector)

  XLR in              XLR out

      1 ------------------ 1

      2  ---R1---+---R2--- 2
                    |
                    R3     
                    |
      3  ---R1---+---R2--- 3             
Component values:
R1 = 300 ohms
R2 = 300 ohms
R3 = 4 ohms
 
Okay...So I finally got some time to get back to this...

Let me preface that REW is a fairly crude method of performing audio measurements...But for lack of better tools (and the sake of staying consistent with Aaron) here are some new graphs.

For those playing along at home...I'm sending out a -20dbFS signal from my converters which are set to be 0dBu at that level.  The NV73 has a "line" input mode that pads down the level post input tranformer.

I've got two NV73's and they measured statistically identical prior to engaging in this, uh...Whatever you want to call it.

Because of the "franken-nature" of the "modded" NV73...It sits outside of the 500 frame on an extender card.  To rule out the possiblity that the extender card and associated cabling are making the differences, I ran the same sweeps against the stock NV73 on the extender as well.  To make things easier to compare I performed minor adjustments in sweep levels to make sure the measured response lined up pretty much the same across the graphs.  I also included the phase response because I think that's important to understanding the reponse that's seen.

Soo...Here goes...
 

Attachments

  • nv73_stock.png
    nv73_stock.png
    50.2 KB · Views: 35
Hey James,

The stock on the extender looks good to me. I also can't speak on the NV73, I've never used it and it's a different PCB Layout... My specific test were geared towards the EZ1073. It could be that all your efforts simply aren't needed in your case and the unit sounds great stock. According to your graphs the first stock sweep looks correct.

As far as REW being a good measurement device. It does give accurate freq and phase response witch is good enough for me. It's certainly no AP but I doubt most DIY people have access to a AP anyway.

Since my last post, I've got another AMS1073 on they way. I can't deny that even after all my efforts I was able to match what the 1084 was doing but I still can't get them to sound as good as my AMS 1073. Extra harmonic mojo in the high end that sounds really nice. Actually I'll post sound samples. The clip is showing the difference between my AMS1073 and AMS1084. To me the 1073 sounds a little better with a harmonic thing going on in the high end the the 1084 isn't doing.

I've found pretty big inconsistencies lately with Neve gear. My 1084 and 1073 sound very different. My EZ1073 and home etch builds sound just like my 1084. However the Neve DPA I heard last week sounded more like my 1073... Not the same but had some sort of mojo going on. Then I heard another DPA in the same studio and it sounded like a blanket was over it.

I'm really thinking there is a inconsistency in the input or output transformers that is causing this... A sound I'm very familiar with now. Almost a boomy, grainy slightly phasey high end. The easiest way to describe it would be when you're aligning the head on a tape deck. You know how it sounds when the alignment is off just a tad, it's subtle but the high end gets loss and also gets phasey. That's the same type of thing I hear going on here.


I'm also not the only one that hears this, I found this post on GS last night and was relieved to hear that someone else was hearing the same thing as me! http://www.gearslutz.com/board/5170933-post57.html


(Remember mediafire compresses and sounds horrible so make sure you download)
http://www.mediafire.com/listen/02p88mgl4hgrzpo/AMS_1073-AMS1084.wav



 
Back
Top