For the sake of 4 Aux sends per channel it strikes me as easier to go active - there’s not many more components than in the passive methods anyway.
new project: small home recording console
- Thread starter Dimitree
- Start date
Help Support GroupDIY Audio Forum:
ruffrecords
Well-known member
Horses for courses. Both are valid solutions. The particulars of the application determine which method best suits. Component count is just one factor.
Cheers
Ian
Figured if OP is doing Neve style summing mix for main outs might as well do the same for the Auxes
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
thanks everyone for your contributions. I'd like to note that I'm going to use VE mixing for everything (PFL bus, AUXs, Groups) except for the main L R mix bus. For this I'd like to use passive mixing with Neve make up gain amp.
some questions came up to my mind:
1) did you determine that I was using passive mixing by something I drew on the schematic on post #19? or was it just a supposition? My believe was that as I draw it, it can't still be determined what kind of mixing system I was going to use, so I'm wondering if I understood that part correctly. I'm still trying to learn about this topic.
2) for passive mixing, if I understand correctly, the bus resistor must be connected direct to 0V when that channel is switched off. So the way I drawn the MIX LR bus switches, is clearly wrong, since it leaves that node floating. If so, should I also change that arrangement for the other buses too, even if those will be using VE mixing?
3) since passive mixing need low source impedance, should I add buffers on MIX LR, or the post panpot opamp IC7 already provides that?
some questions came up to my mind:
1) did you determine that I was using passive mixing by something I drew on the schematic on post #19? or was it just a supposition? My believe was that as I draw it, it can't still be determined what kind of mixing system I was going to use, so I'm wondering if I understood that part correctly. I'm still trying to learn about this topic.
2) for passive mixing, if I understand correctly, the bus resistor must be connected direct to 0V when that channel is switched off. So the way I drawn the MIX LR bus switches, is clearly wrong, since it leaves that node floating. If so, should I also change that arrangement for the other buses too, even if those will be using VE mixing?
3) since passive mixing need low source impedance, should I add buffers on MIX LR, or the post panpot opamp IC7 already provides that?
ruffrecords
Well-known member
Leaving a bus resistor floating when turned off is only an option with VE mixing.
In theory VE nodes can be left floating which will improve noise when fewer channels are selected. However, the floating node is liable to pick up noise if you are not very careful so most people ground them to be safe. In a mixer with a small number of channels the noise penalty is tiny.
Not necessarily. The Neve scheme, for example, slugs the bus so that the bus impedance is reduced to a value low enough to properly match the mix amp transformer input. This also significantly reduces the interaction of controls so that separate buffering is not required (and anyway there is a noise penalty for buffering).
It is not always necessary to use the Neve slugged bus method in a small mixer. Provided the value of the driving pot is about one fifth of the bus feed resistor there will be very little interaction between controls. So 10K pots and 47K bus feed resistors is a typical choice. There is very small noise penalty for using 47K resistors instead of, say, 10K resistors.
Cheers
Ian
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
Since I have an opamp (to implement active panning) between the panpot and the bus feed resistors, and the fader is even before that point, does this means that I just need the bus feed resistor with no pots, and hence there won’t be any interaction between controls?
ruffrecords
Well-known member
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
I'm starting to thinking about the AUX summing amps (before tackling the main LR summing), and browsing various schematics, I noticed this AUX summing amp by Soundcraft Delta.
This approach is explained in the Douglas Self book (I guess he designed that Soundcraft console), but it is not clear to me if this approach can nowadays be equaled (or outperformed) by using a modern opamp in a simpler approach
This approach is explained in the Douglas Self book (I guess he designed that Soundcraft console), but it is not clear to me if this approach can nowadays be equaled (or outperformed) by using a modern opamp in a simpler approach
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
in the meanwhile I started to study the Neve 1272, and this is just a draft for the main mix bus summer. Couldn't find a proper 1272 schematic, but only some modified schematic for use as mic preamp (with input gain selector, that I don't need).
So I don't know if my schematic is correct, I don't think so, mainly because I still haven't understand how to connect pins S T and K.
I guess I need to decide now how many channels will be mixed there, so to determine the make up gain needed.
I'd say I will have 16 channels + 4 groups + 4 aux returns, so 24 total channels for each 1272.
So I don't know if my schematic is correct, I don't think so, mainly because I still haven't understand how to connect pins S T and K.
I guess I need to decide now how many channels will be mixed there, so to determine the make up gain needed.
I'd say I will have 16 channels + 4 groups + 4 aux returns, so 24 total channels for each 1272.
This might help - it’s for the BA283
Last edited:
And this is how it connects in the 1073
MidnightArrakis
Well-known member
As I was reading through this thread and reviewing your various schematic diagrams, I got to wondering.....while you are going through the process of building this "one-off" audio console, will you also be involved with all of the PCB designs, as well as all of the different types of mechanical designs that will be required in order to have this project actually become a "physical reality"??? If so.....there's going to be a whole lot of design engineering work involved!!!
Have you already performed both the mechanical and PCB designs of large-scale electronic equipment before?
While reviewing your schematics, I couldn't help but notice that your component REFERENCE DESIGNATORS do not follow a linear sequential pattern. They jump around all over the place (i.e., R14 is placed before R9, etc.). Small stuff like this only adds to either inadvertent and/or more confusion later on down-the-road. Have you actually done this type of design work on a large project as what you are embarking on here?
Just wondering.....
/
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
thanks for your warnings. I fully understand the complexity of this task. I'm not an electronic engineer as you may have noticed (I'm actually a software engineer), so of course this makes things even more harder. But this is my hobby and over the past 18 years I've been building and designing (many time I failed, many time I succeeded) a lot of audio devices, and probably the largest/most complex were analog and digital synthesizers. All the times I've also been involved in designing their large PCBs and mechanical aspects. I've also built tube guitar amps from scratch (included the CAD and the machining of the chassis and the wood working needed for the cabinet) and for some years I've been building electric guitars. So while I'm not really familiar with consoles, I'm not much scared by mid to large projects.
But that doesn't mean that I'm underestimating the complexity of this project, I understand that a console is a whole different beast, for many reasons. That's why I'm here trying to learn about many aspects that I never had to really pay attention before (you don't have to worry that much about noise, crosstalk, distortion in synthesizers, for example).
I'm aware that this project could become a complete failure at any time, but I'll keep on trying because that's what I like to do with my spare time.
I know and I'm annoyed by that too, but the truth is that the schematics I'm showing here are just drafts that I'm doing with Eagle. Since those drafts are full of little circuit bits or alternate versions of those circuits that, for the sake of clarity, I don't show when posting here, designators make no sense right now. Luckily Eagle has a nice automatic "part renumber" feature, that I will run when I will got the various consolidated (..at least for a prototype) schematics.
MidnightArrakis
Well-known member
[Eagle has a nice automatic "part renumber" feature] -- My CADENCE/OrCAD program has a similar "Automatic Annotation" feature routine within it, but you need to be really careful when using it. These routines run on an X/Y-grid basis and if, for example, on the left-side of your schematic page there is just enough of a gap between two resistor symbols placed vertically above one another.....the top resistor symbol might end up being -- R7 and the resistor symbol directly below it is shown as -- R9 -- while resistor symbol -- R8 -- is shown over on the far right-side of the schematic. This is because the gap between R7 and R9 was just enough that the "Automatic Part Renumber" routine sees the right-side R8 as being the next component to annotate!!!
Personally.....I wait until I have completely finished the drawing of a schematic and moving symbols around until I am totally satisfied with how the schematic looks. THEN.....I will go through the schematic and manually annotate the entire schematic so all of the symbol REF DES and component values are exactly the way that both makes sense and are the way that I want them to be. But.....that's just me.....
Your schematics (mileage) may vary!!!
By the way.....My CADENCE/OrCAD program has an "Import Eagle Schematic Translator" built into it, should you want to share your schematic with me so I can either review it and/or assist with reannotating it for you. > FREE!!! <
/
- Joined
- Dec 10, 2016
- Messages
- 121
This might help. Basic info.
MidnightArrakis
Well-known member
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
that's a great help, I really appreciate it, thanks!
thank you both for those documents, I'm studying!
- Joined
- Dec 10, 2016
- Messages
- 121
Dimitree
Well-known member
- Joined
- Jul 26, 2011
- Messages
- 128
thank you really much! I have indeed a question: how do you use the 1272? as an outboard preamp or as it was used by Neve consoles (post summing amp)? I've seen some vintage 1272 and those had a gain trimmer that could be accessed from the front panel instead of the potentiometer/knob of the BAE 1272. I guess in the consoles the 1272 was used with fixed gain and an actual fader was wired after the unbalanced output of the 1272
- Joined
- Dec 10, 2016
- Messages
- 121
I'm using (or will shortly) the 1272 as a line amp for summing. It's not POST the summing amp, it IS the summing amp. One of the things that makes Neves a little different from other console is that it used a mic input transformer as the 'summing' portion of the bus (Not super correct, but you get the idea). It's a different approach to what is done today but it's part of the Neve thing.
The only place the 1272 was used as a mic pre was for the talkback mic.
On the original console the pot was recessed. If you were constantly futzing around with the trim then there was something wrong. It was meant to be calibrated and left alone.
The fader was between the two amp on the BA283 (preamp and output) using pins K & L. The trim actually adjusted the level going TO the fader.
Here's the path:
routing bus switch to
bus (long copper/vero board) to
1272 mic input transformer (originally the 10468 currently the VTB9045) to
the preamp section of the BA283 (pins N-V) to
the fader (or not, depends on the use) to
the output section of the BA283 (pins A-M) to
the output transformer (originally the LO1166 currently the VTB1148 or equivalent) to
the patchbay (and hence the rest of the world).
On the patchbay the output of the 1272 might be normalled into a 1271 (a line input line amp with fixed gain) as an insert return via a 31267 (currently the VTB9046) transformer then out of the 1271 (via the LO1166) to the rest of the world. So a bus would go out one transformer and back through another line input transformer and out another transformers. That's a lot of iron. Again, not all consoles were set up this way.
Usually the unbalanced output of the 1272 wasn't used for an output, sometimes it was used to feed a meter. You could use it by putting a different type of transformer on the output, up to you.
I also have the Class A/B version of the 1272, the 3415. Same input transformer, different output transformer. While I haven't compared them side by side, my understanding is that the 3415 is a little more 'focused' in the low end, less of the Class-A low end build up that happens. I've talked to a number of well known mixers/producers who all said the same thing, tracking thru a Class-A Neve is great, but when it comes to mix, fire up the high pass filters, you're going to need it. Record on Neve, mix on SSL and use the filters.
YMMV
The only place the 1272 was used as a mic pre was for the talkback mic.
On the original console the pot was recessed. If you were constantly futzing around with the trim then there was something wrong. It was meant to be calibrated and left alone.
The fader was between the two amp on the BA283 (preamp and output) using pins K & L. The trim actually adjusted the level going TO the fader.
Here's the path:
routing bus switch to
bus (long copper/vero board) to
1272 mic input transformer (originally the 10468 currently the VTB9045) to
the preamp section of the BA283 (pins N-V) to
the fader (or not, depends on the use) to
the output section of the BA283 (pins A-M) to
the output transformer (originally the LO1166 currently the VTB1148 or equivalent) to
the patchbay (and hence the rest of the world).
On the patchbay the output of the 1272 might be normalled into a 1271 (a line input line amp with fixed gain) as an insert return via a 31267 (currently the VTB9046) transformer then out of the 1271 (via the LO1166) to the rest of the world. So a bus would go out one transformer and back through another line input transformer and out another transformers. That's a lot of iron. Again, not all consoles were set up this way.
Usually the unbalanced output of the 1272 wasn't used for an output, sometimes it was used to feed a meter. You could use it by putting a different type of transformer on the output, up to you.
I also have the Class A/B version of the 1272, the 3415. Same input transformer, different output transformer. While I haven't compared them side by side, my understanding is that the 3415 is a little more 'focused' in the low end, less of the Class-A low end build up that happens. I've talked to a number of well known mixers/producers who all said the same thing, tracking thru a Class-A Neve is great, but when it comes to mix, fire up the high pass filters, you're going to need it. Record on Neve, mix on SSL and use the filters.
YMMV
