Sonic Penalties for 2 gain stage vs 3 stages in Neve type Circuit (1073/1272)?

Help Support GroupDIY:

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,886
Location
Nashville, TN
Hello, everyone,

I think I know the answer to this, but I would like to ask here to receive verification/correction. In the preamp section of the 1073, there are three gain stages, with the first one feeding the second one at higher gain (or vice versa, I can't remember) while the 1272 has two stages. I have found that in the 1073 preamp circuit, I don't have any need for the top 20 dB gain range (60-80dB), so I wind up changing the gain settings to 0-65. A single BA283AV card (1272) with it's two stages has a theoretical gain of 81dB, so it seems that this would be plenty for what I need and I can scale down the real estate needed in a case. The question I alluded to earlier is does the absence of one of the gain stages affect the sound characteristics even when coupled with the same I/O transformers? From what I can see/know, the third gain stage is just to achieve the higher gain, so minus a gain stage shouldn't affect the subjective qualities of the sound, and a trimmer can still be used attenuate the output so the input can be driven harder if so desired. Plus, a single gang potentiometer could be used for gain, simplifying the build. Is there anything I'm missing as far as what to expect how the two stage setup will sound versus the three gain stage layout? I'm not looking to build anything historic or make claims to authenticity, just a great sounding circuit in the Neve style family of designs. I have no reason to believe it won't nor sound any different than my 1073 builds, but it doesn't hurt to ask here first. At least not too much.

Thanks!

Paul
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,517
Location
Norfolk - UK
The normal Neve internal operating level is -8dBu (so as to achieve 26dB headroom).  You want 65dB of gain so you need 57dB of gain  to reach -8dBu. You get 6dB in the mic transformer so the first stage needs to provide 51dB. However, flat out, the first stage of a BA283 will provide 48dB gain and that is asking a lot from a basic two transistor topology. There will be less NFB and distortion will rise. Notice that in the 1073, the third stage is switched in from gains of 55dB upwards so with a single BA283 you really should not try to get more than 50dB or so of gain from a single BA283 (no matter how much gain you can get).

The 1272 was designed as a bus amp. Since bus level was rarely less than -48dBu, the first amp in the BA 283 was normally expected to provide about 34dB of gain.

Cheers

Ian
 

Rob Flinn

Well-known member
Joined
Jun 3, 2004
Messages
4,501
Location
Sussex, UK
Ian is right in what he says.

Like you I rarely find myself needing more than 55dB of gain.    Quite a few people have built boxes using single BA283 for mic amps in the way that you say, but the result will provide a maximum gain of 55dB.    This behaves just like a 1073 style amp but the gain is limited to 55dB.    There is a mod one  can do that changes a gain setting resistor that basically twists the nuts of the BA283 to give you a bit more gain if you need it but this now operates outside of what Neve did.  Joe Malone used to have a couple of diagrams on his page with ways to do this, but they don't seem to be there now.

 

Tubetec

Well-known member
Joined
Nov 18, 2015
Messages
2,981
I  built the Neve output stage as a standalone unit  (just 3 transistors) I made the the gain trimable over a small range , even though of course this does effect performance  . Ive used it with higher ratio step up transformers at the input where levels suited . I also found some DOA mic pre's from a big Yamaha console ,they are  ideal when I just need clean transparent gain on top of Neve style output. It also makes a handy unbalanced -10 to balanced +4 level adjuster or you could put a 10k fader at the input .

As previously mentioned gain staging is important, if your convertor only accepts +10 or +12 dbu for 0dbfs,you might need to pad down the output of the Neve to make sure the transformer is driven into the zone where nice things start to happen .
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
20,480
Location
Hickory, MS
Potato Cakes said:
Hello, everyone,

I think I know the answer to this, but I would like to ask here to receive verification/correction. In the preamp section of the 1073, there are three gain stages, with the first one feeding the second one at higher gain (or vice versa, I can't remember) while the 1272 has two stages. I have found that in the 1073 preamp circuit, I don't have any need for the top 20 dB gain range (60-80dB), so I wind up changing the gain settings to 0-65. A single BA283AV card (1272) with it's two stages has a theoretical gain of 81dB, so it seems that this would be plenty for what I need and I can scale down the real estate needed in a case. The question I alluded to earlier is does the absence of one of the gain stages affect the sound characteristics even when coupled with the same I/O transformers? From what I can see/know, the third gain stage is just to achieve the higher gain, so minus a gain stage shouldn't affect the subjective qualities of the sound, and a trimmer can still be used attenuate the output so the input can be driven harder if so desired. Plus, a single gang potentiometer could be used for gain, simplifying the build. Is there anything I'm missing as far as what to expect how the two stage setup will sound versus the three gain stage layout? I'm not looking to build anything historic or make claims to authenticity, just a great sounding circuit in the Neve style family of designs. I have no reason to believe it won't nor sound any different than my 1073 builds, but it doesn't hurt to ask here first. At least not too much.

Thanks!

Paul
I do not have first hand experience but will answer in general.

When dealing with cascaded gain stages the S/N is generally dominated by first stage, since later gain stages amplify that first stage's output signal "and" noise floor.

Distortion and less obviously phase shift is degraded by inadequate loop gain margin (total open loop gain minus closed loop gain). Trying to get too much closed loop gain from one stage can result in compromised linearity performance. 

I have nothing practical to share about those specific modules regarding how much one gain stage can deliver cleanly.

JR
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,886
Location
Nashville, TN
I left out the part of the plan to wire the transformers on the in and out to do some of the lifting for achieving gain. I also have a schematic for getting a bit more gain out of this card adding a resistor and capacitor to mimic R44 and C5 from the main 1073 schematic. The noise floor is some thing I have not taken into consideration. I guess I'm just going to have to wire something up and give it a listen.

Thanks!

Paul
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,517
Location
Norfolk - UK
Potato Cakes said:
I left out the part of the plan to wire the transformers on the in and out to do some of the lifting for achieving gain. I also have a schematic for getting a bit more gain out of this card adding a resistor and capacitor to mimic R44 and C5 from the main 1073 schematic. The noise floor is some thing I have not taken into consideration. I guess I'm just going to have to wire something up and give it a listen.

Thanks!

Paul

The in and the out transformers already do some gain lifting,....  the input transformer has 6dB of gain and was taken into account in my original comments about the first stage.

Cheers

ian
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,886
Location
Nashville, TN
I managed to get back to this testing. It seems with just the AV circuit I am able to get plenty of gain with just doing test with a mic. There doesn't seem to be much noise even with the board just out on the bench. I'll have to redo some measurements but with the input transformer wired to give me about 6dB and the output 4dB, I'm getting about 60dB of gain, which puts me right at the limit of Ian's recommendation. This is just with a single pole potentiometer which much simpler to implement than the infamous EK2003 switch (I made one of these for fun but don't know how fun it would be if I had to make 100 of them). I still have more work to do, but the initial "plug it in and see if it works" test seems to produce the desired results. And it sounds like I would expect. I have yet to compare this to any of the other circuits of this caliber that I have built, but I think with the same in and out transformers it will be hard to distinguish any subjective characteristics. So far I am lead to believe that my initial hypothesis is correct in regards to not sacrificing the perceived quality of sound between two and three stages. When I get one more properly wired up and tested I'll come back and share.

Thanks!

Paul

Thanks!

Paul
 
Top